It's being touted as such on some other forums. A paper to be presented this month by Meridian at the AES Convention in LA.
(NB it has been posted once before on HA in the midst (http://www.hydrogenaud.io/forums/index.php?showtopic=106785&view=findpost&p=873842)of a thread...but perhaps we can use this thread to discuss it, once it is available)
http://www.aes.org/events/137/papers/?ID=4058 (http://www.aes.org/events/137/papers/?ID=4058)
P14-3 The Audibility of Typical Digital Audio Filters in a High-Fidelity Playback System—Helen M. Jackson, Meridian Audio Ltd. - Huntingdon, UK; Michael D. Capp, Meridian Audio Ltd. - Huntingdon, UK; J. Robert Stuart, Meridian Audio Ltd. - Huntingdon, UK
This paper describes listening tests investigating the audibility of various filters applied in high-resolution wideband digital playback systems. Discrimination between filtered and unfiltered signals was compared directly in the same subjects using a double-blind psychophysical test. Filter responses tested were representative of anti-alias filters used in A/D (analog-to-digital) converters or mastering processes. Further tests probed the audibility of 16-bit quantization with or without a rectangular dither. Results suggest that listeners are sensitive to the small signal alterations introduced by these filters and quantization. Two main conclusions are offered: first, there exist audible signals that cannot be encoded transparently by a standard CD; and second, an audio chain used for such experiments must be capable of high-fidelity reproduction.
Convention Paper 9174
I recall reading over some of their previous information regarding listeners favoring the filters that Meridian was using against several others. I'm quite appreciative of them for share their findings with the audio community. It will be interesting to revisit that topic to see if Meridian has come to new conclusions. Thank you for posting this.
It's being touted as such on some other forums. A paper to be presented this month by Meridian at the AES Convention in LA.
(NB it has been posted once before on HA in the midst (http://www.hydrogenaud.io/forums/index.php?showtopic=106785&view=findpost&p=873842)of a thread...but perhaps we can use this thread to discuss it, once it is available)
http://www.aes.org/events/137/papers/?ID=4058 (http://www.aes.org/events/137/papers/?ID=4058)
P14-3 The Audibility of Typical Digital Audio Filters in a High-Fidelity Playback System—Helen M. Jackson, Meridian Audio Ltd. - Huntingdon, UK; Michael D. Capp, Meridian Audio Ltd. - Huntingdon, UK; J. Robert Stuart, Meridian Audio Ltd. - Huntingdon, UK
This paper describes listening tests investigating the audibility of various filters applied in high-resolution wideband digital playback systems. Discrimination between filtered and unfiltered signals was compared directly in the same subjects using a double-blind psychophysical test. Filter responses tested were representative of anti-alias filters used in A/D (analog-to-digital) converters or mastering processes. Further tests probed the audibility of 16-bit quantization with or without a rectangular dither. Results suggest that listeners are sensitive to the small signal alterations introduced by these filters and quantization. Two main conclusions are offered: first, there exist audible signals that cannot be encoded transparently by a standard CD; and second, an audio chain used for such experiments must be capable of high-fidelity reproduction.
Convention Paper 9174
I'm expecting yet another trip through the land of testmanship.
First person to post a link to the preprint wins!
this is just a shocking bit of sophistry and rhetorical sleight-of-hand, even from Amir. He's now claiming (without quite saying it outright) that M&M 2007 wasn't peer reviewed. His evidence is....
http://www.avsforum.com/forum/91-audio-the...ml#post28499689 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1598417-avs-aix-high-resolution-audio-test-take-2-a-4.html#post28499689)
Yes, the Journal has "scholarly" papers in it. But no, just because something is in the Journal, it does NOT mean it is a scholarly "peer reviewed paper." The proof is actually in the link you provided. Before we look at that, let's see how AES itself describes the journal it publishes: http://www.aes.org/journal/ (http://www.aes.org/journal/)
The Journal contains state-of-the-art technical papers and engineering reports; feature articles covering timely topics; pre and post reports of AES conventions and other society activities; news from AES sections around the world; Standards and Education Committee work; membership news, patents, new products, and newsworthy developments in the field of audio.
As you see, "state-of-the-art technical papers" is only one of the items in the journal. Those papers are indeed selected to be scholarly and properly reviewed to make sure the writer knows the audio theory and science before acceptance. Such is the Bob Stuart paper which is some 20-30 pages long, is full of audio theory and analysis.
The Meyer and Moran is the second type I have highlighted in red: Engineering Report. How do we know that? It says right on top in the link you provided!
Audibility of a CD-Standard A/DA/A Loop Inserted into High-Resolution Audio Playback
[Engineering Report] Claims both published and anecdotal are regularly made for audibly superior sound quality
The write-up says exactly the same thing on top of the title page so that there is no confusion whatsoever that what is about to come is not a scholarly paper. It is a report on "engineering" work done which in this case is a listening test. The "peer" review that was performed on it was likely the fact that it was not a sighted test. Outside of that, there is no discussion of theory or science in the write-up to have been "peer reviewed." And for sure, nobody in AES verified the core content of the write-up: the listening test. All the work was taken at face value and published. Only the organizers of the work are accountable for its accuracy, not the AES.
I realize understanding the nature of the Journal is beyond what is expected from people not in the audio research/industry. So mistaking their work for a peer reviewed paper is understandable. What is not understandable is why after I explain what it is, you chime in to still on saying otherwise. Hopefully this more detailed explanation puts posts like yours behind us and from here on, we don't exaggerate or misstate the nature of audio work in front of us.
No scientific journal tries to reproduce the result of a paper before it is published. A "peer reviewed" paper is pre-screened by experts in the field for obvious errors, but it has never been my experience that they attempt actual verification of the results either. The real verification is performed after publication when others either corroborate or refute the original paper, and then publish their own evidence.
I think that having seven years elapse without anyone, AFAIK, presenting strong evidence to contradict M&M is in itself a corroboration.
It's just striking that this
"The Journal contains state-of-the-art technical papers and engineering reports; "
would be boldly interpreted to mean 'technical papers are peer reviewed, but engineering reports are not'.
And of course no one ever claimed that every article, news item, letter, etc in JAES is peer reviewed. That's no more true in JAES than it is in, say, Nature or Science.
It's possible, but not obvious from anything Amir has quoted, that 'engineering reports' are not reviewed, while 'technical papers' are (or even vice versa) . The thing to do, of course, is to consult the JAES instructions to authors (which requires a login), or an editor at the journal, to see what sorts of articles are peer-reviewed or not. Amir appears not to have done that.
No scientific journal tries to reproduce the result of a paper before it is published. A "peer reviewed" paper is pre-screened by experts in the field for obvious errors, but it has never been my experience that they attempt actual verification of the results either.
Actually, I've done exactly that when the paper is from someone credible but the claimed results are suspicious.
It's just striking that this "The Journal contains state-of-the-art technical papers and engineering reports; " would be boldly interpreted to mean 'technical papers are peer reviewed, but engineering reports are not'. And of course no one ever claimed that every article, news item, letter, etc in JAES is peer reviewed. That's no more true in JAES than it is in, say, Nature or Science. It's possible, but not obvious from anything Amir has quoted, that 'engineering reports' are not reviewed, while 'technical papers' are (or even vice versa) . The thing to do, of course, is to consult the JAES instructions to authors (which requires a login), or an editor at the journal, to see what sorts of articles are peer-reviewed or not. Amir appears not to have done that.
It seems pretty clear to me "Engineering Reports" comes under the banner of, or is a subset of "All submissions", and therefore are by definition peer reviewed:
"JAES Publication PolicyThe AES Journal seeks original unpublished research papers and engineering reports of archive quality on subjects related to the audio domain. All submissions will go through a peer review process to check their suitability for JAES."
http://www.aes.org/journal/authors/guidelines/ (http://www.aes.org/journal/authors/guidelines/)
Convention papers, such as that new Meridian thing, aren't [necessarily] "peer reviewed" nor are they "published in the Journal as a paper", however they may be submitted at a later *date* to be published. [And I don't know if Meridian has submitted their thing] At least that's how I understand it.
More distinctions here:
http://www.aes.org/journal/authors/manuscript_types/ (http://www.aes.org/journal/authors/manuscript_types/)
P.S. I typically ignore that troll in that other forum but anyone wishing to use any of this, by all means do so, and no need to reference me. [Also keep in mind he's quite likely reading all of this.]
Nice work. Seems pretty normal to me: experimental papers/reports go through peer review. Even letters to the editor (responding to a published article) are at least 'reviewed' by the author of the original article.
Amir's take on peer-reviewed scientific publication is so bizarre that I have to wonder if he has ever authored one.
Nice work.
Agreed.
(his) take on peer-reviewed scientific publication is so bizarre that I have to wonder if he has ever authored one.
IME you've set the bar too high. I'd put the bar at" "Read one and really understood it".
I will represent myself as someone who has for the last 3 years struggled mano-a-mano with that man as much as anybody, if not more.
His mental state and behavior are fairly predictable to me. No way does it seem to me to involve what most of us consider to be accurate perceptions.
I might add there are some who will say the same about me! ;-) One is no doubt JA. I'm am quite pleased with that.
I will represent myself as someone who has for the last 3 years struggled mano-a-mano with that man as much as anybody, if not more. His mental state and behavior are fairly predictable to me. No way does it seem to me to involve what most of us consider to be accurate perceptions.
He's now posting his own screen capture of a Meridian white paper image in the related High-Res Audio Panels at AES 2014 thread, from his private smugmug photo account, showing the threshold of human hearing superimposed over the noise floor of 16 bit audio, without (of course) the advantages of noise shaping since it betrays his/Meridians agenda and CLAIMS it also appears in this JAES paper, perhaps attempting to give it more street cred than the corporate ad he actually took it from:
http://www.aes.org/e-lib/browse.cfm?elib=7140 (http://www.aes.org/e-lib/browse.cfm?elib=7140)
I don't doubt its legitimacy but I can't currently substantiate his SOURCE claim without buying the paper, (and I'm cheap ) so I'm asking for help Arny, since you seem to have battled him on the origins of his version of the Meridian image before. Can you verify it truly exists in the JAES paper he claims it also appears in?
Also, if you don't come back to help counter this commercial troll/ high-end audio lobbyist, I think I may have to throw in the towel myself. As Princess Leia once said: "Please come back, Obiwan K'. Arny; you're our only hope!"
I will represent myself as someone who has for the last 3 years struggled mano-a-mano with that man as much as anybody, if not more. His mental state and behavior are fairly predictable to me. No way does it seem to me to involve what most of us consider to be accurate perceptions.
He's now posting his own screen capture of a Meridian white paper image in the related High-Res Audio Panels at AES 2014 thread, from his private smugmug photo account, showing the threshold of human hearing superimposed over the noise floor of 16 bit audio, without (of course) the advantages of noise shaping since it betrays his/Meridians agenda and CLAIMS it also appears in this JAES paper, perhaps attempting to give it more street cred than the corporate ad he actually took it from:
http://www.aes.org/e-lib/browse.cfm?elib=7140 (http://www.aes.org/e-lib/browse.cfm?elib=7140)
I don't doubt its legitimacy but I can't currently substantiate his SOURCE claim without buying the paper, (and I'm cheap ) so I'm asking for help Arny, since you seem to have battled him on the origins of his version of the Meridian image before. Can you verify it truly exists in the JAES paper he claims it also appears in?
Also, if you don't come back to help counter this commercial troll/ high-end audio lobbyist, I think I may have to throw in the towel myself. As Princess Leia once said: "Please come back, Obiwan K'. Arny; you're our only hope!"
I don't have any access to the recent AES paper in question that you don't have. My usual sources of such PDFs have grown quiet.
However, superpositions of the threshold of hearing and the 16 bit noise floor are old news. The 1997 Stuart he quotes from here http://www.avsforum.com/forum/91-audio-the...ml#post28510553 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1731009-high-res-audio-panels-aes-2014-a-3.html#post28510553) are flawed by the fact that Stuart ignores noise shaping even though Meridian has proudly sold devices that implement noise shaping since no later than the early 1990s.
I think a PDF of some version of the 1997 paper is on the Meridian web site.
What can one say about someone who ignores his own products when they would destroy his agenda?
I don't doubt its legitimacy but I can't currently substantiate his SOURCE claim without buying the paper, (and I'm cheap ) so I'm asking for help Arny, since you seem to have battled him on the origins of his version of the Meridian image before. Can you verify it truly exists in the JAES paper he claims it also appears in?
I think Amir quoted from an old paper "Coding High Quality Digital Audio" that is available on the Meridian website (https://www.meridian-audio.com/ara/coding2.pdf)
This previously unpublished paper is a more complete version of the article published by Audio Magazine in March 1998. This paper was in turn a subset of the landmark paper 'Coding Methods for High-Resolution Recording Systems' presented at AES, New York in 1997 - copies of which can be obtained from AES as preprint #4639
[/size]
Stuart;s paper was an AES Convention Paper (http://www.aes.org/e-lib/browse.cfm?elib=7140) in 1997, then an Audio magazine article in 1998, then a JAES Paper (http://www.aes.org/e-lib/browse.cfm?elib=12986) in 2004. There is no reason to question its legitimacy on that score.
The white paper (referencing no papers later than 1997) on Meridian's site is another incarnation. Its provenance is explained there
Coding High Quality Digital Audio by Bob Stuart. This previously unpublished paper is a more complete version of the article published by Audio Magazine in March 1998. This paper was in turn a subset of the landmark paper 'Coding Methods for High-Resolution Recording Systems' presented at AES, New York in 1997 - copies of which can be obtained from AES as preprint #4639
As I said, he's been flogging this for a
long time.
Meyer and Moran's 2007 paper was a direct response to the 2004 JAES paper, and was preceded by a critical letter to the editor or JAES, published in December 2004 (http://www.aes.org/journal/online/JAES_V52/12/), signed by David Moran, Roy Allison, and E. Brad Meyer.
Amir may think some of us don't know this history, but we do.
Amir may think some of us don't know this history, but we do.
Just glanced at the thread and it appears my BFF "likes" the Dunning-Kruger gangs rejection of DBT (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1598417-avs-aix-high-resolution-audio-test-take-2-a-5.html#post28479610)
Other critical issues include fundamental flaws in DBT itself as a test method
Which makes sense given Bluescreen mans own methods (http://www.avsforum.com/forum/86-ultra-hi-end-ht-gear-20-000/1136745-establishing-differences-10-volume-method-14.html#post16385882) for demonstrating the disorder.
It seems those stricken are ok touting papers....using a method they reject.
cheers,
AJ
It seems pretty clear to me "Engineering Reports" comes under the banner of, or is a subset of "All submissions", and therefore are by definition peer reviewed: "JAES Publication Policy The AES Journal seeks original unpublished research papers and engineering reports of archive quality on subjects related to the audio domain. All submissions will go through a peer review process to check their suitability for JAES." http://www.aes.org/journal/authors/guidelines/ (http://www.aes.org/journal/authors/guidelines/)
Plus here is Moran himself stating it was peer reviewed:
"We got pretty heavily peer-reviewed, actually, compared with other AESJ papers I have been involved with.
I notice no one here objects to the totally untested, unsubstantiated adulatory characterization of hi-rez that licensor Bob Stuart published in the AESJ. Which is what got us motivated to do the test.
If you feel our result is disprovable ("wrong") and that hi-rez is audibly better in some way, well, do your own proper test."
-- David Moran http://www.audioasylum.com/cgi/t.mpl?f=prophead&m=37062 (http://www.audioasylum.com/cgi/t.mpl?f=prophead&m=37062)
Is it possible any actual test of audibility is
, by definition an "engineering report" and research papers are more of the self congratulatory sort of papers of Meridian's Stuart, where he pontificates about WHY his $20,000 SACD machine sounds so great must be due to X, Y, and Z?
It would make sense to me to make some differentiation between papers which are "test results of humans" vs. "guessing what they must hear", right?
On a different note, how long until this conference paper gets added to the AES' online archive? Will it come with the October issue of the journal?
On a different note, how long until this conference paper gets added to the AES' online archive?
IME, varies
Will it come with the October issue of the journal?
Unlikely, but probably never.
The Stuart AES Convention paper is now available (http://www.aes.org/e-lib/browse.cfm?elib=17497)for purchase/download at the AES library
The Stuart AES Convention paper is now available (http://www.aes.org/e-lib/browse.cfm?elib=17497)for purchase/download at the AES library
I sure hope it's not an engineering report, anchors were used extensively and Bob has his credentials in order.
The Stuart AES Convention paper is now available (http://www.aes.org/e-lib/browse.cfm?elib=17497)for purchase/download at the AES library
I sure hope it's not an engineering report, anchors were used extensively and Bob has his credentials in order.
It's a convention paper -- so, not even peer reviewed. (It's also a completely in-house affair -- all three authors work for Meridian).
Thus, I expect Our Man from Madrona will reject it forcefully, using at least three if not four different font colors.
Krabapple, you bought the paper? I see your quote in the AVS forum of more than just the abstract, for example.
Did he give an explanation why rectangular dither was used instead of triangular, considering it is know to introduce noise modulation which triangular is immune from?
Krabapple, you bought the paper? I see your quote in the AVS forum of more than just the abstract, for example.
Yes. I've bought lots of AES papers over the years.
Did he name the sampling rate convertor model number? If so, if in its owner's manual it mentions, say: "Be advised the optional use of rectangular dither [RPDF or "Rectangular Probability Density Function,"] instead of TPDF, may cause audible noise modulation", then he's dead meat.
How about absolute SPL used for the tests? Source songs? Source machine? Does he/they mention checking for time misalignment which we now know some SRCs introduce? Level matching after the conversion? Remember the SRC used by AIX for the AVS forum test introduced BOTH level and time alignment problems which I proved is audible.
Did he name the sampling rate convertor? If so, if in its owner's manual it mentions, say: "Be advised the optional use of rectangular dither [RPDF or "Rectangular Probability Density Function,"] instead of TPDF, may cause audible noise modulation", then he's dead meat.
How about absolute SPL used for the tests? Source songs? Source machine?
C'mon, it's $20 bucks to buy a copy. If you're going to interrogate it this closely, you should do that.
Fair enough. I probably will later.
It's a convention paper -- so, not even peer reviewed. (It's also a completely in-house affair -- all three authors work for Meridian).
Oh noes, big fail according to shyster criteria.
Well, if it's completely free of actual blind listening tests, perhaps Stereophile may have some use for it?
Thus, I expect Our Man from Madrona will reject it forcefully, using at least three if not four different font colors.
Quite so.
I wonder what he thinks about the credentials of the guy who said this? (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1114557-10-biggest-lies-audio-11.html#post15931333):
Why DVD-A, SACD? Good question. There is no good evidence that they sound different to people over the age of 10 or so.
Can we assume, despite the illogic and all the antics, that he is over the age of 10?
cheers,
AJ
Just got copies of 9174 and 9197. This should be interesting...
Did he give an explanation why rectangular dither was used instead of triangular, considering it is know to introduce noise modulation which triangular is immune from?
Interesting. Meridian used to know better than to use rectangular dither:
From their web site:
https://www.meridian-audio.com/download/Han...ies/518user.pdf (https://www.meridian-audio.com/download/Handbooks/500_Series/518user.pdf) Page 37:
"It is well established that applying TPDF dither... ...contributes no noise modulation."
I think that I predicted here about a month ago that testmanship would be resorted to in this paper. Bingo!
Since i am a noob with such AES papers can someone do a simple summary explaining the main findings or claims?
Werner at pinkfishmedia once did this with paper number 8101 in a way even me grasped.
http://www.pinkfishmedia.net/forum/showpos...amp;postcount=9 (http://www.pinkfishmedia.net/forum/showpost.php?p=2307803&postcount=9)
^The abstract (summary) at the beginning of the paper should tell you everything you need to know. If some aspect of it is unclear to you, ask.
http://www.aes.org/e-lib/browse.cfm?elib=14195 (http://www.aes.org/e-lib/browse.cfm?elib=14195)
^The abstract (summary) at the beginning of the paper should tell you everything you need to know. If some aspect of it is unclear to you, ask.
http://www.aes.org/e-lib/browse.cfm?elib=14195 (http://www.aes.org/e-lib/browse.cfm?elib=14195)
I thought we talk about that one meanwhile and for that the summary is very vague.
http://www.aes.org/e-lib/browse.cfm?elib=17497 (http://www.aes.org/e-lib/browse.cfm?elib=17497)
oops, wrong paper, sorry.
I haven't read the new one yet.
In the "old" paper 0 dBFS is referenced to 120 dB spl when audibility of (dither) noise is calculated. This seems very loud to me, especially with modern loudness-war pop music. Is this spl reference still used in the new paper ? (which I refuse to buy, pretending I'm not an audio forum addict)
In the "old" paper 0 dBFS is referenced to 120 dB spl when audibility of (dither) noise is calculated. This seems very loud to me, especially with modern loudness-war pop music. Is this spl reference still used in the new paper ? (which I refuse to buy, pretending I'm not an audio forum addict)
I'd rather buy a couple CDs off Amazon with the $20 myself, than contribute to this silly nonsense.
I see Bob Stuart is an old topic (http://www.hydrogenaud.io/forums/index.php?showtopic=96475&st=75) too.
Wonder if jj knows his old boss finally has that long missing whit of evidence, now being touted on several forums?
cheers,
AJ
It's amusing how *fragile* this perception of difference between hi rez and CD is....it can barely withstand ABX testing. Here's Stuart's* new iteration of the anti-ABX stace, from the Introduction section of his new Convention paper
There is a more general problem with listening tests
of this kind, which concerns the testing procedure.
ABX tests are viewed as the "gold standard" for
objective measures of listening. In an ABX test, a
listener is required to listen to two reference sounds,
sound A and sound B, and then to listen to sound
X, and to decide whether sound X is the same as
sound A or sound B. ABX tests have a high sensi-
tivity, that is, the proportion of true-positive results
out of total positive results is high. However, ABX
tests also have low specifi city, meaning that the pro-
portion of true-negative results out of total negative
results can be spuriously low. Translating this into
outcomes in psychophysical tests, the proportion of
the time that a listener scores well on an ABX test
by chance is low, but the proportion of the time that
a listener can score poorly on a test in spite of being
able to discriminate the sounds is high. An ABX
test requires that a listener retains all three sounds
in working memory, and that they perform a min-
imum of two pair-wise comparisons (A with X and
B with X), after which the correct response must be
given; this results in the cognitive load for an ABX
test being high.
The rest of the intro, btw, is profusely referenced....as you see, this paragraph hasn't got even one. Curious, that.
(*actually Jackson, Capp, and Stuart. FWIW, first and last authors have 'pride of place' on a scientific paper)
It's amusing how *fragile* this perception of difference between hi rez and CD is....it can barely withstand ABX testing. Here's Stuart's* new iteration of the anti-ABX stace, from the Introduction section of his new Convention paper
I always wonder why they don't submit their critique of double-blind testing to Journal of Psychology: Human Perception. Imagine the fame that would be achieved by overturning the gold standard of perceptual testing. I'm sure it would cruise through the peer review.
From krabapple's quote of the convention paper:
ABX tests are viewed as the "gold standard" for objective measures of listening.
Well at least they got something right!
the proportion of the time that a listener can score poorly on a test in spite of being able to discriminate the sounds is high.
And we know they "in truth" can discriminate the sounds by taking their word for it? Or by using sighted tests? Really?
An ABX test requires that a listener retains all three sounds in working memory,
False. The way I use the test is to see if the sound I am hearing seems different than the one I heard 1/10th of a second ago when I switched. If you count that as "memory" the grand total is two things, not three. [And that's also assuming what you are currently hearing is counted as "memory"] Invoking the reader to think "three things" is purposefully used here to make the task seem difficult and confusing, as a form of scaremongering by the writers.
and that they perform a min- imum of two pair-wise comparisons (A with X and B with X), after which the correct response must be given; this results in the cognitive load* for an ABX test being high.
Also not necessarily true. You aren't forced to listen to the anchors, A and B, AT ALL if you don't wish to. For each trial you can go straight to listening to X and then immediately cast your vote, since after all, the difference is said to be "night and day" so in such instances why waste your time listening to a sound you know so well? Remember the identity of A and B is known to the listener at all times and only should they WISH to refresh their memory are they required to hit the A and B buttons. I've taken complete tests having never touched them at all.
*More scary words. The only "cognitive load" or burden is to hit a button, whenever you want and however many times you want, and answer one simple question: "Did it make a differenece?" THAT'S the only "burden".
The aspect of memory on testing has been discussed to death, AFAIC, with the outcome of such discussions never seeming to benefit the anti-DBT crowd.
However, why are people who are not even present in the discussion being debated here?
The aspect of memory on testing has been discussed to death, AFAIC, with the outcome of such discussions never seeming to benefit the anti-DBT crowd.
However, why are people who are not even present in the discussion being debated here?
All we have to do is get one or more of the authors of that paper to deign to post here. I've never been able to get anybody at Meridian to even respond to a email request at all, so the best of luck with that!
I look at the possible good. While they laid a bad rap on ABX apparently they used it and obtained a positive result. The positive result seems to deny the quoted text in the paper, and I'd call that a good thing!
However, why are people who are not even present in the discussion being debated here?
Indeed. Thanks for noting that .
I came here to see what HA's take is on Stuart's latest work and most of what I read is about me??? Is the technical topic not interesting enough by itself?
Wonder if jj knows his old boss finally has that long missing whit of evidence, now being touted on several forums?
cheers,
AJ
Hey AJ. How the heck have you been kid? I see in your signature says you still trying to sell speakers? I can't think of a worse business than selling audio equipment. Was hoping by now you would have seen the merits of that.
As to JJ, I honestly don't understand the constant name dropping there. You don't know him personally, right? You wouldn't be able to hold a conversation with him for a minute. Right? Why keep throwing his name around? And in this context? That somehow you know his position in audio better than me? I don't think so .
But yes, this latest evidence from Stuart is quite strong. It is the first professional test of this type vs hobby work that Moran and Meyer performed. And the paper shows it. It is far more than "whit of evidence." If you were pushing Meyer and Moran around in forums, I suggest pulling back some or you will have your hat handed to you. In other words, stop being yourself AJ.
I came here to see what HA's take is on Stuart's latest work and most of what I read is about me??? Is the technical topic not interesting enough by itself?
Welcome amir!
Yes, that was intentionally done. I of course knew you were reading/lurking, so I figured I could (would) eventually flush you out.
Forgive my methods, despite the effectiveness.
Regarding your recent success with the corrupt/gameable/non-proctored online amateur AVS files and now the Stuart paper, have you enlightened jj with these 2014 revelations, given his 2009 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1114557-10-biggest-lies-audio-11.html#post15931333) and 2012 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=96475&view=findpost&p=805186) positions on the subject?
Why DVD-A, SACD? Good question. There is no good evidence that they sound different to people over the age of 10 or so
I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener.
Based solely on credentials, I'm leaning towards jj and away from Stuart, who has pecuniary interests in the matter. Do you disagree?
You also said the Stuart paper was "available for free on the Meridian website". Link?
cheers,
AJ
P.S. I typically ignore that troll in that other forum but anyone wishing to use any of this, by all means do so, and no need to reference me. [Also keep in mind he's quite likely reading all of this.]
I was not but I am now . What you wrote is correct but unnecessary. Steven (Krab) misunderstood my post. I didn't say Meyer and Moran test was not peer reviewed. I said it was not a "scholarly paper" as posted by m.zillch ( you are not him are you? your alias is similar to his). I showed that their work was an engineering report as documented every which way in their text and AES web site. That they themselves and other members refer to it as a journal paper is incorrect. I am sure you have read the report and know that there is nothing in there other than results of their test. Journal papers don't look anything like that. And they are marked as journal paper.
I don't know who did the review of Meyer and Moran. But I can tell you with confidence that they lacked domain knowledge to as to how to perform these tests properly. Instead of repeating what I have said over and over again on AVS, seeing Arny here, I will go ahead and post his list of what it takes for a reliable test: http://www.avsforum.com/forum/91-audio-the...ml#post24774498 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1533798-double-blind-amplifier-testing.html#post24774498)
Here are some guidelines to follow:
Ten (10) Requirements For Sensitive and Reliable Listening Tests
(1) Program material must include critical passages that enable audible differences to be most easily heard.
(2) Listeners must be sensitized to a audible differences, so that if an audible difference is generated by the equipment, the listener will notice it and have a useful reaction to it.
(3) Listeners must be trained to listen systematically so that audible problems are heard.
(4) Procedures should be "open" to detecting problems that aren't necessarily technically well-understood or even expected, at this time. A classic problem with measurements and some listening tests is that each one focuses on one or only a few problems, allowing others to escape notice.
(5) We must have confidence that the Unit Under Test (UUT) is representative of the kind of equipment it represents. In other words the UUT must not be broken, it must not be appreciably modified in some secret way, and must not be the wrong make or model, among other things.
(6) A suitable listening environment must be provided. It can't be too dull, too bright, too noisy, too reverberant, or too harsh. The speakers and other components have to be sufficiently free from distortion, the room must be noise-free, etc..
(7) Listeners need to be in a good mood for listening, in good physical condition (no blocked-up ears!), and be well-trained for hearing deficiencies in the reproduced sound.
(8) Sample volume levels need to be matched to each other or else the listeners will perceive differences that are simply due to volume differences.
(9) Non-audible influences need to be controlled so that the listener reaches his conclusions due to "Just listening".
(10) Listeners should control as many of the aspects of the listening test as possible. Self-controlled tests usually facilitate this. Most importantly, they should be able to switch among the alternatives at times of their choosing. The switchover should be as instantaneous and non-disruptive as possible.
Arny had this list on his now defunct ABX web site years before Meyer and Moran performed their test. He also was an advocate of following the more detailed and complete list of above in the form of ITU Rec. BS-1116.
Clearly then whoever did the review didn't read his work, or worked in this specialized field to not catch the simplest errors as documented in Arny's and BS-1116 and countless other papers.
Still not convinced? Here is one of the references in the Meyer and Moran engineering report:
[4] D. Blech and M. Yang, “DVD-Audio versus SACD:
Perceptual Discrimination of Digital Coding Formats,”
Here is a reference in that paper:
[3] ITU Radiocommunication Assembly. 1997.
Methods for the subjective assessment of small
impairments in audio systems including
Multichannel Sound Systems. Recommendation
ITU-R BS. 1116-1: 1994-1997.
Did the authors or the reviewers not note the absence of said reference in Meyer and Moran's test? Or compliance with it?
I could go on but that would cement your case that I am a troll so I will stop here .
P.S. Please excuse the English errors. Fixed a few but I am sure there are tons more
Ooops, some cross posting here.
Hey AJ. How the heck have you been kid? I see in your signature says you still trying to sell speakers? I can't think of a worse business than selling audio equipment. Was hoping by now you would have seen the merits of that.
Doing fine Amir, thanks for asking. Yep, still selling audio equipment (speakers), like you....for fun. It is not, nor will be, "my business", other than a side (ad)venture.
Few here will know that our rivalry dates back to our days growing up in Kenya. But lets keep things on track with the actual "Hi Rez" discussion.
As to JJ, I honestly don't understand the constant name dropping there. That somehow you know his position in audio better than me? I don't think so .
Unlike yourself or Stuart, jj appears to be a "credentialed" expert in the field. I've direct quoted him twice above. You feel your ad-hoc revelations...and now Stuarts paper, upends his previous statements?
He is a member here and I'm sure, quite capable of speaking for himself.
But yes, this latest evidence from Stuart is quite strong. It is the first professional test of this type vs hobby work that Moran and Meyer performed. And the paper shows it. It is far more than "whit of evidence." If you were pushing Meyer and Moran around in forums, I suggest pulling back some or you will have your hat handed to you. In other words, stop being yourself AJ.
Well, the jury is still out on the Stuart paper, so to speak.
Out of curiosity, what exactly is your position on using blind/controlled tests for audio?
cheers,
AJ
Welcome amir!
Yes, that was intentionally done. I of course knew you were reading/lurking, so I figured I could (would) eventually flush you out.
Forgive my methods, despite the effectiveness.
Ah, the clever you! Not!
Regarding your recent success with the corrupt/gameable/non-proctored online amateur AVS files and now the Stuart paper, have you enlightened jj with these 2014 revelations, given his 2009 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1114557-10-biggest-lies-audio-11.html#post15931333) and 2012 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=96475&view=findpost&p=805186) positions on the subject?
Why DVD-A, SACD? Good question. There is no good evidence that they sound different to people over the age of 10 or so
I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener.
First, if the test is corrupt, let's see the results of you passing it AJ. You can use the cheats but however you do it, let's see the output of foobar. If you cannot pass it, then we have proven a very significant thing: that our abilities to listen differ. This means listening tests, unless utilizing expert/trained listeners, cannot be representative of everyone's listening ability. And this is the big failing of Meyer and Moran where they had no process for identifying trained listeners and utilizing them in the test.
And remember, there are two ways to get an A on an exam: cheating and actually knowing the material perfectly . Just because someone figured out a way to cheat doesn't mean he has shown everybody else got there the same way.
Based solely on credentials, I'm leaning towards jj and away from Stuart, who has pecuniary interests in the matter. Do you disagree?
Well, by that logic I have better credentials than you in this area but you are not going to take that as an argument, are you? Stuart is one of the luminaries of our industry. He is an AES fellow and one of the few people who has both knowledge of audio engineering/science, especially signal processing, and psychoacoustics. That he is a founder of Meridian did not bring down his position in AES and should not do differently here.
As to JJ, he does not have a listening test we can use for this topic. Stuart has now published one. And that test is the subject of this thread. The posts you show from JJ are prior to publication of this test so I am not sure what relevance they have in the specific. 8 testers with small amount of training across 160 trials in Stuart's listening test managed with statistical reliability to tell the difference between 24 bit and 16 and 44.1 Khz and higher sampling rate. We don't get to throw that out by quoting JJ's comment prior to publication of this test. If there is bias that caused Stuart to create faulty results, let's demonstrate that instead of implying it and ignoring the data.
You also said the Stuart paper was "available for free on the Meridian website". Link?
I meant the original paper, not this latest test. The original is here: https://www.meridian-audio.com/ara/coding2.pdf (https://www.meridian-audio.com/ara/coding2.pdf)
First, if the test is corrupt, let's see the results of you passing it AJ.
Bad logic.
Whatever results anybody else obtains do not necessarily validate or invalidate any particular set of results. There are many relevant influences that can affect the outcome of a given listening test, and all the many ways that the the basic test can be corrupted is only a tiny subset of them. These tests are for example dependent on the listener's hearing acuity and I have publicly tried to exclude my results on those grounds. I know for sure that my hearing isn't it was 5, 10, 30 years ago. That's natures way of telling... I seem to recall that that frank and painful admission on AVS set me up for more bullying. I feel like I'm back in grade school!
The above attack suggests bullying to me, and bullying based on false logic is common when people have something to hide.
I understand that the Foobar2000 ABX plug-in has been updated in such a way that it will put evidence about more critical test parameters into the log file, and there is now a facility for validating log files.
By no means does it validate everything that can invalidate a test, but it provides more evidence.
You might want to repeat your more exceptional results using the new plug in. ;-)
First, if the test is corrupt, let's see the results of you passing it AJ. You can use the cheats but however you do it, let's see the output of foobar. If you cannot pass it, then we have proven a very significant thing: that our abilities to listen differ.
No, it would only prove that one of us would not stoop low enough to cheat. Nor have a strong pecuniary interest in the matter.
It would also prove that one of us has no clue between an ad-hoc test and one that adheres to ITU-BS1116...or BS.1534-2, that are the bible of testing for small impairments. Let me quote some for you:
"It should be understood that the topics of experimental design, experimental execution, and statistical analysis are complex, and that only the most general guidelines can be given in a Recommendation such as this. It is recommended that professionals with expertise in experimental design and statistics should be consulted or brought in at the beginning of the planning for the listening test."
Would you care to explain how the corrupt/gameable/non-proctored online amateur AVS files "test" adhered to ITU-BS1116...or BS.1534-2?
They
did not and thus the results
must be dismissed as ad-hoc garbage, per a highly reputable source.
Well, by that logic I have better credentials than you in this area but you are not going to take that as an argument, are you?
Nope and I agree it would be idiotic to dismiss arguments based on "credentials" or appeal to authority. Like dismissing Meyer and Moran's tests due to "credentials".
Stuart is one of the luminaries of our industry.
"Our"? Yours maybe.
My interest is sound reproduction, not high fashion jewelry.
He is an AES fellow and one of the few people who has both knowledge of audio engineering/science, especially signal processing, and psychoacoustics. That he is a founder of Meridian did not bring down his position in AES and should not do differently here.
Right, which means per you above, jack squat, the only thing that matters is the content of the paper.
Let's hope he didn't resort to pathological science, cherry pick components and adhered to BS.1534-2.....
cheers,
AJ
Such as non-representative pathological samples?
...or use rectangular dither, knowing that it could have a stronger influence over the outcome?
Such as non-representative pathological samples?
And those were what?
...or use rectangular dither, knowing that it could have a stronger influence over the outcome?
What type of dither did Meyer and Moran use? Whatever was in the A/D and D/A converters of the CD-ROM recorder, right? Here is Stuart on his latest test:
This [playback] level was
chosen for comfort, and because it was high enough
for details to be audible but also low enough that 16-
bit RPDF dither would be inaudible at the listening
position [30].
And
We chose to use undithered quantization as
a probe and | although we would normally rec-
ommend TPDF dither for best practice -- we con-
sidered rectangular dither to be more representative
of the non-ideal dither or error-feedback processing
found in some commercial A/D and D/A filters.
Seems reasonable to me. After all, who knows what dither if any was used in the music that people are shipping to us in 16/44.1? It is not like they know what you and I know about dither types.
And those were what?
I don't know, hence my use of a question mark. Why don't you tell us?
This [playback] level was chosen for comfort, and because it was high enough for details to be audible but also low enough that 16- bit RPDF dither would be inaudible at the listening position [30].
I'm not buying it.
We chose to use undithered quantization as a probe and although we would normally recommend TPDF dither for best practice -- we considered rectangular dither to be more representative of the non-ideal dither or error-feedback processing found in some commercial A/D and D/A filters.
This doesn't pass the smell test either.
After all, who knows what dither if any was used in the music that people are shipping to us in 16/44.1?
...or what was the true noise floor and dynamic range.
These are all poor excuses not to employ best practices, if you ask me.
What type of dither did Meyer and Moran use?
Red Herring dither. Now stick to what BS used in his landmark tests, the only relevant one, remember?
Here is Stuart on his latest test:
This [playback] level was
chosen for comfort, and because it was high enough
for details to be audible but also low enough that 16-
bit RPDF dither would be inaudible at the listening
position [30].
Very vague. What was the measured levels? One more time Amir, did BS's test adhere to ITU-BS1116 or BS.1534-2?
What hidden reference anchor was used?
I actually like to measure the ABX/HR box also. Such boxes are used in all of these tests we hang our hat on yet none of us have any measurements or objective evaluations of them.
We chose to use undithered quantization as
a probe and | although we would normally rec-
ommend TPDF dither for best practice -- we con-
sidered rectangular dither to be more representative
of the non-ideal dither or error-feedback processing
found in some commercial A/D and D/A filters.
Ah, so he cherry picked. Nice.
So this has nothing to do with whether 16/44 is transparent as a playback method, but more about cherry picking a pathological case.
Yes, I can really see this flipping JJ's position....
cheers,
AJ
And those were what?
I don't know, hence my use of a question mark. Why don't you tell us?
??? You said: Such as non-representative pathological samples? I am asking what was non-representative and pathological. If you don't know what was used why characterize it that way?
This [playback] level was chosen for comfort, and because it was high enough for details to be audible but also low enough that 16- bit RPDF dither would be inaudible at the listening position [30].
I'm not buying it.
Fine. He is not selling it either.
We chose to use undithered quantization as a probe and although we would normally recommend TPDF dither for best practice -- we considered rectangular dither to be more representative of the non-ideal dither or error-feedback processing found in some commercial A/D and D/A filters.
This doesn't pass the smell test either.
It was a listening test, not cooking.
After all, who knows what dither if any was used in the music that people are shipping to us in 16/44.1?
...or what was the true noise floor and dynamic range.
These are all poor excuses not to employ best practices, if you ask me.
I guess I need to get calibrated on member points of view. What is your position on Meyer and Moran since we both have the report in hand. Followed best practices?
I can see that you clearly aren't worth my time or effort, amirm.
What type of dither did Meyer and Moran use?
Red Herring dither. Now stick to what BS used in his landmark tests, the only relevant one, remember?
The thread name is "follow-up to Meyer and Moran." What plot are you following AJ? We are comparing the two reports to see which one is valid seeing how they contradict themselves.
Here is Stuart on his latest test:
This [playback] level was
chosen for comfort, and because it was high enough
for details to be audible but also low enough that 16-
bit RPDF dither would be inaudible at the listening
position [30].
Very vague. What was the measured levels?
It is in the report.
One more time Amir, did BS's test adhere to ITU-BS1116 or BS.1534-2?
The only BS test I know is the Meyer and Moran where they had not even heard of BS116 it seems. If you mean Stuart's, yes, the followed good bit of it. They deviated in some areas such as using non-expert listeners as to not get accused of results only being valid for that population.
What hidden reference anchor was used?
If something is hidden, you wouldn't be able to see it, right? Part of the answer was in my post.
I actually like to measure the ABX/HR box also. Such boxes are used in all of these tests we hang our hat on yet none of us have any measurements or objective evaluations of them.
When did you learn to measure stuff AJ? You didn't go and learn something about audio in the last year or two, did you?
But no, they did not use a "box" to have measurements as such. As the line in the movie Fifth Element goes, the switching mechanism was "perfect."
We chose to use undithered quantization as
a probe and | although we would normally rec-
ommend TPDF dither for best practice -- we con-
sidered rectangular dither to be more representative
of the non-ideal dither or error-feedback processing
found in some commercial A/D and D/A filters.
Ah, so he cherry picked. Nice.
He was trying to please you AJ. Don't say he didn't do anything for you.
So this has nothing to do with whether 16/44 is transparent as a playback method, but more about cherry picking a pathological case.
I think the only thing pathological is interactions between you and I on forums!
Are you going to offer anything informative AJ or just keep being yourself?
I can see that you clearly aren't worth my time or effort, amirm.
I am sorry I made you feel that way. I didn't see anything constructive in your reply to me. If you say something is pathological I like to see what you mean by that. When I provide actual quotes from the paper, you dismiss them out of hand with not passing smell tests and such. What am I supposed to do with that reply? There is nothing technical there. So when you ask me to quote more from the paper, I am not inclined to do so and get the same dismissal. Instead of being direct and telling you all of this, I resorted to a bit of sarcasm and humor. Would you have preferred this version of my reply instead? If so, well, you have both now .
Such as non-representative pathological samples?
And those were what?
...or use rectangular dither, knowing that it could have a stronger influence over the outcome?
What type of dither did Meyer and Moran use? Whatever was in the A/D and D/A converters of the CD-ROM recorder, right? Here is Stuart on his latest test:
This [playback] level was
chosen for comfort, and because it was high enough
for details to be audible but also low enough that 16-
bit RPDF dither would be inaudible at the listening
position [30].
And
We chose to use undithered quantization as
a probe and | although we would normally rec-
ommend TPDF dither for best practice -- we con-
sidered rectangular dither to be more representative
of the non-ideal dither or error-feedback processing
found in some commercial A/D and D/A filters.
Seems reasonable to me. After all, who knows what dither if any was used in the music that people are shipping to us in 16/44.1? It is not like they know what you and I know about dither types.
If someone were an advocate of so-called high resolution recordings then of course they would like to see the CD format hobbled by suboptimal dither and suboptimal noise shaping.
Furthermore highly atypical recordings were used: "The recording was of _very_ large dynamic range, roughly 16 bits worth."
I've made recordings of live performances using every trick I could except gain riding, and 15 bits was about as far as I could get.
A typical live recording made in a concert hall has dynamic range on the order of 12-13 bits.
A commercial recording with 12 bit dynamic range is actually quite exceptional, even before the days of hypercompression.
So to summarize, hobble the 16/44 recording, and use excluded middle recordings, and you just might be able to get positive results.
So what is known about the monitoring system and room used for playback? I'll bet it wasn't your typical listening room with a 40 dB SPL noise floor...
What plot are you following AJ?
The HA one where we discuss the topic of BS's "follow up" AES paper, not Amirs Red Herrings, Ad Hominems and other assorted dives into the rabbit hole, like on AVS and WTF??? forums.
You do know this is HA, not Kansas?
If something is hidden, you wouldn't be able to see it, right? Part of the answer was in my post.
IOW, no, BS didn't use MUSHRA. We know you "like" statements such as standard ABX is "fatally flawed". Hopefully this BS test wasn't.
But no, they did not use a "box" to have measurements as such. As the line in the movie Fifth Element goes, the switching mechanism was "perfect."
IOW, no measurements or objective evaluations of the switching apparatus to hang our credentialed hats on. Very amateurish.
He was trying to please you AJ.
Right. BS tests are for daydream believers and shysters who have pecuniary interests. Thus cherry picking and pathological methods becomes a necessity.
Are you going to offer anything informative AJ or just keep being yourself?
Both.
cheers,
AJ
Seems reasonable to me. After all, who knows what dither if any was used in the music that people are shipping to us in 16/44.1? It is not like they know what you and I know about dither types.
It is true that the choice of dither is up to the people producing the recording. Over the years various producers have advertised that they were using high performance dithering techniques.
An example of such a product could be JVC XRCDs.
Sony's Super Bit mapping seems to specifically claim the use of special noise shaping.
Anybody with better-than-average converters in their production facility can do this with off-the-shelf software such as Audition.
Just make the recording with good 24 bit ADCs with > 106-116 dB dynamic range (readily available off-the-shelf) and do the 24->16 bit conversion in software where the required dithering algorithms are readily available.
The tricky part is actually making a recording with enough dynamic range of its own for these steps to yield audible improvements. Doing that in a studio or concert hall is not easy, and is in fact unlikely unless very special care is taken with the choice of recording venue, the performance, micing and mixing techniques. As I've said I've done it and it implies a number of artificial constraints.
Or, the production staff can cheat and process the live recording with a dynamics expander, or simply ride the gains during mixing.
Furthermore highly atypical recordings were used
What? Nothing as 16/44-insufficient and highly demanding as Neil freaking Young? No Floyd either?? Say it aint so.
The type of the music the VAST majority of audio show going "Hi Rez" advocates listen to and "hear" benefits for?
Hmmm.
...
Would you care to explain how the corrupt/gameable/non-proctored online amateur AVS files "test" adhered to ITU-BS1116...or BS.1534-2?
They did not and thus the results must be dismissed as ad-hoc garbage, per a highly reputable source. ...
I have followed the execution and discussion of several comparative tests recently, starting with Arnie's keys. All of the tests proved to have samples that were defective. I believe this is the issue that needs to be discussed, not the results of the tests. GI = GO.
A typical live recording made in a concert hall has dynamic range on the order of 12-13 bits.
Your computation there is psychoacoustically blind Arny as we have discussed. You cannot use a single digit SPL number for your room noise. You must look at its spectrum and see how much noise you have relative to threshold of hearing as I show in my article, building on Fieldler's work: http://www.madronadigital.com/Library/RoomDynamicRange.html: (http://www.madronadigital.com/Library/RoomDynamicRange.html:)
.
IOW, no, BS didn't use MUSHRA.
And?
We know you "like" statements such as standard ABX is "fatally flawed". Hopefully this BS test wasn't.
There is no such statement or position in Bob's paper. Nor does it reflect mine.
You are however reflecting the common but mistaken assumption among hobbyists that the term "ABX" brings with it assurance of good results. Folks are blinded by those three letters thinking a gift from above is being handed to them. This is not true at all. The quality of the test determines whether the results are reliable. Not what it is called. You can stick an ABX box in a test, make a mistake of having both A and B be the same input and your results would be totally garbage. And this is what is wrong with people who carry the Meyer and Moran test as their bible to every thread and forum.
But no, they did not use a "box" to have measurements as such. As the line in the movie Fifth Element goes, the switching mechanism was "perfect."
IOW, no measurements or objective evaluations of the switching apparatus to hang our credentialed hats on. Very amateurish.
I am afraid the plot is lost AJ. I mentioned why the question is moot but your myopic view of how these tests can be performed is causing to run into a ditch. Here is another try: do you need to measure or perform objective evaluation if someone used Foobar's ABX comparator?
What? Nothing as 16/44-insufficient and highly demanding as Neil freaking Young? No Floyd either?? Say it aint so.
Hush you. Hi rez is what matters. As long as it's hi rez, it's better than Redbook...and not just better, but 'even my wife can hear it ' better.
That's the audiophile program.
Get with it. Amirm is.
After all, who knows what dither if any was used in the music that people are shipping to us in 16/44.1?
That has been demonstrated by the AIX samples in the AVS test. Note that this is no different for hi-res content, where the end user has no idea what happened in the production path.
We chose to use undithered quantization as
a probe and | although we would normally rec-
ommend TPDF dither for best practice -- we con-
sidered rectangular dither to be more representative
of the non-ideal dither or error-feedback processing
found in some commercial A/D and D/A filters.
This paragraph is talking about the
downconversion to 16/44.1, and yet includes this strange statement. It makes no sense. No one is making native 16/44.1 recordings, so no modern recording uses an A/D running at 16/44.1. 20/48 was ubiquitous more than a decade ago, 24/96 A/D is consumer (even commodity!) level now. Hence talking about "non-ideal dither or error-feedback processing found in some commercial A/D filters" in respect of 16/44.1 is a complete red herring.
D/As don't run at 16-bits or 44.1kHz either.
Coming from a 24/96 master, what matters in this respect is the downconversion...
After all, who knows what dither if any was used in the music that people are shipping to us in 16/44.1? It is not like they know what you and I know about dither types.
You can detect what was used on many CDs by analysing the digital bits on them. Even in the venerable Cool Edit, you have an accurate waveform view, a highly configurable spectral view, the ability to zoom/cut/amplify/speed-change as you wish, and your own ears. Noise shaped dither is easily detectable. Rectangular dither (or basically sub-optimal dither) during fades to silence is detectable. With a little MATLAB script I can plot the distribution of sample values - that can reveal some strange practicies.
It's a while since I've found a problem with the
final conversion to 16/44.1. It's typically gentle noise-shaped dither.
Such audio forensics often reveal problems further up the chain. Samples, tracks, or entire albums re-mastered from a previous generation 16/44.1 (or worse) version, lossy coding, and the many symptoms of the loudness war.
If you want to enjoy the worst that's found on CDs (and hi-res), throw all those things in. If you want to understand what restriction "CD quality" places on a decent master, downconvert it properly. We've known rectangular PDF dither was the wrong way to do it for 30 years.
I wouldn't normally claim that the difference between the right dither and the wrong dither would be or even could be audible, but it seems possible that this test has been so carefully set up that it just might be.
Cheers,
David.
A typical live recording made in a concert hall has dynamic range on the order of 12-13 bits.
Your computation there is psychoacoustically blind Arny as we have discussed.
There was no discussion, just excluded-middle assertions.
You cannot use a single digit SPL number for your room noise.
I didn't. I gave the room noise number in terms of bits. Obviously, I'm responding to a post that appears to come from someone with severe perceptual difficulties - someone who quotes "bits" and responds to "dB SPL"
Who doesn't know the difference between bits and dB SPL? Are they credible?
You must look at its spectrum and see how much noise you have relative to threshold of hearing as I show in my article, building on Fieldler's work: http://www.madronadigital.com/Library/RoomDynamicRange.html: (http://www.madronadigital.com/Library/RoomDynamicRange.html:)
Fielder makes some poorly-documented measurements under unrealistic circumstances in some SOTA venues. Who is stupid enough to use them to characterize actual performances and recordings made typical performance spaces?
One obvious flaw with these measurements is that they don't appear to include the presence of living, moving, breathing musicians. A symphony orchestra often has 100 or more of them. They contribute nothing to the noise floor? GMAB!
And?
Problem is, they (M&M) missed the next level of understanding. Let me quote International Telecommunication Union (ITU) document that is the bible of testing for small impairments in BS1116
According to one self assessed expert (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1585994-avs-aix-high-resolution-audio-test-ready-set-go-7.html#post26617417), non-adherence to BS1116 is a problem. The BS test failed to adhere, no cherry picking about "some" adherence can remedy that.
I am afraid the plot is lost AJ.
No, I'm quite clear on the "plot".
The BS test is a complete farce and fabrication of results in a desperate attempt to justify $$$ales of "Hi Rez" which of course nosed dived once people realized the scam, confirmed by M&Ms AES peer reviewed tests of actual audiophools, their hardware and purported "Hi Rez" media, the EXACT conditions they and the scam industry claimed to be able to "hear" differences.
The scam industry does not require the audiophool be "trained", the "Hi Rez" equipment/system to be certified, the room to have a specific noise floor, or the music content be cherry picked and doctored as in the BS test.
Believers and shysters will latch on to this BS test (despite most of them abhorring and utterly rejecting
Delusion
Blocked
Testing as a listening valid method -see JAs latest Stereophile edition).
The audio fashion jewelry industry will continue it's death spiral, rejecting real audio advances like PSR et al, while craving "more resolution" from the 10% 2channels of info, because they know the audiomoron has long ago ended any pretense of desire for the "real thing".
cheers,
AJ
...
Would you care to explain how the corrupt/gameable/non-proctored online amateur AVS files "test" adhered to ITU-BS1116...or BS.1534-2?
They did not and thus the results must be dismissed as ad-hoc garbage, per a highly reputable source. ...
I have followed the execution and discussion of several comparative tests recently, starting with Arnie's keys. All of the tests proved to have samples that were defective. I believe this is the issue that needs to be discussed, not the results of the tests. GI = GO.
Since there is a lack of agreement, without a clear detailed statement of what constitutes a defective sample, the above post is GI=GO.
Even if there is a clear statement but it requires compliance with some arbitrary standard that is well below perceptual limits, it is still GI=GO.
.
Apparently misleading answers is the order of the day. I see no information related to the AES paper being discussed, just some very old AES papers that are fully discredited at this point.
On-topic responses would appear to need to relate to the recent AES Conference paper titled: "Audibility of Typical Digital Filters in a Hi-Fi Playback System".
The conferencing system tries to help those of us who are literate by putting that information on the title line...
I wouldn't normally claim that the difference between the right dither and the wrong dither would be or even could be audible, but it seems possible that this test has been so carefully set up that it just might be.
Cheers,
David.
This staged test reminds me a lot of the infamous Clarity Cap white toilet paper (http://www.hificollective.co.uk/pdf/claritycapmr.pdf) farce across the pond, where they "proved" capacitor sound, using ITU-BS1116 of course, all dressed up in "University research". It was long ago pulled from their website but still exists in cyberspace.
Did they simply test some cheapo caps vs their magic ones? Nope. They fabricated a "singing" cap themselves, then compared to their own non-singing boutique cap.
Audiophools do like their doctored blind tests....when the results satiate the disorder.
cheers,
AJ
Good morning Arny.
You cannot use a single digit SPL number for your room noise.
I didn't. I gave the room noise number in terms of bits.
What is this then Arny:
So what is known about the monitoring system and room used for playback? I'll bet it wasn't your typical listening room with a 40 dB SPL noise floor...
The 40 db number without any kind of spectrum analysis is psychoacoustically blind. Isn't it?
Obviously, I'm responding to a post that appears to come from someone with severe perceptual difficulties - someone who quotes "bits" and responds to "dB SPL"
I responded to both having had the same argument with you on AVS where you keep subtracting 40 db type numbers from your peak to arrive at your dynamic range and then bits.
I get the feeling that "Arnold B. Krueger", "ajinfla", and "amirm" know each other - maybe in the real world, and certainly in cyberspace.
For the benefit of the rest of us, and maybe to prevent history from being re-run here on HA, can you guys(?) explain very briefly who you are, what you've done, and what baggage you're bringing to this discussion.
I know it's OT - maybe start a new thread if you like - but I think it would be very helpful for the rest of us.
(While people are of course free to hide behind a single username here on HA, I'm guessing that you guys are very far from anonymous and know full well who you are, so it would help if the rest of us knew too. You don't have to answer, but I'm asking nicely: please do :-) I bet you've all got websites with resumes on anyway, so just a link will be fine...)
Cheers,
David.
I get the feeling that "Arnold B. Krueger", "ajinfla", and "amirm" know each other - maybe in the real world, and certainly in cyberspace.
It is off topic, but if you must know, Amir and I grew up together and Arnie is our godfather.
Btw David, user profiles contain info towards who/what, as does mine.
And now back to our regular programming, "Audibility of
Atypical Digital Filters in
a Hi-Fi Playback System"....
cheers,
AJ
It's a bit like watching The View. I don't know what the UK equivalent would be.
You cannot use a single digit SPL number for your room noise.
I didn't. I gave the room noise number in terms of bits.
What is this then Arny:
So what is known about the monitoring system and room used for playback? I'll bet it wasn't your typical listening room with a 40 dB SPL noise floor...
The 40 db number without any kind of spectrum analysis is psychoacoustically blind. Isn't it?
It's a different post, in a different context. In the immediate context of the exact post that you claimed in your post to be responding to Amir, the noise level SPL was given in bits. This is getting tedious.
Furthermore, it is common practice to give background noise level SPLs as a single dB number:
http://www.sengpielaudio.com/TableOfSoundPressureLevels.htm (http://www.sengpielaudio.com/TableOfSoundPressureLevels.htm)
(just one of very many on the web)
I get the feeling that "Arnold B. Krueger", "ajinfla", and "amirm" know each other - maybe in the real world, and certainly in cyberspace.
It is off topic, but if you must know, Amir and I grew up together and Arnie is our godfather.
Btw David, user profiles contain info towards who/what, as does mine.
I stand behind my HA user profile.
I just checked Amir's and its blank.
Amirm is Amir Majidimehr who came up through the worlds of Unix IT and Video editing. If memory serves he has a EE from Florida Atlantic University. He was MS's VP of Windows Digital Media whose relationship with MS ended about the same time they cancelled one of his major projects, HD DVD.
AJ, Amir and I previously spent the most time together at the AVS forums.
It's a bit like watching The View. I don't know what the UK equivalent would be.
Loose Women (http://en.wikipedia.org/wiki/The_View_(U.S._TV_series)#International_variations). I've never seen it. It's daytime TV. I assume you're being insulting
Cheers,
David.
More like this (https://www.google.com/search?q=3+way+cage+match&client=firefox-a&hs=83x&rls=org.mozilla:en-US%3a%6ffficial&channel=sb&source=lnms&tbm=isch&sa=X&ei=D5ZjVN_cJOqCsQTI6oDgBQ&ved=0CAoQ_AUoAw&biw=1143&bih=546&dpr=0.9#facrc=_&imgdii=_&imgrc=mJyUFgj3NiAzXM%253A%3BPiQdp0TTVwQxVM%3Bhttp%253A%252F%252Fwww.wwe.com%252Ff%252Fwysiwyg%252Fimage%252F2012%252F06%252Fcena_no_way_out_mcmahon.jpg%3Bhttp%253A%252F%252Fwww.wwe.com%252Fshows%252Fnowayout%252F2012%252Fjohn-cena-big-show%3B642%3B361).
We hardly all agree on anything (speaker/room/treatments, passive biamping, digital disorders, etc). The only relevant thing you have to know David, is that I'm always right.
I guess this BS test shows that's it's possible to doctor up any results you desire, be it cables, amps, caps or digital.
Says nothing about the transparency of 16/44 playback itself.
Which isn't really a concern of rational people.
cheers,
AJ
It's a bit like watching The View. I don't know what the UK equivalent would be.
Loose Women (http://en.wikipedia.org/wiki/The_View_(U.S._TV_series)#International_variations). I've never seen it. It's daytime TV. I assume you're being insulting
I vote "Big Bang Theory": a bunch of clever guys with a talent for immature social behavior
How many people here have actually bought the paper?
Here's a rundown of some of the relevant things I don't think have been mentioned yet (or if they have they've largely been glossed over).
1. The aggregate result for correctly identifying a sample, across all 160 trials, is 56.25% correct.
2. When they low-passed the signal *without* quantizing it to 16-bits, they got a statistically insignificant result. Quelle surprise. Needless to say, this did not stop them declaring that the digital filters they used were clearly corrupting the sound. I shall refrain from speculating as to why they would possibly want to spread FUD about digital filters.
3. The difference between filters (when they are talking about 16-bit samples the listeners could identify) is right on the edge of statistical significance, and seems to show exactly the opposite of what they expected - their listeners seemed to find low-passing *closer* to the audioband very, very slightly less audible. They speculate that this is because the filter that cuts higher is longer, and will therefore ring more and smear things temporally.
4. They seem to think that the sampling rate imposes a time-domain resolution limitation akin to the time between two adjacent samples, which is, unless I'm very confused, a giant pile of crap.
5. Did I mention the occasionally speculatory tone?
Seriously, all this paper seems to show is that if you don't dither at all, or dither improperly, a group of carefully trained listeners will, on some sample material, be able to distinguish the 16-bit content from the original content, albeit with a fair-to-great amount of difficulty (they did much better worse with rectangular dither than with no dither). Some small differences with distinctly diminished sample sizes are then used to come to some very strange conclusions. There's all sorts of guff about temporal smearing, and a ramble through some decidedly dubious literature (yes, Ooashi et al make an appearance) to portray the 'prima facies' case for hi-res as a reasonably solid one.
all this paper seems to show is that if you don't dither at all, or dither improperly, a group of carefully trained listeners will, on some sample material, be able to distinguish the 16-bit content from the original content
Right, and most of us were already aware this was possible.
temporal smearing
I imagine they gotta keep selling the case for magical apodizing filters.
Amirm is Amir Majidimehr who came up through the worlds of Unix IT and Video editing. If memory serves he has a EE from Florida Atlantic University. He was MS's VP of Windows Digital Media whose relationship with MS ended about the same time they cancelled one of his major projects, HD DVD. AJ, Amir and I previously spent the most time together at the AVS forums.
Arny, you forgot to mention an important career move of his that 2Bdecided might consider relevant, namely that Amir founded an audio retail store which sells, among other things, so-called "high end" amps, processors, and I would assume the cables to hook them up. His AVS forum posts mention he is "retired", perhaps in an attempt to mitigate any potential claims of him having a pecuniary interest in promoting his high end industry and/or its ideologies, however the webpage of his company which his posts' signature directly links to (over 18,000 posts, by the way) makes no mention of him being "retired" from that particular company: http://www.madronadigital.com/about/About.html (http://www.madronadigital.com/about/About.html)
I have followed the execution and discussion of several comparative tests recently, starting with Arnie's keys. All of the tests proved to have samples that were defective. I believe this is the issue that needs to be discussed, not the results of the tests. GI = GO.
aka 'train wrecks'.
It appears the wreckage is still far from being cleared. Some even seem to just spread the wreckage around, rather than clean it up.
Right.
Ok, we've made introductions and are aware of the history. Future posts will be on the topic covered in the first post.
@armirm:
I'm sorry you missed this latest round, but the window to share your background, etc. is over. Whether or not you feel you've been given a fair shake is not the forum's problem.
@all:
Segues into amps, speakers, cables, personalities, backgrounds, who said what when and other proxy discussions (including the AVS debacle!) will be binned. Continued discussion will result in a curtailment of posting privileges.
If you wish to argue other topics, start a new discussion in the appropriate forum if one doesn't already exist. The same rules will apply there as well.
How many people here have actually bought the paper?
(raises hand)
Back on topic: for those who missed it, a test of filter audibility has already been attempted right here on HA...
http://www.hydrogenaud.io/forums/index.php?showtopic=68524 (http://www.hydrogenaud.io/forums/index.php?showtopic=68524)
Any new takers?
Cheers,
David.
2. When they low-passed the signal *without* quantizing it to 16-bits, they got a statistically insignificant result.
Did they specify the low-pass filter(s) for that test ?
How many people here have actually bought the paper?
I am an AES member and so I have the paper.
Here's a rundown of some of the relevant things I don't think have been mentioned yet (or if they have they've largely been glossed over).
I am pretty sure there are more relevant things than what you listed .
Seriously, all this paper seems to show is that if you don't dither at all, or dither improperly, a group of carefully trained listeners will, on some sample material, be able to distinguish the 16-bit content from the original content, albeit with a fair-to-great amount of difficulty (they did much better worse with rectangular dither than with no dither).
You just gave away the farm with that summary. If we are sensitive to such things as dither types then all the talk about any and all small distortions not being audible is proven to be wrong. These are small differences. Right? If we can hear them then we can hear a lot of other artifacts we shove under the rug.
As to "carefully trained listeners," that is a sign of merit for any such tests. Using untrained listeners goes against the best industry practices as outlined in ITU BS-1116. The training here was very limited to this one test alone. The listening group were not trained experts. They were allowed to listen to the music and hear a controlled FIR filter test.
Some small differences with distinctly diminished sample sizes are then used to come to some very strange conclusions. There's all sorts of guff about temporal smearing, and a ramble through some decidedly dubious literature (yes, Ooashi et al make an appearance) to portray the 'prima facies' case for hi-res as a reasonably solid one.
Seems like you read it in an emotional state of mind given this description. I would love to see your read of Meyer and Moran so that I get calibrated on how you judge such things. Is there a link for that?
@armirm:
I'm sorry you missed this latest round, but the window to share your background, etc. is over. Whether or not you feel you've been given a fair shake is not the forum's problem.
Hmmm. I am bombarded with posts addressing me specifically in multiple threads. I am trying to do my best to answer them. I see David's request now but didn't realize there was a time limit on answering it. Closing the opportunity for me to answer them is forum's problem. But it is your forum, you get to set the rules and my job is to follow them. And so shall I .
all this paper seems to show is that if you don't dither at all, or dither improperly, a group of carefully trained listeners will, on some sample material, be able to distinguish the 16-bit content from the original content
Right, and most of us were already aware this was possible.
I haven't seen this awareness in other forums. Is there a link here where folks have said that in double blind tests, people can tell the difference between how 24 bit files are or are not dithered to 16 bits?
temporal smearing
I imagine they gotta keep selling the case for magical apodizing filters.
And what is our theory of how listeners could tell filters apart in this blind test?
If we are sensitive to such things as dither types then all Strawman the argument to talk about any and all small distortions not being audible is proven to be wrong.
If we Red Herring the argument to can hear them then we can hear a lot of other artifacts we shove under the rug
I would love Red Herring the argument to to see your read of Meyer and Moran so that.....
FIFY.
Back on topic: for those who missed it, a test of filter audibility has already been attempted right here on HA...
http://www.hydrogenaud.io/forums/index.php?showtopic=68524 (http://www.hydrogenaud.io/forums/index.php?showtopic=68524)
Any new takers?
Cheers,
David.
Excellent! Two positive outcomes. I stand corrected. You all have tested filters and know the differences can be heard in ABX blind tests. So why all the angst over Stuart's formal test of the same?
And why is it that when the same posters come to AVS forum they don't acknowledge such data? I may have missed it of course but prior to this round of testing, no one ever had posted outputs of foobar with positive detection of small differences such as we are talking about.
But kudos to you and HA forum for creating the data and opportunity.
I gave it a try on the same clip MLXXX had done:
foo_abx 1.3.4 report
foobar2000 v1.3.5
2014/11/13 08:40:05
File A: C:\Users\Amir\Music\HA Forum Tests\limehouse\limehouse_maximum_phase_100.wav
File B: C:\Users\Amir\Music\HA Forum Tests\limehouse\limehouse_reference.wav
08:40:05 : Test started.
08:40:37 : 00/01 100.0%
08:42:19 : 00/02 100.0%
08:43:22 : 00/03 100.0%
08:44:21 : 01/04 93.8% <--- Difference found.
08:45:14 : 02/05 81.3%
08:45:21 : 03/06 65.6%
08:45:34 : 04/07 50.0%
08:45:43 : 05/08 36.3%
08:45:52 : 06/09 25.4%
08:46:00 : 07/10 17.2%
08:46:10 : 08/11 11.3%
08:46:20 : 09/12 7.3%
08:46:29 : 10/13 4.6%
08:46:39 : 11/14 2.9%
08:46:51 : 12/15 1.8%
08:47:00 : 13/16 1.1%
08:47:10 : 14/17 0.6%
08:47:18 : 15/18 0.4%
08:47:26 : 16/19 0.2%
08:47:34 : 17/20 0.1%
08:47:42 : 18/21 0.1%
08:47:49 : 19/22 0.0%
08:47:55 : Test finished.
----------
Total: 19/22 (0.0%)
As an aside, I really enjoy reading your posts. Clearly you know your stuff and stay out of the emotional components.
And here is the same using the new Foobar ABX plug-in with signatures:
2014-11-13 09:16:06
File A: limehouse_maximum_phase_100.wav
SHA1: 722dc26db8d4ce666dc03875b2c8d4570d22b521
File B: limehouse_reference.wav
SHA1: e8ad96830d23cad4bba5bf822ce875ae452b9e7c
Output:
DS : Primary Sound Driver
09:16:06 : Test started.
09:16:48 : 01/01
09:16:56 : 02/02
09:17:04 : 03/03
09:17:14 : 04/04
09:17:21 : 05/05
09:17:29 : 06/06
09:17:38 : 07/07
09:17:45 : 08/08
09:17:52 : 09/09
09:18:02 : 10/10
09:18:08 : 11/11
09:18:14 : 12/12
09:18:20 : 13/13
09:18:28 : 14/14
09:18:36 : 15/15
09:18:42 : 16/16
09:18:42 : Test finished.
----------
Total: 16/16
Probability that you were guessing: 0.0%
-- signature --
5b42b06c414b6ba77a3998695bf119a2d57663c0
Your computation there is psychoacoustically blind Arny as we have discussed. You cannot use a single digit SPL number for your room noise. You must look at its spectrum and see how much noise you have relative to threshold of hearing as I show in my article
I agree with the need to look at the spectrum, but disagree with your conclusions.
http://www.hydrogenaud.io/forums/index.php...st&p=871182 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=106552&view=findpost&p=871182)
http://www.hydrogenaud.io/forums/index.php...st&p=871270 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=106552&view=findpost&p=871270)
It's
really hard to find source material that has audible noise when you use 16-bit noise shaped dither. The current (Bob Stuart) test apparently used no noise shaping, and the wrong dither PDF.
Cheers,
David.
And verification of the signature:
Signature matches; the log appears to be valid.
For some reason cutting and pasting from the forum doesn't work. I had to feed it the text file that was stored. Maybe it is me adding formatting to it.
08:44:21 : 01/04 93.8% <--- Difference found.
What does it sound like?
And verification of the signature:
Signature matches; the log appears to be valid.
That's fair enough, but anyone who suspects cheating is unlikely to think that edited ABX logs are the way someone would do it.
Cheers,
David.
Your computation there is psychoacoustically blind Arny as we have discussed. You cannot use a single digit SPL number for your room noise. You must look at its spectrum and see how much noise you have relative to threshold of hearing as I show in my article
I agree with the need to look at the spectrum, but disagree with your conclusions.
http://www.hydrogenaud.io/forums/index.php...st&p=871182 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=106552&view=findpost&p=871182)
http://www.hydrogenaud.io/forums/index.php...st&p=871270 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=106552&view=findpost&p=871270)
It's really hard to find source material that has audible noise when you use 16-bit noise shaped dither. The current (Bob Stuart) test apparently used no noise shaping, and the wrong dither PDF.
Cheers,
David.
Sorry David. I read your link and it almosts reads word for word like my post and article. What is it that I am missing?
Meanwhile, the subject is even more complex than my brief write-up. Noise that comes from the source is played through point sources of our speakers. Listening tests show that we can hear point noise over broad noise that is resident in the room. This is no doubt due to evolution of human beings. A dangerous noise coming from a specific point trumps broad noises of the environment. There would be fewer of us if that had not been true and more of our ancestors had gotten eaten .
As to your comment regarding content, it is fine for such existence to be rare. We can afford to preserve all that can be for ALL content for no cost today. I don't want to analyze each piece of music I get to determine whether something was stepped on. We can get a format that is transparent (> great than CD specs) and be done with all of these technical arguments. One only has to understand this topic if it is to justify lowering the resolution of the content. We are in no need of such lowering technologically and from business point of view.
Meanwhile, the subject is even more complex than my brief write-up. Noise that comes from the source is played through point sources of our speakers.
Raising the question of how does one distinguish a particular noise in the source from any of the many sources of noise in the source? They all come to our ears from the same speakers...
Also, modern audio systems are surround systems, which reproduce sound from up to 11 or more separate point sources spread throughout the room. Their purpose is to recreate a sonic image that existed in the performance space here the recording was made, which means that they must blend together into a continuous sound feel that duplicates what one would hear in a room. How does one reliably distinguish that continuous sound field from the recording from the one created by various noise sources in the room?
I suspect that the well-known asymmetries in the recent Meridian tests may be tacit admission that without adding them, the listening tests would have been expected by the paper's perpetrators to produce null reslts.
Your computation there is psychoacoustically blind Arny as we have discussed. You cannot use a single digit SPL number for your room noise. You must look at its spectrum and see how much noise you have relative to threshold of hearing as I show in my article
I agree with the need to look at the spectrum, but disagree with your conclusions.
http://www.hydrogenaud.io/forums/index.php...st&p=871182 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=106552&view=findpost&p=871182)
http://www.hydrogenaud.io/forums/index.php...st&p=871270 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=106552&view=findpost&p=871270)
It's really hard to find source material that has audible noise when you use 16-bit noise shaped dither. The current (Bob Stuart) test apparently used no noise shaping, and the wrong dither PDF.
Cheers,
David.
Sorry David. I read your link and it almosts reads word for word like my post and article. What is it that I am missing?
That 16-bit noise shaped dither is sufficient in all but pathological samples? That doing all the calculations properly and fairly delivers the conclusion: 16-bits is sufficient for a home delivery audio format?
You demand a peak level that causes instant hearing damage = unreasonable
You assume no noise shaping = unreasonable
You assume the best possible noise figures at home = exceptional but reasonable
You assume the best possible noise figures in the recording venue = exceptional but reasonable
The "problem" with this scientific justification is "the elephant in the room" (from another thread): the people claiming audible superiority aren't claiming to hear it only on exceptional material recorded and reproduced in exceptional rooms. They're claiming it sounds better, always. Whereas the ABX results back up the very limited and specific scientific justification.
I'm sure you realise the problem here: people take an article like yours and ABX results like Bob Stuart's, and say "here is scientific and double-blind proof that the fantastic audible improvement I hear all the time with hi-res is real and scientifically proven."
I'm fairly sure you understand the resentment here on HA: I think most people here believe that this delusion is exactly what such "science" is intended to perpetuate, hence it's not science: it's marketing, treading the line of implying something that's not true without actually saying it.
Meanwhile, the subject is even more complex than my brief write-up. Noise that comes from the source is played through point sources of our speakers. Listening tests show that we can hear point noise over broad noise that is resident in the room.
I am aware of this (e.g. see Chapter 6 (http://mp3-tech.org/programmer/docs/Robinson_thesis.pdf)). The Binaural Masking Level Difference is well documented. I haven't seen papers examining it near the absolute threshold of hearing, but they may exist. This is only relevant where the dither noise is higher than the noise in the original recording venue and above the absolute threshold of hearing at some frequency, and only when this is not masking by the music.
As to your comment regarding content, it is fine for such existence to be rare. We can afford to preserve all that can be for ALL content for no cost today. I don't want to analyze each piece of music I get to determine whether something was stepped on. We can get a format that is transparent (> great than CD specs) and be done with all of these technical arguments. One only has to understand this topic if it is to justify lowering the resolution of the content. We are in no need of such lowering technologically and from business point of view.
We risk spreading essentially the same discussion over several threads, but I'll put it in this one for now.
I am asking "why?" You are asking "why not?" You want to reverse the burden of proof. Accepting that approach means we haven't finished yet. We will have 384kHz 32-bit next because "why not?", and that's still not the end of it. This is not engineering. This is lunacy. You want to believe there's no downside to this. I don't agree. The problems include:
1. We "improve" the wrong thing, to the detriment of other things that would introduce a real/greater audible improvement.
2. We lead the market to believe that
this is what defines audio quality. We end up with idiocy like a 128kbps 384kHz 32-bit lossy format promoted as being superior quality to CD, because high sample rates are so important. Or idiocy like recordings from truly wrecked old analogue tapes, or dynamically crushed modern recordings, marketed as "high resolution" because high sample rates are so important. And get some people to spend money on this.
3. The kind of people who care enough to want decent sound, but don't have a lifetime to devote to separating reality from bullsh*t, will just give up and go and enjoy some other hobby instead.
4. Audio drowns in its own bullsh*t.
I am conceding (as is everyone else here) that under certain pathological situations 16-bits is not quite transparent. You are saying this means we should use more than 16-bit for all content, because you don't want to check whether it's necessary. Let me save you the bother. Parts of the audio world have been looking for samples on commercially released material that require more than 16-bits for well over a decade. We can't find them.
The evidence is, in the entirety of recorded music commercially released to date, 16-bits with noise shaped dither is sufficient. Someone may well argue "let's have more, just in case" but when we can't find anything to justify the "just in case", then enquiring minds will suspect there's another motivation.
If higher sample rates sound different, despite changing nothing in the audible range, why does this happen? Some people are of the opinion that if you do something, and it makes something sound better, just do it; no need to understand why. That's not progress. If the unexpected improvement reveals some unknown truth about human hearing or audio reproduction, there's more work to do and more improvements to chase once we understand it.
With the kinds of quantities of difference that we find in blind tests, it is
very hard to prove these are not due to IM distortion.
I am interested in all of this, and I do retain an open mind. However, I think taking some imagined or real-but-tiny improvement, and selling it as the next big thing, will be bad for the audio industry. Remember last time? "SACD+DVD-A fought a war, and the iPod won."
Cheers,
David.
We talk about filters and PCM here.
Keep in mind the modern audiophile has no doubt DSD sounds better all the time as PCM because its differernt filter behaviour.
That 16-bit noise shaped dither is sufficient in all but pathological samples? That doing all the calculations properly and fairly delivers the conclusion: 16-bits is sufficient for a home delivery audio format?
I am on record agreeing with that. I hope you also agree with me that assuming "all the calculations are done properly" is not a given. Nor is it a given that is what the content producer did. MLXXX put up a version of Scott/Mark test on AVS forum where the conversion was done with Audacity. The converted version had incredibly audible noise. You would have had to been deaf to not hear it. See: http://www.avsforum.com/forum/91-audio-the...ml#post28872521 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1598417-avs-aix-high-resolution-audio-test-take-2-a-15.html#post28872521).
Now he used an older version of Audacity but who knows how it does what it does? I use Adobe Audition CC. There is no documentation on its signal processing. It for example has a quality slider. What does do exactly? It doesn't say. Scott/Mark used Sonic Solutions. They guard their resampling as proprietary and an advantage. How do you know correctness?
All of this is immaterial anyway. As I have said, you guys are still fighting a forgotten war. At the risk of repeating my position:
I am not here to prove that there is benefit to high resolution. I am here to prove that if I get the original stereo master, I don't have to worry about what you just said. Everything you say relies on the signal processing being correct and listening tests proving the same. Everything I just said requires none of that but simple logical reasoning. The original bits are available before conversion to 16/44.1. I have been having this argument for months with people but no one has shown why I am better off with someone else downgrading the bits before I get them.If there is a benefit to down conversion, I can do it myself with full knowledge of the field and not leave it to someone in music production knowing what all of this means.
Please let me know why I am facing all of this resistance to the above position.
You demand a peak level that causes instant hearing damage = unreasonable
I a disappointed to see that forum argument with someone who has so much understanding of this field. There is no risk of hearing damage. We are not talking about a steady state tone at 120 db. We are talking about peaks that last few milliseconds in music. And that is critical because hearing damage has two components: level
and duration.Here is the US occupational standards with respect to workplace noise:
You assume no noise shaping = unreasonable
Per my other post, unless you are the one producing all the music I am buying, this is not a justification for downconversion of 24 bit content to 16.
You assume the best possible noise figures at home = exceptional but reasonable
I don't want to re-buy my music just because I build a quieter listening space.
You assume the best possible noise figures in the recording venue = exceptional but reasonable
I don't want additional noise added to my bits when I can have them prior to that process. Practically every bit of recording is done in 24 bit these days. I want to have those bits. Plain and simple.
I use Adobe Audition CC. There is no documentation on its signal processing.
...
All of this is immaterial anyway. As I have said, you guys are still fighting a forgotten war. At the risk of repeating my position: I am not here to prove that there is benefit to high resolution. I am here to prove that if I get the original stereo master, I don't have to worry about what you just said.
If you distrust unknown signal processing, but hope to avoid it by buying 192/24 instead of 44.1/16, that's a pretty misplaced sense of safety you're clinging on to. Unless you're talking about a purist direct-to-stereo unprocessed audiophile recording, the downconversion is the last and least complex of many many audio signal processing options. Even in that case, it it sometimes an upconversion (e.g. 48kHz to 96kHz), and sometimes a cross conversion (e.g. DSD to 96kHz). This is no less problematic.
Please let me know why I am facing all of this resistance to the above position.
I think I explained the problem with "why not" as well as I can in my previous most. Others my have other opinions.
Cheers,
David.
The "problem" with this scientific justification is "the elephant in the room" (from another thread): the people claiming audible superiority aren't claiming to hear it only on exceptional material recorded and reproduced in exceptional rooms. They're claiming it sounds better, always. Whereas the ABX results back up the very limited and specific scientific justification.
I think you continue to think I am a different person than I am. I was instructed to not tell you my background so we are stuck with you putting me in a different bucket than I am in.
I am not here to advocate any other person's point of view. I have not taken the above position. So writing that to me and asking me to respond to it is not logical.
My position which I keep repeating completely bypasses all of these arguments and your position above. It says let's get the original bits. We can do that now as we are not bound by the CD format. We can download them. I downloaded your 24/96 Khz samples with ease. What is wrong with downloading my next album that way? And let's remember that I can download high resolution content on demand but can't do that with CDs.
AVS members who are here know this to be my position. But they are still up in arms. They are ready to hang themselves in their bedroom with this latest set of outcomes from my testing and Stuart's AES paper. Why? Because the real elephant in the room. That is, if you concede my position, you may inadvertently hand over a point to the other side. And we can't have that. Because who knows what they are going to do next if it is proven that we were wrong. That there can never be a double blind test that shows a difference between high resolution audio and reduced version to 16/44.1. I have been saying for years that such a possibility exists. But folks didn't want to hear it. So we have the mess on our hands where we said one thing, and the very standard of proof we require, proved it otherwise.
That is the real elephant in the room. And no one explains it better than a strong fellow objectivist on AVS and WBF Forums in the very first discussion thread I had with Arny:
well, in that case let me thank you [Amir] for your contributions. I KNOW I could not have kept my patience as you have, let alone maintained a sense of humour! It's funny how hard *we* can go to maintain our rightness, and how quickly that line is crossed where we no longer wish to learn (despite our objections to the contrary) where we fight tooth and nail...usually because we know our position is so tenuous that the slightest 'loss' means the whole game is over.
FFS, Amir has sat here page after page and SHOWN how, and under what possible conditions jitter may be audible. Hey, if it were a cable debate, and we showed with maths and sims that there could not possibly be a difference, well that would have proved it no? So why the **** in an 'argument' where the shoe is on the other foot does it suddenly become irrelevant what the
science says??
My take on what the fear might be is the worry of what might happen if we concede a point of argument. The 'other side' will drive a frickin lorry thru the door if we do. I mean, there only has to be ONE person who hears a power cord (for sake of illustration) in what seems to be a proper test and the whole frickin lot of the rest of them will claim it as proof that they too can hear it.
No they can't, 'one in a million' means just that. But we KNOW every single one of them thinks they can hear it, using that person as proof,
and even less urge to test the truth properly. After all it has been shown. So, we had better clamp down HARD on the one ever coming out, if only to keep the lid on the rest.
So, move on to something far less controversial than PCs, but as long as it falls into audiofool territory we had better clamp down on that too. It is just safer that way, keep each and every genie in the bottle. So the need to put amir in his place, and keep the lid hammered on tight. Because the ramifications of this little argument go waaaay past it's tiny borders.
""Oh, but amir has not given any evidence of audibilty"" (apart from the science you mean? The science that would be perfectly acceptable in a different argument, that the one we are talking about???).
Be totally honest here. If he told you that he had found, to his satisfaction, that turning the front panel on and off on his thingamabob had an audible difference, would you accept that? What then his findings of jitter? We know you would not accept his results, the genie is too terrifying to contemplate.
So don't come back at me with 'amir has yet to show audibility' ok? It is a definitional thing you know. Some things, by definition, are inaudible.
Bit like cancer, it cannot be cured hence any cure of cancer is untrue (why we are always then exhorted to donate to cancer research is beyond me). All of you could be right, it may be completely inaudible. But you sure as hell have not shown it by your arguments. Unless 'nanah nanah nah' counts as an argument.
Give it to an Aussies to say it so directly .
I'm sure you realise the problem here: people take an article like yours and ABX results like Bob Stuart's, and say "here is scientific and double-blind proof that the fantastic audible improvement I hear all the time with hi-res is real and scientifically proven."
I'm fairly sure you understand the resentment here on HA: I think most people here believe that this delusion is exactly what such "science" is intended to perpetuate, hence it's not science: it's marketing, treading the line of implying something that's not true without actually saying it.
Ah, you did get there . Yes, I understand the reaction. But I could not care less. I am an objectivists. That calls for fairness. It calls for being unbiased at all times. If there are problems with what we, our camp does, I volunteer it. I don't let it come out from another channel and make us look like crooks that were hiding it.
If the people here have a position of hanging together instead of hanging with science and objective view of audio, then you are right. I won't get along with them. My hope though is that there are many who are not that way. Who like to see transparency. After all, we enjoy that in real life. Here is a quick story I have told on AVS (you will learn that I often tell such stories ).
I was on a jury pool for a criminal case. I get invited to come and sit in the jury box while they were deciding which one of us to keep. I expected the defendant to be in his sunday clothes, looking like anyone but a criminal. But no. This guy was sitting there and if you took a picture of him, you would swear he is guilty and there would be no need for the trial. I look at the judge and attorneys but they absolutely do not reflect any of this bias. I mean the guy was slouching in his chair full of contempt for the court and the judge. Yet the judge was referring to him as he would to the most upstanding citizen. I am sure deep down he has seen so many cases as to know the guilt or innocence of the guy. But he knew he had to be fair. And put aside all bias. It gave me a lot of comfort that heaven forbid, if I am mistakenly accused and sitting in that chair, I would have a shot at a fair trial no matter how the circumstances looked.
That is me in a nutshell in these topics. I don't like it when I ask Arny to show us any documented double blind tests and he says the only one is from 30 years ago and I have to go and buy the magazine. And I buy it and see that it is an amplifier test where positive outcome was achieved. Isn't that embarrassing? Is this the way we want to appear in front of general public? And let's remember that this forum is indexed by Google and open to all to read, not just forum members.
So no, please don't ask me to worry about the clicks among forum members here or elsewhere. Folks want good side of me? They need to stick to science rather than to some camp whose goal is to ridicule our fellow audiophiles.
I am interested in all of this, and I do retain an open mind. However, I think taking some imagined or real-but-tiny improvement, and selling it as the next big thing, will be bad for the audio industry. Remember last time? "SACD+DVD-A fought a war, and the iPod won."
I already addressed this David. There is no repeat of SACD+DVD-A. There is no format war. There is no new physical format. Labels are offering high resolution stereo masters and people are buying them. Folks here can cry all they want, talk about "train wrecks" and such and it won't matter. Yes, the difference may be small or nonexistent. Your view that this data does not come out is incorrect. With advent of free Audacity, every kid is performing spectrum analysis and posting results. And on forums like WBF, people are frank and quick to slam HD downloads that don't sound better.
P.S. Gosh. This forum software gives you no chance to fix spelling errors before declaring that you have edited your post! But that is all I did .
You demand a peak level that causes instant hearing damage = unreasonable
I a disappointed to see that forum argument with someone who has so much understanding of this field. There is no risk of hearing damage. We are not talking about a steady state tone at 120 db. We are talking about peaks that last few milliseconds in music. And that is critical because hearing damage has two components: level and duration.
I understand this, but we are putting the noise shaped dither at the threshold of human hearing, and shaped to it. At this replay level, I can now hit 120dB SPL peak. Anything that gets anywhere close to digital full scale is extremely loud. If you need more bits, it's to go even louder. I understand instantaneous peaks vs RMS level very well - but hitting 126dB (17-bits), 132dB (18-bits), 138dB (19-bits), 144dB (20-bits) even briefly - come on. I hate dynamic range compression more than most, but at that kind of level I would welcome it!
Cheers,
David.
I use Adobe Audition CC. There is no documentation on its signal processing.
...
All of this is immaterial anyway. As I have said, you guys are still fighting a forgotten war. At the risk of repeating my position: I am not here to prove that there is benefit to high resolution. I am here to prove that if I get the original stereo master, I don't have to worry about what you just said.
If you distrust unknown signal processing, but hope to avoid it by buying 192/24 instead of 44.1/16, that's a pretty misplaced sense of safety you're clinging on to.
I think you misunderstood what I wrote. I want the master prior to conversion and remastering to 16/44.1. If that is the same fidelity, so be it. If it is better, that much the better .
AVS members who are here know this to be my position. But they are still up in arms. They are ready to hang themselves in their bedroom with this latest set of outcomes from my testing and Stuart's AES paper. Why? Because the real elephant in the room. That is, if you concede my position, you may inadvertently hand over a point to the other side. And we can't have that.
I would be delighted to run a fair double-blind test proving that hi-resolution audio delivers an audible advantage. There is no shame in new ABX tests which prove audible differences where previously they could not be found. Heck...
http://home.provide.net/~djcarlst/abx_md.htm (http://home.provide.net/~djcarlst/abx_md.htm)
...there would be no HA if we all believed that one!
However, "extraordinary claims require extraordinary evidence." A few of the results posted, if they were obtained fairly, are the start of some useful evidence. Many are just the result of badly set-up or performed tests. I haven't seen you discount these faulty tests, but then I haven't trawled through all the relevant threads.
And on forums like WBF, people are frank and quick to slam HD downloads that don't sound better.
You know, people who really are objective about this stuff are very rare, anywhere. On the whole, where people care about this stuff (99% don't!) I see about 70% claiming something sounds amazingly night-and-day my-wife-heard-the-difference-from-the-kitchen better just because it's an SACD, 10% reciting something they've read here or from Monty without really understanding it about why it can't possibly sound any better, 10% trying to figure out which version really sounds better, and 10% crying at the cost and effort of figuring out which of 7 different compromised versions of their favourite album is most worth listening to.
Cheers,
David.
AVS members who are here know this to be my position.
Amir, speaking just for myself, I know nothing of the sort. I actually don't know what reasonable position to ascribe to you because of your habitual self-contradiction and ignorance of posts containing reasonable questions.
I read the post by 2Bdecided @ Nov 14 2014, 02:56 (http://www.hydrogenaud.io/forums/index.php?act=findpost&pid=880696) and agreed with it as being a reasonable characterization based on my AVS experiences.
But now it has all been denied. I don't know what to think. In that context the claim "AVS members who are here know this to be my position." is simply not true.
So let's leave such arguments to AJ, mzil, Arny's of the world and keep the level of discussion at high technical level.
Yeah, at a high technical level like the +/- 10% volume method (http://www.avsforum.com/forum/86-ultra-hi-end-ht-gear-20-000/1136745-establishing-differences-10-volume-method-14.html#post16385079) and "hearing" SPDIF cables, "power regenerators", etc.
That's pretty high.
Practically every bit of recording is done in 24 bit these days.
Right, in
2ch. Because the scammers and shysters have abandoned any pretense of wanting fidelity to original soundfields, which the extra file size could be used for (via added channels) and have conned the audiomorons into believing that 99% of distributed music, like 5db dynamic range Neil Young studio electronics recordings, need anything more than 16/44 for fidelity to the real thing (whatever imaginary construct that may be).
cheers,
AJ
AVS members who are here know this to be my position.
Amir, speaking just for myself, I know nothing of the sort. I actually don't know what reasonable position to ascribe to you because of your habitual self-contradiction and ignorance of posts containing reasonable questions.
Well, your memory is as bad as mine. Fortunately the Internet keeps these things around forever. On the master thread on AVS on Monty's write-up, this is what I said on page #1 after your posts: http://www.avsforum.com/forum/91-audio-the...ml#post21750075 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1398397-24-192-music-downloads-why-they-make-no-sense.html#post21750075)
The goals for setting a standard here shouldn't be what is adequate but what has some safety margin as to give us high confidence of inaudibility. In that regard, we need to also allow for less than optimal implementations. To that end, Bob Stuart has published a much more authoritative version of this report at AES. Here is an online copy: http://www.meridian-audio.com/w_paper/Coding2.PDF (http://www.meridian-audio.com/w_paper/Coding2.PDF). These are his recommendations:
"This article has reviewed the issues surrounding the transmission of high-resolution digital audio. It is
suggested that a channel that attains audible transparency will be equivalent to a PCM channel that
uses:
· 58kHz sampling rate, and
· 14-bit representation with appropriate noise shaping, or
· 20-bit representation in a flat noise floor, i.e. a ‘rectangular’ channel"
So as we see, the CD standard somewhat misses the mark on sampling rate. And depending on whether you trust the guy reducing the sample depth from 24-bit to 16 bits, we may be missing the right spec there too.
Ultimately, I think to the extent bandwidth and storage have become immaterial for music, it is best to get access to the same bits the talent approved when the content was produced. For a high-end enthusiast, there is no need for them to shrink down what they recorded before delivery. Let the customer have the same bits and then there is no argument one way or the other (http://www.avsforum.com/avs-vb/images/smilies/smile.gif).
You spent the next month arguing with me that the above was not right starting with this response: http://www.avsforum.com/forum/91-audio-the...ml#post21750677 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1398397-24-192-music-downloads-why-they-make-no-sense.html#post21750677)
Interestingly enough this paper is neither an AES conference paper or a JAES article." At least I can't find it published that way. It appears to be a rewrite of a 1988 (24 year old!) article in the now-long-departed Audio magazine. It's a corporate white paper that has no standing as an industry standard or recommendation.
This paper is arguably part of the support for SACD and DVD-A which are now known to be failed technical initiatives that failed to make it in the mainstream consumer marketplace.
[...]
The paper in question is full of unsupported assertions. Probably the most honest statement it contains is:
[... quotes Monty's graph]
The key sentence above is: "... unnecessary reproduction of ultrasonic content diminishes performance."
Enough said, eh? ;-)
Apparently not since we are still discussing it. As you hopefully recall now, I showed that Bob Stuart's paper was a Journal paper and nothing remotely like "corporate white paper that has no standing as an industry standard or recommendation" that you said.
But now it has all been denied. I don't know what to think. In that context the claim "AVS members who are here know this to be my position." is simply not true.
My bad. I thought you would remember such argumentative threads that go on for so long. No worries. It is all linked here now. That thread was refreshed this year and I think it would be fun to see what else you had written in the past: http://www.avsforum.com/forum/91-audio-the...ml#post24395228 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1398397-24-192-music-downloads-why-they-make-no-sense-26.html#post24395228)
Arny Krüger wrote:
> "Tushar" wrote in message
> news:803qsp$2uj$1@nnrp1.deja.com...
> > Could someone explain in laymans terms how the dsd technology used
> in
> > the SACD format is different from the PCM used in the CD and if it
> is
> > superior what are the reasons.
>
> I would like to do that, but the technical literature that I've
> been able to pull together from various sources so far lacks the
> detail I feel I need to reliably do so.
>
> Reading between the lines and speculating wildly, SACD seems to me
> to be a bit stream-oriented digital data coding technique, one that
> effectively uses data words of various lengths for different parts
> of the audio spectrum and/or sound levels.
There seem to be claims
> that such data that is transmitted is not subject to lossy
> compression, but if, as I may erroneously or correctly infer,
> different parts of the frequency and/or amplitude domains are coded
> with different length data words, then it SACD is in fact a form of
> perceptual (lossy) coding. FWIW, HDCD seems to have implemented a
> subset of these benefits.It doesn't get worse than confusing HDCD with SACD. Or saying SACD is a form of lossy compression.
TLDNR
and
TSDNR
Practically every bit of recording is done in 24 bit these days.
Right, in 2ch. Because the scammers and shysters have abandoned any pretense of wanting fidelity to original soundfields, which the extra file size could be used for (via added channels) and have conned the audiomorons into believing that 99% of distributed music, like 5db dynamic range Neil Young studio electronics recordings, need anything more than 16/44 for fidelity to the real thing (whatever imaginary construct that may be).
You are suffering from double confusion on this matter AJ. First you did not understand my post where I was talking about recording being in 24 bits. Just about everyone uses 24 bits there and if you want to dispute that, I don't know what to tell you.
On adding more channels, the market has spoken. The music buying industry is young people who want their tunes with headphones and such so we are stuck with stereo. Multi-channel music's era came and gone. Sure, there is a path through Blu-ray but highly limited. If you want to boil the ocean by advocating that the whole market transform around your wishes, go ahead. I don't like hitting my head against the wall nearly as much as you do.
However, "extraordinary claims require extraordinary evidence."
They don't really. This line came from a couple of devotees in the skeptic camp and has no other foundation. I don't know why it keeps getting repeated in these discussions as if it is a universal truth.
That aside, who is to say what is extraordinary? To my wife, it is extraordinary claim that blu-ray looks better than DVD. She thinks they look the same. Do I need to come up with extraordinary proof to show she is wrong? I do not.
To many lay people me passing the 320 kbps MP3 test is extraordinary. Are we going to let them set the standard and make me jump through a thousand hoops to prove otherwise? Not logical yet that is what I had to do for Steven as he kept insisting that he knows more than me.
A few of the results posted, if they were obtained fairly, are the start of some useful evidence. Many are just the result of badly set-up or performed tests. I haven't seen you discount these faulty tests, but then I haven't trawled through all the relevant threads.
I will do so now. I discount all of them. Every test has flaws. But we are told that if we pass a "DBT ABX" folks would believe. Well, we passed them and passed them all. Yet, it has managed to do absolutely nothing. Now they are all out to discredit validity of these tests. Which way is up now? We have schmucks now saying we have to have witnesses to prove we did not doctor up some log to win a point online. Really?
I did not say that we can prove anything with these tests. I was told passing would make people believe so I participated. Now folks want to distance themselves from the very tests/results. Clearly they were misguided in thinking I could not pass the tests, right?
I don't care to prove to anyone here high resolution is better. I want to show them that attempting to bully people into submission by demanding ABX tests and throwing stuff they have read online about audio science can backfire and backfire big. They have two choices now. Change their views or stay with them and look totally illogical and biased to the core. The latter is what they are doing.
Multi-channel music's era came and gone. Sure, there is a path through Blu-ray but highly limited.
Really? The majority of homes have 2ch vs MCH systems? Ever been to a movie theater?
The MCH systems are already in place. Whats missing is the sanely done encoding/
content/decoding (like PSR!!!)....which the "High end" market
should be leading the charge for...instead of "Hi-Rez" (or heaven forbid analog) 2ch audio fashion jewelry moronic stupidity.
If you want to boil the ocean by advocating that the whole market transform around your wishes, go ahead.
Like your "Hi-Rez" original 2ch file distribution demands? That no one can hear in supervised, non-doctered up tests, as your friend and real credentialed expert JJ keeps telling you? What % of the "whole market" is demanding "Hi-Rez"?
I don't like hitting my head against the wall nearly as much as you do.
By now you should know I much prefer wailing your head with the weapons/ammo you always provide me with (like ITU-BS1116, thanks, even if it flies over Davids head ).
cheers,
AJ
The answer has been given - good content representative of what consumers listened to was apparently not used. Tricked up laboratory freak recordings were apparently used.
Do we actually know that to be true though, meaning MUSIC recordings were used, and not some concocted test generated signal? Although you made a good point that it would be easy to use a dynamic range expander to gimmick an existing music recording to make it seem to have an unusually wide dynamic range, which strains 16/44 (especially if it's been purposefully hobbled), it has always struck me that the paper's abstract reads: "Two main conclusions are offered: first, there exist audible signals that cannot be encoded transparently by a standard CD..." Note they used the word "signals", not "music".
Multi-channel music's era came and gone. Sure, there is a path through Blu-ray but highly limited.
Really? The majority of homes have 2ch vs MCH systems? Ever been to a movie theater?
I have not in a long time. Last time I went though was for watching a movie, not listening to music. I used to live in Florida in 1980s. Do they now play music in the theaters there?
The MCH systems are already in place. Whats missing is the sanely done encoding/content/decoding (like PSR!!!)...
No. Really? That is what I said and you thought I was talking about equipment. Bad AJ. Bad!
which the "High end" market should be leading the charge for...instead of "Hi-Rez" (or heaven forbid analog) 2ch audio fashion jewelry moronic stupidity.
Leading the charge? Now that would be stupid. High-end customers make such a tiny portion of the market that no one would care if they all went on hunger strike. But since it is important to you let's see you lead the charge. Go ahead and I will be behind you (not really but I am here for moral support).
If you want to boil the ocean by advocating that the whole market transform around your wishes, go ahead.
Like your "Hi-Rez" original 2ch file distribution demands? That no one can hear in supervised, non-doctered up tests, as your friend and real credentialed expert JJ keeps telling you? What % of the "whole market" is demanding "Hi-Rez"?
You seem very agitated AJ. Everything OK in life? When was your last physical? Measuring blood pressure once in a while is a good idea.
I don't like hitting my head against the wall nearly as much as you do.
By now you should know I much prefer wailing your head with the weapons/ammo you always provide me with (like ITU-BS1116, thanks, even if it flies over Davids head ).
cheers,
AJ
There. You got it out of your system. You can do your victory dance now.
The answer has been given - good content representative of what consumers listened to was apparently not used. Tricked up laboratory freak recordings were apparently used.
Do we actually know that to be true though, meaning MUSIC recordings were used, and not some concocted test generated signal? Although you made a good point that it would be easy to use a dynamic range expander to gimmick an existing music recording to make it seem to have an unusually wide dynamic range, which strains 16/44 (especially if it's been purposefully hobbled), it has always struck me that the paper's abstract reads: "Two main conclusions are offered: first, there exist audible signals that cannot be encoded transparently by a standard CD..." Note they used the word "signals", not "music".
Actually I have far better knowledge now. I reactivated my AES membership and obtained a PDF of the convention paper for the princely sum of $5.00. ;-)
Here's what is known about the playback system:
"To minimise unknown variations in the signal, we
implemented our test lters at 192 kHz, so that orig-
inal and test signals were presented as 192 kHz 24-
bit PCM format to the system independent of test
condition. Signals were sent to a Meridian 818v2
Reference Audio Core using a high-quality USB ca-
ble, which was connected to a pair of Meridian
DSP7200SE digital loudspeakers. This signal path
was digital from the computer into the loudspeak-
ers, and entered the analogue domain in the indi-
vidual 192 kHz D/A converters feeding each loud-
speaker drive unit. The playback system had a wide
frequency response (up to 40 kHz determined by the
tweeter), had very low di erential group delay, and
showed a compact impulse response with insigni -
cant ringing.
"The volume control of the playback system enabled
listeners to adjust playback level in 1 dB steps, pre-
sented via a UI showing relative volume scale of 0-
99 dB. During the tests, listeners were asked to listen
with the volume control in the region of 74{76 dB,
which they could adjust for comfort. For a system
gain of 75 dB, the loudest peak passage was mea-
sured as 102 dB SPL at the listening position, some-
what lower than the level we would expect from a
live performance at a distance of 3 m. This level was
chosen for comfort, and because it was high enough
for details to be audible but also low enough that 16-
bit RPDF dither would be inaudible at the listening
position [30]. The system acoustic gain of 105 dB
SPL for 0 dBFS was con rmed using an octave-wide
band of pink noise centred at 1 kHz with an average
level of -20 dBFS, which gave a level of 85 dB SPL
at the listening position.
Here's what is known about the recording
"The recordings used were selections from The signals
used were extracts from Haydn's String
Quartet Op.76 No.5 in D \Finale, Presto" from
\Nordic Sound (2L Sampler)" 1, issued with a sample
rate of 192 kHz using 24-bit PCM23.
This recording can be downloaded in a number of different formats from here:
https://shop.klicktrack.com/2l/35847 (https://shop.klicktrack.com/2l/35847)
Price in 24/192 format: $6.00
I have it on hand and it seems to be a fairly typical string quartet recording, kinda phasey sounding and lots of room reverb, with what appears to be room tone that is about 70 dB below peak recorded levels.
The recording contains over 6 seconds of room sound before and after the playing of the musical selection. There are also brief segments when no music is being played at appropriate intervals within the musical selection.
"This piece of music was selected for its suitability of
recording quality, dynamic range, instrumental tex-
ture and ease of performing the test, based on pre-
liminary data that we do not include here. Subjec-
tively, this recording was clean and had low back-
ground noise. The recording was apparently free
from modulation noise and contained spectral con-
tent up to at least 40 kHz.
This is what is known about the test procedures:
"The recording was divided into musical phrases to
reduce possible distraction. The mean length of
phrase was 11.9 s with a standard deviation of 1.8 s.
The whole recording was used. A table giving the
start and end times of the sections used here is given
in an Appendix to this paper.
Can we stick to the facts at hand rather than digging up dirt, assassinating each others character, and other ad hominem tactics which stand only to antagonize rather than progress the discussion?
My use of a question mark was rhetorical (this time).
The recording was apparently free
from modulation noise and contained spectral con-tent up to at least 40 kHz.
40 kHz at how many dB down though, and was there nearby HF sound to perceptually mask it? [Pretending for the moment those frequencies are even audible, of course.] String instruments aren't exactly known for having strong ultrasonics in the absence of nearby maskers.
"The whole recording was used. A table giving the start and end times of the sections used here is given in an Appendix to this paper."
Since the WHOLE recording was used, then by definition some of those segments contain musical silence or at least the musical lulls you spoke of. This would make noise floor comparisons quite easy if conducted in a dead quiet listening room, despite his claims "there was no rectangular dither problem, trust me" perhaps, and enough to show statistical significance.
The volume control had "1dB increments" but I sure hope that wasn't the same contraption to do the level matching! Does he say to what precision that was accomplished?
Did the listeners hold an A/B toggle and could switch at will or were they told, "This one is A, the computer display will show when it switches to B automatically for you, etc." ?
Unless the speakers were used near field, then their actual room SPL at 40 kHz comes into question since the polar response pattern of tweeters [one inch dome, "semi horn loaded"] up at that frequency becomes extremely narrow and they shoot out like a laser, or say a spot light instead of flood light, and the actual acoustical power in the far field is steeply reduced at that frequency. Does the paper explicitly say the 40 kHz was MEASURED at the seated position [and with what distance, mic, and with what SPL meter that goes THAT high?] or is his claim of 40kHz actually just the 1m response as typically measured anechoicly, as I suspect?
1. We "improve" the wrong thing, to the detriment of other things that would introduce a real/greater audible improvement.
Bingo. While Amirm and Stuart and Neil Young are flogging this technology, we're still getting 'loudness wars' mastering. And we are still stuck ina 2-channel paradigm. And listening in rooms that are often horrendous, acoustically.
What the hi rez cheerleading squad is doing is akin to trying to focus all attention on the
font of a document that is too often contains gibberish.
The question is why...why would this be the focus, when *all of them* know where the actual 'obvious and audible and known about for years' problems lie?
Signals were sent to a Meridian 818v2
Reference Audio Core using a high-quality USB ca-
ble, which was connected to a pair of MeridianDSP7200SE digital loudspeakers (https://www.meridian-audio.com/en/collections/products/special-edition-dsp7200-digital-active-loudspeaker/48/specification/).
For a system gain of 75 dB, the loudest peak passage was mea-
sured as 102 dB SPL at the listening position, some-
what lower than the level we would expect from a
live performance at a distance of 3 m.
Did you know that Beryllium domes can have significant oil can resonances well above 22.05Khz?
(https://www.madisoundspeakerstore.com/images/products/D2908_7140-graph.jpg)
I wonder what would happen if you drove such a direct radiator dome to > 108db, when direct radiator domes start exhibiting serious non-linearities >95db or so, without the benefit of a 22k low pass filter?
Could it result in lower band IM products?
Is that why Soundstage never tests direct radiator domes >95db (http://www.soundstagenetwork.com/index.php?option=com_content&view=article&id=1043:nrc-measurements-magico-s5-loudspeakers&catid=77:loudspeaker-measurements&Itemid=153) when they start exhibiting a full 3db of compression?
I'm very curious what types of measurements BS & Co did here.
cheers,
AJ
Their higher end speaker's response above these amazingly can defy Nyquist with a 44.1 or 48kHz sampling rate input!
"Specifications DSP8000 SE
PERFORMANCE
## Frequency response in-room 20Hz – >32kHz ±3dB (for inputs at 44.1kHz or 48kHz)."
Unless the speakers were used near field, then their actual room SPL at 40 kHz comes into question since the polar response pattern of tweeters [one inch dome, "semi horn loaded"] up at that frequency becomes extremely narrow and they shoot out like a laser, or say a spot light instead of flood light, and the actual acoustical power in the far field is steeply reduced at that frequency. Does the paper explicitly say the 40 kHz was MEASURED at the seated position [and with what distance, mic, and with what SPL meter that goes THAT high?] or is his claim of 40kHz actually just the 1m response as typically measured anechoicly, as I suspect?
Normal, sane speaker reviewers never show us what the polar response is at 40 kHz, since it is immaterial to human perception, but look how much narrower it is already at 30 kHz, compared to 10 kHz:
(http://player.slideplayer.us/1/3320/data/images/img23.jpg)
40kHz would be even worse and narrower still, so the room's actual power response at 40kHz, measured in the far field, would be tiny. [Plus ultrasonic content diminishes with distance more quickly, to a degree, and doesn't reflect off of typical room surfaces as well. It tends to dissipate and scatter more than "bounce".]
"
To actually hear ultrasonics the listener must be very carefully aligned with the driver,
both horizontally and vertically...This precision in alignment is unlikely in music listening.". - Griesinger
PPT slide on Ultrasonic Directivity: http://slideplayer.us/slide/3320/ (http://slideplayer.us/slide/3320/)
40kHz would be even worse and narrower still, so the room's actual power response at 40kHz, measured in the far field, would be tiny. [Plus ultrasonic content
diminishes with distance more quickly, to a degree, and doesn't reflect off of typical room surfaces as well. It tends to dissipate and scatter more than "bounce".]
"To actually hear ultrasonics the listener must be very carefully aligned with the driver, both horizontally and vertically...This precision in alignment is unlikely in music listening.". - Griesinger
PPT slide on Ultrasonic Directivity: http://slideplayer.us/slide/3320/ (http://slideplayer.us/slide/3320/)
Meridian doesn't explicitly claim that the differences that their successful listeners heard between filters, was due to hearing ultrasonics.
Of course, that still leaves the question of what it is they *did* hear. The paper is rather vague on that score. 'Temporal smearing' is offered as one hypothesis.
"This level was
chosen for comfort, and because it was high enough
for details to be audible but also low enough that 16-
bit RPDF dither would be inaudible at the listening
position [30]."
@Arny, what reference does "30" lead to? Notice he wrote, "so 16 bit RPDF dither WOULD be inaudible" so apparently an actual test of humans, for noise audibility, in situ, was NOT performed. Had it been he would have presumably said so, and written "so 16 bit RPDF was kept to an inaudible level". Instead, by the use of the word "would", his assertion that it was kept to an inaudible level seems based on theory and conjecture. [Pending whatever "[30]" refers to].
New topic. Mark Waldrep from AIX records gave us the redbook vs. Hi res files in those AVS forum threads, to compare, having never checked them for level matching nor time alignment after his expensive, high end studio gear did the down res conversion, so if Stuart et al. has made the exact same mistake, that could be the reason why his subjects found audible differences. [90 correct answers in 160 trials, as I understand it] So Arny, does the paper make ANY mention of level matching or testing for time alignment (I presume in the digital domain) and if so how it was done?
[30] J. R. Stuart. Noise: methods for estimating
detectability and threshold. AES 93rd Convention, Berlin, 1993.
I am unable to find a mention of whether/how level-matching and time alignment were verified.
@Arny, what reference does "30" lead to? Notice he wrote, "so 16 bit RPDF dither WOULD be inaudible" so apparently an actual test for this, in situ, was NOT performed. Had it been he would have presumably said so, and written "so 16 bit RPDF was kept to an inaudible level". Instead, by the use of the word "would", his assertion that it was kept to an inaudible level seems based on theory and conjecture. [Pending whatever "[30]" refers to].
J. R. Stuart. Noise: methods for estimating
detectability and threshold. AES 93rd Conven-
tion, Berlin, 1993.
The problem is that AFAIK Stuart has never before based any of his claims on relable listening test. Any DBT that is criticallly contingent on potentially flawed principles is obviously highly suspect.
New topic. Mark Waldrep from AIX records gave us the redbook vs. Hi res files in those AVS forum threads, to compare, having never checked them for level matching nor time alignment after his expensive, high end studio gear did the down res conversion, so if Stuart et al. has made the exact same mistake, that could be the reason there are audible differences. So Arny, does the paper make ANY mention of level matching or testing for time alignment (I presume in the digital domain) and if so how it was done?
I think that the general answer to that question and questions like it is to try to duplicate the Stuart results using procedures and principles that are at our far higher level of scientific orthodoxy.
I have what the footnotes to the Stuart paper claims to be the music they used, and I have their list of cut points and statements about which are the most diagnosis segments. Each segment is about 15 seconds long and is thus AFAIK suitable to public posting without concerns about copyrights. I could put a few of them up and interested parties can listen with Foobar2000 R2.
Comments?
This recording can be downloaded in a number of different formats from here:
https://shop.klicktrack.com/2l/35847 (https://shop.klicktrack.com/2l/35847)
$38 for the MCH version of album.
$42 for the "eXtreme Definition" 352kHz/24bit....2ch.
Requires special iso-ward room, doctored dither, direct radiator beryllium domes driven >105db, to be just barely distinguished by trained listeners.
I think I'll stick to used CDs on Amazon for $2....
cheers,
AJ
I could put a few of them up and interested parties can listen with Foobar2000 R2. Comments?
So that in a week from now we can read some pompous, self serving propaganda in another forum brag about how "Not only myself but even some of the doubters at HA had to admit THEY could even distinguish the difference between Arny's new files, proving Stuart's paper is correct!" And by "others" that'll mean me, simply because I found an artifact you later had to correct for? No thanks.
When that manipulation happens, as I believe you are aware has already occurred with the earlier tests, it's an abusive misrepresentation which soils my reputation by falsely aligning my results and conclusions with that of a manipulative con artist for his own benefits, and stopping this established pattern of abuse (which makes me feel stepped on) is completely impossible. Post them if you wish, but I probably won't be participating.
Also, you do realize that foobar ABX v2 is still in beta, a work in progress, and even with its new "verified signature" thing, which a computer wiz could forge, there are still numerous, easy, free OTHER ways to post fake results.
Any failure to detect differences will be dismissed by our detractors simply because none of us are using the correct $20K+ Meridian speakers, etc., plus none of us have appropriate listening environments with single digit SPL background noise levels. Hearing the modulated noise from the rectangular dither would be dependent on that.
Regarding Stuart's reference to his own study, I agree it seems to be based on theory, predictions, and models, not actual human studies.
[It also is about steady state noise, not modulated noise, I believe.]
@Arny, what reference does "30" lead to? Notice he wrote, "so 16 bit RPDF dither WOULD be inaudible" so apparently an actual test for this, in situ, was NOT performed. Had it been he would have presumably said so, and written "so 16 bit RPDF was kept to an inaudible level". Instead, by the use of the word "would", his assertion that it was kept to an inaudible level seems based on theory and conjecture. [Pending whatever "[30]" refers to].
J. R. Stuart. Noise: methods for estimating
detectability and threshold. AES 93rd Conven-
tion, Berlin, 1993.
That is a superb paper by the way. It is the closest thing to figuring out how to correlate signal noise to threshold of hearing that was determined by Fletcher/Munson's listening tests which used test tones. I am not the only one who thought so. The paper was elevated to higher status and published in the Journal of AES: http://www.aes.org/e-lib/browse.cfm?elib=6959 (http://www.aes.org/e-lib/browse.cfm?elib=6959)
Noise: Methods for Estimating Detectability and Threshold
JAES [Journal of Audio Engineering Society] Volume 42 Issue 3 pp. 124-140; March 1994
The problem is that AFAIK Stuart has never before based any of his claims on relable listening test. Any DBT that is criticallly contingent on potentially flawed principles is obviously highly suspect.
So now your position is that the Journal of AES publishes papers that are based on flawed principals?
When had you read this paper and where has it gone wrong? Seeing how the paper overlays modeling and predictions on top of the Fletcher and Munson listening tests, I am at a loss as to why you think it did not include such.
Regardless, he has now demonstrated his analysis in the current listening test. He has both types of evidence. But apparently to you all more is needed???
Seems to me if pope showed up here, you all would question him on his christianity if you happen to disagree with his views!
You guys live and die by listening tests and "peer review." Both are present here. Neither is now good enough. The only thing that is good enough is your lay opinion of these matters it seems. No one else can know more with a different view.
I could put a few of them up and interested parties can listen with Foobar2000 R2. Comments?
So that in a week from now we can read some pompous, self serving propaganda in another forum brag about how "Not only myself but even some of the doubters at HA had to admit THEY could even distinguish the difference between Arny's new files, proving Stuart's paper is correct!" And by "others" that'll mean me, simply because I found an artifact you later had to correct for? No thanks.
That's right. It is all about forum battles. Heaven forbid you learn something from another listening test. I am not afraid of all of us failing to tell the files apart. Because I know we will learn something and that is what I value. Not winning or losing a battle. Let me know if and when you ever find yourself in that situation.
Unbelievable....
You forgot independent verification.
You forgot independent verification.
There was none for Meyer and Moran. Or are you referring to some other test I don't know about???
What gave you the impression that I was referencing any specific test???
So you weren't making a general comment about HA?
What gave you the impression that I was referencing any specific test???
I don't know. I just work here .
This thread is comparing and contrasting the work of Stuart vs Meyer and Moran. You put fault at the feet of Stuart for lack of independent verification. Is it unreasonable to ask if you damn the Meyer and Moran test the same way?
So you weren't making a general comment about HA?
No. I am referring to the vocal few who think we have a "train wreck" on our hands, that a published test by respected industry members is to be ignored and hobby test run by Meyer and Moran believed. I am talking about people who roam many audio forums beating people into submission demanding ABX tests and then when the ABX test is performed, turn around and find every which way, including calling the ethics of the person who bothered to do as they say into question. The list includes Arny, Steven (Krab), Mzil and AJ.
And please forgive me for being blunt but I now include you in that list given your personal commentary about me and the work that I had produced. But not the forum at large. Not at all.
So you weren't making a general comment about HA?
No.
I was.
This thread is
You'll have to run that by krab. He has a bad habit of poorly titling and/or poorly defining his discussions as well as using other forums to fight his battles via proxy. This one is no exception.
Since we can't manage to shame him into stopping, the best I can do is contain it.
...and now you're here.
So now your position is that the Journal of AES publishes papers that are based on flawed principals?
So it is your position that with its papal infallibility still intact, the AES has never published a paper containing a technical error? ;-)
Seems to me if pope showed up here, you all would question him on his christianity if you happen to disagree with his views!
Yes. It's what many Christians inside and outside of the Roman Church have done for millennia. Ever hear of Martin Luther? ;-)
You guys live and die by listening tests and "peer review." Both are present here. Neither is now good enough. The only thing that is good enough is your lay opinion of these matters it seems. No one else can know more with a different view.
One of my very recent posts on the matter called for people to try to duplicate Meridian's results, only using greater scientific rigor. Got a problem with that, Amir?
(I guess I missed the white smoke that went up when Stuart was elected pope of all things audio.);-)
One of my very recent posts on the matter called for people to try to duplicate Meridian's results, only using greater scientific rigor. Got a problem with that, Amir?
He's just going to deflect this by raising M&M again.
This thread is comparing and contrasting the work of Stuart vs Meyer and Moran.
Nope. This thread is about the BS paper.
You insist on comparing to Meyer & Moran, because you are Amir the illogical and thus constantly Red Herring, Ad Hominem (their "credentials", etc, etc) arguments.
The BS paper abstract is how to dither doctor concoct 16/44 vs Hi-Rez audibility, with direct radiator Beryllium dome tweeters at >105db in an iso-ward. We'll see how that pans out.
The M&M paper was about lots of Hi-Rez audiophiles, using their available Hi-Rez recordings, on their high end systems to (not) hear 16/44 vs Hi-Rez audibility in their listening rooms.
IOW, reality vs a questionable abstract concoction.
One the Hi-Rez audiophile buying public can hang their hat on, the other for desperate shills with vested pecuniary interests.
No comparison, really.
cheers,
AJ
So now your position is that the Journal of AES publishes papers that are based on flawed principals?
So it is your position that with its papal infallibility still intact, the AES has never published a paper containing a technical error? ;-)
We are not talking about AES. We are talking about the Journal of AES. I am told no matter how many faults Meyer and Moran test had, the fact that it appeared in the Journal, made it immune to such criticism. You on the other hand with a broad brush dismissed a Journal paper where peer review actually has some meaning, and said
"potentially flawed principles is obviously highly suspect."What does "potentially" mean? You either know the principal and can tell us what is wrong with it. Or you can't.
So state what you think those principles are. Tell us what is wrong with them. And give us a reference to when you had actually read the paper. Here is mine in the WSR article I wrote a couple of years ago: http://www.madronadigital.com/Library/RoomDynamicRange.html (http://www.madronadigital.com/Library/RoomDynamicRange.html).
Absence of that I am going to state that you have not read the paper, nor are aware of what principals it is using. So your comment about it was without foundation.
One of my very recent posts on the matter called for people to try to duplicate Meridian's results, only using greater scientific rigor. Got a problem with that, Amir?
No, I chimed in to support it. It was mzil who said should we hear the same differences, the game will be over and there will be lots more crying and riots in streets.
(I guess I missed the white smoke that went up when Stuart was elected pope of all things audio.);-)
Compared to your knowledge of signal processing, that of mzil, Steven and Krab, he could not have higher standing. He is one of the luminaries in this field. He is an AES Life Fellow. This is why he got it: http://www.aes.org/awards/?ID=1510 (http://www.aes.org/awards/?ID=1510)
The AES recognizes and honors those who have made outstanding contributions to the field of audio in engineering, technology, service, and the arts through its Awards Program.
In 1996, Robert Stuart was presented with the AES Fellowship Award for important insights into the models of auditory perception and their applications to audio technology.In other words, he was recognized as a thought leader in the very field that you say his paper has flawed principals.
So yes, he absolutely trumps you Arny. Doesn't mean you can't find faults in his work. You can. But absence of that your blanket statement means absolutely nothing whatsoever against one of the luminaries in the industry.
Nice argument from authority. The funny thing is it wasn't necessary, but you couldn't help yourself and put it in anyway.
So, who is Steven?
He is one of the luminaries in this field. He is an AES Life Fellow.
In other words, he was recognized as a thought leader in the very field that you say his paper has flawed principals.
Hey Amir,
Completely theoretical, but, if you were say, a MS exec and had to hire such a luminary in the fields of signal processing and perception, who would you chose, BS or JJ (a Hi-Rez heretic)?
cheers,
AJ
I imagine it depends on what you're trying to sell.
I am here to prove that if I get the original stereo master, I don't have to worry about what you just said.
The above says nothing about logic and reason, so it must be about irrational fears. Why should other audiophiles care about your irrational fears?
$42 for the "eXtreme Definition" 352kHz/24bit....2ch.
I thought you were kidding. Then I clicked the link.
So now your position is that the Journal of AES publishes papers that are based on flawed principals?
So it is your position that with its papal infallibility still intact, the AES has never published a paper containing a technical error? ;-)
We are not talking about AES. We are talking about the Journal of AES.
Actually my statement talks about both the AES and the Journal of the AES, so your answer is totally non-responsive.
I am told no matter how many faults Meyer and Moran test had, the fact that it appeared in the Journal, made it immune to such criticism.
By whom? Me? Not in this life! I am of the opinion that the Meyer and Moran paper is highly flawed because they innocently took industry claims including specific claims by Stuart at face value.
Amnir, you've got quit making up false claims about other people and then arguing with them.
Nice argument from authority. The funny thing is it wasn't necessary, but you couldn't help yourself and put it in anyway.
So, who is Steven?
Steven is Krabapple.
On the first comment, that is a forum debating technique which has no parallel in real life. In real life my doctor's medical training and degree trumps my lay opinion without being called argument from authority. In a patent dispute I can bring expert witnesses and they are not being dismissed as arguing from authority.
In forums we like to pretend that such real life experience has no value and hence our lay opinion matters as much. We build up a portfolio of these phrases we throw at people and stand back as if it have real meaning. They don't. Stuart is an authority in signal processing and psychoacoustics. That authority has value way, way above someone like Arny just dismissing what he has written as being faulty principals. Stating that he is wrong by showing Stuart's credentials, is logical, appropriate and cannot be dismissed with your comment.
Ditto for other phrases like "strawman," "extraordinary claims require extraordinary proof," etc. All are nonsense. So please don't use them against me or I will write a much longer response next time.
Nice argument from authority. The funny thing is it wasn't necessary, but you couldn't help yourself and put it in anyway.
So, who is Steven?
you know: me. Shameless bad-title-maker and proxy-battler.
Nice argument from authority. The funny thing is it wasn't necessary, but you couldn't help yourself and put it in anyway.
So, who is Steven?
you know: me. Shameless bad-title-maker and proxy-battler.
Why don't you tell them what your field of study is and how hobbyists routinely present data that you use in your everyday profession.
He called you out twice. I guess you're doubly-important.
So please don't use them against me or I will write a much longer response next time.
The question remains as to whether it would be worth reading.
This thread is comparing and contrasting the work of Stuart vs Meyer and Moran.
Nope. This thread is about the BS paper. You insist on comparing to Meyer & Moran, because you are Amir the illogical and thus constantly Red Herring, Ad Hominem (their "credentials", etc, etc) arguments.
This thread was originally started and titled (by me) as 'A follow up to Meyer & Moran?'. BS & Co's convention paper casts itself as a rebuttal to M&M, which is mentioned/cited profusely therein.
This new convention paper is merely the latest salvo in a battle that goes back the late 1990s in AES, as previously documented on HA here (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&view=findpost&p=878502)
I get a sense that the warring parties perhaps do not think very highly of each other.
So please don't use them against me or I will write a much longer response next time.
The question remains as to whether it would be worth reading.
Usually not but may want to flip the coin once in a while and do so....
I might get 10 out of 10 if I do it long enough and you can congratulate yourself on a job well done. I may or may not mean my reading roughly half of your posts.
So please don't use them against me or I will write a much longer response next time.
The question remains as to whether it would be worth reading.
Than answer is: it depends on how much you like going down rabbit holes.
But even then maybe worth it sometimes. Or at least being aware of the rabbit holes. What you call 'battle by proxy' isn't just a personal vendetta. The narrative pushed by the Amirs and Atkinsons and Stuarts of the world is what we at HA, and people like Monty, and members of AES who concur with us, get as blowback when we try to inject sense into audio. Because you can bet for sure that
they -- the Amirs, Atkinsons, Stuarts -- will
not be the ones cautioning audiophiles and marketers not to oversell the benefits of hi rez.
So if you care about this stuff at all it's good to be aware of what story is being propagated/propagandized on other audio forums.
In real life my doctor's medical training and degree trumps my lay opinion without being called argument from authority.
That authority has value
Fantastic Amir, we agree!
Ok, now as your favorite audiophile disorder doctor, I prescribe you this:
Let's start: "how do you even imagine that one can hear a difference between two systems, one with noise 98 dB down and the other 146dB down, when the level is set to peak at 96dB?"
How do you explain how "obviously difference" fails to show up in even the worst kind of ABX test?
I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener. Have you any, bearing in mind that citing non-blind-testing proves nothing but the incompetence, the complete and total incompetence, of the person citing it as evidence.
Bear in mind the hard evidence for the persistance of loudness memory while you're at it.
I have my doubts that SACD or DVDA are much, if any, of an improvement, but the test is just blisteringly hard to run, and more likely to respond to artifacts, either positively or negatively, than it is to actual differences. Time alignment, level alignment, frequency response in-band can all throw it positive, lack of training, bad test environment, bad time alignment, etc, can also cause false negatives. Subject verification, likewise, is an important issue.
So, I remain undecided, but I note that I own a lot of CD's and not a single SACD or DVDA, except for some people have sent me.
Amir, hopefully this meets your luminary-authority requirements sufficiently as not to question it?
Quien es mas macho, BS o JJ?
cheers,
AJ
He called you out twice. I guess you're doubly-important.
For the record, the first time (http://www.hydrogenaud.io/forums/index.php?showtopic=90464) was when he was grabbed a graph from a Hydrogenaudio wiki and grossly overselling/misinterpreted it on AVS Forum.
He was asked but refused to come here to answer for that.
So if you care about this stuff at all it's good to be aware of what story is being propagated/propagandized on other audio forums.
My takeaway is that this further underscores the need for the parties involved to agree to adhere to a pre-chosen and carefully defined topic, and call-out those who try to steer it away to an alternate narrative.
Failing to do this can and will lead to train wrecks, especially when taking into account the history of the participants.
I bin off-topic posts like this for a reason:
http://www.hydrogenaud.io/forums/index.php?showtopic=107484 (http://www.hydrogenaud.io/forums/index.php?showtopic=107484)
Dog forbid, I turn my blunt scalpel on this discussion. Maybe I can re-write history like what is done on other forums and then you guys can point to it while decrying that the internet is indelible.
He called you out twice. I guess you're doubly-important.
For the record, I meant in this post:
http://www.hydrogenaud.io/forums/index.php...st&p=880881 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=880881)
graph from a Hydrogenaudio wiki
I'm glad that graph stirred things up and I was able to convince someone to remove it.
No matter how well-intentioned, that ill-advised graph should have never been created.
So if you care about this stuff at all it's good to be aware of what story is being propagated/propagandized on other audio forums.
I'm also aware that people are left to resorting to accusing people of cheating.
Then we have people who are intentionally cheating but get upset when you call a spade a spade.
Then we have people who refuse to consider that the test files in question are fallible while insisting that other test files must be.
But hey, the fb2k ABX plugin has and has always had a bug, so all results ever obtained by using it are officially null and void.
Down is Up, I guess!
I'm glad that graph stirred things up and I was able to convince someone to remove it.
No matter how well-intentioned, that ill-advised graph should have never been created.
Agree 100% . But in hindsight it serves as a good example of how ill-advised 'evidence' can be misused.
But hey, the fb2k ABX plugin has and has always had a bug, so all results ever obtained by using it are officially null and void.
Down is Up, I guess!
The amusing thing is, that would apply only to *positive* results. (At least, I can't imagine how the bug would bias toward false negatives)
(But neither Meyer & Moran nor Meridian used foo_abx)
In real life my doctor's medical training and degree trumps my lay opinion without being called argument from authority.
That authority has value
Fantastic Amir, we agree!
If we do on anything, I will have to go and hang myself.
Let's start: "how do you even imagine that one can hear a difference between two systems, one with noise 98 dB down and the other 146dB down, when the level is set to peak at 96dB?"
How do you explain how "obviously difference" fails to show up in even the worst kind of ABX test?
I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener. Have you any, bearing in mind that citing non-blind-testing proves nothing but the incompetence, the complete and total incompetence, of the person citing it as evidence.
Bear in mind the hard evidence for the persistance of loudness memory while you're at it.
I have my doubts that SACD or DVDA are much, if any, of an improvement, but the test is just blisteringly hard to run, and more likely to respond to artifacts, either positively or negatively, than it is to actual differences. Time alignment, level alignment, frequency response in-band can all throw it positive, lack of training, bad test environment, bad time alignment, etc, can also cause false negatives. Subject verification, likewise, is an important issue.
So, I remain undecided, but I note that I own a lot of CD's and not a single SACD or DVDA, except for some people have sent me.
Amir, hopefully this meets your luminary-authority requirements sufficiently as not to question it?
The "whit of evidence" and non-sighted test manifested itself in the form of Stuart's paper and us passing the forum ABX tests. JJ is on private industry thread where we are discussing these latest developments for some time now. He has not objected to anything we have been saying. Maybe he is dumbing things down for folks like you in forums. Go live what, buster!
Now where is the youtube video for how to make a noose....
@Arny, what reference does "30" lead to? Notice he wrote, "so 16 bit RPDF dither WOULD be inaudible" so apparently an actual test for this, in situ, was NOT performed. Had it been he would have presumably said so, and written "so 16 bit RPDF was kept to an inaudible level". Instead, by the use of the word "would", his assertion that it was kept to an inaudible level seems based on theory and conjecture. [Pending whatever "[30]" refers to].
J. R. Stuart. Noise: methods for estimating
detectability and threshold. AES 93rd Conven-
tion, Berlin, 1993.
That is a superb paper by the way.
Motherhood and apple pie.
To those of us who are already well-informed in the area it is a tutorial for newbies.
If you are a newbie, then it is probably pretty impressive.
If we do on anything, I will have to go and hang myself.
Well, do find that Youtube noose video, since you somehow managed to get bi-amping and speaker/room treatments, etc. right.
You seem to be afflicted mainly with "digital" disorder, this "Hi-Rez" business, "hearing" SPDIF, etc, etc.
I'll assume the "Power Regenerator" stuff was a prank.
I have my doubts that SACD or DVDA are much, if any, of an improvement, but the test is just blisteringly hard to run, and more likely to respond to artifacts, either positively or negatively, than it is to actual differences.
The "whit of evidence" and non-sighted test manifested itself in the form of Stuart's paper
Yes, so it seems and now here we are, examining that "whit" of dither doctored, direct radiator beryllium domes driven >105db evidence. Not quite the home run you were insinuating, but a lot more "lets see what happened here when those with pecuniary interests run a test". Artifacts?
Btw, 3rd time, do you have the (overlayed?) measurements/listening test data demonstrating the transparency of the BS paper switch, speakers, etc?
...and us passing the forum ABX tests.
Yes, like this (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=106442&view=findpost&p=877811), demonstrating there worthlessness to all but shysters who insist otherwise.
JJ is on private industry thread where we are discussing these latest developments for some time now.
Yet another of your insinuations, vs actual verifiable, public statements, like the ones I quoted above.
cheers,
AJ
Nice argument from authority. The funny thing is it wasn't necessary, but you couldn't help yourself and put it in anyway.
So, who is Steven?
Steven is Krabapple.
On the first comment, that is a forum debating technique which has no parallel in real life.
Say what? Argument from authority has no parallel in real life? In what alternative unverse? It is very common in advertising. A rather amusing example of argument from authority is as follows:
(http://171.67.24.121/tobacco_web/images/tobacco_ads/doctors_smoking/more_doctors_smoke_camels/medium/camels_doctors_whiteshirt.jpg)
In real life my doctor's medical training and degree trumps my lay opinion without being called argument from authority.
Formal training in medicine?
Congratulations Amir for keeping that a secret.
Unfortunately, the expertise of medical doctors even on medical topics is not looking so good right now...
In a patent dispute I can bring expert witnesses and they are not being dismissed as arguing from authority.
Expert testimony in court has no special legal weight.
http://legal-dictionary.thefreedictionary....xpert+Testimony (http://legal-dictionary.thefreedictionary.com/Expert+Testimony)
"Generally speaking, the law of evidence in both civil and criminal cases confines the testimony of witnesses to statements of concrete facts within their own observation, knowledge, and recollection."
Expert testimony outside of facts relevant to the case and even of those facts has the same legal weight as any other evidence. What does the jury do when experts disagree?
Recommended instructions to the jury from the judge are as follows:
https://www.justia.com/criminal/docs/calcrim/300/332.html (https://www.justia.com/criminal/docs/calcrim/300/332.html)
"If the expert witnesses disagreed with one another, you should weigh each opinion against the others. You should examine the reasons given for each opinion and the facts or other matters on which each witness relied. You may also compare the experts' qualifications."
In forums we like to pretend that such real life experience has no value and hence our lay opinion matters as much.
We build up a portfolio of these phrases we throw at people and stand back as if it have real meaning. They don't.
Please speak for yourself.
Stuart is an authority in signal processing and psychoacoustics.
It appears that Stuart has played an amateur's game for decades when it comes to psychoacoustics and subjective audio testing. AFAIK the article we are discussing is the first time he's put his name on anything but the results of sighted evaluations. If he is such an expert abut psychoacoustics why have we waited all these decades for even the pretense of him following some kind of recognizable attempt at bias controlled listening test. Blind tests were the gold standard in pyschoacoustics 40 years ago.
If these guys are such experts why this set of major gaffes?
"In an ABX test, a listener is required to listen to two reference sounds, sound A and sound B, and then to listen to sound X, and to decide whether sound X is the same as sound A or sound B."
"An ABX test requires that a listener retains all three sounds in working memory, and that they perform a minimum of two pair-wise comparisons (A with X and B with X), after which the correct response must be
given; this results in the cognitive load for an ABX test being high."
In fact the above account of the listener's behavior in an ABX is completely false.
(1) Its up to the listener to chose the order of listening to reference sounds and sound X. IME for the past 30+ years most listeners choose orders such as AXBXAXB... They are of course not restricted to the order ABX.
(2) The listener can base his decision on the comparison between A and X or the comparison B and X or both or neither - it is all up to him.
(3) The listener need not retain all three sounds in working memory. He need on retain just two sounds like any other comparison test. The listener usually does the pair-wise comparisons one at a time. When he compares the unknown to the second reference he only has to maintain just two sounds in his memory.
(4) A correct response need not be given until the listener is ready to give it. All that is required is a response - the listener can base it on what he chooses to base it on, including making intuitive guesses.
I couldn't have paid these guys to screw up worse! ;-)
@Arny, what reference does "30" lead to? Notice he wrote, "so 16 bit RPDF dither WOULD be inaudible" so apparently an actual test for this, in situ, was NOT performed. Had it been he would have presumably said so, and written "so 16 bit RPDF was kept to an inaudible level". Instead, by the use of the word "would", his assertion that it was kept to an inaudible level seems based on theory and conjecture. [Pending whatever "[30]" refers to].
J. R. Stuart. Noise: methods for estimating
detectability and threshold. AES 93rd Conven-
tion, Berlin, 1993.
That is a superb paper by the way.
Motherhood and apple pie.
To those of us who are already well-informed in the area it is a tutorial for newbies.
If you are a newbie, then it is probably pretty impressive.
That is interesting. You had just said this just a few hours ago:
J. R. Stuart. Noise: methods for estimating
detectability and threshold. AES 93rd Conven-
tion, Berlin, 1993.
The problem is that AFAIK Stuart has never before based any of his claims on relable listening test. Any DBT that is criticallly contingent on potentially flawed principles is obviously highly suspect.
How did it go from "highly suspect and flawed principles" to a tutorial that you already knew Arny?
Here is an excerpt from the paper:
(http://amirviews.smugmug.com/photos/i-DBbNKjN/0/O/i-DBbNKjN.png)
Which newbie do we want to pick to read this as a tutorial? Where would we see you describing the same anywhere online?
Would you like to take back what you said Arny?
Say what? Argument from authority has no parallel in real life?
No. It is a phony tactic used on forums. I have been to countless technical conferences. Not once have I seen anyone stand up and say, "Sir, that is appeal to authority." Or "that is a strawman." Or "excluded middle."
We have gotten so used to deploying these tactics to avoid having a technical discussion that we have confused ourselves to the point of thinking this is reality. You should have left these phrases in high-school debating class where they belong. They have no use in a professional discussion.
It appears that Stuart has played an amateur's game for decades when it comes to psychoacoustics and subjective audio testing.
The only amateurs are folks like yourself writing forum posts Arny with no education or work experience in psychoacoustics. You are a hobbyist. This man is not: http://www.aes.org/events/137/presenters/?ID=2425 (http://www.aes.org/events/137/presenters/?ID=2425)
AES Los Angeles 2014 Presenter or Author
J. Robert Stuart
Primary Affiliation: Meridian Audio Ltd. - Huntingdon, UK
AES Member Type: Life Fellow
J. Robert (Bob) Stuart was born in Belfast, Northern Ireland in 1948. He received a B.Sc. in electronic engineering from the University of Birmingham and an M.Sc. in operations research from Imperial College, London. While at Birmingham he studied psychoacoustics under Professor Jack Allinson, which began a lifelong fascination with the subject.
In 1977 he co-founded Meridian and is now chairman and technical director of Meridian Audio.
His professional interests are the furthering of analogue and digital audio and developing understanding of the human auditory perception mechanisms relevant to live and recorded music. His specialities include the design of analogue and digital electronics, loudspeakers, signal processing and optical disc players.
Mr Stuart has contributed to DVD-Audio and BluRay standards and has served on the technical committees of the National Sound Archive and the ADA (Japan). He has a deep interest in music and spends a good deal of time listening to live and recorded material. He is a fellow of AES and a member of JAS, ASA, IEEE and the Hearing Group at Cambridge.
This is a luminary. This is an expert you learn from. Please don't elevate yourself above him when you don't have a single accolade, educational or work experience he has in this field Arny. Please apply some common sense before writing such things.
Edit: fixed the formatting.
Dog forbid, I turn my blunt scalpel on this discussion.
I was quite enjoying the humorous approach...
This thread is
You'll have to run that by krab. He has a bad habit of poorly titling and/or poorly defining his discussions as well as using other forums to fight his battles via proxy. This one is no exception.
Since we can't manage to shame him into stopping, the best I can do is contain it.
...and now you're here.
On the master thread on AVS on Monty's write-up, this is what I said on page #1 after your posts: http://www.avsforum.com/forum/91-audio-the...ml#post21750075 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1398397-24-192-music-downloads-why-they-make-no-sense.html#post21750075)
[color="#0000FF"]The goals for setting a standard here shouldn't be what is adequate but what has some safety margin as to give us high confidence of inaudibility. In that regard, we need to also allow for less than optimal implementations. To that end, Bob Stuart has published a much more authoritative version of this report at AES. Here is an online copy: http://www.meridian-audio.com/w_paper/Coding2.PDF (http://www.meridian-audio.com/w_paper/Coding2.PDF). These are his recommendations:
"This article has reviewed the issues surrounding the transmission of high-resolution digital audio. It is
suggested that a channel that attains audible transparency will be equivalent to a PCM channel that
uses:
· 58kHz sampling rate, and
· 14-bit representation with appropriate noise shaping, or
· 20-bit representation in a flat noise floor, i.e. a ‘rectangular’ channel"
So as we see, the CD standard somewhat misses the mark on sampling rate. And depending on whether you trust the guy reducing the sample depth from 24-bit to 16 bits, we may be missing the right spec there too.
If you have read the paper, you will know that after a lengthy discussion, the 58kHz figures come (not mostly from that discussion, but) from Nyquist sampling the highest frequency that anyone has ever been able to hear, adding 10%, and rounding up to the nearest whole number of kHz.
Following this reasoning, those of us who can't hear 26kHz don't need such a high sample rate.
However, "extraordinary claims require extraordinary evidence."
They don't really. This line came from a couple of devotees in the skeptic camp and has no other foundation.
Wikipedia told me it was popularised by Carl Sagan from an original quote by Marcello Truzzi.
I don't know why it keeps getting repeated in these discussions as if it is a universal truth.
I'm not claiming it's "true", but it's certainly relevant and should give pause for thought.
That aside, who is to say what is extraordinary? To my wife, it is extraordinary claim that blu-ray looks better than DVD. She thinks they look the same. Do I need to come up with extraordinary proof to show she is wrong? I do not.
But the previously known spatial resolution limit of the human eye + the size of your new HDTV + the distance you sit from it says the extra resolution that BluRay delivers in your living room is visible to normal eyes. Whereas the previously known frequency response of the best human ears says the higher frequencies in your hi-res audio files are inaudible.
They're opposite situations. The benefits claimed by one match known science. The benefits claimed by the other contradict known science.
A few of the results posted, if they were obtained fairly, are the start of some useful evidence. Many are just the result of badly set-up or performed tests. I haven't seen you discount these faulty tests, but then I haven't trawled through all the relevant threads.
I will do so now. I discount all of them. Every test has flaws. But we are told that if we pass a "DBT ABX" folks would believe. Well, we passed them and passed them all. Yet, it has managed to do absolutely nothing. Now they are all out to discredit validity of these tests. Which way is up now? We have schmucks now saying we have to have witnesses to prove we did not doctor up some log to win a point online. Really?
I am differentiating between the tests where, unless someone cheated, a positive result means something - and the tests where a positive result could mean nothing because the tests themselves included fundamental flaws which could deliver a positive result even if both files contained CD quality audio.
If you won't draw a distinction between these, then it's a bit rich to accuse others of looking "totally illogical and biased to the core".
Cheers,
David.
On adding more channels, the market has spoken. The music buying industry is young people who want their tunes with headphones and such so we are stuck with stereo. Multi-channel music's era came and gone. Sure, there is a path through Blu-ray but highly limited. If you want to boil the ocean by advocating that the whole market transform around your wishes, go ahead. I don't like hitting my head against the wall nearly as much as you do.
One minute you're advocating an audio standard which requires a 10 year old's hearing* and/or a $100k+ acoustically insulated room* to appreciate it, but when someone suggests more channels, we can't have that because we have to limit what we deliver to what can be appreciated by folks listening on their iPods.
I think we need a minute's silence to let the size of that contradiction settle in...
btw, if you have head tracking, then multiple source channels, or (better still) complete sound fields, become very useful as sources for headphone listening. It's only without head tracking that two channels is enough, and without head tracking headphones and binaural perception are pretty fragile.
Cheers,
David.
P.S. * = exaggerated for effect, but I think think you can still see the serious point here.
Say what? Argument from authority has no parallel in real life?
No. It is a phony tactic used on forums.
Like "credentials".
And Red Herrings for breakfast lunch and dinner.
Welcome to Amirworld, now available in full resolution on HA
Now Amir, in these double secret probation private industry discussions, are any non-pecuniary interest industry insiders saddened by the use of rectangular dither to doctor the BS tests? Is there some concern about the levels used with the direct radiator beryllium domes driven >105db speakers with and without band limiting? Is anyone demanding system transparency (Time alignment, level alignment, frequency response in-band, switching, speakers, etc) data?
Inquiring outsider minds would like to know, TIA.
cheers,
AJ
However, "extraordinary claims require extraordinary evidence."
They don't really. This line came from a couple of devotees in the skeptic camp and has no other foundation.
The genesis of this phrase is attributed to:
http://en.wikipedia.org/wiki/Marcello_Truzzi (http://en.wikipedia.org/wiki/Marcello_Truzzi)
Marcello Truzzi and Carl Sagan, not exactly chopped liver.
Marcello and Carl have their own articles in Wikipedia and many other encyclopedias and reference works.
Heck, Wikipeida even mentions my name and credits me for inventing something fairly important in audio... I believe that Stuart said something about "Gold standard" in reference to it.
Of course this is just an appeal to authority and who would be dumb enough to do that? ;-)
I don't know why it keeps getting repeated in these discussions as if it is a universal truth.
Something about making sense... ;-)
Oh and by the way thanks for completely ignoring the critical arguments that I presented. That's a win!
On adding more channels, the market has spoken. The music buying industry is young people who want their tunes with headphones and such so we are stuck with stereo. Multi-channel music's era came and gone. Sure, there is a path through Blu-ray but highly limited. If you want to boil the ocean by advocating that the whole market transform around your wishes, go ahead. I don't like hitting my head against the wall nearly as much as you do.
One minute you're advocating an audio standard which requires a 10 year old's hearing* and/or a $100k+ acoustically insulated room* to appreciate it, but when someone suggests more channels, we can't have that because we have to limit what we deliver to what can be appreciated by folks listening on their iPods.
I think we need a minute's silence to let the size of that contradiction settle in...
Neither one is my position. Once more:
I am advocating getting the bits prior to remastering for CD. Whatever was produced before conformance with the CD format and commercialization of music to teenage girls is what I want.
I have said nothing about wanting higher frequencies than 20 Khz.
And what I want is already happening. What AJ wants, multi-channel music, is not. I stated the market dynamics. I know the business and technology side of the audio market. You want to ignore the former, sure, go ahead but don't ask me to participate.
P.S. * = exaggerated for effect, but I think think you can still see the serious point here.
I did. Disappointing to see word arguments rather that substantive discussion.
...as if you played no role in the former.
JJ is on private industry thread where we are discussing these latest developments for some time now.
Yet another of your insinuations, vs actual verifiable, public statements, like the ones I quoted above.
As it happens I'm graciously allowed on a mailing list frequented by recognizable 'names' in audio ...JJ posts there, too... and I'm very sure it's not
that one, at least. Though to see how that crowd would respond to Madrona's fancy footwork, and to Meridian's paper, would be...interesting. I expect the latter at least will eventually get some scrutiny there.
But maybe 'what happens in private industry, stays in private industry'?
Oh and by the way thanks for completely ignoring the critical arguments that I were presented. That's a win!
Indeed! Hope you don't mind my speading the credit around, as it wasn't all owed to you.
'Multichannel music is not happening' but I just happened to find these two new multichannel music releases in my mailbox this week:
Yes Relayer in BluRay 5.1
http://www.amazon.com/gp/product/B00NAYYSS8/ (http://www.amazon.com/gp/product/B00NAYYSS8/)
Pink Floyd Endless River in BluRay 5.1
http://www.amazon.com/Endless-River-Deluxe.../dp/B00NQKWAIQ/ (http://www.amazon.com/Endless-River-Deluxe-Blu-ray-Casebook/dp/B00NQKWAIQ/)
I did. Disappointing to see word arguments rather that substantive discussion.
So Amir, when you resort to word arguments as you frequently do (many times on this thread alone) do you feel a sense of disappointment?
I did initially after conversing with you for the first few times, but I learned how you work, and I'm over all that!
Oh and by the way thanks for completely ignoring the critical arguments that I were presented. That's a win!
Indeed! Hope you don't mind my speading the credit around, as it wasn't all owed to you.
Not at all.
Some credit goes to HA moderation staff that isn't bamboozled by all of the word games and dancing around questions that he can't answer. Happens all the time at you-know-where #1 & #2.
My elementary school teacher said: "every class needs a clown. you just want to make sure you are not the one."
That advice seems to gone the same way as the basics of electronics usually related to a degree in electrical engineering that seemed to have gone missing by the first time I encountered you on AVS.
I have never corrected anybody in my life for as many very basic mistakes, even total laymen.
All I can hope to do is keep this shit on-topic and above board while allowing you guys your fun.
The last thing I want is for placebophiles in anti-DBT forums to see this thing get closed and declare a win, even though it would have been the result of it going sideways (like the other one).
It's krab's thread; he defined the topic. Those who wish to discuss something else should do him the courtesy start a new discussion.
The paper seems to make two big omissions, both of relating to the fact that their choice of program material and listening room make their experiment almost 100% non-comparable with say Meyer and Moran or most work that proceeded it.
That is the true. It is the difference between hobby work (Meyer and Moran) and that of professionals (Stuart).
Interesting that up until this paper the so called professional had AFAIK only authored AES papers that failed to substantiate
their exceptional claims with professional listening tests. Please show the BS1116 compliant tests by Stuart that had been substantiating Stuart's many exceptional claims prior to this one.
"Exceptional claims" don't make it as papers to the Journal of AES. But let's say it does have exceptional claims. That it was accepted means that those "exceptional claims came with exceptional proof." Right? Right.
Interesting that ABX for audio was raised up by hobbyists while the rest of the AES (the so-called hobbyists were AES members) were publishing papers in the Journal that were based on sighted evaluations.
Much of psychoacoustics/acoustics research is done with sighted analysis Arny. Almost all industry research is done using sighted testing. I can show you modern papers published in ASA that are done sighted. You live in this fictitious world that is called an audio forum where you have convinced yourself the onlything valid is something with the words "ABX DBT" stamped on it. That is not reality. Go get a job in the industry instead of being a hobbyist and you quickly realize that is not how it works.
Double blind test are used sparingly because they take so long to administer and are so resource intensive. And when they are done, they are not ABX.
Please don't confuse hobbyists performing sighted listening tests and forum arguments for real life. In real life, you want the real answer and you want it efficiently. You don't sit there trying to please Arny's of the world.
You mention ITU BS1116 which of course our hobbyists friends, Meyer and Moran did not use or reference. That aside, there is absolutely no reference to ABX in BS1116. Here is what it does stay about the testing methodology:
The “double-blind triple-stimulus with hidden reference” method has been found to be especially
sensitive, stable and to permit accurate detection of small impairments.The tests use a grading system and not binary forced choice like ABX. So not sure why you reference BS1116 when it is not supportive of your view.
In fact the program material that M&M used was in an audio format that met or exceeded every criteria that Stuart had called for.
It didn't remotely seeing how they did not even test to see if their content was high resolution. ITU BS1116 says this about the proper test content:
* When selecting the programme material, it is important that the attributes which are to be assessed are precisely defined.
The responsibility of selecting material shall be delegated to a group of skilled subjects with a basic knowledge of the
impairments to be expected.
Meyer and Moran score: FAIL. They did not delegate the selection to skilled subjects with the knowledge of impairments. They self appointed themselves as the experts with no understanding of audio theory.
Stuart: PASS: As an expert in the field, and evidenced by their test results, absolutely was qualified to pick the test material.
* Prior to formal grading, subjects must be allowed to become thoroughly familiar with the test facilities, the test
environment, the grading process, the grading scales and the methods of their use. Subjects should also become
thoroughly familiar with the artefacts under study. For the most sensitive tests they should be exposed to all the material
they will be grading later in the formal grading sessions. During familiarization or training, subjects should be preferably
together in groups (say, consisting of three subjects), so that they can interact freely and discuss the artefacts they detect
with each other.
Meyer and Moran Score: FAIL. Nothing remotely like this was done. No control/degraded reference so that they could familiarize themselves with what they were supposed to hear.
Stuart: PASS. They followed the above to the letter. A degraded positive control was used to train and familiarize testers.
So if you are in favor of compliance with BS1116, then you need to back the outcome of Stuart's test. Since you are not, I imagine having BS1116 listed on your now defunct ABX web site was a farce.
Say what? Argument from authority has no parallel in real life?
No. It is a phony tactic used on forums. I have been to countless technical conferences. Not once have I seen anyone stand up and say, "Sir, that is appeal to authority." Or "that is a strawman." Or "excluded middle."
I attribute that to several strong influences:
(1) Conference speakers who know enough about basic rhetoric to not commit those gaffes.
(2) Audiences who are too polite and wise to waste everybody's time and patience pointing those gaffes out when poor speakers and politicians make them.
(3) Audiences who are already bored by the crappy speaker and know better than to do anything to extend the boredom.
However if you step out of the public show presentations and enter the (now mostly figurative) smoke filled rooms, well stuff happens.
I guess you never got invited that far inside... I have.
The paper seems to make two big omissions, both of relating to the fact that their choice of program material and listening room make their experiment almost 100% non-comparable with say Meyer and Moran or most work that proceeded it.
That is the true. It is the difference between hobby work (Meyer and Moran) and that of professionals (Stuart).
The subjective testing by Stuart up until lately wasn't even up to the level of Meyer and Moran until just lately as I have documented.
Interesting that up until this paper the so called professional had AFAIK only authored AES papers that failed to substantiate
their exceptional claims with professional listening tests. Please show the BS1116 compliant tests by Stuart that had been substantiating Stuart's many exceptional claims prior to this one.
"Exceptional claims" don't make it as papers to the Journal of AES.
Very many exceptional claims made it into the Journal back in the days right before ABX. One topic that was rife with them related to Slew Induced Distortion. The very rapid acceptance of ABX by the AES was expedited due to displeasure among top AES officers over how much floobydust was making its way into the Journal.
Interesting that ABX for audio was raised up by hobbyists while the rest of the AES (the so-called hobbyists were AES members) were publishing papers in the Journal that were based on sighted evaluations.
Much of psychoacoustics/acoustics research is done with sighted analysis Arny.
Of course, and it is completely appropriate in many of those cases.
Almost all industry research is done using sighted testing.
Of course, and again it is completely appropriate in many of those cases.
I can show you modern papers published in ASA that are done sighted.
Of course, and again it is completely appropriate in many of those cases.
However the ASA was publishing a great many papers based on the JASA version of ABX long before the AES even mentioned the AES version of ABX. People who read journals and don't just drop their names know such things. They also know the nature of the differences between the two. One of the items in the committee review of Clark's AES ABX paper included clarifying these things because enough people in the AES knew about the ASA version of ABX whcih came much earlier. So did we. BTW the letters ABX in the AES were originally a play on my initials.
(This particular forum has a fairly low limit on quoting and rather than running afoul of it I will just skip over the many errors and misrepresentations that follow in the post I am responding to. I know that you will ignore most of my answers because you cannot muster any convincing response to them.)
... And listening in rooms that are often horrendous, acoustically. ...
The phrase "The elephant in the room" is being used a lot in these discussions.
It might be more appropriate to say "The elephant
is the room."
... And listening in rooms that are often horrendous, acoustically. ...
The phrase "The elephant in the room" is being used a lot in these discussions.
It might be more appropriate to say "The elephant is the room."
One ironic fact is that while the recent Meridian test's organizers no doubt have a vast array of acoustical test gear at their disposal, their report appears to contains no measurements of the listening room that they used. Not even its dimensions - I guess their tape measures went missing!
This is the description of the listening room: "All signals were presented via loudspeakers. Subjects sat in a silent and soundproof listening room with near-optimal acoustic properties." Compliance with well-known recommendations for the listening room (BS 1116) don't seem to be even hinted at.
There's even a likely spot to report some of this information (Figure 3), but I see nothing about the room.
The text's description of Figure 3 is: "Fig. 3 shows the frequency analysis of the test signal on the same axes as behavioural thresholds MAF (minimum audible eld) and UEN (uniformly-exciting noise) threshold [29, 30]."
The Stuart AES Convention paper is now available (http://www.aes.org/e-lib/browse.cfm?elib=17497)for purchase/download at the AES library
I sure hope it's not an engineering report, anchors were used extensively and Bob has his credentials in order.
It's a convention paper -- so, not even peer reviewed. (It's also a completely in-house affair -- all three authors work for Meridian).
Thus, I expect Our Man from Madrona will reject it forcefully, using at least three if not four different font colors.
You were right about the last sentence. This answer will be in technicolor. Why? Because you are misstating such simple facts and it is important to get the right data out.
This is what is on the title page of Stuart and crew's listening test paper:
----------
Winner of the AES 137th Convention Best Peer-Reviewed Paper AwardAudio Engineering Society Convention Paper
Presented at the 137th Convention
2014 October 9-12 Los Angeles, USA
The audibility of typical digital audio lters in a high-delity playback systemHelen M. Jackson, Michael D. Capp, and J. Robert Stuart
Meridian Audio Ltd, Huntingdon, Cambridgeshire, PE29 6YE, England
----------
Winner of best *peer-reviewed* paper award. You say you have the paper but didn't see that? And positioned it as not being peer-reviewed?
Of course any read of the paper would convinced you of its high quality and in no need of any award.
... And listening in rooms that are often horrendous, acoustically. ...
The phrase "The elephant in the room" is being used a lot in these discussions.
It might be more appropriate to say "The elephant is the room."
I like it!
Source of the BS paper tracks.
https://shop.klicktrack.com/2l/35847 (https://shop.klicktrack.com/2l/35847)
16/44 (FLAC) tracks: $16 Score: PASS Reasonable, on par with B&M store
HiRez Scam tracks: up to $42 Score: FAIL Highly unreasonable for no whit of evidence of audibility in purchaser system
Website contains no purchaser requirements for:
Listener training: Score FAIL
Iso-ward listening room: Score FAIL
Speakers with DR metal domes driven to near maximum levels that might create audible artifacts without band limiting FAIL
Use of atypical rectangular dither filters in Hi-Fi playback system FAIL
Luckily, there is a specific market that craves scams like no other.
cheers,
AJ
An independent, non-pecuniary interest outsider grading of the BS paper:
Adherence to ITU BS1116: BS Paper score: FAIL: "some" adherence, not full.
BS expert in the fields covered by paper score: FAIL: Red Herring, only thing that matters are methods and results, not "expertise".
Test results, BS Paper score: FAIL: Deliberate use of Rectangular dither instead of their own recommended best practices Triangular, to fabricate positive results.
I add several other concerns/grades about the BS paper:
No system transparency (Time alignment, level alignment, frequency response in-band, switching, speakers, etc) data, score: FAIL
Levels used with the direct radiator beryllium domes driven >105db speakers with and without band limiting possibly the cause of artifacts, score: FAIL
I certainly hope those with pecuniary interests in Hi Rez don't hang their hats on this Farce, but I guess we will see in the long run.
cheers,
AJ
To borrow a communication style I've seen used recently:
So let me get this straight...
If I were to spend exorbitant amounts of money on speakers, I might possibly hear a difference between 24/96 content and an improperly dithered 16/44 version.
Can the differences always be attributed to the actual content and not the equipment and/or SRC algorithm?
If yes:
What assurances do I have that spending exorbitant amounts on content will provide a difference in my listening experience?
What do these differences sound like, and under what circumstances in the content will I hear them?
If no:
How am I to know for certain when the differences are only attributable to the content?
To borrow a communication style I've seen used recently:
So let me get this straight...
If I were to spend exorbitant amounts of money on speakers, I might possibly hear a difference between 24/96 content and an improperly dithered 16/44 version.
There was an overall 0.56 fraction of correct identifications.
"We analysed the percent-correct scores for
each of sections 1-17 across all conditions, and found
that some sections yielded a higher ratio of correct
results than others. For example, sections 2, 6 and
17 gave correct-to-total ratios of 0.714, 0.710 and
0.769 respectively, whereas sections 1, 7 and 10 gave
correct-to-total ratios of 0.541, 0.455 and 0.509 re-
spectively..."
Winner of best *peer-reviewed* paper award. You say you have the paper but didn't see that? And positioned it as not being peer-reviewed?
The copy of the paper I just obtained from the AES web site starts out:
"This Convention paper was selected based on a submitted abstract and 750-word precis that have been peer reviewed
by at least two quali ed anonymous reviewers. The complete manuscript was not peer reviewed. This convention
paper has been reproduced from the author's advance manuscript without editing, corrections, or consideration by the
Review Board. The AES takes no responsibility for the contents."
To borrow a communication style I've seen used recently:
So let me get this straight...
If I were to spend exorbitant amounts of money on speakers, I might possibly hear a difference between 24/96 content and an improperly dithered 16/44 version.
This would be a hypothesis to be tested in another listening test. This one found differences in the system they tested. The outcome says nothing one way or the other about lesser systems.
FYI, I passed all of my double blind ABX tests using my laptop and headphones. Does that mean everyone with every laptop and headphone can hear the same? Of course not.
These points aside, if a very expensive speaker managed to do something everyday speakers cannot do with respect to bring out small differences in how the content was encoded, that by itself is quite significant. That high-end hardware brings with it better resolution of audio and with it, potentially lend claim that if someone hears a difference on a high-end system and you cannot, it could very well be due to the system cost/performance.
Can the differences always be attributed to the actual content and not the equipment and/or SRC algorithm?
The paper analyzes the content and demonstrates that this is indeed very good content as far as having a noise floor that is well below the threshold of hearing in the playback system/test. Does that prove that it played a critical role in this test? We don't know. We can either validate or invalidate the null hypothesis. We cannot validate alternative hypothesis without running another set of tests.
The SCR algorithm is very specifically documented in the paper so I leave it to everyone to read it. And I answered the bit about equipment above.
If yes, what assurances do I have that spending exorbitant amounts on content will provide a difference in my listening experience?
On Amazon I routinely see MP3 albums being cheaper than CD. I buy the CD. So the "exorbitant" difference in cost is not a scientific question but a personal choice. Either you care about the differential or you don't. Either you think you will always hear the same or you won't.
No one is twisting your or my arm to buy higher resolution audio for higher price. I don't always do that. It depends on what the music is. Giving a video example, for some movies I have no issue watching the DVD my wife buys. But if she comes home with the next Star Trek movie in DVD instead of Blu-ray, I will be grumpy. Very grumpy.
So the cost difference is orthogonal to the discussion. That is a economic consideration between as the buyer and the content provider/distributor. We don't judge if a four bedroom house brings more happiness to you than three, so don't ask us about the same difference in music fidelity .
And no, there is no guarantee. That is like asking for a guarantee that you will like some music better. You have to be a smart shopper. Have good ears that can detect such distortions and if your theory is right, proper hardware to play it.
If no, how am I to know for certain when the differences are only attributable to the content?
I don't know. If I just hand you a CD what can you tell about its quality as it sits there? Can you swear it will sound better than the MP3 to you in all cases?
What do these differences sound like, and under what circumstances will I hear them?
No one can tell you that. How you will hear the difference is something you must experience. You know how to do that with compressed music, right? If you had not, would the concept of pre-echo have any meaning to you if I wrote it? It wouldn't right?
Same here. Instead of hoping to win a written argument, spend some time seeing if you too hear the differences some of us have reliably reported through ABX testing. If you do, you will then speak first hand about what we are talking about. And if you do not, then you can sign off from this thread knowing that increased fidelity beyond CD is of no value to you.
===========
I answered your questions as you put to me. But none of it may matter! Not at all. Because the high resolution stereo master may very well NOT be subject to loudness compression. If so, then the difference will be "huge." Night and day and that sort of thing. Before you ask me again if that is guaranteed, no. It is not. Seek out reviews on WBF Forum, Computer Audiophile, etc. web site where people discuss such things including posting spectrums and such, and you will walk into your purchase with open mind.
The other benefit is that you can download the high resolution master, but you cannot with the CD. I for example downloaded the L2 sampler that was used in the Stuart tests. Lovely, lovely music. That is why I bought it. I sampled it, it was beautiful music. 1.3 gigabyte download that happened in a few minutes in the background. Done. No trip to store. No ripping when my laptop doesn't even have an optical disc reader.
If these things are not worth it to you, then ignore it. But my hope is that you don't keep throwing word arguments at me and not looking at the larger argument.
Winner of best *peer-reviewed* paper award. You say you have the paper but didn't see that? And positioned it as not being peer-reviewed?
The copy of the paper I just obtained from the AES web site starts out:
"This Convention paper was selected based on a submitted abstract and 750-word precis that have been peer reviewed
by at least two quali ed anonymous reviewers. The complete manuscript was not peer reviewed. This convention
paper has been reproduced from the author's advance manuscript without editing, corrections, or consideration by the
Review Board. The AES takes no responsibility for the contents."
Hi Arny. That is just a boilerplate AES puts on every conference paper these days. The Stuart paper is different and has this additional line I post:
Winner of the AES 137th Convention Best Peer-Reviewed Paper Award
That is the result of this (rather) new category of paper submission to the conference: http://www.aes.org/events/137/authors/137thCallForPapers.pdf (http://www.aes.org/events/137/authors/137thCallForPapers.pdf)
Authors may submit proposals in three categories:
1. Complete-manuscript peer-reviewed convention papers (submit at www.aes.org/137th_authors)
2. Abstract-precis-reviewed convention papers (submit at www.aes.org/137th_authors)
3. Synopsis-reviewed engineering briefs (submit at www.aes.org/137th_ebriefs)
For the
complete-manuscript peer-reviewed convention papers (category 1), authors are asked to submit papers of 4–10 pages to the
submission site. Papers exceeding 10 pages run the risk of rejection without review. These complete-manuscript papers will be reviewed
by at least two experts in the field, and authors will be notified of acceptance by 2014 June 18. Final manuscript with revisions
requested by the reviewers have to be submitted before 2014 July 9. If rejected as a convention paper (Cat. 1), the proposal may still be
accepted for categories 2 or 3.
Award selection is then made from these complete, peer-reviewed papers. Winning the award opens the door to the paper being published in the journal.
I think we have slammed the door shut on any statement regarding standing of this paper in the eyes of professional AES community. It is excellent work. It is the most careful double blind test of small differences due to coding of audio that has been published. It is an inconvenient truth to be sure for many, but hopefully the search for knowledge and learning trumps that negative emotion.
This one found differences in the system they tested. The outcome says nothing one way or the other about lesser systems.
Less expensive is not "lesser" system. It's less expensive. If it does not generate the artifacts of the DR beryllium domes speaker driven >105db without band limiting, it's far from "lesser" of a system, unless you mean "lesser" artifacts. Some of us who actually value fidelity prefer that. Less artifacts.
FYI, I passed all of my double blind ABX tests using my laptop and headphones.
Unsupervised, pointed out multiples times as useless and unrelated to the BS tests under examination, which were supervised and submitted to AES for review.
These points aside, if a very expensive speaker managed to do something everyday speakers cannot do with respect to bring out small differences in how the content was encoded, that by itself is quite significant. That high-end hardware brings with it better resolution of audio and with it, potentially lend claim that if someone hears a difference on a high-end system and you cannot, it could very well be due to the system cost/performance.
A "high end" system that generates artifacts is not "better resolution". Be that speakers, amps or cables. Artifacts and distortions are not "better resolution".
So the cost difference is orthogonal to the discussion.
You just stated it wasn't for system costs. Now it is for content?
The other benefit is that you can download the high resolution master, but you cannot with the CD.
Benefit because the label offers downloads in every format but 16/44 WAV? They offer it in 16/44 FLAC, which is the equivalent of CD.
There is not a whit of evidence that any audiophile, anywhere, can hear any "benefit" with bigger files, in their system, in their room. Unless you mean "artifacts" like oil can resonances without band limiting. I certainly concede that possibility of "more resolution".
cheers,
AJ
Benefit because the label offers downloads in every format but 16/44 WAV? They offer it in 16/44 FLAC, which is the equivalent of CD.
To the extent a high resolution file does exist, yes they do offer the downgraded versions. Those versions however are not the CD master. At least I don't think they are .
There is not a whit of evidence that any audiophile, anywhere, can hear any "benefit" with bigger files, in their system, in their room.
There is no evidence that they are not either. So?
Unless you mean "artifacts" like oil can resonances without band limiting. I certainly concede that possibility of "more resolution".
You are talking about this in the stereophile review?
Although the tweeter's "oil-can" dome resonance results in a narrow spike 12dB above the reference level at 1kHz, this can be seen to occur at a high 26kHz. So your contention is that people heard the 26 Khz resonance in the formal listening test? What data do you have on audibility of 26 Khz AJ?
What type of dither did Meyer and Moran use? Whatever was in the A/D and D/A converters of the CD-ROM recorder, right? Here is Stuart on his latest test:
This [playback] level was
chosen for comfort, and because it was high enough
for details to be audible but also low enough that 16-
bit RPDF dither would be inaudible at the listening
position [30].
And
We chose to use undithered quantization as
a probe and | although we would normally rec-
ommend TPDF dither for best practice -- we con-
sidered rectangular dither to be more representative
of the non-ideal dither or error-feedback processing
found in some commercial A/D and D/A filters.
Seems reasonable to me. After all, who knows what dither if any was used in the music that people are shipping to us in 16/44.1? It is not like they know what you and I know about dither types.
So they calibrate the test for a particular system parameter but change that parameter during the actual test? And they give no citation to justify the change, but instead say "we consider"? This section should not have passed peer review (if it really did).
There is not a whit of evidence that any audiophile, anywhere, can hear any "benefit" with bigger files, in their system, in their room.
There is no evidence that they are not either. So?
Seriously? Doh!
So they calibrate the test for a particular system parameter but change that parameter during the actual test? And they give no citation to justify the change, but instead say "we consider"? This section should not have passed peer review (if it really did).
I don't follow you. There is no system calibration. There is no parameter change during the test. The type of dither or not were independent tests. The peers read the whole paper and not just that snippet.
To the extent a high resolution file does exist, yes they do offer the downgraded versions. Those versions however are not the CD master.
Downsampled, not "downgraded", which implies otherwise. This is
end user product, so all that (should) matter, is the
sound, not the "performance" or other weasel words that have nothing to do with end user
sound that they
hear.
I'll remind you amir, there is not a whit of evidence to suggest 16/44
playback is non-transparent in home audio end use.
Dubious dither doctored tests and unsupervised online games are not evidence against this.
There is no evidence that they are not either. So?
Argument from ignorance. The onus falls on your "side" to present evidence of proof.
So your contention is that people heard the 26 Khz resonance in the formal listening test? What data do you have on audibility of 26 Khz AJ?
Nope. My contention is that purported "experts" better look at
all factors that may generate
artifacts in the (human)
audible range (<20k) when driving DR metal dome speakers near their limits without the benefit of 22K band limiting.
It would be a very amateur hobbyist type mistake, to overlook factors that create false positive results in any test. It's bad enough that one was deliberately introduced (RPDF) to induce positives. Overlooked ones might cause unnecessary embarrassment.
cheers,
AJ
So they calibrate the test for a particular system parameter but change that parameter during the actual test? And they give no citation to justify the change, but instead say "we consider"? This section should not have passed peer review (if it really did).
I don't follow you. There is no system calibration. There is no parameter change during the test. The type of dither or not were independent tests. The peers read the whole paper and not just that snippet.
They said that the playback level was selected so that TPDF dither would not be audible (calibration) but then used dither known to be more audible (a change). There is no way to determine if the audibility difference is due to the known more audible dither, the higher resolution sampling, or a combination of both.
Doing something because it sounds reasonable to do so is not sufficient in a scientific test. Any test choice must be rigorously justified to insure that the stated hypothesis is actually being tested. The dither choice was not adequately justified. (I notice that they said "some" commercial filters in the text but in the abstract they say the filters were "representative" of filters used. Does "some" mean "representative"?)
We still have no idea whether any differences, supposing that they were detected reliably, were the result of the content or the hardware.
If it was the content (in the case of the Meridian test), we do not know what would have happened if the files had not been intentionally compromised with rectangular dither.
The only response that directly addressed the gist of my post was, "we don't know."
You are talking about this in the stereophile review?
No, as a speaker designer who deals with band limiting/suppressing breakup resonances with many rigid drivers.
But now that you mention it, I see there was a SP test of the 8000, the 7200SE obviously being a downgraded, lesser system version.
(http://www.stereophile.com/images/archivesart/merdsp80001.jpg)
Although the tweeter's "oil-can" dome resonance results in a narrow spike 12dB above the reference level at 1kHz, this can be seen to occur at a high 26kHz. With CD material, this resonance will have no audible consequences, as it will not be excited by the band-limited source, even when upsampled by the Meridian 800. The resonance will be excited by wide-bandwidth SACD and DVD-Audio material, but I heard nothing untoward when using these sources.
-JA
The question is, if JA repeated the test in an iso-ward, with cherry picked and dithered material, then driven the speakers to near their maximum limits of 112db......would he still have heard nothing "untoward"? In a controlled, blind fashion of course.
Lot's of unanswered questions about this BS test. Including the transparency of the switching software, given the recent online fiascoes.
cheers,
AJ
...
"We analysed the percent-correct scores for each of sections 1-17 across all conditions, and found that some sections yielded a higher ratio of correct results than others. For example, sections 2, 6 and 17 gave correct-to-total ratios of 0.714, 0.710 and 0.769 respectively, whereas sections 1, 7 and 10 gave correct-to-total ratios of 0.541, 0.455 and 0.509 respectively..."
Do I detect a faint aroma of fresh-picked cherries?
Or is it that...
mmmm, fresh cherries.
...
"We analysed the percent-correct scores for each of sections 1-17 across all conditions, and found that some sections yielded a higher ratio of correct results than others. For example, sections 2, 6 and 17 gave correct-to-total ratios of 0.714, 0.710 and 0.769 respectively, whereas sections 1, 7 and 10 gave correct-to-total ratios of 0.541, 0.455 and 0.509 respectively..."
Do I detect a faint aroma of fresh-picked cherries?
Or is it that...
mmmm, fresh cherries.
Do they give the p-values for the data in each section?
I'll remind you amir, there is not a whit of evidence to suggest 16/44 playback is non-transparent in home audio end use.
AJ, you are typing in a thread where we are discussing an award winning, peer-review paper showing such differences can be audible. The time for making that statement has come and gone. You should look for other talking points, battles to fight. This one is done.
Dubious dither doctored tests and unsupervised online games are not evidence against this.
I hear you. You are wrong. But I hear you. Buy the paper, read it and try again.
So your contention is that people heard the 26 Khz resonance in the formal listening test? What data do you have on audibility of 26 Khz AJ?
Nope. My contention is that purported "experts" better look at all factors that may generate artifacts in the (human) audible range (<20k) when driving DR metal dome speakers near their limits without the benefit of 22K band limiting.
Demonstrate your point with measurements and data please.
It would be a very amateur hobbyist type mistake, to overlook factors that create false positive results in any test. It's bad enough that one was deliberately introduced (RPDF) to induce positives. Overlooked ones might cause unnecessary embarrassment.
Well, the odds are against a hobbyist like you knowing about these points than the luminaries in the industry and the peer review experts who awarded them the award. But not impossible. So make your case in detail without just claiming it AJ.
We still have no idea whether any differences, supposing that they were detected reliably, were the result of the content or the hardware.
The thesis of the paper is neither. Those were preconditions for the test. The test hypothesizes that filtering from 192 khz down and conversion with or without dither may be audible. And their listening test showed that to better than 95% confidence, they are.
If it was the content (in the case of the Meridian test), we do not know what would have happened if the files had not been intentionally compromised with rectangular dither.
Please demonstrate who the files were compromised with rectangular dither.
The only response that directly addressed the gist of my post was, "we don't know."
You asked a number of questions whose answer was that. You were hypothesizing this and that and the answer is we don't know to all of that. The core thesis of the paper we do know and I stated it again above.
Yes, and obvious flaws exist which were touched upon, which you have not been able to dismiss successfully.
If dither wasn't a problem, which hasn't been shown to my satisfaction*, then you're still left to contend with differences caused by artifacts in hardware/software; to which you also don't have a satisfactory answer.
So, sure, the thesis *may* have been supported by the results, but with severe caveats.
This isn't particularly earth-shattering when it has already been conceded that 16 bits is not audibly transparent under all circumstances. I'm pretty sure most here will also accept that not all SRCs will guarantee transparency, either.
However, this topic is more about what people will attempt to make of the results and the predictable posturing they will take.
Did I get that right, krab?
(*) which will likely require independent verification with more complete transparency regarding the test conditions, which you are unable to provide at this point in time, so don't bother with this dead horse.
mmmm, fresh cherries.
...and a nice recipe for cherry crumble.
Yes, and obvious flaws exist which were touched upon, which you have not been able to dismiss successfully.
Both statements are without foundation. Anyone can declare anything like you are. You need to make an attempt at a technical argument. And as a minimum read the technical paper. Not keep writing these information-free posts, passing judgement.
I know this topic. I am in objectivity camp. And I am telling that your case, our case, doesn't hold water. There are no obvious flaws. How would a paper with obvious flaws get the best paper award by peer review board of AES?
But sure, so that we actually discuss something technical, please in your own words state what you think is an obvious flaw.
This isn't particularly earth-shattering when it has already been conceded that 16 bits is not audibly transparent under all circumstances. I'm pretty sure most here will also accept that not all SRCs will guarantee transparency, either.
I don't understand this tendency to use third party ghosts in such references. Who are you speaking on behalf? Let's have that list of people that agree with what you said.
And what do you think the lack of transparency will sound like?
And how do you know that is what was heard in Stuart's paper? You have some back technical details to add to that or is that a generic assumption?
Finally, how come such understood problems did not get uncovered in Meyer and Moran?
Ok, the paper is flawless and on solid ground.
So...
Rectangular dither doesn't matter because the noise floor on the 16-bit version was far enough below that of the listening environment over all frequencies. This rules out quantization noise. So we're left with differences elsewhere.
Is this correct so far?
There is not a whit of evidence that any audiophile, anywhere, can hear any "benefit" with bigger files, in their system, in their room.
There is no evidence that they are not either. So?
Talk about shooting yourself in the foot with a rocket launcher. Your failing to understand how to formulate a valid null-hypothesis puts any analysis of the paper in question in serious doubt. I didn't read the paper, yet, but if the authors use the same "scientific" approach we're done here.
To make it clear, the only meaningful, valid null-hypothesis is that all systems sound the same until proven otherwise.
I think we have slammed the door shut on any statement regarding standing of this paper in the eyes of professional AES community. It is excellent work. It is the most careful double blind test of small differences due to coding of audio that has been published. It is an inconvenient truth to be sure for many, but hopefully the search for knowledge and learning trumps that negative emotion.
It is tragic that such high accolades are being heaped on a document that libels and/or ignores well known previous technology (ABX testing, BS1116 recommendations, etc.)
The fact that the published document on the AES web site is apparently mislabeled speaks to excess haste and rush to judgement.
I am advocating getting the bits prior to remastering for CD. Whatever was produced before conformance with the CD format and commercialization of music to teenage girls is what I want.
I have said nothing about wanting higher frequencies than 20 Khz.
And what I want is already happening. What AJ wants, multi-channel music, is not. I stated the market dynamics. I know the business and technology side of the audio market. You want to ignore the former, sure, go ahead but don't ask me to participate.
If I'm asking for anything, it's for reflection on this "why not" attitude. "Why not" get the "original bits" from the recording studio? If it were only that, who could object? Yet you allude to one problem already: because it drives the market towards a small or non-existent improvement, away from a real and easily audible one.
In the short term, that near-imagined improvement might be profitable for a small niche of the audio market. In the longer term, and for the rest of the market, it's a distraction at best, and a joke (making audio itself a joke) at worst.
It wouldn't matter so much if it was marketed and believed to be "just in case" as you claim here for the sake of argument on this forum. But we both know that's not the case: Some people are selling (and some people are buying) hi-res as the next big thing.
P.S. * = exaggerated for effect, but I think think you can still see the serious point here.
I did. Disappointing to see word arguments rather that substantive discussion.
You ignored 12 sentences from me (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=150&p=880953&#entry880953) and kept 3 (even removing the most important one from the middle of them! (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=880958)). You can't be that disappointed at a lack of "substanetive discussion" otherwise you would have replied to it.
Cheers,
David.
Much of psychoacoustics/acoustics research is done with sighted analysis Arny.
If you really think that, get someone to buy you a subscription to JASA for Christmas.
http://scitation.aip.org/content/asa/journal/jasa (http://scitation.aip.org/content/asa/journal/jasa)
I don't know about all acoustic experiments, because some are hard or impossible to blind, but most psychoacoustic experiments have been routinely blind for nearly a century, and routinely double-blind since mechanised, and then computer controlled, and then fully computerised experiments became the norm.
Cheers,
David.
There are no obvious flaws.
http://robertwannamaker.com/writings/ieee.pdf (http://robertwannamaker.com/writings/ieee.pdf)
Page 31 figure 5 (b) shows the noise modulation due to rectangular dither.
Page 34 figure 7 shows the lack of noise modulation with triangular dither.
Wannamaker, Lipshitz, and Vanderkooy have spread this information, with varying levels of detail, across the world of audio since the 1980s. I am sure that the authors of the BS paper know and understand this information.
The BS paper accurately reports what they did, and tells you (at least partly) why they intentionally introduced this flaw into the test.
Saying "there are no obvious flaws" is just silly.
Cheers,
David.
AJ, you are typing in a thread where we are discussing an award winning, peer-review paper showing such differences can be audible.
Amir, save the "award winning" crap for your upside down DAC crowd, it's irrelevant here, where only the
content of the BS paper is relevant. No one with a functional brain (my "side") has or would ever claim that 16/44 can't be
made to be audible, any more than wires, caps or amps. That's the whole foundation of the your "side" of the industry, pathological audio, where everything that should be inaudible, is made specifically not to be, including digital.
Dubious dither doctored tests and unsupervised online games are not evidence against this.
I hear you. You are wrong. But I hear you. Buy the paper, read it and try again.
That is the entire basis of my analysis...and anyone else who can read. The BS paper deliberately doctored the test with RPDF to ensure positives. But there are other factors which may have created that result, none addressed in the paper - system transparency, switching software transparency etc, etc.
So there's no guarantee the deliberate dither doctoring was the actual cause of the positives.
Demonstrate your point with measurements and data please.
Demonstrate some logic please. Burden of proof is on the BS paper creators. It is they who must show (beyond the pathological dither) proof of validity.
cheers,
AJ
It is tragic that such high accolades are being heaped on a document that libels and/or ignores well known previous technology (ABX testing, BS1116 recommendations, etc.)
Come on Arny, it's a good paper. It accurately reports the lengths they went to in order to get some barely significant results.
You can't blame the AES, or the paper's authors, for the frankly embarrassing way that amirm is lauding it about, or the way he's combining the results with some other seriously flawed tests to imply that hi-res audio is now easily and routinely ABXed.
Cheers,
David.
FYI, I passed all of my double blind ABX tests using my laptop and headphones.
You probably think you're getting a hard time here, but you should take it as a compliment that no one has challenged you on this. The assumption must be that, given your employment history, you must know what you're doing. Anyone else who arrived on HA and made this claim would get a really hard time, due to a long history of drivers, sound cards, and transducers that don't take kindly to ultrasonic content. ABXing hi-res vs 16/44.1 on such crappy equipment is easy, and it's not because hi-res is audibly better.
Cheers,
David.
Are there some details on the system amir used? Which operating system/version, which headphones, which soundcard, which sound API, which DirectSound common sampling rate was configured (if applicable) ... ?
We still have no idea whether any differences, supposing that they were detected reliably, were the result of the content or the hardware.
The thesis of the paper is neither. Those were preconditions for the test. The test hypothesizes that filtering from 192 khz down and conversion with or without dither may be audible. And their listening test showed that to better than 95% confidence, they are.
As I read the paper there was no conversion of the data:
"We wanted to emulate the frequency and impulse responses of such converters while remaining in the 192 kHz domain. Both un ltered and ltered audio samples were sent to
the replay system at 192 kHz..."
Umm Amir due to this and other false claims like it I've read here lately I gotta ask - did you ever actually read the paper?
There are no obvious flaws.
Wishful think fallacy.
1) Test deliberately doctored with (RPDF) dither against BS own recommended best practices (TPDF) to fabricate possible positives.
2) Positives may have been the results of system artifacts, loudspeakers, switching software, etc., not just the pathological dither doctoring.
The test
may have shown what it set out to do (fabricate positives with pathological dither). Or may it not have.
cheers,
AJ
I'll remind you amir, there is not a whit of evidence to suggest 16/44 playback is non-transparent in home audio end use.
AJ, you are typing in a thread where we are discussing an award winning, peer-review paper showing such differences can be audible. The time for making that statement has come and gone. You should look for other talking points, battles to fight. This one is done.
Dubious dither doctored tests and unsupervised online games are not evidence against this.
I hear you. You are wrong. But I hear you. Buy the paper, read it and try again.
Amir I've done exactly that, bought the paper, loaded it on my Kindle and several PCs, and read it any number of times, quoted the paper repeatedly in this thread and watched my quotes and comments be totally ignored by you.
Furthermore, I've posted evidence that suggests that you yourself don't properly grasp the content of the paper, particularly the provenance of the samples that the listeners listened to. You said that they were "Converted" but its quite clear from the paper's text that they were presented at the same sample rate and format as the 24/192 source material, and were never converted to any other sample rate or format.
For another example of well documented critiques that you have ignored Amir, the paper seems to almost completely obfuscate the listening environment. Over on AVS you harshly critiqued the Meyer-Moran tests for their non-compliance with BS 1116, but this paper has the same fault and yet you continue to praise it. At least Meyer-Moran have published photographs of their listening environment (which you also trashed).
Finally Amir, I documented how this paper completely distorted ABX testing, and again I see no response from you.
The test hypothesizes that filtering from 192 khz down and conversion with or without dither may be audible.
Well, I'm glad we all agree now.
But did we really need this extensive BS paper effort to demonstrate what we all already knew?
cheers,
AJ
It is tragic that such high accolades are being heaped on a document that libels and/or ignores well known previous technology (ABX testing, BS1116 recommendations, etc.)
Come on Arny, it's a good paper. It accurately reports the lengths they went to in order to get some barely significant results.
Agreed.
BTW I don't know exactly how significant the results were because I can't find any clear numeric tally of the number of trials. I do find the following:
"
2.4. Listeners
Eight listeners took part in the test, seven of whom
were male. Most of the listeners were audio engi-
neers, and their ages ranged from 25 to 65. All re-
ported normal hearing, although this was not tested
formally.
2.5. Procedure, trials and blocks
The procedure used was intended to combine high
statistical sensitivity and speci ficity with ease of use
by listeners.
In a given trial, listeners were presented with the
same extract of music in two intervals, and were
asked to decide whether or not the extracts had been
processed identically. The fi rst interval always con-
tained an un ltered extract, the so-called \reference
sound", and the second interval contained \sound
X", which was either identical to the reference sound
(un ltered), or had been ltered in some way. Lis-
teners were not limited in the number of times they
could listen to each extract before deciding. Visual
feedback was given as to whether the answer given
was correct or incorrect.
The extract presented in each trial was selected ran-
domly from the 17 sections into which the piece had
been divided based on musical phrases. Twelve trials
were presented within a block", with the results of
the last 10 being counted; the two uncounted trials
were included in order to familiarise listeners with
the task and processing before beginning the test.
For each block, the type of ltering used was the
same. Each listener completed 2 blocks for each con-
dition, giving a total of 12 blocks per subject.
"
I'm still trying to decode the above. Can anybody help me?
You can't blame the AES, or the paper's authors, for the frankly embarrassing way that Amir is lauding it about, or the way he's combining the results with some other seriously flawed tests to imply that hi-res audio is now easily and routinely ABXed.
Again agreed. The overall percentage of correct trials was very low - near 50%. I know from personal experience that what most people consider to be audible effects and even many fairly subtle ones can usually be detected more than 90% of the time. The glossed over asymmetries between the Meridian test environment and a normal or even exceptionally good residential listening system make any possible positive results irrelevant to the world of audiophilia.
As I understand it 8 listeners completed 12 blocks each.
Each block consists of 12 A-X trials (where X=A, the original, or X=B, the processed A) but the first 2 trials were always (?) dismissed.
Visual feedback for each trial if listener chose correctly.
Each trial had a randomly assigned segment from a song.
Within each block, B would be processed the same way.
Now they say "2 blocks for each condition". What is a condition, are those the different processing setups (like different lowpass filters)? I guess so.
There is not a whit of evidence that any audiophile, anywhere, can hear any "benefit" with bigger files, in their system, in their room.
There is no evidence that they are not either. So?
Talk about shooting yourself in the foot with a rocket launcher. Your failing to understand how to formulate a valid null-hypothesis puts any analysis of the paper in question in serious doubt. I didn't read the paper, yet, but if the authors use the same "scientific" approach we're done here.
Thank you for your comment. The day you learn to debate AJ is the day you give me advice. He doesn't come from a logical point of view or interest in the topic. He comes after you personally to get a rise out of you and have fun doing it. While I am answering him with a straight face, I am hoping you see the reality of this. If not, then when you are dealing with him you can try to use the "scientific approach" and see how far that gets you.
To return the favor, you should know I have very little appreciation for people who enter a discussion about a paper but think they don't need to first read it before commenting on it, and instead focus on personal remarks of what the person knows or doesn't know. See below on best example of why you don't want to be in that group:
To make it clear, the only meaningful, valid null-hypothesis is that all systems sound the same until proven otherwise.
And that is what the Stuart's study did: to better confidence than 95%, it showed that in 7 out of 8 tests of different filters and dither strategies, the difference was heard. The null hypothesis is rejected.
These are differences that heretofore were said to be below JND. If challenged, demand was made of a double blind ABX test. That demand was met with running the tests offered. That was not enough. They wanted published tests. Well we have a published test now in the form of Stuart's test. The game of Calvinball then turned into, "oh, but it is not peer reviewed." As I showed, not only was it peer reviewed but won an award among all the others presented at this year's AES conference. An evidence that was visible on the first page of the paper which you say you have not read.
Seems like we have no interest in learning something new. This is a thread about a new paper and listening test. The remarks instead are judgemental about which party should be supported on principal and personal bias. That is not how science works. One of the core tenets of scientific method is being unbiased and neutral at all times. Don't take sides and let that cloud your judgement. Here is a great video from one of my favorite Physics professors explaining what the scientific method is:
https://www.youtube.com/watch?v=IX2NPyvYz3w (https://www.youtube.com/watch?v=IX2NPyvYz3w)
If you are too impatient to watch a 12 minute vide, jump to minute 6 and see the point I just made.
Amir I've done exactly that, bought the paper, loaded it on my Kindle and several PCs, and read it any number of times, quoted the paper repeatedly in this thread and watched my quotes and comments be totally ignored by you.
Yet you missed the fact it was peer reviewed and won an award before calling it and its author a bunch of names. More below
Furthermore, I've posted evidence that suggests that you yourself don't properly grasp the content of the paper, particularly the provenance of the samples that the listeners listened to. You said that they were "Converted" but its quite clear from the paper's text that they were presented at the same sample rate and format as the 24/192 source material, and were never converted to any other sample rate or format.
You need to go and it again then Arny. The sample rate was kept constant just like our tests on AVS Forum. The signal processing in Matlab performed the filtering/dither while keeping the sample rate the same. Just like you simulating 44.1/16 by giving us upsampled files to 24/96, they have done that using Matlab and 192 Khz sampling of the original tracks.
For another example of well documented critiques that you have ignored Amir, the paper seems to almost completely obfuscate the listening environment. Over on AVS you harshly critiqued the Meyer-Moran tests for their non-compliance with BS 1116, but this paper has the same fault and yet you continue to praise it. At least Meyer-Moran have published photographs of their listening environment (which you also trashed).
And you learned what from that photo Arny?
I have already compared and contrasted this work to that of Meyer and Moran with respect to BS1116. I am happy to do a much more complete one. You know Meyer and Moran will lose and lose big, yes?
Finally Amir, I documented how this paper completely distorted ABX testing, and again I see no response from you.
I intend to create a new thread on some of your comments there per TOS #5 and instructions from our moderator.
There are no obvious flaws.
Wishful think fallacy.
1) Test deliberately doctored with (RPDF) dither against BS own recommended best practices (TPDF) to fabricate possible positives.
Explain how that is an obvious flaw AJ. Can you do that?
2) Positives may have been the results of system artifacts, loudspeakers, switching software, etc., not just the pathological dither doctoring.
Sure, anything is possible. Your job is to show it is probable and the way to do that is with data. Otherwise, it is a FUD tactic to put a cloud on an award winning, peer reviewed, scientific paper, published and presented at AES conference, by an AES Life Fellow, with proper engineering and research experience. Yes, I have learned to talk like you guys.
You want to counter all of that with just your opinion AJ?
The test may have shown what it set out to do (fabricate positives with pathological dither). Or may it not have.
Rectangular dither is pathological dither? Where would I read about that AJ?
Explain how that is an obvious flaw AJ. Can you do that?
Two words: noise modulation.
Now I don't think that the choice of dither usually matters, but in critical evaluations of the limitations of a format I would
at least expect triangular dither.
The day you learn to debate AJ is the day you give me advice. He doesn't come from a logical point of view or interest in the topic. He comes after you personally to get a rise out of you and have fun doing it. While I am answering him with a straight face, I am hoping you see the reality of this. If not, then when you are dealing with him you can try to use the "scientific approach" and see how far that gets you.
In Amirworld this passed as logic. But out here in the real world of HA, that is classic fallacy of focusing on the arguer, rather than argument, which is the BS paper.
Amir, please keep your focus on discussion of the
paper, not me. Thanks.
Now I don't think that the choice of dither usually matters, but in critical evaluations of the limitations of a format I would at least expect triangular dither.
At least. Someone wanting to give 16-bits the best chance might use noise shaping, as advocated by a certain J.R.Stuart in his presentations at the 96th AES convention in Amsterdam (1994) and the 100th AES convention in Copenhagen (1996).
(preprints 3501 and 3871 respectively)
amirm, this really is getting embarrassing. The paper is very clear: they wanted to test bad downconversion. They
intentionally introduced a flaw. Please stop pretending otherwise. They didn't feel the need to pretend that this was a state-of-the-art test; why do you?
Cheers,
David.
Explain how that is an obvious flaw AJ. Can you do that?
Rectangular dither is pathological dither? Where would I read about that AJ?
Here:
2) Positives may have been the results of system artifacts, loudspeakers, switching software, etc., not just the pathological dither doctoring.
Sure, anything is possible. Your job is to show it is probable and the way to do that is with data.
Wrong. Your job is to learn logic and understand that the burden of proof lies squarely on those creating this BS paper and its claims. They need to show that system transparency - speakers, switching software, etc, etc....aren't the source of the positives....which would make them false (despite all the dither doctoring efforts).
Btw, that's in his previous paper also, warnings about false positives. Just like JJ cautioned...
You really should listen to those experts Amir.
cheers,
AJ
I thought my previous "Seriously? Doh. (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881060)" response was enough to show the absurdity of such a position, but I guess I was wrong.
There is no evidence that unicorns don't exist, so it's reasonable to assume that they do? That they are likely to exist?
And on the latter post: If I make an experiment that shows telepathic abilities, it is your job to show that the experiment did indeed show exactly that? It is your job to ensure that bias in the experiment was eliminated, variables were controlled ....? What?!
As I understand it 8 listeners completed 12 blocks each.
Each block consists of 12 A-X trials (where X=A, the original, or X=B, the processed A) but the first 2 trials were always (?) dismissed.
Visual feedback for each trial if listener chose correctly.
Each trial had a randomly assigned segment from a song.
Within each block, B would be processed the same way.
Now they say "2 blocks for each condition". What is a condition, are those the different processing setups (like different lowpass filters)? I guess so.
A condition appears to be a test condition;;
Condition number Filter cutoff frequency (Hz) Further processing
1 21591-22050 None
2 21591-22050 16-bit quantization
3 21591-22050 16-bit quantization and rectangular dither
4 23500-24000 None
5 23500-24000 16-bit quantization
6 23500-24000 16-bit quantization and rectangular dither
Table 1: Details of each condition tested.
The above cutoff frequency numbers seem to relate to the low pass filter's transition frequency.
"
The frequencies of the transition bands were 23500-
24000 Hz and 21591-22050 Hz, corresponding to the
standard sample rates of 48 kHz and 44.1 kHz re-
spectively.4 Fig. 2 shows the amplitude and energy
of the impulse response for the 48-kHz fllter.
"
The following statement appears to raise more critical questions:
"These parameters were chosen to off er a reasonable match to the downsampling fi lters used in good- quality A/D converters or in the mastering process;"
I looked at the spec sheets for a number of modern good quality DACs, and found that the 500 Hz wide transition bands chosen by Meridian for their tests seem to vary considerably from the approximate 2 KHz or wider transition bands that I found in these commercial products operating with the same corner frequencies.
This is interesting because FIR filters with narrow transition bands were used for listener training, with the comment that "This fi lter was chosen as it would have been straightforward for most listeners to identify diff erences introduced by its application."
So, there you go - yet another asymmetry between the test conditions and the real world.
BTW, there is a lesson here. The Quality slider on software resamplers such as CoolEdit Pro/Audition is a tuning knob for transition band width. Higher quality means a narrower transition band.
If you push it to the far right (99%) the resulting transition band became exceedingly narrow in my tests - just a few Hz, which puts it in the same realm as the signal that Meridian used for listener training.
If you push it to the far left the transition band is more like 1.6 KHz and better approximates a real world converter.
All of the sample rate ABX test files I've circulated lately were made with quality set for the maximum. They are therefore not representative of real world circumstances. One word: Invalid.
My guess is he ignores this and goes for low hanging fruit, or revert to jabbing at AJ, which he knows is not appropriate.
Or insist the paper is a slam dunk technical demonstration as to why people should feel justified in choosing hi-res because it won an award, which now seems to be the new default fallback position when he can't convince anyone that they shouldn't be concerned that the results may have been skewed by the hardware or choice of processing; let alone whether this is all that representative of real-world conditions beyond the world of selling compromised designs to the gullible eager to part with their cash.
The sample rate was kept constant just like our tests on AVS Forum.
I clearly said that I converted those test files from 24/96 to 44/16 and then back again to 24/94. You can take that literally - the "4416" test files existed for a while as 44/16 files. The Meridian files never did.
By continuing to argue with me it is clear that you remain confused about the facts of the matter which is yet another win.
Explain how that is an obvious flaw AJ. Can you do that?
Two words: noise modulation.
And the data that shows it to be in effect in this test is where?
Now I don't think that the choice of dither usually matters, but in critical evaluations of the limitations of a format I would at least expect triangular dither.
If it usually doesn't matter then it is not an "obvious flaw."
The sample rate was kept constant just like our tests on AVS Forum.
I clearly said that I converted those test files from 24/96 to 44/16 and then back again to 24/94. You can take that literally - the "4416" test files existed for a while as 44/16 files. The Meridian files never did.
It is the same thing in Meridian test but using an advanced methodology (for lay people) using Matlab real-time processing. David, do you want to explain this to Arny? Clearly he doesn't want to listen to me.
By continuing to argue with me it is clear that you remain confused about the facts of the matter which is yet another win.
No Arny. The paper has won an award from its peers. Don't think of pedantic mistakes. Any that you think you find is due to not understanding how professional audio tests are performed thee days.
Explain how that is an obvious flaw AJ. Can you do that?
Two words: noise modulation.
And the data that shows it to be in effect in this test is where?
Right here, Amir.
http://www.hydrogenaud.io/forums/index.php...st&p=881116 (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&view=findpost&p=881116)
which links this excellent paper:
http://robertwannamaker.com/writings/ieee.pdf (http://robertwannamaker.com/writings/ieee.pdf)
Page 31 figure 5 (b) shows the noise modulation due to rectangular dither.
Page 34 figure 7 shows the lack of noise modulation with triangular dither.
The post was addressed to you Amir. You've responded to posts made since it went up.
It is now obvious that you couldn't properly comprehend its meaning and relevance. Another win.
My guess is he ignores this and goes for low hanging fruit, or revert to jabbing at AJ, which he knows is not appropriate.
Or insist the paper is a slam dunk technical demonstration as to why people should feel justified in choosing hi-res because it won an award, which now seems to be the new default fallback position when he can't convince anyone that they shouldn't be concerned that the results may have been skewed by the hardware or choice of processing; let alone whether this is all that representative of real-world conditions beyond the world of selling compromised designs to the gullible eager to part with their cash.
Disappointing to see yet another commentary devoid of any technical insight. Would you like to say everything that is on your chest so that we can be done with it?
Disappointing to see yet another commentary devoid of any technical insight. Would you like to say everything that is on your chest so that we can be done with it?
You mean another commentary devoid of any technical insight like this one?
http://www.hydrogenaud.io/forums/index.php...st&p=881152 (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&view=findpost&p=881152)
There are no obvious flaws.
http://robertwannamaker.com/writings/ieee.pdf (http://robertwannamaker.com/writings/ieee.pdf)
Page 31 figure 5 (b) shows the noise modulation due to rectangular dither.
Page 34 figure 7 shows the lack of noise modulation with triangular dither.
Wannamaker, Lipshitz, and Vanderkooy have spread this information, with varying levels of detail, across the world of audio since the 1980s. I am sure that the authors of the BS paper know and understand this information.
The BS paper accurately reports what they did, and tells you (at least partly) why they intentionally introduced this flaw into the test.
Saying "there are no obvious flaws" is just silly.
Cheers,
David.
Looks like the plot is lost. You are giving me a tutorial on noise modulation and dither. That has nothing to do with the topic at hand or the question I asked.
The test has been called to have a major flaw. You need to demonstrate using the paper and data within, how it was that people heard noise modulation and not effect of filtering/dither. Do you have that?
I also like to see people go on record and state that anytime a 24 bit file is converted to 16 bit using rectangular dither, the conversion is highly flawed and lossy.
There are no obvious flaws.
http://robertwannamaker.com/writings/ieee.pdf (http://robertwannamaker.com/writings/ieee.pdf)
Page 31 figure 5 (b) shows the noise modulation due to rectangular dither.
Page 34 figure 7 shows the lack of noise modulation with triangular dither.
Wannamaker, Lipshitz, and Vanderkooy have spread this information, with varying levels of detail, across the world of audio since the 1980s. I am sure that the authors of the BS paper know and understand this information.
The BS paper accurately reports what they did, and tells you (at least partly) why they intentionally introduced this flaw into the test.
Saying "there are no obvious flaws" is just silly.
Cheers,
David.
Looks like the plot is lost.
More indications of poor reading comprehension. The plots are just where the post says they are and they speak loudly to any person who is even vaguely informed on this topic.
You are giving me a tutorial on noise modulation and dither. That has nothing to do with the topic at hand or the question I asked.
Even more indications of poor reading comprehension. The issue being discussed is exactly noise modulation and dither, more specifically the well known (by most of us) adverse effects of rpdf dither on noise modulation and other potentially audible badness.
To summarize: the recent Meridian tests sabotaged 16/44 and 16/48 processing by
intentionally using
(1)
suboptimal dither PDF,
(2)
suboptimal dither PSD,
(3) and
suboptimal low pass filter transition band width.
All fully documented in this thread.
Even more indications of poor reading comprehension. The issue being discussed is exactly noise modulation and dither, more specifically the well known (by most of us) adverse effects of rpdf dither on noise modulation and other potentially audible badness.
To summarize: the recent Meridian tests sabotaged 16/44 and 16/48 processing by intentionally using
(1) suboptimal dither PDF,
(2) suboptimal dither PSD,
(3) and suboptimal low pass filter transition band width.
All fully documented in this thread.
Those are claims. This is reality from the paper:
(http://amirviews.smugmug.com/photos/i-VwtbpfR/0/O/i-VwtbpfR.png)
Now explain why you dispute the line I have highlighted in yellow and expressed very clearly on this graph from the paper.
And I like to see you go on record that dither shown at those levels is audible, and a major flaw in any conversion from 24 bits to 16.
It is the same thing in Meridian test but using an advanced methodology (for lay people) using Matlab real-time processing.
"Real-time"?
You need to demonstrate using the paper and data within, how it was that people heard noise modulation and not effect of filtering/dither.
Since noise modulation
is the effect of using that incorrect dither, I don't know how to answer that self-contradictory question.
I thought xnor had already explained the other part pretty well. Many signals are sufficiently self-dithering for this to be academic, but with carefully chosen samples and sufficient gain you can demonstrate that the choice of dither can be audible. None of this is news to anyone.
Go on - remind me again how you want to pay extra to get the original bits from the studio to avoid this - because the way to make sure that the rest of the astronomically-more-complex recording chain is OK is obviously to find some guy who you can't even trust to leave the "dither" box ticked, and pay him extra money because he can't be trusted I mean, honestly, what could possibly go wrong?!
Cheers,
David.
P.S. You said something about the CD version being the dumbed down version for teenage girls. Have you heard the amount of dynamic range compression on
some hi-res recordings?! But of course, you have the "original bits", so they must be wonderful? Not that you are daft enough to believe this, but plenty are.
Now explain why you dispute the line I have highlighted in yellow and expressed very clearly on this graph from the paper.
Do you understand the significance of the text immediately before the part you highlighted in yellow? "Noise spectral density is derived in a 1-Hz bandwidth"?
I think if I had time to do the calculation properly, I might find there's enough dither noise in a critical band to touch the threshold of hearing. That's the kind of thing that some people might just be able to detect.
Cheers,
David.
Is amir deliberately obtuse? Is he trolling? Or is that just the way he is?
amir, do you have the tracks used in this experiment? The algorithms used for processing? The software/hardware used to do the AB switching, listening ... ?
You need to demonstrate using the paper and data within, how it was that people heard noise modulation and not effect of filtering/dither.
Since noise modulation is the effect of using that incorrect dither, I don't know how to answer that self-contradictory question.
I thought xnor had already explained the other part pretty well. Many signals are sufficiently self-dithering for this to be academic, but with carefully chosen samples and sufficient gain you can demonstrate that the choice of dither can be audible. None of this is news to anyone.
I post the relevant data on this in my response to Arny. Please show using the gain in their setup how noise modulation was easily audible. This is what I meant when I said to back your statements using the report. You can't critique it in absentia.
Alternatively, there is only one track in this test. It should be easy then to demonstrate how dither noise modulation caused a major flaw and was readily audible. Or else it would not be a major flaw, right?
For now, these are the subjective observations we have from the paper from the test subjects:
It was reported that filtering gave "softer edges" to
the instruments, and "softer leading edges" to musi-
cal features with abrupt onsets or changes. Echoes,
when audible, were identi ed as being affected the
most clearly by the ltering. It was felt that some
of the louder passages of the recording were less ag-
gressive after filtering, and that the inner voices (sec-
ond violin and viola) had "a nasal quality". Over-
all, the filtered recording gave a "smaller and
flatter auditory image", and speci cally the physical space
around the quartet seemed smaller.
Listeners described that quantization gave a "rough-
ness" or "edginess" to the tone of the instruments,
and that quantization had a significant impact on
decay, particularly after homophonic chords, where
"decay was sustained louder for longer and then died
suddenly". This could be an effect of quantization
distortion; it is interesting that this was audible even
in a 24-bit system, and is consistent with the hy-
potheses of Stuart [29] that 16 bits are not sucient
for inaudible quantization.
Would you say these are consistent with audible effects of noise modulation?
Edit: fixed the OCR problems in cut and paste of the quote above.
Is amir deliberately obtuse? Is he trolling? Or is that just the way he is?
More disappointing personal commentary. Looks like HA forum is much less professional in this manner than I thought prior to joining the forum.
amir, do you have the tracks used in this experiment? The algorithms used for processing? The software/hardware used to do the AB switching, listening ... ?
I have the tracks. They are at 96 Khz and I bought them because I liked the music.
I do not have the algorithms.
I do have Matlab. But don't have the GUI front-end they wrote to perform the test.
How about you? Do you have the paper and read it? Downloaded the tracks?
Explain how that is an obvious flaw AJ. Can you do that?
Two words: noise modulation.
And the data that shows it to be in effect in this test is where?
The argument is that rectangular dither may have affected the results. If you're insisting that the argument has been refuted, which appears to be the case (the insistence, mind you) then, you are essentially claiming that dither was not responsible. This is the claim that requires supporting evidence. It's easy enough to test if that was the intention of the authors, so where is the hard data?
I also like to see people go on record and state that anytime a 24 bit file is converted to 16 bit using rectangular dither, the conversion is highly flawed and lossy.
I love the smell of moldy straw in the morning...
More disappointing personal commentary.
Well, if several people have to point out to you several times basic stuff (such as burden of proof), but you keep repeating nonsense... (which is very disappointing when reading the thread) at some point people will question if you're up to the discussion.
I do not have the algorithms.
I guess it's the same for the hardware. Then why do you expect others to have the software/hardware used, so they could reproduce the experiment accurately (a basal requirement in science, btw) and pinpoint flaws?
Explain how that is an obvious flaw AJ. Can you do that?
Two words: noise modulation.
And the data that shows it to be in effect in this test is where?
The argument is that rectangular dither may have affected the results. If you're insisting that the argument has been refuted, which appears to be the case, you are essentially claiming that dither was not responsible. This is the claim that requires supporting evidence. It's easy enough to test if that was the intention of the authors, so where is the hard data?
Here it is again:
(http://amirviews.smugmug.com/photos/i-VwtbpfR/0/O/i-VwtbpfR.png)
And the comments from listeners.
I also like to see people go on record and state that anytime a 24 bit file is converted to 16 bit using rectangular dither, the conversion is highly flawed and lossy.
I love the smell of moldy straw in the morning...
No, it is the stink of not wanting to admit noise modulation at such low levels is audible, yet was an obvious/audible flaw in the test.
More disappointing personal commentary. Looks like HA forum is much less professional in this manner than I thought prior to joining the forum.
Haha, you've always been a kidder amir.
Now, regarding (BS) commentary, what do you make of this:
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
Why does BS use TPDF in their own products for "transparent processing", then use (seemingly) non-transparent RPDF to fabricate desired results?
But don't have the GUI front-end they wrote to perform the test.
Nor the speakers?
So you are purely assuming transparency (wishful thinking fallacy?), since there is no indication in the paper either was tested.
You're choosing to ignore both BS and JJ warnings regarding false positives due to the strong possibility of artifacts with any such test? Why?
cheers,
AJ
Yes, and obvious flaws exist which were touched upon, which you have not been able to dismiss successfully.
If dither wasn't a problem, which hasn't been shown to my satisfaction*, then you're still left to contend with differences caused by artifacts in hardware/software; to which you also don't have a satisfactory answer.
So, sure, the thesis *may* have been supported by the results, but with severe caveats.
This isn't particularly earth-shattering when it has already been conceded that 16 bits is not audibly transparent under all circumstances. I'm pretty sure most here will also accept that not all SRCs will guarantee transparency, either.
Indeed. No one , including Meyre and Moran, has claimed 16bit is never audible. No one has claimed that SRC *cannot* be audible. No one has said that Redbook is transparent to *all* audible signals.
The issue, for me, is entirely this: since the dawn of hi rez, the reviews have been replete with reports of
obvious improvement over Redbook. You need merely listen to hear the obvious improvement (if you don't, you're deaf, or 'your equipment is not resolving enough') . Do any factors uncovered in DBT account for such widespread reports, gleaned from all sorts of different systems, using all sorts of different recordings, and all sorts of listeners, united only in their firm conviction that hi rez 'sounds better'?
However, this topic is more about what people will attempt to make of the results and the predictable posturing they will take.
Did I get that right, krab?
Partly. I started this thread as a discussion forum for the paper:
(NB it has been posted once before on HA in the midst of a thread...but perhaps we can use this thread to discuss it, once it is available)
The posturing is just the usual , though I'm glad HA is getting to see the Dancing Man's performance. Everyone should see it once.
hard data?
Here it is again:
And the comments from listeners.
"Should" and the interpretation of anecdotes do not constitute hard data.
It is simple enough to test, so where is the evidence that rectangular dither didn't affect the results of the test?
I think we have slammed the door shut on any statement regarding standing of this paper in the eyes of professional AES community. It is excellent work. It is the most careful double blind test of small differences due to coding of audio that has been published. It is an inconvenient truth to be sure for many, but hopefully the search for knowledge and learning trumps that negative emotion.
It is tragic that such high accolades are being heaped on a document that libels and/or ignores well known previous technology (ABX testing, BS1116 recommendations, etc.)
The fact that the published document on the AES web site is apparently mislabeled speaks to excess haste and rush to judgement.
- we
don't know how fierce the competition was for the award
- we
don't know who the two reviewers were
- we
do know that Amir has dismissed peer review before , when it suited him (re: Meyer & Moran 2007)
So we can be sure he won't make
too much of this convention paper having been peer reviewed for a convention presentation. I mean, he'd just look silly.
Right?
Even more indications of poor reading comprehension. The issue being discussed is exactly noise modulation and dither, more specifically the well known (by most of us) adverse effects of rpdf dither on noise modulation and other potentially audible badness.
To summarize: the recent Meridian tests sabotaged 16/44 and 16/48 processing by intentionally using
(1) suboptimal dither PDF,
(2) suboptimal dither PSD,
(3) and suboptimal low pass filter transition band width.
All fully documented in this thread.
Those are claims. This is reality from the paper:
(http://amirviews.smugmug.com/photos/i-VwtbpfR/0/O/i-VwtbpfR.png)
Now explain why you dispute the line I have highlighted in yellow and expressed very clearly on this graph from the paper.
And I like to see you go on record that dither shown at those levels is audible, and a major flaw in any conversion from 24 bits to 16.
Amir, you obviously did not comprehend the part of the paper which clearly said that in a 44 KHz channel, the RPDF dither was approximately 2 times or 6 dB higher.
Footnote 5 page 6:
"Since the total power of RPDF dither in a 16-bit channel is constant, spectral density in a 192-kHz channel is approximately 6 dB lower than in a 44.1-kHz channel." The graphs shown are obviously for 192 KHz sampling, since they extend a great deal beyond the Nyquist frequency of 44.1 KHz sampling, or 22.05 KHz. Another win!
You also must not have comprehended the graph and text in the now twice cited Wannamaker/Lipschitz paper (p32):
"The error variance, on the other hand, (for rpdf dither) is clearly signal dependent, so that the noise power in the signal varies with the input. This is sometimes referred to as noise modulation and is
undesirable in audio or video signals."
A bit of friendly advice - just because we call a process Noise Shaping or an undesired signal Quantization Noise doesn't mean that they only apply to noise (random or stochastic signals). Quantization Noise can be caused by coherent and even deterministic signals such as music, and involve pseudorandom noise, in which case it can be deterministic and might even be coherent. Then, its audibility follows the rules for coherent signals which raises the JND threshold considerably. The cited graph in the Meridian paper does not apply.
I am reminded that people who are ultra-quick to play the amateur/professional card most vigorously are often some the biggest posers around. ;-)
You can't blame the AES, or the paper's authors, for the frankly embarrassing way that amirm is lauding it about, or the way he's combining the results with some other seriously flawed tests to imply that hi-res audio is now easily and routinely ABXed.
Cheers,
David.
Well, remember what he did with that HA wiki graph 'showing' that 320kbps is not transparent?
It's how he do.
And you know he's not just doing it
here. And he's not the only one doing.
So, just as the hi-end crowd had to contend with Meyer & Moran since 2007 (there' s even a letter about M&M in *this month's*,
Stereophile -- very conveniently I would add, since it gives the eds an excuse to tout the Meridian paper*) , now 'we' will have to deal with this, and with parties like Amir flogging it around the Internets like cheerleaders on Red Bull.
*there's also particularly( but not surprisingly) obtuse essay by Art Dudley, on the ever-popular theme of
why double blind testing doesn't work for audio (http://www.stereophile.com/content/listening-143). Which, given that Meridian used DBT in their convention paper, is amusing.
The posturing is just the usual , though I'm glad HA is getting to see the Dancing Man's performance.
You mean attempt to shift the burden of proof and demand evidence to counter the presentation of data that doesn't support his position?
Or ignore posts that are deserving of a response?
Or argue a different topic?
Or engage in personal attacks?
No, this hasn't gone unnoticed.
The posturing is just the usual , though I'm glad HA is getting to see the Dancing Man's performance.
You mean attempt to shift the burden of proof and demand evidence to counter the presentation of data that doesn't support his position?
Or ignore posts that are deserving of a response?
Or argue a different topic?
Or engage in personal attacks?
No this, hasn't gone unnoticed.
You left out his po-faced concern trolling when a response to him get too 'personal'.
It's such a well-rehearsed dance of many steps, I'm guessing it's the work of a lifetime -- I don't want to see any of it neglected.
Can someone upload a 30 second sample to the uploads section please? Include a loud bit and a quiet bit. Thanks.
Cheers,
David.
FYI, I passed all of my double blind ABX tests using my laptop and headphones.
You probably think you're getting a hard time here, but you should take it as a compliment that no one has challenged you on this. [...] Anyone else who arrived on HA and made this claim would get a really hard time, due to a long history of drivers, sound cards, and transducers that don't take kindly to ultrasonic content. ABXing hi-res vs 16/44.1 on such crappy equipment is easy, and it's not because hi-res is audibly better.
I can re-open the train wreck discussion if you like, though it will get shut down again if arguments that belong here spill over into there like they did before.
FYI, I passed all of my double blind ABX tests using my laptop and headphones.
You probably think you're getting a hard time here, but you should take it as a compliment that no one has challenged you on this. [...] Anyone else who arrived on HA and made this claim would get a really hard time, due to a long history of drivers, sound cards, and transducers that don't take kindly to ultrasonic content. ABXing hi-res vs 16/44.1 on such crappy equipment is easy, and it's not because hi-res is audibly better.
I can re-open the train wreck discussion if you like, though it will get shut down again if arguments that belong here spill over into there like they did before.
Two obfuscation contests at the same time? You must enjoy Dancing with the stars.
*there's also particularly( but not surprisingly) obtuse essay by Art Dudley, on the ever-popular theme of why double blind testing doesn't work for audio (http://www.stereophile.com/content/listening-143). Which, given that Meridian used DBT in their convention paper, is amusing.
Yes, I'm not quite sure how the Hi-Rez peddlers can utilize this BS paper.
The scam addicts vehemently reject the notion of blind tests, so how do they hang their hats on this one?
Quite the dilemma, unless of course....
cheers,
AJ
Amir, you obviously did not comprehend the part of the paper which clearly said that in a 44 KHz channel, the RPDF dither was approximately 2 times or 6 dB higher.
Footnote 5 page 6:
"Since the total power of RPDF dither in a 16-bit channel is constant, spectral density in a 192-kHz channel is approximately 6 dB lower than in a 44.1-kHz channel." The graphs shown are obviously for 192 KHz sampling, since they extend a great deal beyond the Nyquist frequency of 44.1 KHz sampling, or 22.05 KHz. Another win!
Yes, another win for science, and a fail for laymen trying to interpret the same. The test is run at 192 Khz Arny. Effect of filtering down to 44.1 Khz simulated within (again, just like your tests of upsampling back the 44.1 to 96 Khz). As such, the graph properly shows its audibility:
(http://amirviews.smugmug.com/photos/i-VwtbpfR/0/O/i-VwtbpfR.png)
But go ahead and move the green line up 6 db if it makes you feel better. It is still below the threshold of hearing.
Remember,
you will lose a much larger battle if you stay on this path Arny. That such noise levels is audible in a 105 db SPL playback. And that they are so audible as to be an obvious flaw in the test.So keep going at your peril.
Yes, another win for science, and a fail for laymen trying to interpret the same.
We'll hang on to this one for a while.
Amir, you obviously did not comprehend the part of the paper which clearly said that in a 44 KHz channel, the RPDF dither was approximately 2 times or 6 dB higher.
Footnote 5 page 6:
"Since the total power of RPDF dither in a 16-bit channel is constant, spectral density in a 192-kHz channel is approximately 6 dB lower than in a 44.1-kHz channel." The graphs shown are obviously for 192 KHz sampling, since they extend a great deal beyond the Nyquist frequency of 44.1 KHz sampling, or 22.05 KHz. Another win!
Yes, another win for science, and a fail for laymen trying to interpret the same. The test is run at 192 Khz Arny.
The fail is yours Amir. When downsampling with dither per test conditions, the quantization error is generated at the point of downsampling, and therefore distributed over the narrower bandwidth which leads to the higher amplitude.
But go ahead and move the green line up 6 db if it makes you feel better. It is still below the threshold of hearing.
Prove it Amir, by drawing in the threshold of hearing for continuous tones.
Finally Amir, you keep forgetting that the reliance on flat PSD dither is one of the serious flaws in the recent Meridian tests. The use of perceptually shaped PSD dither rather spectacularly drops the noise level, particularly in the range of frequencies where the ear is most sensitive.
Prove that you even know what perceptually shaped PSD dither is Amir, by drawing it in as well.
In my last post I was guilty of using the royal we, though I can think of a few members who'd see at least a little irony in what I quoted.
Since I'm on the subject of we...
I haven't seen this awareness in other forums. Is there a link here where folks have said that in double blind tests, people can tell the difference between how 24 bit files are or are not dithered to 16 bits?
I don't understand this tendency to use third party ghosts in such references. Who are you speaking on behalf? Let's have that list of people that agree with what you said.
More disappointing personal commentary. Looks like HA forum is much less professional in this manner than I thought prior to joining the forum
It is not our responsibility to acquaint you with our community. We have a search function. No one is prohibiting you from using it prior to posting.
Guys, there is a difference between linear spectral density (measured in V/sqrt(Hz)) and linear spectrum (measured in V).
The density is obviously lower for a wider bandwidth, since it is "spread" across the wider bandwidth.
sqrt(22050)/sqrt(96000) = -6.38 dB
For months on AVS Forum we have been discussing the audibility of high resolution audio. After passing ABX DBT after ABX DBT, the claim always went back, "well, Meyer and Moran didn't find this difference and it was peer reviewed so this data must be wrong."
I guess you didn't get the memo Amir (silliy I should expect you to comprehend it, given that it is in a recent post to this thread), but the ABX tests I posted on AVS had a built in flaw that I only very recently became aware of which is that the transition band of my downsampling filter was way too narrow per this certain peer reviewed paper from Meridian.
It was covered in the second part of this post: http://www.hydrogenaud.io/forums/index.php...st&p=881145 (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&view=findpost&p=881145)
We know that the other set of files had other serious flaws of their own.
It thus becomes improper for anybody to say: "After passing ABX DBT after ABX DBT" because there were no proper tests.
the ABX tests I posted on AVS had a built in flaw
Any reasonable, objective person would see a flawed test as flawed and thus the results flawed, be they online games or manufacturer doctored up results.
Such a position is incompatible with objectivity, if one accepts flawed, unsupervised-gameable and deliberately doctored results like this BS test.
It must be something else.
cheers,
AJ
For months on AVS Forum we have been discussing the audibility of high resolution audio. After passing ABX DBT after ABX DBT, the claim always went back, "well, Meyer and Moran didn't find this difference and it was peer reviewed so this data must be wrong."
I guess you didn't get the memo Amir (silliy I should expect you to comprehend it, given that it is in a recent post to this thread), but the ABX tests I posted on AVS had a built in flaw that I only very recently became aware of which is that the transition band of my downsampling filter was way too narrow per this certain peer reviewed paper from Meridian.
It was covered in the second part of this post: http://www.hydrogenaud.io/forums/index.php...st&p=881145 (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&view=findpost&p=881145)
We know that the other set of files had other serious flaws of their own.
It thus becomes improper for anybody to say: "After passing ABX DBT after ABX DBT" because there were no proper tests.
I appreciate you falling on your sword this way as to invalidate your own test that you had put out there for years to prove high resolution audio is not differentiated against CD rate. But it is not necessary as your theory is not correct here. As you know, the fidelity of your resampler was questioned repeatedly and you defended it. This is one such example with me asking the same question: http://www.avsforum.com/forum/91-audio-the...ml#post26134194 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1532092-debate-thread-scott-s-hi-res-audio-test-86.html#post26134194)
Why are you asking me? Won't I be a biased source? ;-)
If you read my posts you know that the SRC I use was the one in Cool Edit Pro 2.1 with pre/post filtering, and you know about the SRC reviews at Infinite Wave. Adobe Audition 2.0 is a close descendant of Cool Edit Pro 2.0 which is testsed at Infinnite Wave. I could tell you to go figure.
However, I can also tell you that I've run the tests that Infinite Wave used on all those SRCs on the Cool Edit Pro 2.1 SRC with very similar results. I also did some tests that are far more difficult than what Infinite Wave used, and found it has a weakness. The weakness is that if you upsample or downsample with really high ratios such as upsampling from 44.1 to several, maybe 10 megahertz, its digital filter's bandpass filtering seems to loose some of its precision. The circumvention is to do the upsampling in steps.
Strange that such obvious problems as to invalidate your test were not brought out in the above testing. Perhaps your testing wasn't comprehensive enough then?
You say that the fault lay with using the quality slider in advanced settings and pushing it to max. As it turns out, I had tested it on the default 50% setting as I reported here: http://www.whatsbestforum.com/showthread.p...ll=1#post281829 (http://www.whatsbestforum.com/showthread.php?15255-Conclusive-quot-Proof-quot-that-higher-resolution-audio-sounds-different&p=281829&viewfull=1#post281829)
Arny's files never had sync problems. But yes, I did resample them using latest version of Audition CC and I could still tell them apart.
============
foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/24 20:27:41
File A: C:\Users\Amir\Music\Arnys Filter Test\keys jangling amir-converted 4416 2496.wav
File B: C:\Users\Amir\Music\Arnys Filter Test\keys jangling full band 2496.wav
20:27:41 : Test started.
20:28:07 : 00/01 100.0%
20:28:25 : 00/02 100.0%
20:28:55 : 01/03 87.5%
20:29:02 : 02/04 68.8%
20:29:12 : 03/05 50.0%
20:29:20 : 04/06 34.4%
20:29:27 : 05/07 22.7%
20:29:36 : 06/08 14.5%
20:29:44 : 07/09 9.0%
20:29:55 : 08/10 5.5%
20:30:00 : 09/11 3.3%
20:30:07 : 10/12 1.9%
20:30:16 : 11/13 1.1%
20:30:22 : 12/14 0.6%
20:30:29 : 13/15 0.4%
20:30:36 : 14/16 0.2%
20:30:41 : 15/17 0.1%
20:30:53 : 16/18 0.1%
20:31:03 : 17/19 0.0%
20:31:07 : Test finished.
----------
Total: 17/19 (0.0%)
So I managed to differentiate the files with a much newer implementation of your SRC, at less than maximum setting and still managed to easily tell the files apart in DBT ABX test.
------
But let's put all of this aside and presume you are right. This solidly makes the case for getting the original bits prior to down conversion to 16/44.1. After all, if you with all of your experience would produce an obviously faulty conversion as to have clear audible distortions, and said problem was setting a "quality slider to 100%," what hope is there for any content producer creating transparent content? Are we to assume they know far more than you do as to not have made this mistake Arny?
Let me leave you with this note from Adobe Audition CC's Help file: http://help.adobe.com/en_US/audition/cs/us...0c2c5-7f52.html (http://help.adobe.com/en_US/audition/cs/using/WS58a04a822e3e5010548241038980c2c5-7f52.html)
In the Advanced section, drag the Quality slider to adjust the quality of the sampling conversion.
Higher values retain more high frequencies, but the conversion takes longer. Lower values require less processing time but reduce high frequencies.
Use higher Quality values whenever you downsample a high rate to a low rate. When upsampling, higher values have little effect.
This puts it at odds with your post in the above link:
BTW, there is a lesson here. The Quality slider on software resamplers such as CoolEdit Pro/Audition is a tuning knob for transition band width. Higher quality means a narrower transition band.
If you push it to the far right (99%) the resulting transition band became exceedingly narrow in my tests - just a few Hz, which puts it in the same realm as the signal that Meridian used for listener training.
If you push it to the far left the transition band is more like 1.6 KHz and better approximates a real world converter.
All of the sample rate ABX test files I've circulated lately were made with quality set for the maximum. They are therefore not representative of real world circumstances. One word: Invalid.
But what do I know.
... But what do I know.
That question appears to have been answered quite comprehensively, at least in the subjects covered by this thread.
But what do I know.
Hopefully that unsupervised gameable online computer tests (including those by ex-MS execs) are not the subject of the thread and don't prove anything related to 16/44 transparency.
Speaking of which, Amir how do you explain this:
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
Amir, which Meridian/BS statement is false, the one above from their website about "achieving 16/44 transparency" , or the new BS test just released claiming (apparent) 16/44 non-transparency (as we both know due to RPDF dither doctoring or possible setup artifacts)?
They can't both be right.
cheers,
AJ
It is not our responsibility to acquaint you with our community. We have a search function. No one is prohibiting you from using it prior to posting.
Ah, I didn't realize that. Thanks for the suggestion. So I searched for noise modulation audibility and landed on this thread: http://www.hydrogenaud.io/forums/index.php?showtopic=16963 (http://www.hydrogenaud.io/forums/index.php?showtopic=16963). Yet another set of excellent posts by David. You guys have a gift there. He has super hands on experience with signal processing. I hope he doesn't mind me reaching out to him as I often run into start-ups I am advising that could benefit from his expertise and my Microsoft resources are too busy to help out.
Anyway, here is the key post by another member I sadly don't know but seems to also be knowledgeable:
If I'm not wrong (2Bdecided probably knows better than me), mathematically, with rectangular dither, 1 bit amplitude is enough to totally remove distortion, but noise modulation remains. With triangular dither, 1 bit is enough to totally remove both, distortion and noise modulation. With noiseshaping dither, the amount depends on the shaping of the noise floor, but is usually below 1 bit. I don't know of a way of calculating this, but maybe the mentioned paper does. At the tests I did on the mentioned thread I used subjective listening tests for the noise modulation issue, using a critical test signal, amplified several tenths of dBs to make dither failure audible, and also frequency analysis (FFT) to detect distortion.
So when you said most everyone here agrees noise modulation is audible, you meant it under these conditions? FFT analysis, and amplifying the signal by several tens of dBs? Is it your impression that Stuart's test include such overamplification and FFT analysis was performed by expert listeners who understand what to look for?
AJ, is this the audible problem you think should have been avoided in the test?
So when you said most everyone here agrees noise modulation is audible
Since I'm the only other person you quoted, I'd like to see where I said this.
AJ, is this the audible problem you think should have been avoided in the test?
I think you should quit evading/dancing around the doctored up dither BS test dilemma and obvious conflict with BS statements
direct from Meridian above. Both can't be right.
Plus the whole artifacts issue with unconfirmed system switching software/speaker transparency, just like BS and JJ warn about.
But I know you. ..and now so does HA.
cheers,
AJ
So when you said most everyone here agrees noise modulation is audible
Since I'm the only other person you quoted, I'd like to see where I said this.
I was going to say you could do your own search but thought it would not be in good form . So here it is: http://www.hydrogenaud.io/forums/index.php...mp;#entry880450 (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=75&p=880450&#entry880450)
all this paper seems to show is that if you don't dither at all, or dither improperly, a group of carefully trained listeners will, on some sample material, be able to distinguish the 16-bit content from the original content
Right, and most of us were already aware this was possible.[/color]
Still stand by that position given how it is at odds with the post I linked to on this forum?
About the audibility of improper dither, yes.
About all that the paper seems to show, no; though this has been made clear already.
Perhaps you should read it more carefully noting the qualifications on the statement (or lack thereof).
About the audibility of improper dither, yes.
Yes what? That it takes tens of dbs of level amplification and FFT examination to tell?
http://www.hydrogenaud.io/forums/index.php...st&p=881162 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881162)
(first part only)
Goodbye for now, Amir.
About all that the paper seems to show, no; though this has been made clear already.
Nothing remotely has been accomplished there let alone be clear.
(http://amirviews.smugmug.com/photos/i-VwtbpfR/0/O/i-VwtbpfR.png)
Please explain how that green line is jumping up to exceed the threshold of hearing.
Please explain likelihood of peer review not realizing what you have made clear in this thread.
Please explain how these the manifestations of the problem you see as reported in the paper by the subjects taking a blind test:
It was reported that filtering gave "softer edges" to
the instruments, and "softer leading edges" to musi-
cal features with abrupt onsets or changes. Echoes,
when audible, were identi ed as being affected the
most clearly by the ltering. It was felt that some
of the louder passages of the recording were less ag-
gressive after filtering, and that the inner voices (sec-
ond violin and viola) had "a nasal quality". Over-
all, the filtered recording gave a "smaller and
flatter auditory image", and speci cally the physical space
around the quartet seemed smaller.
Listeners described that quantization gave a "rough-
ness" or "edginess" to the tone of the instruments,
and that quantization had a significant impact on
decay, particularly after homophonic chords, where
"decay was sustained louder for longer and then died
suddenly". This could be an effect of quantization
distortion; it is interesting that this was audible even
in a 24-bit system, and is consistent with the hy-
potheses of Stuart [29] that 16 bits are not sucient
for inaudible quantization.
Perhaps you should read it more carefully noting the qualifications on the statement (or lack thereof).
Perhaps. But you need to be specific and quote the section of the paper. And indicate that you are in possession of the paper and understand its content.
http://www.hydrogenaud.io/forums/index.php...st&p=881162 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881162)
(first part only)
Goodbye for now, Amir.
Same to you. Good night.
FFS, we each don't communicate in a vacuum. You've already been schooled by others on the matter.
Any test scores showing rectangular dither didn't affect the results?
To summarize: the recent Meridian tests sabotaged 16/44 and 16/48 processing by intentionally using
(1) suboptimal dither PDF,
(2) suboptimal dither PSD,
(3) and suboptimal low pass filter transition band width.
All fully documented in this thread.
I agree that all of these things are sub-optimal, but I think we* should at least have the humility to accept that we* wouldn't normally expect any of them to be audible.
(we* = people who read+post at HA a lot.)
For each separate issue, it is of course possible that the exact processing parameters and the exact playback parameters have combined to bring these issues up to the threshold of audibility. I'm not yet clear whether we have sufficient information to make this judgement, but I think that's the judgement we should attempt to make.
I'm especially interested to see you criticising the narrow filter transition band Arny. If the width of the filter transition band matters
when that filter is acting at 24kHz, then that raises all kinds of interesting questions. I know why dither matters if it's at an audible level: you can hear it. I don't know why filter ringing matters when it's at an ultrasonic frequency. I know various possible mechanisms by which it might become audible, but that's different from saying that it is.
Cheers,
David.
We talk about filters and PCM here.
Keep in mind the modern audiophile has no doubt DSD sounds better all the time as PCM because its differernt filter behaviour.
Interestingly enough, the AES Conference paper that centerpieces this thread pretty well trashes DSD:
"
Another potential problem with the setup used by
Meyer and Moran [4] concerns the source material.
SACD uses 1-bit DSD (Direct-Stream Digital) en-
coding, which supports a wide dynamic range for
frequencies below about 20 kHz by \noise shaping",
that is, pushing the noise in the signal to frequencies
higher than 20 kHz. As a result, the signal-to-noise
ratio for frequencies higher than 20 kHz decreases
quickly. Therefore SACD is not a suitable candi-
date for the kind of listening test intended to estab-
lish any contribution to perception from frequencies
higher than those encoded by a conventional CD. In
addition, it is customarily recommended that SACD
players include a high-order low-pass lter in the re-
gion of 30-50 kHz to reduce the high-frequency noise
introduced to ampli fiers and tweeters; such fil lters
may have audible impact.
"
Talk about falling on your sword!
Of course Meridian profits from so-called high resolution PCM with their data compression patents (MLP), not DSD.
So it is not their sword that they are falling on, it is Sony et al that takes the fall.
To summarize: the recent Meridian tests sabotaged 16/44 and 16/48 processing by intentionally using
(1) suboptimal dither PDF,
(2) suboptimal dither PSD,
(3) and suboptimal low pass filter transition band width.
All fully documented in this thread.
I agree that all of these things are sub-optimal, but I think we* should at least have the humility to accept that we* wouldn't normally expect any of them to be audible.
(we* = people who read+post at HA a lot.)
For each separate issue, it is of course possible that the exact processing parameters and the exact playback parameters have combined to bring these issues up to the threshold of audibility. I'm not yet clear whether we have sufficient information to make this judgement, but I think that's the judgement we should attempt to make.
I'm especially interested to see you criticizing the narrow filter transition band Arny.
Don't get me wrong. I don't know what to think. I'm just reading "Peer reviewed" AES papers...
There are many assertions in the recent Meridian paper that IMO need far better support. This is just one of them.
The main point is that if Meridian critiizes flat PSD dither and non-TPDF dither when they are selling digital processors, why are they ramming them down our throats when they are testing other people's products?
If they use files created by low pass filters with too-narrow transition bands to train their listeners to hear filtering artifacts, why are they ramming them down our throats when they are testing other people's products?
Looks like meat for some ABXing.
If the width of the filter transition band matters when that filter is acting at 24kHz, then that raises all kinds of interesting questions. I know why dither matters if it's at an audible level: you can hear it. I don't know why filter ringing matters when it's at an ultrasonic frequency. I know various possible mechanisms by which it might become audible, but that's different from saying that it is.
That all goes through my mind every time I try to think about their apparently unsupported claims about transition bands.
But it must be true - the AES review board signed off on it and gave the paper their highest reward, right? ;-)
Here's what is known about the recording
"The recordings used were selections from The signals
used were extracts from Haydn's String
Quartet Op.76 No.5 in D \Finale, Presto" from
\Nordic Sound (2L Sampler)" 1, issued with a sample
rate of 192 kHz using 24-bit PCM23.
This recording can be downloaded in a number of different formats from here:
https://shop.klicktrack.com/2l/35847 (https://shop.klicktrack.com/2l/35847)
Price in 24/192 format: $6.00
No one has mentioned something really funny.
amirm wants the original bits.
This is reportedly a 352.8kHz/24bit/5.1 channel recording.
Check out the available formats. You can't buy the original 352.8kHz/24bit/5.1. The "best" available options are downgraded to stereo, "downgraded" to 96kHz, and "downgraded" to DSD. You can't buy the original bits. At most, you get something with 1/3rd the number of bits, or less than half the number of channels.
The BluRay (which is cheaper) includes 192kHz/24bit/5.1.
Is anyone going to post a 30 second sample?
Cheers,
David.
Now I don't think that the choice of dither usually matters, but in critical evaluations of the limitations of a format I would at least expect triangular dither.
If it usually doesn't matter then it is not an "obvious flaw."
Let me rephrase so even you can understand it:
In an evaluation of the limitations of a format, which normal listening is not, you shouldn't go for anything subpar, be it dither, filters, playback hard/software ...
Now I don't know if dither was a problem. But it is reasonable to have doubts when these people have written papers on dither and noise shaping and how it can be audible with 16 bits!
Yes, another win for science, and a fail for laymen trying to interpret the same. The test is run at 192 Khz Arny.
And since you have the algorithms at hand, you know exactly how they did or didn't account for noise bandwidth in their 44.1/16 simulation within a 192 kHz format, right? Right?
No..
Also, I've already explained it, but do you know the difference between spectral density and an amplitude/linear spectrum?
You can plot the spectral density of rectangular dither with 192/16 at -144 dB (which is what they did in that fig. 3), but the SNR is still only ~95 dB.
So if you plot the dither noise in fig. 3 in terms of human-audible significance (approx. 1/3rd octave wide noise-gathering bandpass filters above 1 kHz), you will see that around ~4 kHz we are down roughly -100 dB relative to the mean acoustic gain, so above even the hearing threshold for single tones.
And that doesn't account for additional attention-drawing due to noise modulation.
Please explain how these the manifestations of the problem you see as reported in the paper by the subjects taking a blind test:
[...]
Listeners described that quantization gave a "rough-
ness" or "edginess" to the tone of the instruments,
and that quantization had a significant impact on
decay, particularly after homophonic chords, where
"decay was sustained louder for longer and then died
suddenly". This could be an effect of quantization
distortion; it is interesting that this was audible even
in a 24-bit system, and is consistent with the hy-
potheses of Stuart [29] that 16 bits are not suffcient
for inaudible quantization.
First of all, the above is subjective and anecdotal. But anyway, some of it doesn't sound too far fetched. I guess you do not know what noise modulation even is.
Secondly, Stuart has written papers that show that 52 kHz / 11 bits are the absolute minimum PCM channel
using noise shaping, capable of replicating the information received by the ear. 14 bits "ought to be adequate" to offer enough headroom. That would be 18.2 bits for a rectangular channel (TPDF, no noise shaping).
If proper shaping can make a few bits difference, then why do you dismiss the idea that RPDF could be inadequate in such a test given the 16-bit quantization?
Here's what is known about the recording
"The recordings used were selections from The signals
used were extracts from Haydn's String
Quartet Op.76 No.5 in D \Finale, Presto" from
\Nordic Sound (2L Sampler)" 1, issued with a sample
rate of 192 kHz using 24-bit PCM23.
This recording can be downloaded in a number of different formats from here:
https://shop.klicktrack.com/2l/35847 (https://shop.klicktrack.com/2l/35847)
Price in 24/192 format: $6.00
No one has mentioned something really funny.
amirm wants the original bits.
This is reportedly a 352.8kHz/24bit/5.1 channel recording.
Check out the available formats. You can't buy the original 352.8kHz/24bit/5.1. The "best" available options are downgraded to stereo, "downgraded" to 96kHz, and "downgraded" to DSD. You can't buy the original bits. At most, you get something with 1/3rd the number of bits, or less than half the number of channels.
The BluRay (which is cheaper) includes 192kHz/24bit/5.1.
Is anyone going to post a 30 second sample?
.
Is http://www.hydrogenaud.io/forums/index.php...st&p=881260 (http://www.hydrogenaud.io/forums/index.php?showtopic=107527&view=findpost&p=881260) a good place?
For months on AVS Forum we have been discussing the audibility of high resolution audio. After passing ABX DBT after ABX DBT, the claim always went back, "well, Meyer and Moran didn't find this difference and it was peer reviewed so this data must be wrong."
I guess you didn't get the memo Amir (silliy I should expect you to comprehend it, given that it is in a recent post to this thread), but the ABX tests I posted on AVS had a built in flaw that I only very recently became aware of which is that the transition band of my downsampling filter was way too narrow per this certain peer reviewed paper from Meridian.
It was covered in the second part of this post: http://www.hydrogenaud.io/forums/index.php...st&p=881145 (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&view=findpost&p=881145)
We know that the other set of files had other serious flaws of their own.
It thus becomes improper for anybody to say: "After passing ABX DBT after ABX DBT" because there were no proper tests.
Arny's files never had sync problems. But yes, I did resample them using latest version of Audition CC and I could still tell them apart.
The claim above is false - there were never any audible synch problems.
The problem in question had nothing to do with synch.
It had to do with transition bands Were the alleged proper 4416 file available for download I would review them and grant an opinion.
Good morning there. I must say, that was the shortest goodbye I have seen .
FFS, we each don't communicate in a vacuum. You've already been schooled by others on the matter.
You say that as if it is supposed to be something negative. It better not be. I live for days I learn something from these forum interactions as opposed to dealing with male insecurities to get personal.
You have some very knowledgeable members here who for sure could school me. Alas, as it happens, they are the ones without a copy of the paper so can’t speak properly to what the test is all about. As it turns out the people with a copy of the paper don't have expertise in this field so we are honestly reading gibberish from them.
Not saying the above to put folks down. It is the reality of the work in front of us. It is not a school project like Meyer and Moran. Even though the paper describes a listening test, it is extremely dense with jargon of audio signal processing, statistics, and analysis. It is way beyond the knowledge level of a typical forum member who doesn't do this work professionally. I hope to remedy this situation below.
Any test scores showing rectangular dither didn't affect the results?
If you are asking this question, it means that there is still no understanding of what the paper was about some 12 pages and 300 posts later. So please allow me to give you the summary as to facilitate proper schooling, whichever direction it will go.
Let's start at the top. This is the title of the paper:The audibility of typical
digital audio filters in a high-fidelity playback system
No, I am not being pedantic . Please pay attention to what I have highlighted: digital audio filters. That is the mission of the paper. They start with 192 Khz music samples, then apply two filters to it. One is to represent CD sampling rate of 44.1 Khz and the other, 48 Khz. Here is the paper itself in technicolor to make sure these key points are not lost:
2.3. Signal processing and test conditionsTwo kinds of linear-phase FIR (fnite impulse re-
sponse) filter were used, both of which operated at
192 kHz and
both of which were implemented us-
ing TPDF (triangular probability density function)dither at the 24th bit. For both FIR filters,
the
ripple depth over the passband was a maximum of
0.025 dB, and the stopband attenuation was 90 dB.The frequencies of the transition bands were 23500-
24000 Hz and 21591-22050 Hz, corresponding to the
standard sample rates of 48 kHz and 44.1 kHz re-
spectively. Fig. 2 shows the amplitude and energy
of the impulse response for the 48-kHz filter.
I have highlighted two key things:
1. The "ideal" TPDF filtering was used contrary to the impression left so far in this thread but I will get to where it is not used and source of confusion in a bit.
2. The filtering is the kind of textbook response that we always say is as good as “perfect” and hence inaudible: a fraction of irregularity over the audio band and strong rejection of out of band spectra.
This is what the test sets out to find out. Remember the title again: it is all about digital audio *filtering*. Key thing to note is that the above transformations are in 24 bit as is the source. Further, the sampling rate in all cases remains at 192 Khz. We are simply looking at what would happen if the bits were filtered down to lower bandwidth that 44.1 and 48 Khz would entail while keeping everything else the same.
Our camp’s position is that no double blind test would ever demonstrate anything an audible difference to statistical confidence. As otherwise it would say the mere conversion of higher bandwidth to CD’s (and that of 48 Khz) is a lossy audible conversion. Let’s see how it turned out.
The paper explains six (6) ways this filtering was tested:
(http://amirviews.smugmug.com/photos/i-cfWrnnX/0/O/i-cfWrnnX.png)
Translating into English, listing tests 1 and 4 (“none”) examine exactly what I described above. Filtering of a 192 Khz/24-bit file down to what it would have been at 44.1 Khz and 48 Khz sampling while remaining in 24-bit mode. Again please allow me to remind you that in both cases TPDF dither is used, not rectangular. So the answer you ask in this regard is “no.” RPDF was not used so it did not interfere with this part of the test.
The other test conditions deal with what would happen if we converted the filtered 24 bit output to 16 bit samples using two different methods. One by simply truncating it or the so called “self-dither.” The extra resolution bits are simply discarded. This is tests #2 and #5.
The other method and what folks have been fixated on, namely conversion of 24 bit to 16 rectangular PDF dither. The theory says this is superior to doing nothing. Results here are surprising as with the filtering.
I hope everyone is with me so far. That we have six (6) independent tests, not just one like Meyer and Moran performed. This allows us to investigate the effect of each processing separately and together.
The ResultsHere they are:
(http://amirviews.smugmug.com/photos/i-FKcswgv/0/M/i-FKcswgv-M.png)
Remember, we had three tests for each filtering: no dither (24 bit), truncation to 16 bits, and dithered to 16 bits. And that is what is represented in the above graph.
The horizontal line at 56% shows the 95% level high confidence bar. Please don’t be swayed by the small numerical value of “56%.” The statistics are not for lay people. Just remember that this is the same standard we require of all such tests. If we achieve better than 95% results, we trust the outcome.
We see that five (5) out of six (6) listening tests comfortably cleared the 95% interval both in their mean and standard deviation. Oddly, the 24-bit, 48 Khz sampling did not do so fully when you account for the error range. A further look at the results focusing on critical/more revealing music segments dealt with that by bringing that score above 95%.
So what do we learn?
All processing was distinguished from the original to statistics confidence! Filtering with or without conversion to 16 was audibly different to listeners (to statistical confidence). None of this was expected by our camp’s vocal members. And certainly counter to Meyer and Moran’s test indicating these are all transparent conversions.
The rest of the paper theorizes as to why these differences were heard. I won’t get into that now. But I hope it is clear that this test easily schools all of us. That our knowledge of psychoacoustics with respect to digital audio filters/quantization is not complete. All of these tests should have failed to invalidate the null hypothesis but they did not. I am confident this is one of the reasons that peer reviewers were impressed with this work. It is groundbreaking in documenting audible differences this way.
Circling back to your question, if you like to throw out the results of 16 bit conversion with or without dither, do. But then explain why simple filtering of spectrum to 22.05 and 24 Khz resulted in audible differences despite near perfect frequency domain response of this type of anti-aliasing filter.
Great, and we had the conditions already posted before a few pages back. The focus on the dither was indeed derailing a bit.
What we really need are the original algorithms and software used to do the processing and switching. And someone into statistics really needs to explain what is going on.
I have a question about the test methodology. The "same or different" choice is subject to statistical bias unless care is taken since human guessing is biased. (Humans tend to pick "c" on multiple choice questions and "different" on "same or different" questions.) Were the number of "same" cases equal to the number "different" cases in each test case? If not, the results might be biased. Can you tell from the paper itself?
Good morning there. I must say, that was the shortest goodbye I have seen .
FFS, we each don't communicate in a vacuum. You've already been schooled by others on the matter.
You say that as if it is supposed to be something negative. It better not be. I live for days I learn something from these forum interactions as opposed to dealing with male insecurities to get personal.
You have some very knowledgeable members here who for sure could school me. Alas, as it happens, they are the ones without a copy of the paper so can’t speak properly to what the test is all about. As it turns out the people with a copy of the paper don't have expertise in this field so we are honestly reading gibberish from them.
Not saying the above to put folks down. It is the reality of the work in front of us. It is not a school project like Meyer and Moran. Even though the paper describes a listening test, it is extremely dense with jargon of audio signal processing, statistics, and analysis. It is way beyond the knowledge level of a typical forum member who doesn't do this work professionally. I hope to remedy this situation below.
Any test scores showing rectangular dither didn't affect the results?
If you are asking this question, it means that there is still no understanding of what the paper was about some 12 pages and 300 posts later. So please allow me to give you the summary as to facilitate proper schooling, whichever direction it will go.
Let's start at the top. This is the title of the paper:The audibility of typical digital audio filters in a high-fidelity playback system
No, I am not being pedantic . Please pay attention to what I have highlighted: digital audio filters. That is the mission of the paper. They start with 192 Khz music samples, then apply two filters to it. One is to represent CD sampling rate of 44.1 Khz and the other, 48 Khz. Here is the paper itself in technicolor to make sure these key points are not lost:
2.3. Signal processing and test conditions
Two kinds of linear-phase FIR (fnite impulse re-
sponse) filter were used, both of which operated at
192 kHz and both of which were implemented us-
ing TPDF (triangular probability density function)
dither at the 24th bit. For both FIR filters, the
ripple depth over the passband was a maximum of
0.025 dB, and the stopband attenuation was 90 dB.
The frequencies of the transition bands were 23500-
24000 Hz and 21591-22050 Hz, corresponding to the
standard sample rates of 48 kHz and 44.1 kHz re-
spectively. Fig. 2 shows the amplitude and energy
of the impulse response for the 48-kHz filter.
I have highlighted two key things:
1. The "ideal" TPDF filtering was used contrary to the impression left so far in this thread but I will get to where it is not used and source of confusion in a bit.
2. The filtering is the kind of textbook response that we always say is as good as “perfect” and hence inaudible: a fraction of irregularity over the audio band and strong rejection of out of band spectra.
The paper goes on to say:
"After filtering with either FIR lter, the signals
were either unchanged or were quantized to 16-
bit. The quantization either included RPDF (rect-
angular probability density function) dither or did
not.
So while the low pass filtering used TPDF dither applied to the 24th bit which due its tiny size is far less important, the requantization to 16 bits which was highly important and in fact one of the two main points that this test was all about used RPDF dither.
So, what we have in the comments above that I am responding to, is yet another failure of reading comprehension.
But I hope it is clear that this test easily schools all of us.
Yes, it teaches how far those with pecuniary interest will spin rigged tests that appear to show some desired results, but with no verification of system transparency/artifacts (that BS and JJ warn about).
But then it creates this dilemma for your side Amir:
cheers,
AJ
You missed the target a little bit, Arny. You too, AJ.
Taken at face-value, the data clearly shows that 16-bit dither did not matter for the 22 kHz-treated version. It doesn't prove the choice of dither couldn't have affected the results for the 24kHz-treated version, however.
It also doesn't address the other issues raised about the hardware and unknown(?) filter algorithms.
...and now we have at least one sample that appears suspect, as well as statistical chop suey to wade through.
Thank you, Amir, for finally presenting something based on the actual test data.
I wonder about one simple thing. Please don't shoot me because i am just one of these hobby noobs around here.
If that tweeter for playback really has such a strong resonance shouldn't music with content at that resonance frequency trigger some unwanted behaviour that even sounds different as with music without content there?
As a sidenote, i really love magnetostats.
Thank you, Amir, for finally presenting something based on the actual test data.
I could post more graphics from the paper if there was some easy cheap way to host them. Some forums host uploaded graphics. Does this one?
Yes. Just dump them in the thread you created in the uploads forum and link to them here.
I wonder about one simple thing. Please don't shoot me because i am just one of these hobby noobs around here.
Modesty speaks . You are more than qualified to discuss this topic.
If that tweeter for playback really has such a strong resonance shouldn't music with content at that resonance frequency trigger some unwanted behaviour that even sounds different as with music without content there?
As a sidenote, i really love magnetostats.
The system measurement in the paper does not show that resonance from stereophile review. It has rather smooth response extending to 50 Khz (vertical scale is too squished to read accurately but this is more or less the picture). There is certainly no 12 db peak at 26 Khz.
It also doesn't address the other issues raised about the hardware and unknown(?) filter algorithms.
There is good bit of discussion of the filters in use including both frequency and impulse responses. They can definitely be duplicated although the person has to be considerably less lazy than me to do so .
Thank you, Amir, for finally presenting something based on the actual test data.
My pleasure.
The paper goes on to say:
"After filtering with either FIR lter, the signals
were either unchanged or were quantized to 16-
bit. The quantization either included RPDF (rect-
angular probability density function) dither or did
not.
So while the low pass filtering used TPDF dither applied to the 24th bit which due its tiny size is far less important, the requantization to 16 bits which was highly important and in fact one of the two main points that this test was all about used RPDF dither.
So, what we have in the comments above that I am responding to, is yet another failure of reading comprehension.
I will take another shot at summarizing the paper but if that doesn't work, someone else has to explain it to you Arny.
Almost every bit of recording done today is in 24 bit and at sampling rates > CD. In conversion to CD, two steps need to be performed.
1) Reducing the bandwidth down to 22.05 Khz (44.1 Khz sampling). This is a low pass filter and what is tested in the report.
2) Convert the bits from 24 to 16.
The report shows with high confidence that the transformation in #1 is audible. If so, it doesn't matter what #2 is. The sum of #1 and #2 by definition would be audible since the first step is audible. To make a case that CD is transparent, you need to demonstrate why #1 is wrong. Since that filtering used TPDF, then you have no argument with respect to dither.
On #2, they found audible differences between all three methods they tested: no dither, truncation and rectangular dither. And paradoxically using rectangular dither that on paper does much better than truncation, did the opposite, scoring lower. So based on this report, there is little reason to think that using TPDF dither would have improved the situation.
Edit: fixed typos.
Again, feel free to discard #2 and answer how #1 is consistent with your views and that of prior tests such as Meyer and Moran.
I should have read more carefully. ajinfla already noticed the resonance in the dome may influence its sound.
Since the dome itself because of its hard material still has this resonance but a linear frequency response there is very likely some notch filter working inside the active speaker.
I still can imagine there is interaction of music signals coming in at the notch filters frequency or not.
Here's the spectral plot of the constant subsonic rumble in the third excerpt Arny has posted from the last second where the main music has faded away. it shows here as being 50 dB down, or so, but compared to the actual music content of this short sample it is only 30 dB down from the average peak levels, by my eye, hence rather loud.
[Let's see if this uploading of attachments from that sub forum section works for me. Here goes.]
Arny, whatever mic made these recordings has an impressive range from subsonic to ultrasonic! I'm not as experienced as you, so tell me, is this fairly loud LF rumble which seems to peak at a frequency even lower than 16 Hz (the lowest our spectrum analyzer shows) just an artifact or rather a real, I believe you call it "room tone"?
Note, although it is present in the entire passage of your third posted segment, I show it here in isolation where the music has faded away at the very last second or so. It's actually only 30 dB down from the musical peaks of the passage itself, or unusually loud IMHO.
Haydn_string_quartet_in_D__Op76_No5_finale_presto_Arny_segment_final_second.jpg (http://www.hydrogenaud.io/forums/index.php?act=attach&type=post&id=8069)
I'm no mic expert but does the <16hz to >33kHz capability narrow it down to only a handful? [I see Arny has answered that he suspects HVAC in the attachment thread before I could even figure out how to post this!]
I'm no mic expert but does the <16hz to >33kHz capability narrow it down to only a handful?
The company 2L uses various models of DPA microphones (http://www.dpamicrophones.com/en/products.aspx?c=Catalog&category=234). Freq.response of small diaphragm condenser (sdc) microphones is in general from <20Hz to >30 kHz. Other brands have similar specs. DPA is quite popular amongst classical engineers though. (I've been using them for over 30 years)
Arny's files never had sync problems. But yes, I did resample them using latest version of Audition CC and I could still tell them apart.
The claim above is false - there were never any audible synch problems.
Hi Arny. I say your files never had sync problems and you repeat the same but add that my claim is false? How is that possible?
It had to do with transition bands Were the alleged proper 4416 file available for download I would review them and grant an opinion.
And I post results from my version of Audition CC which is a few generations more recent than yours and representative of what is in Stuart's paper, with the filter at 50% setting. I still managed to tell the files apart with the same ease as your version. So your theory that you created the files wrong is well, wrong itself .
But I can live with you saying that moving a slider in resampler is enough to create reliable audible differences. So don't stop on my account .
But I hope it is clear that this test easily schools all of us.
Yes, it teaches how far those with pecuniary interest will spin rigged tests that appear to show some desired results, but with no verification of system
AJ, once more the paper has won peer-review award. It would have to be rigged pretty good to fool the minimum of two people who reviewed the work, yes?
And what do mean by verification? The system frequency and impulse response is documented in the paper. Have you read the paper or keep posting these things devoid of said knowledge?
But then it creates this dilemma for your side Amir:
Your question has been answered. See the summary of the test I post. Every step of conversion to 16/44.1 was found to be audible to high confidence. Yes, they didn't try TPDF in the quantization step. You can try to prove how adding that to filtering that had audible effect would keep things transparent. You can do that AJ?
I should have read more carefully. ajinfla already noticed the resonance in the dome may influence its sound.
Since the dome itself because of its hard material still has this resonance but a linear frequency response there is very likely some notch filter working inside the active speaker.
I still can imagine there is interaction of music signals coming in at the notch filters frequency or not.
Hi Wombat,
As it turns out those peddling the BS paper don't have expertise in this field so we are honestly reading gibberish from them.
The possibility of speaker generated artifacts is way beyond the knowledge level of a typical forum member who doesn't do this work professionally.
You can't look at the terribly scaled on axis FR data to determine this as I will teach to hobbyists in my further response.
(http://i216.photobucket.com/albums/cc73/AJinFLA/DSP7200SEFR.jpg)
cheers,
AJ
AJ, once more the paper has won peer-review award. It would have to be rigged pretty good to fool the minimum of two people who reviewed the work, yes?
The "award" is meaningless to the validity of the paper and no, it simply depends on who reviewed and awarded, which you have no clue.
Having read it and knowing your interests, unsurprisingly, you are greatly spinning the results.
And what do mean by verification? The system frequency and impulse response is documented in the paper. Have you read the paper or keep posting these things devoid of said knowledge?
Before you were assuming I hadn't, now you don't seem so sure.
I'm not surprised when hobbyists view things, they are different from those who work professionally. The FR and impulse are the same thing, just a different view and shed zero information regarding the concerns of artifacts at the extreme levels (near max) they drove the system. That "squished" FR is on axis only and practically useless, other than to show the response does indeed extend out to 40KHz. Not shown is the off axis, which may reveal the resonance around 30k:
(https://www.madisoundspeakerstore.com/images/T29B001-curve.jpg)
Actually, I strongly suspect these (http://www.seas.no/index.php?option=com_content&view=article&id=495:e0058-06-t29b001&catid=50&Itemid=360)
are the DSP7200 tweeters. What is needed is some distortion data for that tweeter to see if IM products are present downstream
in the audible range when it is driven without band limiting, at extreme levels for a DR dome.
Absolutely nothing in this regard is presented in the paper. But hobbyists or those in over their heads can't be cognizant of these things.
But then it creates this dilemma for your side Amir:
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
Is that blatantly false info being perpetrated by BS/Meridian and what recourse do owners have about such false info it the BS Test shows otherwise?
No, it is correct information consistent with Stuart's previous journal paper work. You would do well to read more of that manual as it will teach you a ton about signal processing. It makes your AVR manual look like chicken scratchings on toilet paper: https://www.meridian-audio.com/download/Han...ies/518user.pdf (https://www.meridian-audio.com/download/Handbooks/500_Series/518user.pdf)
As expected, complete obfuscation.
Yes, they didn't try TPDF in the quantization step. You can try to prove how adding that to filtering that had audible effect would keep things transparent. You can do that AJ?
You still have no clue, I don't have to prove a damn thing.
Now having read the BS paper, I am more convinced it's bollocks. I've seen JJ skewer folks over the "smearing" nonsense they keep referencing and the fact that they "emulated" filtering rather than use an actual converter, have not a single thing showing system transparency, switching, etc.
You're lucky he won't come on here anymore to
publicly comment on this paper. Whose conclusions are vastly different from yours.
cheers,
AJ
AJ, once more the paper has won peer-review award. It would have to be rigged pretty good to fool the minimum of two people who reviewed the work, yes?
The "award" is meaningless to the validity of the paper and no, it simply depends on who reviewed and awarded, which you have no clue.
Having read it and knowing your interests, unsurprisingly, you are greatly spinning the results.
It is the old bear story AJ. How fast do you need to run from the bear? Faster than the guy with you.
By the same token, yes, peer review does not in at all guarantee correctness of the paper. But it sure as heck blows away your theory of a "rigged test. " You are making a hit and run empty accusation. The peer review easily trumps anonymous hobbyists on a forum saying otherwise.
If these DSP7200 were used, the tweeter indeed reacts a lot on axis response with its hefty resonance.
page 16:
http://www.meridian-audio.info/public/meri...1%5B3099%5D.pdf (http://www.meridian-audio.info/public/meridian1%5B3099%5D.pdf)
Wombat that's the DSP7200, non SE (special edition) version, with a different tweeter made of aluminum, among other things, although there may (or may not) be similarities.
I'm not surprised when hobbyists view things, they are different from those who work professionally. The FR and impulse are the same thing, just a different view and shed zero information regarding the concerns of artifacts at the extreme levels (near max) they drove the system.
You are wrong in the specific here but let's put that aside. You say that I don't know this fact. Well, read this article I wrote on WBF Forum with respect to relationship of Time vs Frequency in acoustic measurements: http://www.whatsbestforum.com/showthread.p...e-and-Frequency (http://www.whatsbestforum.com/showthread.php?8203-Acoustic-Measurements-Understanding-Time-and-Frequency). Do you have a write-up like that so that we can see you know the topic and not just repeating what you have read in Dr. Toole's research?
On page #2 I say this: http://www.whatsbestforum.com/showthread.p...ll=1#post141558 (http://www.whatsbestforum.com/showthread.php?8203-Acoustic-Measurements-Understanding-Time-and-Frequency&p=141558&viewfull=1#post141558)
"In layman terms, this means that the amplitude fully describe system behavior and hence, modification of it must by definition also modify its time response since the two define the same signal. They are two sides of the same coin."So you have shared nothing new despite boasting about professional experience. Since when a hobbyist putting speakers together become a professional anyway?
That "squished" FR is on axis only and practically useless, other than to show the response does indeed extend out to 40KHz.
Which in this case is something to be verified! Remember we don't have anything remotely like that from Meyer and Moran. No measurements of any sorts were provided.
Not shown is the off axis, which may reveal the resonance around 30k:
(https://www.madisoundspeakerstore.com/images/T29B001-curve.jpg)
Actually, I strongly suspect these (http://www.seas.no/index.php?option=com_content&view=article&id=495:e0058-06-t29b001&catid=50&Itemid=360) are the DSP7200 tweeters. What is needed is some distortion data for that tweeter to see if IM products are present downstream in the audible range when it is driven without band limiting, at extreme levels for a DR dome. Absolutely nothing in this regard is presented in the paper. But hobbyists or those in over their heads can't be cognizant of these things.
In other words, this is pure speculation. You don't even know if this is the tweeter they used. You don't know if it is that it generates IM distortion in-band.
The paper includes specific references to intermodulation:
An experiment conducted by Ashihara and Kiryu
[27] found that at least some of the work demonstrat-
ing auditory sensitivity to the presence of ultrasonics
could have su ered from intermodulation distortion
introduced by the loudspeakers used to present the
signals. Intermodulation distortion caused by the in-
teraction of high-frequency components played from
the same loudspeaker could have provided a cue to
the presence of the ultrasonics.Your theory is that with full knowledge of issues surround intermodulation, they entered the test blind and fell victim to it anyway?
If these DSP7200 were used, the tweeter indeed reacts a lot on axis response with its hefty resonance.
page 16:
http://www.meridian-audio.info/public/meri...1%5B3099%5D.pdf (http://www.meridian-audio.info/public/meridian1%5B3099%5D.pdf)
The SE version uses a different tweeter. They say it is their own design: https://www.meridian-audio.com/en/campaigns...sary/upgrades/# (https://www.meridian-audio.com/en/campaigns/anniversary/upgrades/#)
Beryllium-dome Tweeter - for clearer highs
[NEW MERIDIAN-DESIGNED
BERYLLIUM-DOME TWEETER]
The stiffness of the tweeter cone helps determine how pure and uncoloured the sound is at high frequencies. Beryllium is many times stiffer than aluminium, the material usually used in tweeter cones. Being stiffer, the dome acts as a more perfect piston – delivering more accurate transients and lower colouration. In addition, its extended high frequency response is ideal for getting the best out of today’s high-resolution recordings.(https://www.meridian-audio.com/static/oscar/img/campaigns/anniversary/upgrades/beryllium-dome.jpg)
The stiffness of the tweeter cone helps determine how pure and uncoloured the sound is at high frequencies. Beryllium is many times stiffer than aluminium, the material usually used in tweeter cones. Being stiffer, the dome acts as a more perfect piston – delivering more accurate transients and lower colouration. In addition, its extended high frequency response is ideal for getting the best out of today’s high-resolution recordings.
This stiffness is the problem. Some years back all extreme hard domes had a resonance. Unfortunately i can't find a neutral measurement of these 7200SE
Imagine BS just released the most comprehensive paper ever published about audibility of IM in typical metal dome tweeters
The stiffness of the tweeter cone helps determine how pure and uncoloured the sound is at high frequencies. Beryllium is many times stiffer than aluminium, the material usually used in tweeter cones. Being stiffer, the dome acts as a more perfect piston – delivering more accurate transients and lower colouration. In addition, its extended high frequency response is ideal for getting the best out of today’s high-resolution recordings.
This stiffness is the problem. Some years back all extreme hard domes had a resonance. Unfortunately i can't find a neutral measurement of these 7200SE
I hear you. Ultimately the paper does document the performance of that speaker and it shows no on-axis resonance in semi-anechoic measurements.
AJ's hypothesis hinges on off-axis resonance. What's our theory of that surviving the room absorption by fair bit?
I hear you. Ultimately the paper does document the performance of that speaker and it shows no on-axis resonance in semi-anechoic measurements.
AJ's hypothesis hinges on off-axis resonance. What's our theory of that surviving the room absorption by fair bit?
I don't want to present speculation as fact, but
if there is a measurable resonance, that indicates the presence of a real mechanical property that would usually have wide-ranging implications. These could be somewhat (but not completely) ameliorated by making the on-axis response "flat" using DSP. (A gifted designer might be able to achieve a similar effect mechanically, damping the resonance sufficiently on axis, but leaving traces off axis.) The effect of the resonance may still be detectable off axis, and/or in certain time domain measurements, and/or in certain distortion measurements. The result of IMD (for example 24+26kHz), being at a much lower frequency, may have a different polar response, and in any case has no trouble bouncing off the walls.
I have no idea of the facts or reality of the situation, but we don't know nearly enough to discount this. It would be foolish to claim that this particular effect must be audible, but it would be almost as foolish to claim that it must not be.
Chance would give you 50% correct. The overall results were 56% correct. Whatever effect we're looking for here is a very small one.
Cheers,
David.
Chance would give you 50% correct. The overall results were 56% correct. Whatever effect we're looking for here is a very small one.
Just commenting on this part, my read of this number is that it was the threshold of 95% confidence interval, not the actual outcome. Here is the graphics again:
.
I hear you. Ultimately the paper does document the performance of that speaker and it shows no on-axis resonance in semi-anechoic measurements.
AJ's hypothesis hinges on off-axis resonance. What's our theory of that surviving the room absorption by fair bit?
I don't want to present speculation as fact, but if there is a measurable resonance, that indicates the presence of a real mechanical property that would usually have wide-ranging implications. These could be somewhat (but not completely) ameliorated by making the on-axis response "flat" using DSP. (A gifted designer might be able to achieve a similar effect mechanically, damping the resonance sufficiently on axis, but leaving traces off axis.) The effect of the resonance may still be detectable off axis, and/or in certain time domain measurements, and/or in certain distortion measurements. The result of IMD (for example 24+26kHz), being at a much lower frequency, may have a different polar response, and in any case has no trouble bouncing off the walls.
I have no idea of the facts or reality of the situation, but we don't know nearly enough to discount this. It would be foolish to claim that this particular effect must be audible, but it would be almost as foolish to claim that it must not be.
Absence of data is not data David. It is not logical to say that we don't know if you are or are not a bank thief and have it create the impression that you might be.
As I quoted, the authors are very well aware of previous test of this nature that was potentially invalidated due to ultrasonic intermodulation distortion. It would be really odd to note this and then go on to make the same mistake. Should this turn out to be a repeat of the same issue, it would reflect very poorly on the authors/work.
The real measurements are posted for on-axis and do not show this type of resonance or at least not an extreme one. So the prima facie evidence is that our guess of extreme resonance here is not supported.
On off-axis where no measurement is provided (although some of it is included in the semi-anechoic measurement), I ask again: how does that survive at nearly 30 Khz in room? Absorption of such short wavelengths will be quite high and very different than sticking a mic right in front of the driver in anechoic chamber.
I want to be clear: I am not defending the performance of Meridian speakers. I am just saying if this is all we can hang our hat on, it is weak argument given the data we have (or more correctly, don't have).
What do you think the 56% represents?
Please try not to sprain your wrists; they already look a little swollen.
Re: Regardless of whatever they didn't address, we must assume it was done correctly.
I see you're misplacing the burden of proof, yet again.
So much for being in the objective camp.
What do you think the 56% represents?
I explained that already.
I see you're misplacing the burden of proof, yet again.
Come again? We have a double blind study showing with high confidence there are differences. The burden of proof that it is not valid lies on shoulder's people who just want to throw guesses out there and see what sticks. And do so with extreme prejudice. I am calling them on their lack of supporting evidence.
So much for being in the objective camp.
Same to you mister. When will you start to act objectively? There is a report in front of us. You can't only focus on negative hypothesis, and say you are being objective.
I am not telling you this report is definitive. I am saying that we relied on hobby work before (Meyer and Moran) and now we have a much more proper test. The report has won an award backing my impression of it being good work. This is the objective view. The notion that the test must be wrong is the non objective view.The notion that Meyer and Moran was the definitive truth was the non objective view.
So please don't lecture me on what is objective. I know what it is and it has nothing to do with what you all do in these forums where emotions and prejudice rule ahead of knowledge. Just look at the first two pages of this thread. So there. We are even.
Edit: typos.
Eye-balling that plot suggests the overall arithmetic mean was about 62% correct then - is that how you interpret it?
That's still a small effect.
FWIW I can remember of a couple of ABX tests here on HA that have gone to that level (running over 100 tests to hit statistical significance), but I bet I'm not the only one who has ABXed differences so small I thought I might be imagining them, yet passed an ABX test much more easily than this. Hence in my mind (rather than mathematically) I can't begin to grasp how small this difference must be.
It's a strange thing to launch a new product on, isn't it? Doesn't the commercial world usually go for bigger wins? 2x or 4x subjective improvements to launch a new format? AM to FM. VHS to DVD. SD to HD. I don't recall a 15 wait before someone finally managed to spot the improvement in a double-blind test with those. I know we're not talking about physical formats any more, but still; the commercial realities of what people will pay to replace/upgrade must still apply. Maybe in 2-channel audio formats there isn't a 2x or 4x improvement left.
Don't you think it's strange that you do so much better with your laptop and headphones on whatever sample someone throws at you than the inventors of MLP did when they got to pick the sample, conversion settings, speakers and listeners? Did your work at Microsoft reveal you had sensitive hearing right from the start, or did your ability to hear artefacts improve with training and exposure to them during that time, or did it happen since?
Cheers,
David.
Does he always take so much time to make his pointless?
You are wrong in the specific here but let's put that aside. You say that I don't know this fact. Well, read this article I wrote on WBF Forum with respect to relationship of Time vs Frequency in acoustic measurements.
Whooosh.
This is way over your loudspeaker hobbyist/non-professional head Amir. You implied the paper showed
relevant loudspeaker data, it does not (save for the FR extension).
It shows the same useless thing twice! Your (WTF? forum) obfuscation can't help there.
Completely absent is the distortion data for the tweeter, showing whether the >22k resonance creates artifacts
within the audio band, when driven to 105+ db, with and without 22k band limiting. Or not.
Which in this case is something to be verified!
BINGO!! Yes, this possible artifact generating confounder is completely absent in the BS paper. Your side better address it.
Remember we don't have anything remotely like that from Meyer and Moran.
Remember, that's a complete Red Herring to the BS paper. Even if you can't be cognizant of it, due to lack of basic comprehension of logic.
In other words, this is pure speculation. You don't even know if this is the tweeter they used. You don't know if it is that it generates IM distortion in-band.
Correct, because it is clumsily absent from your BS paper. Zero data on speaker transparency to the test goal. You really should pay heed to JJ's warnings about possible artifacts creating false positives in these type tests, even if you have no understanding of them.
Once again, the burden of proof is not on me to "prove" the BS system is free of audible artifacts, speaker, software or otherwise.
Your theory is that with full knowledge of issues surround intermodulation, they entered the test blind and fell victim to it anyway?
Nope. Not my theory.
cheers,
AJ
AJ's hypothesis hinges on off-axis resonance. What's our theory of that surviving the room absorption by fair bit?
Bzzzt, whoosh.
It's not the "off axis" >22khz FR that might, or might not be audible. Room absorption has ZERO to do with this.
cheers,
AJ
This stiffness is the problem. Some years back all extreme hard domes had a resonance. Unfortunately i can't find a neutral measurement of these 7200SE
If the SEAS is confirmed as the tweeter in the 7200SE (Meridian forum give hints), there are measurements.
**However**, rational people generally don't perform the specific measurement we need.
When high end manufacturers 1" DR beryllium domes (Scan-Speak in this case (http://www.soundstagenetwork.com/index.php?option=com_content&view=article&id=1043:nrc-measurements-magico-s5-loudspeakers&catid=77:loudspeaker-measurements&Itemid=153)) already exhibit significant deviation from linearity by 95db
(http://www.soundstagenetwork.com/images/stories/loudspeakermeasurements/magico_s5/dev_95db.gif)
Most won't press to 105db, probably from fear of damage/destroying them.
Also, I highly doubt they would do so with test signals with strong ultrasonic content (like the cherry picked recording).
Btw, isn't it curious that no direct downsampled to 16/44 WAV (CD) is available on the 2L site for
direct (Hi Rez vs CD) comparison to the "Hi Rez" versions. (Only a FLAC version, which adds another confounder).
Hmmm.
cheers,
AJ
It's a strange thing to launch a new product on, isn't it? Doesn't the commercial world usually go for bigger wins? 2x or 4x subjective improvements to launch a new format? AM to FM. VHS to DVD. SD to HD. I don't recall a 15 wait before someone finally managed to spot the improvement in a double-blind test with those. I know we're not talking about physical formats any more, but still; the commercial realities of what people will pay to replace/upgrade must still apply. Maybe in 2-channel audio formats there isn't a 2x or 4x improvement left.
The launch of both SACD and DVD-A were due to business needs, not consumer.
SACD came about because the CD patents were about to expire and Philips and Sony needed another proprietary format so they came up with SACD. The format also came out when piracy was seen as a major threat to music labels. So they wrote in their license that you could not even play SACD on a PC let alone rip it! They wanted to "unring the bell" with CD's lack of copy protection.
DVD-A came about because the rest of the CE industry couldn't see another gravy train for Sony/Philips. They had created the DVD standard in DVD Forum (and Audio DVD) so they pushed ahead with a complex and messy system requiring interpreting fancy menus and such. This made building players difficult and very expensive due to high royalties that the fancy menu people demanded. And of course they put in a strong copy protection because DVD was breached and due to piracy factor above, they would have no hope of getting content if they did not rise up to SACD's level of copy protection.
You say that the quality difference was incremental/tiny. Not at all. Music that we remastered for these formats sounded far better than the CD. You can read this in Meyer and Moran report:
Though our tests failed to substantiate the claimed advantages of high-resolution encoding for two-channel audio, one trend became obvious very quickly and held up throughout our testing: virtually all of the SACD and DVD-A recordings sounded better than most CDs—sometimes much better. Had we not “degraded” the sound to CD quality and blind-tested for audible differences, we would have been tempted to ascribe this sonic superiority to the recording processes used to make them.
Plausible reasons for the remarkable sound quality of these recordings emerged in discussions with some of the engineers currently working on such projects. This portion of the business is a niche market in which the end users are preselected, both for their aural acuity and for their willingness to buy expensive equipment, set it up correctly, and listen carefully in a low-noise environment.
Partly because these recordings have not captured a large portion of the consumer market for music, engineers and producers are being given the freedom to produce recordings that sound as good as they can make them, without having to compress or equalize the signal to suit lesser systems and casual listening conditions. These recordings seem to have been made with great care and manifest affection, by engineers trying to please themselves and their peers. They sound like it, label after label. High-resolution audio discs do not have the overwhelming majority of the program material crammed into the top 20 (or even 10) dB of the available dynamic range, as so many CDs today do."So the quality differential was absolutely there. If not in specs, but actual performance.
Why did they fail? Simple: general consumer puts convenience ahead of fidelity. With MP3s they were getting huge convenience over CDs. Many songs in your pocket. And here come these labels and CE companies saying, "here is a new disc and oh, you can't rip it and we are proud of it!" The tiny group of audiophiles who cared about the difference could not save one format let alone two in the midst of a format war.
So both failed and the record execs who claimed this was their savior fired.
Fortunately we are able to take a second bite out of the apple now. By downloading high resolution content and ease with which we can play them, there is no longer a technological barrier. Most wonderfully, the files are copy protection free allowing full portability. This is why the formats are growing despite catastrophic failure before. Everything about them is different this time around.
Edit: typos as usual .
Finally Amir, I documented how this paper completely distorted ABX testing, and again I see no response from you.
OK, here is the thread I promised to create to discuss ABX: http://www.hydrogenaud.io/forums/index.php?showtopic=107540 (http://www.hydrogenaud.io/forums/index.php?showtopic=107540)
Most won't press to 105db, probably from fear of damage/destroying them.
Also, I highly doubt they would do so with test signals with strong ultrasonic content (like the cherry picked recording).
??? Here is the measurements of the track in question:
The stiffness of the tweeter cone helps determine how pure and uncoloured the sound is at high frequencies. Beryllium is many times stiffer than aluminium, the material usually used in tweeter cones. Being stiffer, the dome acts as a more perfect piston – delivering more accurate transients and lower colouration. In addition, its extended high frequency response is ideal for getting the best out of today’s high-resolution recordings.
This stiffness is the problem. Some years back all extreme hard domes had a resonance. Unfortunately i can't find a neutral measurement of these 7200SE
Some may find it ronic that that soft domes can perform far better than hard domes.
This is the FR of the DSP7200 from the paper:
(http://www.hydrogenaud.io/forums/uploads/monthly_11_2014/post-61311-1416418365_thumb.png)
Larger version in this post: http://www.hydrogenaud.io/forums/index.php...ost&id=8070 (http://www.hydrogenaud.io/forums/index.php?act=attach&type=post&id=8070)
Ironically, I have 2 Primus PC351 with their hard dome tweeters replaced with Vifa soft dome tweeters that have similar response up to the 40 KHz limit shown here. The hard domes had the expected resonance and roll off following it.
??? Here is the measurements of the track in question:
Where do you see 105 db of ultrasonic dynamic range AJ?
Bzzzt. Those aren't the droid we're looking for.
At some point, you may or may not figure out why why need to see system transparency (speakers, software, etc) data to avoid the possibility of false positives due to artifacts.
Take your time amir, we'll wait for you.
I asked them why they did that and the answer was: "to make the dude in florida angry" except they used the Nordic term for "dude." Honest!
It's not the 2L folks that have me wondering why that track was picked amir.
Now, before wee get further, perhaps some tidbits
directly copied from the "award winning" BS paper:
This Convention paper was selected based on a submitted abstract and 750-word precis that have been peer reviewed by at least two qualified anonymous reviewers. The complete manuscript was not peer reviewed. This convention paper has been reproduced from the author's advance manuscript without editing, corrections, or consideration by the Review Board. The AES takes no responsibility for the contents.
It should be noted that, although the FIR filters were comparable to those used in converters or mastering, the downsampling to 44.1 kHz or 48 kHz was not carried out; it was not feasible to replicate an A/D or D/A converter exactly since they typically use multi-stage decimation. We wanted to emulate the frequency and impulse responses of such converters while remaining in the 192 kHz domain.
The quantization either included RPDF (rectangular probability density function) dither or did not. We chose to use undithered quantization as a probe and although we would normally recommend TPDF dither for best practice we considered rectangular dither to be more representative of the non-ideal dither or error-feedback processing found in some commercial A/D and D/A filters.
CONCLUSIONS
1. FIR filters that emulate downsampling for sample rates of 44.1 kHz and 48 kHz can have a deleterious effect on the listening experience in a wideband playback system.
2. 16-bit quantization with and without RPDF dither can have a deleterious effect on the listening experience in a wideband playback system.
Seems like a tempest in a teapot rather than anything revelatory.
cheers,
AJ
Don't you think it's strange that you do so much better with your laptop and headphones on whatever sample someone throws at you than the inventors of MLP did when they got to pick the sample, conversion settings, speakers and listeners? ... Cheers, David.
Keep in mind Amirm isn't using just any old headphones; the Etymotic ER4 series effectively gives the listener a state-of-the-art, soundproof room to listen in, quite possibly even better than the special room Stuart et al. actually used, so hearing faint residual noise in the recordings, including the recordings' noise floors, over one's environmental ambient noise [which acts as a masker] is a much easier task (without any need for the use of elevated gain) especially apparent during the musical lulls and fadeouts.
The ER4 series [green curve] suppress environmental room noise where the ear is most sensitive so well that it is literally "off the charts" even when using the expanded scale range seen below. [As reference I've selected a more typical audiophile design, the Sennheiser 800 [blue], and Bose ACTIVE Noise Canceling headphones QC15 [red] for comparison]:
(http://i.imgur.com/bauGnHV.png)
I own several Etymotic earphones as well, they were given to me by the manufacturer, but I rarely use them mostly due to comfort issues. Similar to standing in a UBER expensive anechoic chamber, under quiet conditions when there is no nearby traffic or HVAC noise to contend with, late at night, I can hear my own heart beat through "bone conduction" when wearing them (in the absence of music). It's impressive.
??? Here is the measurements of the track in question:
Where do you see 105 db of ultrasonic dynamic range AJ?
Bzzzt. Those aren't the droid we're looking for.
At some point, you may or may not figure out why why need to see system transparency (speakers, software, etc) data to avoid the possibility of false positives due to artifacts.
So you don't want to answer the question. I will. There was no 105 db tone hitting the tweeter at 30 Khz. The music signal is down to 20 db SPL at that frequency. Any intermodulation would be at far lower level.
If such low level distortions are audible, that gives away the farm more than anything in this report. You would be saying that sub-zero db SPL distortions are not masked and are audible in double blind tests. Is that where you want to go AJ?
You say that the quality difference was incremental/tiny. Not at all. Music that we remastered for these formats sounded far better than the CD. You can read this in Meyer and Moran report:
This is a hilarious step, even for you, DM.
The *mastering* difference was *sometimes* great and therefore obvious (sometimes it was WORSE (http://www.avsforum.com/forum/112-surround-music-formats/1009384-hi-rez-dvd-sometimes.html)). That same great and therefore obvious quality could have been achieved on a CD, had the producers and mastering engineers so chosen.
THAT was M&M 's point. That great and obvious difference did not
require >44kHz sampling or >16 bits in the delivery medium, to achieve.
The 'high resolution delivery format' was immaterial, except in the sense that it sometimes 'obligated' (but did not force) the producers to provide a proper 'audiophile quality' mastering....which was then used to *falsely 'demonstrate' the overblown claims for the HR format itself, to uninformed listeners* --like the ones who buy fancy DACs from you, thinking they will hear 'night and day' difference.
Don't you think it's strange that you do so much better with your laptop and headphones on whatever sample someone throws at you than the inventors of MLP did when they got to pick the sample, conversion settings, speakers and listeners? Did your work at Microsoft reveal you had sensitive hearing right from the start, or did your ability to hear artefacts improve with training and exposure to them during that time, or did it happen since?
No, I started up pretty "deaf." I remember being shocked that I could not hear much difference between 128 kbps MP3 and the source. That bothered me . So I started on a path to become a trained listener and after literally hundreds of hours of testing, and tons of education learning about psychoacoustics and algorithms in audio processing, all of a sudden found myself in a super unique situation of outperforming everyone else around me in such tests. I have never liked the phrase but everyone would call me the "golden ear."
The experience taught me to use my ears as an instrument and separate it from enjoyment of music. Today I can't let go of that skill. It seems to be with me forever. For that reason US DBS broadcasting (eg. XM) bothers me to no end. I love the content but can't listen to it.
And let's distinguish between hearing and listening. My hearing is shot due to age. I am embarrassed to say that I can't even hear 12 Khz well. The reason I do well still I think is that I focus on what I think technically matter and combine that with my training to look for small differences. I suspect people with my training but intact frequency response will be able to do a lot better. I actually ran into one such person at Microsoft who was working with a partner of ours. He was hearing high frequency distortions I could not. We hired him immediately . And I moved off from doing a lot of the listening tests myself.
Let me confess. I had no idea I could do this after so many years after retirement. I just ran the tests because people like Arny kept egging me on so I gave it a try.
Having read the Stuart paper and prior citations, I am starting to think this may be due to time domain impact. It certainly is not frequency domain since I can't hear the ultrasonics. Just as pre-echo is a very audible time domain artifact (although created in frequency domain), maybe these filters act the same way.
THAT was M&M 's point. That great and obvious difference did not require >44kHz sampling or >16 bits in the delivery medium, to achieve.
It did from business point of view. What enabled labels to dump millions of dollars into new masters was the advent of new formats with strong copy protection. If you stayed with the CD you wouldn't get it.
The 'high resolution delivery format' was immaterial, except in the sense that it sometimes 'obligated' (but did not force) the producers to provide a proper 'audiophile quality' mastering....which was then used to *falsely 'demonstrate' the overblown claims for the HR format itself, to uninformed listeners* --like the ones who buy fancy DACs from you, thinking they will hear 'night and day' difference.
Well, you seem to be lost as to what argument we are having.
THAT was M&M 's point. That great and obvious difference did not require >44kHz sampling or >16 bits in the delivery medium, to achieve.
It did from business point of view. What enabled labels to dump millions of dollars into new masters was the advent of new formats with strong copy protection. If you stayed with the CD you wouldn't get it.
Good lord, this is a weak and weasely argument, even for you. Hi rez is not marketed as 'better copy protected!' No one reviews the copy protection of hi rez releases. Its not what you and your ilk rave about.
Here's the facts:
Sub-audiophile mastering was not a development *necessitated* by the CD format. (It's actually a perversion of the original intent of CD)
And you didn't *necessarily* avoid it with 'hi rez' either. I have hi rez releases that are as 'smashed' as their 'modern remastered' CD counterparts. TAKE A LOOK (http://www.avsforum.com/forum/112-surround-music-formats/1009384-hi-rez-dvd-sometimes.html)
So even you can see, surely, how mendacious it is to ascribe 'better sound' to the *hi rez format* ? That's what magicians call *misdirection*.
The fact is, we play the odds when be buy music. The odds of getting an audiophile mastering might be *better* for 'hi rez' but it's still a gamble, and it still just shows how contemptuous the industry is of us -- when the same mastering could be offered on CD or 16/44 download, for less $$.
Well, you seem to be lost as to what argument we are having.
We are *talking about*, at root, the overblown claims for hi rez audibility that have accompanied it since at least the late 1990s. The sequence of reports since then about hi rez sound from Stuart, from Oohashi et al, Meyer & Moran, from Monty, and everyone in between, have *all* been about that, in essence.
*You* don't get to dictate what *my* thread is about, Dancing Man.
*You* don't get to dictate what *my* thread is about
I tried to make this point earlier.
THAT was M&M 's point. That great and obvious difference did not require >44kHz sampling or >16 bits in the delivery medium, to achieve.
It did from business point of view. What enabled labels to dump millions of dollars into new masters was the advent of new formats with strong copy protection. If you stayed with the CD you wouldn't get it.
Good lord, this is a weak and weasely argument, even for you.
Can't you ever post something without personal remarks Steven? It gets tiring to sift through them.
Hi rez is not marketed as 'better copy protected!' No one reviews the copy protection of hi rez releases. Its not what you and your ilk rave about.
Re-read my post. I said that both high resolution formats came about not because of consumer demand but business need. The labels needed better copy protection. Hardware companies wanted to sell new gear and make money from patents. These business motivations funded spending money to remaster content in those formats.
Nothing was said about marketing anything or stating the obvious that the consumer doesn't want or care about copy protection. Dial down the level of angst and read the posts more carefully.
Here's the facts:
Sub-audiophile mastering was not a development *necessitated* by the CD format. (It's actually a perversion of the original intent of CD)
It certainly was. CD did not have copy protection. Labels wanted to get rid of the format so supported two new ones that had such.
So you have full context, what I am positing comes from direct and personal interactions with music labels and CE companies.
And you didn't *necessarily* avoid it with 'hi rez' either. I have hi rez releases that are as 'smashed' as their 'modern remastered' CD counterparts. TAKE A LOOK (http://www.avsforum.com/forum/112-surround-music-formats/1009384-hi-rez-dvd-sometimes.html)
Off-topic. David asked why the small incremental fidelity like we are finding in Stuart's tests justified the new formats. I explained that there were titles with huge difference due to different masters. So the fidelity difference was definitely there in the form of finished goods that the consumer could buy.
So even you can see, surely, how mendacious it is to ascribe 'better sound' to the *hi rez format* ? That's what magicians call *misdirection*.
No, I call you totally confused about what discussion David and I were having. You keep repeating these talking points of yours regarding high res. I hear you. And much of it is true. But off topic. You are jumping in the middle of a discussion thinking it is the larger spec vs spec battle. It is not. We are discussing the business end of two new physical formats entering the market.
The fact is, we play the odds when be buy music. The odds of getting an audiophile mastering might be *better* for 'hi rez' but it's still a gamble, and it still just shows how contemptuous the industry is of us -- when the same mastering could be offered on CD or 16/44 download, for less $$.
Sure. And I gave you even more insidious reasoning as to why high resolution formats came about. Why are you complaining then?
I was not in any way form or fashion supportive of those physical formats. Not because of your tired talking points. But because of the stupidity of setting back convenience from CD.
When Blu-ray copy protection system was being designed, I was at the table as one of the founding companies (AACS). Studios walked in with their list of improved copy protection. I walked in with one: one free rip of the disc included in the price. They asked why. I said everyone could rip DVDs. We could not improve fidelity but then take away that option. We better offer that right. You may know this as "managed copy." I left the industry and unfortunately they never finished that work which resulted in breach of Blu-ray's copy protection by both honest and dishonest users.
This is what I like about today's high resolution offer. It is free of copy protection so can be very portable. It doesn't harm anyone for it to be out there. But satisfies many of us who want it. This infuriates you why? How many angry posts do we need to read from you Steven? What business of yours is it to keep running around creating animosity among members with these battles? The world has changed and you need to change with it. This is not 2007. It isn't. Look it up!
We are *talking about*, at root, the overblown claims for hi rez audibility that have accompanied it since at least the late 1990s. The sequence of reports since then about hi rez sound from Stuart, from Oohashi et al, Meyer & Moran, from Monty, and everyone in between, have *all* been about that, in essence.
*You* don't get to dictate what *my* thread is about, Dancing Man.
Once again, you are totally confused. Go back and start reading from David's post. I have explained it all again above. You are so emotional that you can't even read what is post objectively. Or let totally logical arguments sink in.
*You* don't get to dictate what *my* thread is about
I tried to make this point earlier.
With respect, the only thing you are trying to do is fan the flames of personal battles. Why is it so hard to stay on the technical topic? Why do you have to keep posting taking sides in personal bickering? Is there a shortage of technical things you could say?
Post after post is full of derogatory terms, personal insults, and you go on to praise them as being right? Is this what this forum is about? No objective discussion of technology but encouraging personal fights by running commentary?
There was no 105 db tone hitting the tweeter at 30 Khz. The music signal is down to 20 db SPL at that frequency. Any intermodulation would be at far lower level.
You're continuing to flail wildly is amusing me.
Now who's speculating?
Fact is amir, your assumption that the system was transparent to the object of the BS test is just that, an assumption. No data whatsoever, especially for the critical HF performance.
Then of course there is the rest of the chain, software switching, etc, etc.
Have I mentioned JJ's statement about the difficulty of such tests due to artifacts? Pity you pay no heed to the real experts in the field.
You would be saying that...
Nope, that's Amir doing that saying for me.
CONCLUSIONS
1. FIR filters that emulate downsampling for sample rates of 44.1 kHz and 48 kHz
can have a deleterious effect on the listening experience in a wideband playback system.
2.
16-bit quantization with and without RPDF dither can have a deleterious effect on the listening experience in a wideband playback system.
Yaaawwwwnn
With respect, the only thing you are trying to do is fan the flames of personal battles.
Because you're being called out on multiple occasions for not honoring the topic, I'm fanning the flames of personal battles?
Meanwhile you keep flogging the same chart and tired unconvincing arguments and you wonder why there is a lack of technical discussion?
So let's leave such arguments to AJ, mzil, Arny's of the world and keep the level of discussion at high technical level.
The only BS test I know is the Meyer and Moran where they had not even heard of BS116 it seems
When did you learn to measure stuff AJ? You didn't go and learn something about audio in the last year or two, did you?
Post after post is full of derogatory terms, personal insults, and you go on to praise them as being right? Is this what this forum is about?
Thank goodness you always rise above the fray with professionalism and laser focus on the argument amir.
cheers,
AJ
You left out his po-faced concern trolling when a response to him get too 'personal'.
Indeed!
With respect, the only thing you are trying to do is fan the flames of personal battles.
Because you're being called out on multiple occasions for not honoring the topic, I'm fanning the flames of personal battles?
Meanwhile you keep flogging the same chart and tired unconvincing arguments and you wonder why there is a lack of technical discussion?
So there is no hope of you contributing technically and constructively, taking the high road and encouraging members to act professionally?
Can't you ever post something without personal remarks Steven? It gets tiring to sift through them.
Add 'utterly unself-aware' to the list of your charms.
Hi rez is not marketed as 'better copy protected!' No one reviews the copy protection of hi rez releases. Its not what you and your ilk rave about.
Re-read my post. I said that both high resolution formats came about not because of consumer demand but business need. The labels needed better copy protection. Hardware companies wanted to sell new gear and make money from patents. These business motivations funded spending money to remaster content in those formats.
Typically for your posts, that's not the whole story. DSD was invented as a worry-free archiving format for Sony, easily convertible to Redbook or any multiple sample rate thereof; it was then co-opted into a (highly copy protected) distribution format. Mark Waldrep has a nice 3-part summary (http://www.realhd-audio.com/?p=3772) of the genesis of DSD, though he of course is biased in favor of the competing technology; still, it accord with everything I have read previously about subject.
Here's the facts:
Sub-audiophile mastering was not a development *necessitated* by the CD format. (It's actually a perversion of the original intent of CD)
It certainly was. CD did not have copy protection. Labels wanted to get rid of the format so supported two new ones that had such.
So you have full context, what I am positing comes from direct and personal interactions with music labels and CE companies.
Naturally, your reply doesn't even address what I wrote. And spare me your insider anecdotes and self-puffery.
Hi rez was (and
*continues to be*) sold NOT as the new way for industry to protect its content from piracy. It was (is) sold as markedly better sound for the consumer. ...but *not* just due to better mastering. The claim is that they sound greatly better *due to the format*.
Stuart et al , who of course have a huge investment in DVD-A and its sequelae, are in their own way pushing this line still.
And I don't care what *you * think the topic is, so stop whining about it. What I've seen 2bdecided (David) write here is an eloquent affirmation of what I'm writing too. He too wonders strongly at the disjunction between the 'obvious' sonic difference (improvement!) promised for hi rez by the cheerleaders, and the paltry reality indicated even by the new Meridian report. THAT is the topic.
And I don't care what *you * think the topic is, so stop whining about it. What I've seen 2bdecided (David) write here is an eloquent affirmation of what I'm writing too. He too wonders strongly at the disjunction between the 'obvious' sonic difference (improvement!) promised for hi rez by the cheerleaders, and the paltry reality indicated even by the new Meridian report. THAT is the topic.
Greynol, so I am calibrated, would you kindly advise if Krab is right and this is the topic? What cheer leaders say, what the industry promised, wondering about sonic difference, etc? I am happy to have this discussion but don't want to get told I am being off-topic.
So there is no hope of you contributing technically and constructively, taking the high road and encouraging members to act professionally?
That is entirely up to you. I tried at first, only to be completely put-off by your antics.
Now if you would prefer I moderate, rather than participate, don't be surprised when large portions of the discussion are binned for being off-topic. If it turns out to be at least 5 out of every 9 with a 95% likelihood as not to have occurred by chance, would that constitute a huge win and worthy of an award?
and btw, the hi rez 'even more perfect sound forever' pitch wasn't just to Joe Listener. It was to recording engineers too. Back in the early days of SACD Sony was offering SACD production tech (the Sonoma system ) *free* to studios..and what were they telling them?
Mark Waldrep relates:
I heard about the DSD hype at the time. I spoke to some folks at Sony about the Sonoma system and the theory behind the whole 64fs 1-bit miracle that was going to bring new levels of “analog like” fidelity to the digital world.
No, I started up pretty "deaf." I remember being shocked that I could not hear much difference between 128 kbps MP3 and the source. That bothered me . So I started on a path to become a trained listener and after literally hundreds of hours of testing, and tons of education learning about psychoacoustics and algorithms in audio processing, all of a sudden found myself in a super unique situation of outperforming everyone else around me in such tests. I have never liked the phrase but everyone would call me the "golden ear."
The experience taught me to use my ears as an instrument and separate it from enjoyment of music. Today I can't let go of that skill. It seems to be with me forever. For that reason US DBS broadcasting (eg. XM) bothers me to no end. I love the content but can't listen to it.
And let's distinguish between hearing and listening. My hearing is shot due to age. I am embarrassed to say that I can't even hear 12 Khz well. The reason I do well still I think is that I focus on what I think technically matter and combine that with my training to look for small differences. I suspect people with my training but intact frequency response will be able to do a lot better. I actually ran into one such person at Microsoft who was working with a partner of ours. He was hearing high frequency distortions I could not. We hired him immediately . And I moved off from doing a lot of the listening tests myself.
Let me confess. I had no idea I could do this after so many years after retirement. I just ran the tests because people like Arny kept egging me on so I gave it a try.
Having read the Stuart paper and prior citations, I am starting to think this may be due to time domain impact. It certainly is not frequency domain since I can't hear the ultrasonics. Just as pre-echo is a very audible time domain artifact (although created in frequency domain), maybe these filters act the same way.
After reading this I get an even stronger sense that you must have been cheating in a lot of the logs you posted, or you have some seriously anomalous hearing (regarding detecting lossy compression).
The filters used have linear phase. The pre-echo is no echo but filter ringing at over 21 kHz that only acts on the energy that is up there.
If you can't hear well past 12 kHz then how would you detect the ringing of a filter at over 21 kHz? So please explain why this filter ringing is of concern for you.
Using the transition band from the paper to simulate 44.1 kHz within 192 kHz you get a filter order of 1715+.
A less escapist filter that starts roll-off at 20500 Hz results in an order of 508.
At 16 kHz the order would be only 130. At 12 kHz only 78. So if you "magically" could hear the ringing at 21 kHz, which doesn't seem to make any sense, you could upgrade all your CDs to hi-res quality (ignoring mastering differences) by lowpass filtering them again.
And I don't care what *you * think the topic is, so stop whining about it. What I've seen 2bdecided (David) write here is an eloquent affirmation of what I'm writing too. He too wonders strongly at the disjunction between the 'obvious' sonic difference (improvement!) promised for hi rez by the cheerleaders, and the paltry reality indicated even by the new Meridian report. THAT is the topic.
Greynol, so I am calibrated, would you kindly advise if Krab is right and this is the topic? What cheer leaders say, what the industry promised, wondering about sonic difference, etc? I am happy to have this discussion but don't want to get told I am being off-topic.
What malarkey. This is one of Amir's steps: in all 'politeness', he will parse every word down to its atomic structure, in order to divert from the main topic, if it isn't going his way. "Control the narrative' never found a more fervent practitioner.
I would simply ask respondents: don't necessarily follow his lead, unless you can bring it back to the subject of claims made about the 'sound' of hi rez vs redbook.
We have already discussed to death that NO ONE says there cannot be audible difference. So it does really all come down to a matter of degree. The industry has claimed for well over a decade now that hi rez, in itself, produces an inherent and *not subtle* audible difference -- regardless of source (the first SACDs were all analog/PCM sourced, for example).
Now we have this new report, the latest eruption of a long-bubbling vein, offering mildly positive statistical evidence that 'typical' digital filters used to render 'hi rez' down to Redbook rates can produce audible artifacts under highly specified conditions.
Supposing the results actually hold up, the question is, does this 'explain' in any significant way, the numerous concordant and
bold claims from hi rez proponents over the years, reporting from a vast variety of listening situations, of
bold audible difference due to the formats?
Adding to the ringing post above:
(http://xserv.compress.to/xnor/audio/images/ringingchirp1.png)
The background signal (blue) shows a sweep from 20 kHz to 23 kHz.
The rest show this signal filtered with the following pass band edges:
magenta = 20.5 kHz
cyan = 21.591 kHz
red = 22 kHz
Stop band starts at 22.05 kHz in all cases.
Here the same graph when subtracting the filtered signals from the original:
(http://xserv.compress.to/xnor/audio/images/ringingchirp2.png)
Here's what happens when you mess up the filtering (see cyan line which doesn't null anymore with the original signal): png (http://xserv.compress.to/xnor/audio/images/ringingchirp3.png)
Greynol, so I am calibrated, would you kindly advise if Krab is right and this is the topic?
It is clear to me that this topic is about the Meridian paper and how it will be used in the on-going narrative promoting the benefits of >48/16 audio to the end user.
How the M&M test fits into the narrative is fine too, as is DSD/SACD. The idea that DSD wasn't intentionally pitched to the end user as delivering a sonic improvement is ludicrous.
While I'm pretty sure David sees this the same way, your conversation with him does not take precedence or dictate what is on or off-topic.
The idea that DSD wasn't intentionally pitched to the end user as delivering a sonic improvement is ludicrous.
Very much so. What I can't figure out is why you are attributing that to me.
And thanks for the answer. I will take forum thread titles with a grain of salt from here on.
You say that the quality difference was incremental/tiny. Not at all. Music that we remastered for these formats sounded far better than the CD. You can read this in Meyer and Moran report:
[color="#0000FF"]Though our tests failed to substantiate the claimed advantages of high-resolution encoding for two-channel audio, one trend became obvious very quickly and held up throughout our testing: virtually all of the SACD and DVD-A recordings sounded better than most CDs—sometimes much better. Had we not “degraded” the sound to CD quality and blind-tested for audible differences, we would have been tempted to ascribe this sonic superiority to the recording processes used to make them.
The above text does not appear to be supported by the DBTs that the authors claimed to have performed. While ABX is great for detecting small differences, it is well known to be very suboptimal for ranking sound quality among alternatives that do sound different. In short, I see no use of DBT procedures that are preferred when the goal is ranking preference.
I fail to see the relevance of this line of arguments to the topic at hand.
It is already accepted that different masters can be used for different formats.
If one wants to discuss format limitations, please search for an existing topic.
It's not relevant. It's just annoying ankle-biting.
Remastering differences are what people are typically hearing when they hear differences between 'hi rez' releases and CD. This was M&M's point. It's hardly a world-shaker. Objectively and comparatively, those differences tend to be 'huge': decibel-range differences in EQ and compression and sourcing, after all, not differences in noise floors and ultrasonic content and degree of 'temporal smearing'. It hardly 'requires' a DBT to differentiate them (though for *preference* the proper DBT would be ABC/hr, or similar, not ABX, yadda yadda yadda) Subjectively, they are the drivers of preference too...unless the listener has also been primed to believe that 'hi rez' makes things sound much better, and they know nothing about 'remastering'. In that case, they will ascribe the 'big' difference they hear, to hi rez.
And here we are again:
Are the 'big' differences that fans ascribe to 'hi rez', really due to it? The M&M paper strongly suggests not. The Meridian paper doesn't seem to support a robust 'yes' to that question either. Does *any* work support it?
I guess others (perhaps c|net, Wired, New Scientist, etc.) will report on the Meridian paper's significance (if it has any significance at all, that is). Anyone seen any reports?
We have already discussed to death that NO ONE says there cannot be audible difference. So it does really all come down to a matter of degree. The industry has claimed for well over a decade now that hi rez, in itself, produces an inherent and *not subtle* audible difference -- regardless of source (the first SACDs were all analog/PCM sourced, for example).
Now we have this new report, the latest eruption of a long-bubbling vein, offering mildly positive statistical evidence that 'typical' digital filters used to render 'hi rez' down to Redbook rates can produce audible artifacts under highly specified conditions.
I find it rather interesting that this article is based on listening tests that never came within a country mile of involving anything like the typical digital filters that one finds in a Hi-Fi playback chain. Their operational frequencies were wrong, and their transition bands were wrong. The test also went off-topic as compared to its title and purported to test word length reduction, and got that wrong as well.
I guess others (perhaps c|net, Wired, New Scientist, etc.) will report on the Meridian paper's significance (if it has any significance at all, that is). Anyone seen any reports?
I just asked Google that question, and she found pre-conference mentions at just HA, Gearslutz, HiRezAduioCentral and ComputerAudiophile.
However Google's little webbot seems to be a little behind in its chores, she seems to think that this thread has 25 posts, the last dated 10/6/2014
To be fair, the earliest mainstream-ish reference to M&M I can find is Mixonline's "The Emperor's New Sampling Rate", a good 6 months after M&M published.
To be fair, the earliest mainstream-ish reference to M&M I can find is Mixonline's "The Emperor's New Sampling Rate", a good 6 months after M&M published.
Please try this:
http://www.hydrogenaud.io/forums/index.php?showtopic=57406 (http://www.hydrogenaud.io/forums/index.php?showtopic=57406)
Sep 9 2007, 22:11
I guess others (perhaps c|net, Wired, New Scientist, etc.) will report on the Meridian paper's significance (if it has any significance at all, that is). Anyone seen any reports?
Stereophile officially denounced blind/honesty controls testing as relevant in this months issue (Krab linked earlier), so I would expect this BS paper to be touted by Amirs side in next months issue as a, see, we told you deaf objectivists so.
cheers,
AJ
And let's distinguish between hearing and listening. My hearing is shot due to age. I am embarrassed to say that I can't even hear 12 Khz well.
Thanks, that might be the most refreshingly honest thing I've seen you say.
So if we are to understand this correctly, the artifacts of low-rez 16/44 you "hear" are not due to hearing, but "listening"?
You "listen" to artifacts >12k, but hear only to 12k, or the artifacts you hear with downsampled master >16/44 exist <12k?
I don't recall the ages of the listeners in the BS test being listed and I don't believe their hearing was tested.
cheers,
AJ
I don't recall the ages of the listeners in the BS test being listed and I don't believe their hearing was tested.
"
2.4. Listeners
Eight listeners took part in the test, seven of whom
were male. Most of the listeners were audio engi-
neers, and their ages ranged from 25 to 65. All re-
ported normal hearing, although this was not tested
formally.
"
I guess others (perhaps c|net, Wired, New Scientist, etc.) will report on the Meridian paper's significance (if it has any significance at all, that is). Anyone seen any reports?
I don't think so. None of their reporters will be in a position to understand this research/paper. This is news for professionals in the industry. Even in forums like this people are struggling to come to terms with it.
That said, I might write an article for it .
I guess others (perhaps c|net, Wired, New Scientist, etc.) will report on the Meridian paper's significance (if it has any significance at all, that is). Anyone seen any reports?
I don't think so. None of their reporters will be in a position to understand this research/paper. This is news for professionals in the industry. Even in forums like this people are struggling to come to terms with it.
I that that the above statements show a backwards idea of what constitutes good writing. If the Meridian paper wasn't trying to pander so hard to the golden ears, obfuscate so much questionable science, and sneak so many self-serving commercial messages between its lines, it might have been a lot easier to understand.
Good morning Xnor. If you don't mind I am going to respond to this in two parts: the personal remarks you make about me below and the technical one. The latter is actually quite easy and simple misunderstanding.
I beg our moderator's indulgence that this is on topic seeing how Xnor's post remains and not binned.
No, I started up pretty "deaf." I remember being shocked that I could not hear much difference between 128 kbps MP3 and the source. That bothered me . So I started on a path to become a trained listener and after literally hundreds of hours of testing, and tons of education learning about psychoacoustics and algorithms in audio processing, all of a sudden found myself in a super unique situation of outperforming everyone else around me in such tests. I have never liked the phrase but everyone would call me the "golden ear."
The experience taught me to use my ears as an instrument and separate it from enjoyment of music. Today I can't let go of that skill. It seems to be with me forever. For that reason US DBS broadcasting (eg. XM) bothers me to no end. I love the content but can't listen to it.
And let's distinguish between hearing and listening. My hearing is shot due to age. I am embarrassed to say that I can't even hear 12 Khz well. The reason I do well still I think is that I focus on what I think technically matter and combine that with my training to look for small differences. I suspect people with my training but intact frequency response will be able to do a lot better. I actually ran into one such person at Microsoft who was working with a partner of ours. He was hearing high frequency distortions I could not. We hired him immediately . And I moved off from doing a lot of the listening tests myself.
Let me confess. I had no idea I could do this after so many years after retirement. I just ran the tests because people like Arny kept egging me on so I gave it a try.
Having read the Stuart paper and prior citations, I am starting to think this may be due to time domain impact. It certainly is not frequency domain since I can't hear the ultrasonics. Just as pre-echo is a very audible time domain artifact (although created in frequency domain), maybe these filters act the same way.
After reading this I get an even stronger sense that you must have been cheating in a lot of the logs you posted, or you have some seriously anomalous hearing (regarding detecting lossy compression).
Let me express a big sigh of relief that you did not call me ugly too! A man is not a man without his good looks.
If I may, I like you to step back and think about what you have said/are doing. Our shtick in these discussions has always been a challenge for the other camp to run an ABX test to prove their subjective impressions of audio. Since I am not in the opposing camp, and see value in blind tests, I ran the tests. Not one but many. Not the first version but the follow ups. Not just on AVS but also here and on WBF Forum. What’s has been the response? What you say: either I have some hearing disease I am too dumb to understand, or fabricated test results.
Imagine what impression this leaves in the minds of anyone who googles about this research and lands on this thread and your post. The inescapable conclusion would be to never run an ABX/blind test. Nothing good comes out of it. If they run it and get negative outcome, it would be bad for their position. Even worse is actually running them and getting a positive result! Hell will break loose. The goal post suddenly moves from audio to their own personal integrity.
If our real intentions are to encourage others to believe and run such tests, we have failed and failed badly in that mission. I mean if the “inventor of ABX” test, our beloved Arny, goes and declare his test of high resolution audio “invalid,” clearly no one is qualified to create a valid one. At least not one that has any chance of creating a positive outcome.
Now, let’s dig into your theory. How many such tests have you run? Can you pass any of them? I even allow you to cheat and pass them. Just looking to see any, any indication that believing in blind tests is actually real. So far I have only seen Arny run one test and he said he cheated. When I asked him to re-run it with the latest ABX test with hash signatures, he refuses. Steven (Krab) has not posted any. AJ has not posted any. Mzil has done a couple but he insists to have found a cheat and not really heard a proper difference. I haven’t seen any from our moderator or others arguing with me.
That is a tough place to be. So what to do? What to do? Let’s claim impossibility here. Never mind that we are discussing a published double blind study that was peer reviewed test that found similar outcome. What do you think? They are also fabricating the results or have weird hearing? It can't be that your knowledge of what is or is not audible is incomplete? You have mastered all there is to know? Maybe you link us to what professional or academic research has led you to that position.
My wish for you is to put aside your doubts and run the tests. See if you can focus and find some difference. You may, just may, develop critical listening abilities. And start believing in reality of some people being able to way outperform other listeners. And what they do, is no cheating, or miracle of hearing but like a sport, is due to training.
That is if you care to learn about these experiments.
Even in forums like this people are struggling to come to terms with it.
And you base this on what? Two people losing the actual topic and paper contents out of sight in heated back and forths?
You know, some people don't even own the paper. That doesn't stop people from pointing out potential flaws from the limited information available, or pointing out nonsense written by others.
That said, I might write an article for it .
Doing an interpretation for the audiophile laymen to tell them what they want to hear?
Btw, you still haven't explained why 21+ kHz filter ringing is of concern for your ears. (Not that I believe that your ABX logs are not based on chance or cheating...).
That seems to be the only argument
you have left.
And regarding the paper: there are still unknowns, such as the switching software, so that we cannot draw any reliable conclusions from it.
Stuff like this really should be immediately attempted to be reproduced by some unaffiliated group...
I guess others (perhaps c|net, Wired, New Scientist, etc.) will report on the Meridian paper's significance (if it has any significance at all, that is). Anyone seen any reports?
I don't think so. None of their reporters will be in a position to understand this research/paper. This is news for professionals in the industry. Even in forums like this people are struggling to come to terms with it.
I that that the above statements show a backwards idea of what constitutes good writing. If the Meridian paper wasn't trying to pander so hard to the golden ears, obfuscate so much questionable science, and sneak so many self-serving commercial messages between its lines, it might have been a lot easier to understand.
Good morning. I appreciate your difficulty in getting through it. Fortunately the members of AES peer review board did not have any such difficulties:
(http://amirviews.smugmug.com/photos/i-h4bVPfz/0/X3/i-h4bVPfz-X3.png)
If they could get past all of the things that are bothering you, how come you can't? They found good learning in it or they would not have bestowed this award on them. Right? Why can't you put the resentment aside for a second and let the knowledge and data sink in? Too difficult?
Even in forums like this people are struggling to come to terms with it.
And you base this on what? Two people losing the actual topic and paper contents out of sight in heated back and forths?
Two people? I count a lot more than two people. But sure, why don't you summarize for us what you think the test was and the outcome.
Btw, you still haven't explained why 21+ kHz filter ringing is of concern for your ears. (Not that I believe that your ABX logs are not based on chance or cheating...).
As I mentioned, you are confused about that. I will explain later.
That seems to be the only argument you have left.
You mean other than a boatload of positive ABX test results? You mean other than a peer reviewed double blind listening test that is published by AES with authors that are true experts in the industry? None of that counts but the rhetoric on this forum by constantly declaring the game over in your favor?
And regarding the paper: there are still unknowns, such as the switching software, so that we cannot draw any reliable conclusions from it.
Stuff like this really should be immediately attempted to be reproduced by some unaffiliated group...
Unfortunately I don't recall any of you immediately or at any time trying to repeat Meyer and Moran tests. But now we get religion. Why? Because we don't like or understand the outcome. That aside, here is you saying you don't know the full story still after so many pages. So why object to me stating the same?
I don't think so. None of their reporters will be in a position to understand this research/paper. This is news for professionals in the industry. Even in forums like this people are struggling to come to terms with it.
Please .
It's fodder for the Hi Rez $cam peddlers and the $cam addicts, the only ones buying "Hi-Rez".
They did wisely choose a 2L track with no (Meridian recommended TPDF downsampled) 16/44 version.
.[/quote]
Aimed at industry professional insiders, or $cam addicts who buy this stuff?
cheers,
AJ
I don't think so. None of their reporters will be in a position to understand this research/paper. This is news for professionals in the industry. Even in forums like this people are struggling to come to terms with it.
Please .
It's fodder for the Hi Rez $cam peddlers and the $cam addicts, the only ones buying "Hi-Rez".
Nah. They don't care about our tests AJ. It would be like asking an atheist to swear on a bible god doesn't exist.
It is our camp that is crying in anger. It is kryptonite for people who think small distortions are inaudible. You remember kryptonite, yes?
It is *our* worst nightmare. People in our own camp running tests and passing the impossible ones. As expected, there is a lot of crying. We had that on AVS but it eventually died down. The evidence is nothing like we have faced before. No subjectivist claiming one cable to sound better than other. No. The evidence is the bible we believe: passing double blind ABX tests. We could now turn around say such tests are no good but only makes us look worse. Much worse.
2014 will go down as the end of an era whether you like it or not. The era of believing hobbyist tests just because it got published in the Journal. The era of believing stuff you read online as gospel. Era of people claiming to be objectivists yet be the most afraid to run double blind ABX tests. Era of people thinking by reading stuff online they all of a sudden are audio experts. Era of people using their lay understanding of audio science and confusing it with the real thing.
The storm on this forum shall too pass. It has to. It is all a lot of empty protests at the end of the day. I am here for you AJ. You can cry on my shoulder but don't expect me to do the same .
If I may, I like you to step back and think about what you have said/are doing. Our shtick in these discussions has always been a challenge for the other camp to run an ABX test to prove their subjective impressions of audio.
ABX is just one method that has its flaws. We want to see reliable, valid, objective, reproducible ... evidence for any claim.
Neither of these are given by posting ABX logs, so there is some faith involved in trusting the person and logs.
So what happens is not only skepticism against the original claim, but also the claim that the ABX log is genuine if the result is surprising. (Who cares if someone fudged at 128 kbit mp3 ABX log, right? We know that at this bitrate transparency is not given.)
This leads people to trying to reproduce what the person allegedly heard. Now what if we have many people trying to reproduce the result but failing, and a few that also claim to have heard differences?
These people are either:
- cheating as well
- hearing some artifact of their system that is not related to the actual comparison
- have some anomalous hearing (e.g. psychoacoustic models of lossy encoders expect normal hearing)
...
- have better hearing than everyone else (which you boldly admitted to do not) and genuinely hear a difference
Did I miss something important?
Not one but many. Not the first version but the follow ups. Not just on AVS but also here and on WBF Forum. What’s has been the response? What you say: either I have some hearing disease I am too dumb to understand, or fabricated test results.
No, I'm not saying that.
I'm saying that in some of these tests, the more incredible ones, you either
- didn't genuinely hear a difference (this includes purely guessing right a couple of times, hearing differences that are due to some incompetent person being unable to filter/resample files properly, abusing some artifacts in your soft- and hardware configuration ...) or
- you have anomalous hearing that can somehow magically hear 21 kHz ringing despite a roll-off above 12 kHz or
- you think you genuinely hear a difference, but which is actually an artifact of one of the problems mentioned above
- you actually hear a difference, throwing over quite a bit of what we know about psychoacoustics, physics, hearing ...
Imagine what impression this leaves in the minds of anyone who googles about this research and lands on this thread and your post. The inescapable conclusion would be to never run an ABX/blind test. Nothing good comes out of it. If they run it and get negative outcome, it would be bad for their position. Even worse is actually running them and getting a positive result! Hell will break loose. The goal post suddenly moves from audio to their own personal integrity.
Disillusioning yourself is a negative outcome? If I believe that I can hear a filter far above my hearing range, but fail all tests, I would have been better off not making the test and keeping my false belief? Or even worse, cheating so that the test results conform with my false beliefs? Seriously, wtf?
Again, even if you genuinely get a positive result, there's a chance of it being a false positive. Nobody here accepts ABX logs or the results of any paper blindly, except for trivial stuff. That doesn't mean that ABX logs are never accepted, they often are (see above for reproduction of the results).
Let's try an example:
A guy shows up posting ABX logs, claiming to hear differences between AIFF and WAV files. You look at the files, the audio data is identical, the player decodes them identically ... so you say you don't believe this result is genuine. A few other guys show up posting similar logs. Hundreds of others have also tried, but failed. Do you now believe the claim?
If our real intentions are to encourage others to believe and run such tests, we have failed and failed badly in that mission. I mean if the “inventor of ABX” test, our beloved Arny, goes and declare his test of high resolution audio “invalid,” clearly no one is qualified to create a valid one. At least not one that has any chance of creating a positive outcome.
Why? An honest person would look for the reason for an incredible positive ABX result, trying everything to eliminate the possibility of a false positive.
Positive outcomes are easy to produce. The hard part is eliminating false positives.
Now, let’s dig into your theory. How many such tests have you run? Can you pass any of them? I even allow you to cheat and pass them. Just looking to see any, any indication that believing in blind tests is actually real. So far I have only seen Arny run one test and he said he cheated. When I asked him to re-run it with the latest ABX test with hash signatures, he refuses. Steven (Krab) has not posted any. AJ has not posted any. Mzil has done a couple but he insists to have found a cheat and not really heard a proper difference. I haven’t seen any from our moderator or others arguing with me.
I regularly run ABX tests. I couldn't count them. Yes, I passed many in the past, but have a hard time hearing differences at 21 kHz when my hearing rolls off at 18 kHz.
Have you ever considered the possibility, that people don't post their negative ABX results? For each successful ABX log posted at a public test there are probably tens to hundreds of failed logs that people do not submit.
I won't cheat. If I post a fake log, I clearly state so. I do all the checks I can, to not post false positives for hard tests. I wouldn't want to be seen in the same light as some people see you, who say they trivially pass said tests (which should actually be extremely hard to impossible) and throw logs around.
Never mind that we are discussing a published double blind study that was peer reviewed test that found similar outcome. What do you think? They are also fabricating the results or have weird hearing?
Do you understand the difference between someone reading a paper and not spotting gross flaws vs. someone else actually trying to reproduce the results?
It can't be that your knowledge of what is or is not audible is incomplete? You have mastered all there is to know? Maybe you link us to what professional or academic research has led you to that position.
You surely would make a "good apologist" for your favorite religion. I hope I don't need to explain the utter nonsense that you just wrote. (Scroll back a few pages, where people have explained the basics of logic and science at least 3 times to you.)
My wish for you is to put aside your doubts and run the tests. See if you can focus and find some difference. You may, just may, develop critical listening abilities. And start believing in reality of some people being able to way outperform other listeners. And what they do, is no cheating, or miracle of hearing but like a sport, is due to training.
That is if you care to learn about these experiments.
What tests?
Before I can be honest about ABX testing I need files that were created competently. Otherwise I end up with false positives as I did with some of the files that you use to throw out positive ABX logs.
... and no, hearing 21+ kHz is not a matter of training. It's a matter of having pristine, young ears and being blessed with excellent hearing.
@mods: Please extract this into a new thread if it bothers you.
What you say: either I have some hearing disease I am too dumb to understand, or fabricated test results.
Imagine what impression this leaves in the minds of anyone who googles about this research and lands on this thread and your post.
Amir, please don't blame xnor, I made him aware of this thread (http://www.avsforum.com/forum/86-ultra-hi-end-ht-gear-20-000/1136745-establishing-differences-10-volume-method-14.html#post16386041), so he knows your methods and will expect you to Red Herring the argument into how many tests he/someone else did....
How many such tests have you run? Can you pass any of them? I even allow you to cheat and pass them. Just looking to see any, any indication that believing in blind tests is actually real. So far I have only seen Arny run one test and he said he cheated. When I asked him to re-run it with the latest ABX test with hash signatures, he refuses. Steven (Krab) has not posted any.
Yes, exactly like that!
AJ has not posted any.
I've explained to you multiple times that I have no pecuniary interest in this matter and would not stoop to gaming/cheating on unsupervised Windows online games. Further, your post is false, in the sense that I have partaken in unsupervised Windows online games (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=106825&view=findpost&p=874603), but when I found differences due to cheatability/gaming, I did not parade my "victory" around forums. I reported it to the website.
None of which have one iota to do with the BS paper.
cheers,
AJ
Stuff like this really should be immediately attempted to be reproduced by some unaffiliated group...
There are
way too many vague explanations and details missing for this to be possible.
Switching software and speakers are the tip of the iceberg.
Amir, here is a basic one, at what distance were the listeners from the speakers in this test?
cheers,
AJ
Btw, you still haven't explained why 21+ kHz filter ringing is of concern for your ears. (Not that I believe that your ABX logs are not based on chance or cheating...).
As I mentioned, you are confused about that. I will explain later.
I'm waiting.
You mean other than a boatload of positive ABX test results? You mean other than a peer reviewed double blind listening test that is published by AES with authors that are true experts in the industry? None of that counts but the rhetoric on this forum by constantly declaring the game over in your favor?
You mean the false positive / cheated ABX logs? Well... that was easy.
The paper describes people with better hearing and I guess better hardware (we will never know what system you use to ABX since you don't answer the question posted pages back) and I see this as a separate thing vs. your claims.
I have never claimed game over regarding the paper. Stop strawmanning.
All I did so far are pointing out potential problems, or commented on problems in other people's posts. It is called being skeptical, which I am towards all claims.
And regarding the paper: there are still unknowns, such as the switching software, so that we cannot draw any reliable conclusions from it.
Stuff like this really should be immediately attempted to be reproduced by some unaffiliated group...
Unfortunately I don't recall any of you immediately or at any time trying to repeat Meyer and Moran tests. But now we get religion. Why? Because we don't like or understand the outcome. That aside, here is you saying you don't know the full story still after so many pages. So why object to me stating the same?
Seriously, if I didn't have a headache from your previous posts already, I would bash my head against the wall instead of just facepalming.
Do I really need to give you an introductory course about null and alternative hypotheses, when to reject the null hypothesis, that we never accept it ...? You can't say "Meyer and Moran, therefore no audible difference", so why do you post such a rubbish reply?
It also seems like you are assuming that I'm in one of those two strictly divided camps that you have in your head. Unless you stop that stereotypical thinking I think I've had enough for a while.
The evidence is nothing like we have faced before. No subjectivist claiming one cable to sound better than other. No. The evidence is the bible we believe: passing double blind ABX tests. We could now turn around say such tests are no good but only makes us look worse. Much worse.
No, "we" could just report the test "passing" like
I did to Klippel.
2014 will go down as the end of an era whether you like it or not. The era of believing hobbyist tests just because it got published in the Journal. The era of believing stuff you read online as gospel. Era of people claiming to be objectivists yet be the most afraid to run double blind ABX tests. Era of people thinking by reading stuff online they all of a sudden are audio experts. Era of people using their lay understanding of audio science and confusing it with the real thing.
So they've legalized it in WA too. I had no idea, haven't really been keeping track. It failed an amendment down here.
I am here for you AJ.
I know, I invited you.
I'll just keep pointing out all the many issues with the BS paper, not worry about the personalities.
cheers,
AJ
2014 will go down as the end of an era whether you like it or not. The era of believing hobbyist tests just because it got published in the Journal. The era of believing stuff you read online as gospel. Era of people claiming to be objectivists yet be the most afraid to run double blind ABX tests. Era of people thinking by reading stuff online they all of a sudden are audio experts. Era of people using their lay understanding of audio science and confusing it with the real thing.
So they've legalized it in WA too. I had no idea, haven't really been keeping track. It failed an amendment down here.
This is too damn funny.
Sorry, that's the entirety of my contribution.
I guess others (perhaps c|net, Wired, New Scientist, etc.) will report on the Meridian paper's significance (if it has any significance at all, that is). Anyone seen any reports?
They may well 'report' on it, but will they be in a position to evaluate its significance to home audio?
I don't think so. None of their reporters will be in a position to understand this research/paper. This is news for professionals in the industry. Even in forums like this people are struggling to come to terms with it.
Please .
It's fodder for the Hi Rez $cam peddlers and the $cam addicts, the only ones buying "Hi-Rez".
Nah. They don't care about our tests AJ. It would be like asking an atheist to swear on a bible god doesn't exist.
You know, AJ has posted that Meridian blurb strongly endorsing TPDF dither (as AES gods Lipshitz and Vanderkooy first did long before Meridian)
at least a half dozen times now....admittedly only about half as often as you post your fabulous ABX results, but still, you can't have missed it -- and he's requested your thoughts on it every time, re: the disjunction between it and what was reported in Meridian's convention paper. Do you plan to ever engage?
(If you do, please try to restrain yourself to maybe just one or two twitter-length paragraphs. See if you can do that. Consider it a character-building exercise, by character-deleting.)
No, I'm not saying that.
I'm saying that in some of these tests, the more incredible ones, you either
- didn't genuinely hear a difference (this includes purely guessing right a couple of times, hearing differences that are due to some incompetent person being unable to filter/resample files properly, abusing some artifacts in your soft- and hardware configuration ...) or
- you have anomalous hearing that can somehow magically hear 21 kHz ringing despite a roll-off above 12 kHz or
- you think you genuinely hear a difference, but which is actually an artifact of one of the problems mentioned above
- you actually hear a difference, throwing over quite a bit of what we know about psychoacoustics, physics, hearing ...
Apparently the last one, which I think we'd all call an extraordinary claim, would not require 'extraordinary evidence', that's just some dumb meme Carl Sagan propagated, we are told. (Though I would call the Meridian methods
extraordinary in some senses, if not particularly germane to the standard line on hi rez audio)
The excitable pro hi-rez parties (and here I do *not* include the authors of the Stuart paper, at least not so far) don't seem terribly interested in figuring out what their extraordinary results might actually *derive from* , which is what any scientist would focus in on with laser intensity. Our most excitable party just wants to trumpet it as heralding a New Age in audio hobby discourse.
Hi rez for home audio has been touted since the late 1990s. We had years of tests -- not just M&M, but also those who attempted to replicate Oohashi et al, and of course here-- indicating only the most modest 'difference' if at all, and not really 'easy' to detect by most people, without some 'enhancement' of the difference cues (hell, in the case of Oohashi et al, it required *not* using ABX and *using* an MRI scanner). Now suddenly we have a cluster of virtually slam-dunk 'positives' from a handful of online trials, and another modest positive from Meridian's work. Certain excitable parties (who happen to sell 'hi rez' hardware) find this very exciting. Curious, no?
Have you ever considered the possibility, that people don't post their negative ABX results? For each successful ABX log posted at a public test there are probably tens to hundreds of failed logs that people do not submit.
That's one of the 'train wreck' aspects of the AVS Forum trials. To the test moderators' credit, though, they have asserted multiple times that the test and results there are not scientific.
What you say: either I have some hearing disease I am too dumb to understand, or fabricated test results.
Imagine what impression this leaves in the minds of anyone who googles about this research and lands on this thread and your post.
Amir, please don't blame xnor, I made him aware of this thread (http://www.avsforum.com/forum/86-ultra-hi-end-ht-gear-20-000/1136745-establishing-differences-10-volume-method-14.html#post16386041), so he knows your methods and will expect you to Red Herring the argument into how many tests he/someone else did....
The key phrase that one finds in Amir's first post to that thread is:
"I did not level match anything. However, once I found one source was worse than the other, I would then turn up the volume to counter any effect there. Indeed, doing so would close the gap some but it never changed the outcome. Note that the elevated level clearly made that source sound louder than the other. So the advantage was put on the losing side."
This is just one more reason why when Amir says "Our camp" I wonder WTF he thinks he is talking about.
ABX is just one method that has its flaws.
Them are fighting words for some here!
We want to see reliable, valid, objective, reproducible ... evidence for any claim.
For any claim? How about giving me a link to one that has passed all of those here so that I get calibrated.
Neither of these are given by posting ABX logs, so there is some faith involved in trusting the person and logs.
Well, do we get to say Meyer and Moran's test reports are not reliable? Or do we take it on faith that it is because it had negative outcome? Again, I am trying to get calibrated.
Let's say you don't trust the person. What then? He has to get a live witness? Why on earth would he want to go through all that trouble to convince you? What is in there for him to have an anonymous poster be satisfied in this regard?
And why do you post under an alias? Shouldn't I be able to examine your history and who you are in judging your opinions and any test results you put forward? You have mine. How come you are afraid of transparency on your part?
So what happens is not only skepticism against the original claim, but also the claim that the ABX log is genuine if the result is surprising.
Your skepticism is due to lack of experience and knowledge. You can't turn that around and put the work on me to prove you wrong. We have peer reviewed tests in the form of Stuart's tests. We also have peer reviewed tests in the form of Meyer and Moran. In neither case will I remotely go to the place you are, questioning the integrity of people.
Run these tests yourself. See if you can pass them. There are people on AVS Forum who shocked themselves after following me lead and carefully listening, found differences reliably. Until you demonstrate that you have done that, your suspicions are not an issue I worry about.
(Who cares if someone fudged at 128 kbit mp3 ABX log, right? We know that at this bitrate transparency is not given.)
Oh I would. If they cheated I wouldn't want to have anything to do with them in anything. I have little patience for people with lack of integrity.
Now what if we have many people trying to reproduce the result but failing, and a few that also claim to have heard differences?
These people are either:
- cheating as well
- hearing some artifact of their system that is not related to the actual comparison
- have some anomalous hearing (e.g. psychoacoustic models of lossy encoders expect normal hearing)
...
- have better hearing than everyone else (which you boldly admitted to do not) and genuinely hear a difference
Did I miss something important?
Of course. You missed the most important one: training. Are you really not aware of the concept of trained/expert listeners? You think anyone who does better has better hearing? How come with the same hearing ability I went from not hearing compression artifacts to outperforming all but one person in my entire career at Microsoft?
The other major thing you missed is knowledge of the technology in question. I know what to listen for. When I took David's test I quickly zoomed in on the right note and the game was over. If you don't know that, you be lost in the woods when it comes to dynamic distortion. The difference may be there and audible to that person but if you don't know where it is in a 3 minute song, good luck finding it by randomly clicking here and there. There is a science to what is relevaing and what is not.
Let me give you a real life example. The year is 1999 or early 2000. The music industry is up in arms about piracy. It forms a consortium called SDMI chaired by Leonardo whom you may know has chairman of MPEG to remedy that situation. One of the things they went after was adding audio watermark to DVD audio. They put out a call for proposal and Microsoft Research put one that they were working on forward. I hear about them doing that just before submission. I ask them to let me listen to see if the mark is transparent (big deal as far as acceptance by the labels). They assure me that they had done their listening tests and no one could tell the difference.
I ask them to humor me and give me the tracks which were supplied by the labels. I don't know how much you know about audio watermarks but the concept is data hiding where the encoder attempts to find segments where masking would easily cover the toggling of the bits. Back to the story, they files were 24 bit 96 Khz and 3+ minutes long. I listen and I can't tell the difference. Bothered , I try harder and all of a sudden I detect something in a fraction of a second. It was just a transient that didn't sound right to me. I go back to the Microsoft Research manager and tell him that I thought I had found the difference. With disbelief, he asks me for the timecode where I heard it. I give it to him in seconds and milliseconds . The reaction was golden. He comes back after a bit and says I had indeed found it and was a bug in the code! They fixed it and all was well.
This is what training, critical listening ability and technical knowledge enables. To people without those attributes, it comes across as something impossible. But with objective confirmation of the bug in the code, you can take what I said to bank.
By the same token, unless you can demonstrate to me that you are an expert in this area, your incredulity I am afraid has no weight or importance to me.
The above is one of the most important lessons here. We have proven beyond any doubt that the above skills exist and you cannot take your listening results as a measure of whether someone else can or cannot hear small distortions. This is why I don't care what Meyer and Moran found. They did not have critical listeners so their results don't apply to me.
I will stop here and encourage you to go and take these tests. Try hard to pass them. You might, just might be able to do that. If so, you will learn something valuable.
I'd be curious, Amir, in understanding just what training you undertook while at Microsoft? Myself, I'd prefer to work under the assumption that You didn't cheat. One thing I find just a bit peculiar is that your success at using ABX is at odds with what Stuart et al said regarding that approach, being that it was more difficult. Thoughts?
The paper describes people with better hearing and I guess better hardware (we will never know what system you use to ABX since you don't answer the question posted pages back) and I see this as a separate thing vs. your claims.
Not it doesn't. The listeners were not expert/trained listeners. They received training for the test. That is a different animal than someone who has spent years becoming an expert listener. And further, has the technology experience to know what to listen for as I just explained.
Here is another example, this time from the work of current president of AES, Dr. Sean Olive. Download the software from this link and take the blind test: http://www.whatsbestforum.com/showthread.p...le-For-Download (http://www.whatsbestforum.com/showthread.php?2984-Harman-s-How-to-Listen-Software-is-Now-Available-For-Download)
It is a simple test. Just a set a parametric EQ whose frequency is randomly picked by the software and the Q progressively increased. When I first ran the test I couldn't go past level 3 or so. Then I listened to a bit more and stopped. A while later I am at Harman and Sean is presenting to a group of us the same tool. Up to level 3 most people were getting it right. Past that I was the only one that could keep up with Sean. I went up to level 6 or 7 and my limited training capped out. Sean had no trouble going past that. I think he said the minimum level for their trained listeners is 12. Try it. Only then will you get a firsthand feel for what it means to be good at listening tests, trying to find small differences. And remember, your "hearing" stays constant. It is your listening skills that improve.
So no, they are not similarly situated to me.
And regarding the paper: there are still unknowns, such as the switching software, so that we cannot draw any reliable conclusions from it.
Stuff like this really should be immediately attempted to be reproduced by some unaffiliated group...
Unfortunately I don't recall any of you immediately or at any time trying to repeat Meyer and Moran tests. But now we get religion. Why? Because we don't like or understand the outcome. That aside, here is you saying you don't know the full story still after so many pages. So why object to me stating the same?
Seriously, if I didn't have a headache from your previous posts already, I would bash my head against the wall instead of just facepalming.
Do I really need to give you an introductory course about null and alternative hypotheses, when to reject the null hypothesis, that we never accept it ...? You can't say "Meyer and Moran, therefore no audible difference", so why do you post such a rubbish reply?
The only thing rubbish is this uninformed opinion. Null hypothesis? Are you kidding me? Meyer and Moran violated just about every rule for properly conducting a listening test. I am asking you are not pointing those things out if you have genuine intentions.
It also seems like you are assuming that I'm in one of those two strictly divided camps that you have in your head. Unless you stop that stereotypical thinking I think I've had enough for a while.
You are showing extreme angst and frustration over these findings, both mine and that of stuart. You are being accusatory and emotional. And resentful. These are characteristics I see in my stereotype of the people you talk about. In contrast, I don't see that in David. Yes he is disagreeing with me but he is demonstrating calm, knowledge and measured interactions. So please don't put this at my feet. Think before you let your emotions write the words if you want to see a different impression of you. And remember, you are posting under an alias. So you can't have any expectation of me knowing who you are.
BTW, I would be perfectly fine with you stopping this interaction. I still owe you that technical answer but otherwise, I am getting tired of dealing with another person's emotional outbursts.
I'd be curious, Amir, in understanding just what training you undertook while at Microsoft? Myself, I'd prefer to work under the assumption that You didn't cheat. One thing I find just a bit peculiar is that your success at using ABX is at odds with what Stuart et al said regarding that approach, being that it was more difficult. Thoughts?
Hi Chu. Good to see you here and thanks for chiming in.
Stuart is absolutely right. And would be even more right had he had to endure taking the test using the foobar ABX plug-in.
It is an incredibly difficult challenge when one is detecting small differences. Remembering an image of A in mind, then B and then X when differences are small is very challenging. With discipline I am able to do it better than most but it requires so much concentration because of the test fixture not being ideally suited to finding small differences.
Ironically placebo or better said, nocebo plays a very strong role here. You listen to A and then B. Let's say you hear a valid difference that can be objectively proven. I don't know about others but I can easily "un-hear" that difference! What placebo giveth, nocebo taketh away! I can take fidelity differences that one would value like a beautiful reverb trail and with the right level of doubt, erase it as if it does not exist.
The above is deadly in the tests we are talking about. I found a work-around but it relies even more on memory. What I do is listen to A and B and classify their difference. Once there, I do NOT allow myself to second guess myself when listening to each trial. If I let myself have any doubt, I get lost and fail the rest of the test.
The poor usability of Foobar ABX plug-in just adds to it. Try to create a segment using that slider in a 3 minute song. You will go nuts. Tiniest movement of the mouse jumps ahead many seconds. Take the test once and try to take it again and there is no memory of what segments you had picked. Non-deterministic glitches and popps add to the "excitement." It is like trying to run a marathon while people throw rocks at you. The job is difficult enough without the tool getting in the way.
The newest version takes more steps backward. It seems the goal is to make it harder and harder for people to get positive results. I used it and passed David's test to deal with people's accusations of cheating but it was even more difficult to use than before.
Now, is the alternative of using sighted tests better? Not at all. The magnitude of differences is so small that placebo effect easily dwarfs it. So we are stuck between two non-ideal methods to get to the "truth." This is why I like mathematical analysis such as what Stuart does so beautifully. Or measurements. We take out the untrustworthy listener out of the picture and get truly objective data.
You are showing extreme angst and frustration over these findings, both mine and that of stuart. You are being accusatory and emotional. And resentful. These are characteristics I see in my stereotype of the people you talk about. In contrast, I don't see that in David. Yes he is disagreeing with me but he is demonstrating calm, knowledge and measured interactions. So please don't put this at my feet. Think before you let your emotions write the words if you want to see a different impression of you. And remember, you are posting under an alias. So you can't have any expectation of me knowing who you are.
BTW, I would be perfectly fine with you stopping this interaction. I still owe you that technical answer but otherwise, I am getting tired of dealing with another person's emotional outbursts.
oh, the hilarity
Them are fighting words for some here!
So you agree that ABX logs can be fudged, that there can be false positives due to a large number of variables etc.
Great, we're finally getting somewhere.
For any claim? How about giving me a link to one that has passed all of those here so that I get calibrated.
How about you look at one of those low bitrate multiformat listening tests coordinated in this forum? (They are not perfect, and I'm not claiming they are, but they are pretty solid for a community effort.)
Well, do we get to say Meyer and Moran's test reports are not reliable? Or do we take it on faith that it is because it had negative outcome? Again, I am trying to get calibrated.
No, I take the results for what they are and don't make fallacious leaps in logic that you seem to love to do.
You don't need to get calibrated, you need to take a
basic course on logic, science and statistics.
Let's say you don't trust the person. What then? He has to get a live witness? Why on earth would he want to go through all that trouble to convince you? What is in there for him to have an anonymous poster be satisfied in this regard?
As I said, take a basic 101 course. First of all, the burden of proof is on the one making the claim. It's not about a single anonymous poster but every rational, skeptical person that is not a gullible fool.
We are interested in truth, so if that person is also interested in that he/she will try to present as much evidence as necessary and as much information necessary for others to reproduce the result or even convince them directly.
And why do you post under an alias? Shouldn't I be able to examine your history and who you are in judging your opinions and any test results you put forward? You have mine. How come you are afraid of transparency on your part?
Who says you are not, additionally to "amirm", posting with another alias? Not that I would care... especially not about your history.
You complained so much about personal remarks, but now you're suddenly interested in me personally? A bit hypocritical..
What matters are the claims you make and evidence you present, regardless if you're John Doe or the president. I don't know you. You don't know me. I don't personally care about you more than about any other human being I don't know. (Btw, you can read all my posts in this forum, download my DSP plugins ...)
Your skepticism is due to lack of experience and knowledge. You can't turn that around and put the work on me to prove you wrong. We have peer reviewed tests in the form of Stuart's tests. We also have peer reviewed tests in the form of Meyer and Moran. In neither case will I remotely go to the place you are, questioning the integrity of people.
Run these tests yourself. See if you can pass them. There are people on AVS Forum who shocked themselves after following me lead and carefully listening, found differences reliably. Until you demonstrate that you have done that, your suspicions are not an issue I worry about.
You're talking absolute rubbish. Skepticism is probably the best method to navigate in a world full of woo and nonsense.
Again, take a basic course. Also look up fallacies, like shifting the burden of proof, proving a negative..
I'm not questioning integrity, I'm questioning the tests and test results.
Some of the files of the tests you mention have been shown to have flaws. You keep sweeping that fact under the rug, so you don't have to admit your false positive / botched logs and invalidate entire "tests".
Oh I would. If they cheated I wouldn't want to have anything to do with them in anything. I have little patience for people with lack of integrity.
So you assert and blindly accept that every posted log and tests result and study is genuine and does not contain false positives? That would be gullibility par excellence.
Do you believe in aliens flying around in UFOs visiting earth also? They are videos and first person accounts of people who genuinely believe that they were abducted. Be consistent please...
Of course. You missed the most important one: training. Are you really not aware of the concept of trained/expert listeners? You think anyone who does better has better hearing? How come with the same hearing ability I went from not hearing compression artifacts to outperforming all but one person in my entire career at Microsoft?
I include that in hearing, because training doesn't magically repair your treble roll-off. We're not talking about some burst of noise being inserted by a lossy codec that you can more reliably detect with some training, but filters operating at 21+ kHz.
The other major thing you missed is knowledge of the technology in question. I know what to listen for. When I took David's test I quickly zoomed in on the right note and the game was over. If you don't know that, you be lost in the woods when it comes to dynamic distortion. The difference may be there and audible to that person but if you don't know where it is in a 3 minute song, good luck finding it by randomly clicking here and there. There is a science to what is relevaing and what is not.
I don't care about "David's test", stop evading. Tell us finally what you hear when there's a filter operating at 21+ kHz with your super-trained and super-rolled-off hearing.
The excitable pro hi-rez parties ...
Don't know those people are here. If you mean me I am in the "don't dumb it down" party. I want the bits prior to conversion to 16/44.1. Have no need for that conversion. My equipment plays the upstream bits just fine. I suspect yours does too. This doesn't sit well with you why?
Let me give you a real life example.
No thanks, enough off-topic and self-praise.
By the same token, unless you can demonstrate to me that you are an expert in this area, your incredulity I am afraid has no weight or importance to me.
So another fallacy, ad hominem this time.
Have you never heard "arguments stand or fall on their own merits"? Now if you want to run.. I'm not stopping you.
The above is one of the most important lessons here. We have proven beyond any doubt that the above skills exist and you cannot take your listening results as a measure of whether someone else can or cannot hear small distortions. This is why I don't care what Meyer and Moran found. They did not have critical listeners so their results don't apply to me.
I will stop here and encourage you to go and take these tests. Try hard to pass them. You might, just might be able to do that. If so, you will learn something valuable.
No, you have not shown that any of what you wrote applies to filters operating at 21+ kHz. You make this ridiculously ambiguous claim that you can hear some difference by posting logs.
Please:Show me one of "those tests" that has properly processed files, but you still hear a difference between them.
Then tell me what difference you hear, and where if applicable, and with what equipment and settings (player, DirectSound playback format if that is your API of choice, ...).
If this checks out then I will take a closer look (I did take a look at some of the files mentioned in the train wreck thread before - they were faulty) and ABX them.
This is what people usually do in this forum, by themselves, because they're interested in the truth. Are you?
First of all, the burden of proof is on the one making the claim.
And you are claiming I cheated yet you want me to do the work. What happened to that burden?
How about you look at one of those low bitrate multiformat listening tests coordinated in this forum? (They are not perfect, and I'm not claiming they are, but they are pretty solid for a community effort.)
Not interested in doing more work that will get dismissed. Should have thought of that before insinuating that any data that I put forward could be a cheat. Remember what I said about people reading these things?
For now, I answered the same for your proxy Steven (Krab) on AVS Forum who claimed I couldn't pass 320 kbps MP3 against the original. So I just took the clips in play and ran that test too:
=============
foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/19 19:45:33
File A: C:\Users\Amir\Music\Arnys Filter Test\keys jangling 16 44.wav
File B: C:\Users\Amir\Music\Arnys Filter Test\keys jangling 16 44_01.mp3
19:45:33 : Test started.
19:46:21 : 01/01 50.0%
19:46:35 : 02/02 25.0%
19:46:49 : 02/03 50.0%
19:47:03 : 03/04 31.3%
19:47:13 : 04/05 18.8%
19:47:27 : 05/06 10.9%
19:47:38 : 06/07 6.3%
19:47:46 : 07/08 3.5%
19:48:01 : 08/09 2.0%
19:48:19 : 09/10 1.1%
19:48:31 : 10/11 0.6%
19:48:45 : 11/12 0.3%
19:48:58 : 12/13 0.2%
19:49:11 : 13/14 0.1%
19:49:28 : 14/15 0.0%
19:49:52 : 15/16 0.0%
19:49:56 : Test finished.
----------
Total: 15/16 (0.0%)===============
foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/31 15:18:41
File A: C:\Users\Amir\Music\AIX AVS Test files\On_The_Street_Where_You_Live_A2.mp3
File B: C:\Users\Amir\Music\AIX AVS Test files\On_The_Street_Where_You_Live_A2.wav
15:18:41 : Test started.
15:19:18 : 01/01 50.0%
15:19:30 : 01/02 75.0%
15:19:44 : 01/03 87.5%
15:20:35 : 02/04 68.8%
15:20:46 : 02/05 81.3%
15:21:39 : 03/06 65.6% <--- Difference found
15:21:47 : 04/07 50.0%
15:21:54 : 04/08 63.7% <--- Dog barked!
15:22:06 : 05/09 50.0%
15:22:19 : 06/10 37.7%
15:22:31 : 07/11 27.4%
15:22:44 : 08/12 19.4%
15:22:51 : 09/13 13.3%
15:22:58 : 10/14 9.0%
15:23:06 : 11/15 5.9%
15:23:14 : 12/16 3.8%
15:23:23 : 13/17 2.5%
15:23:33 : 14/18 1.5%
15:23:42 : 15/19 1.0%
15:23:54 : 16/20 0.6%
15:24:06 : 17/21 0.4%
15:24:15 : 18/22 0.2%
15:24:23 : 19/23 0.1%
15:24:34 : 20/24 0.1%
15:24:43 : 21/25 0.0%
15:24:52 : 22/26 0.0%
15:24:57 : Test finished.
----------
Total: 22/26 (0.0%)==============
Let's see you run these same tests. Don't worry, I won't question your integrity by doubting your foobar abx logs.
without comment
foo_abx 2.0 beta 4 report
foobar2000 v1.3.4
2014-11-17 19:56:45
File A: limehouse_maximum_phase_100.wav
SHA1: 722dc26db8d4ce666dc03875b2c8d4570d22b521
File B: limehouse_reference.wav
SHA1: e8ad96830d23cad4bba5bf822ce875ae452b9e7c
Output:
DS : Primärer Soundtreiber
19:56:45 : Test started.
19:57:01 : 00/01
19:57:08 : 00/02
19:57:16 : 00/03
19:57:24 : 00/04
19:57:33 : 00/05
19:57:41 : 00/06
19:58:02 : 00/07
19:58:13 : 00/08
19:58:21 : 00/09
19:58:28 : 00/10
19:58:37 : 00/11
19:58:43 : 00/12
19:58:49 : 00/13
19:58:56 : 00/14
19:59:06 : 00/15
19:59:14 : 00/16
19:59:21 : 00/17
19:59:29 : 01/18
19:59:36 : 02/19
19:59:43 : 03/20
19:59:55 : 04/21
20:00:01 : 05/22
20:00:09 : 06/23
20:00:16 : 07/24
20:00:26 : 08/25
20:00:31 : 09/26
20:00:36 : 10/27
20:00:42 : 11/28
20:00:52 : 12/29
20:00:58 : 13/30
20:01:03 : 14/31
20:01:10 : 15/32
20:01:20 : 16/33
20:01:26 : 17/34
20:01:26 : Test finished.
----------
Total: 17/34
Probability that you were guessing: 56.8%
-- signature --
a146fd5e354e1a7f5df78bae353e11a3356bf601
The paper describes people with better hearing and I guess better hardware (we will never know what system you use to ABX since you don't answer the question posted pages back) and I see this as a separate thing vs. your claims.
Not it doesn't. The listeners were not expert/trained listeners. They received training for the test. That is a different animal than someone who has spent years becoming an expert listener. And further, has the technology experience to know what to listen for as I just explained.
Wow, so now audio engineers who work with audio daily - according to you - are not trained listeners? And yes, they additionally could prepare for each specific test! Audio engineers also have no "technology experience", right. What the actual f?
It seems that all you're interested in at this point is contradicting me, posting even more rubbish than usually.
Here is another example, this time from the work of current president of AES, Dr. Sean Olive.
I've run this application
years ago when it first was released to the public. Still no explanation from your side what this has to do with filters operating at 21+ kHz... Still waiting...
The only thing rubbish is this uninformed opinion. Null hypothesis? Are you kidding me? Meyer and Moran violated just about every rule for properly conducting a listening test. I am asking you are not pointing those things out if you have genuine intentions.
I can only double facepalm so hard. #255 (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=250&p=881162&hl=obtuse&#entry881162)
From your posts everyone can see that you think that M&M disproved hi-res sound. If you had taken the basic science course I asked you to take, you'd see how absolutely ridiculous that sounds.
You are showing extreme angst and frustration over these findings, both mine and that of stuart. You are being accusatory and emotional. And resentful. These are characteristics I see in my stereotype of the people you talk about. In contrast, I don't see that in David. Yes he is disagreeing with me but he is demonstrating calm, knowledge and measured interactions. So please don't put this at my feet. Think before you let your emotions write the words if you want to see a different impression of you. And remember, you are posting under an alias. So you can't have any expectation of me knowing who you are.
BTW, I would be perfectly fine with you stopping this interaction. I still owe you that technical answer but otherwise, I am getting tired of dealing with another person's emotional outbursts.
Angst and frustration regarding the papers findings? What the..? Projecting much?
There is frustration ... with
your rubbish posts which everyone can see. And again, what's with all the ad hominem?
You even admit to stereotyping me!
Btw, I'm completely calm, but still surprised at how you interact and what you post. I mean you're not the average audiophile Joe, well, you shouldn't be.
Anyway, if you finally answer the questions further above I would be highly interested in preparing a 21+ kHz filter test for you where you cannot, let's say, produce false positives as easily. Are you up to putting your money where your mouth is?
The excitable pro hi-rez parties ...
Don't know those people are here. If you mean me I am in the "don't dumb it down" party. I want the bits prior to conversion to 16/44.1. Have no need for that conversion. My equipment plays the upstream bits just fine. I suspect yours does too. This doesn't sit well with you why?
oh, so you *aren't* the one claiming that the recent fabulous outbreak of positive HR/Redbook ABX results heralds a new age of discourse in the hobby, that your AVSF trainwreck results were 'conclusive' proof of something, that Meyer & Moran 2007 was previously held to be flawless proof of a negative, and has now been rendered null and void? All the while houndogging Arny at every conceivable turn? You aren't the guy with quite a 'rep' for self-congratulation and condescension and fancy footwork on certain forums, touting the dangers of jitter, of Redbook, of monoprice cables? That wasn't you? Oh dear, what a mistake I've made. All apologies.
If your new line to the audio faithful is "you'll probably never hear the differences between the bits prior to 16/44 conversion, and the bits after 16/44 conversion, in your normal listening. They're
quite minor differences at best; I myself had to train for years to become a dancing ninja master of millisecond artifact detection to ace those ABX tests. The differences *you* plebes are hearing, then, when you listen at home in your comfy chairs to 'hi rez' and Redbook versions released commercially, are really due to different
mastering, not the formats. Like, duh. We in the 'biz' all know that, and have known for years. So let's all push for
proper mastering, whether it's of Redbook, hi rez, mastered for iTunes , or whatever" -- if THAT is your new line, well then I'd say you've calmed down considerably.
You could even add '"Oh, and if you're
veryworried that your DAC , which upsamples everything, is doing a terrible job of it, then by all means go for the 'upstream bits' version, if it makes you sleep better at night. It's probably overkill though" and I'd still grant that you that mantle of calm.
Will that your new line, so we can be brothers in objectivity?
First of all, the burden of proof is on the one making the claim.
And you are claiming I cheated yet you want me to do the work. What happened to that burden?
No, after I read your posts (especially #347) I said that I believe that it is more probable that you cheated and I also explained that this could include you genuinely believing to hearing differences that are actually flaws in the test.
All you have done so far is evade the only questions I asked (which I'm far more interested in than this incredibly painful back and forth), which as you can imagine does the opposite of convincing me.
Wow, so now audio engineers who work with audio daily - according to you - are not trained listeners?
Of course not. In a hardware company like Meridian, audio engineers design audio equipment. They have nothing to do with critical listening.
From the summary section of Stuart paper:
Differences were demonstrated here in a double-blind test using non-expert listeners who received minimal training.An electrical engineer designing circuits, a PCB layout person, an FPGA programmer/designer, etc. are all "audio engineers" but nothing in their training or work gives them expertise in listening test.
It is a lay assumption that an "audio engineer" is an expert listener. They are not as I explained to Arny earlier.
Please don't keep posting this stuff. Good grief...
How about you look at one of those low bitrate multiformat listening tests coordinated in this forum? (They are not perfect, and I'm not claiming they are, but they are pretty solid for a community effort.)
Not interested in doing more work that will get dismissed. Should have thought of that before insinuating that any data that I put forward could be a cheat. Remember what I said about people reading these things?
What? What's wrong amirm?
I didn't ask you to retroactively participate in these tests that have been long closed .. You asked what to look for and I pointed you to look at these tests.
Now, except if you had anomalous hearing, you
could repeat all of these tests (I'm not asking you to!) with results that would fit in the results of the many participants of those tests. That's what you asked for, right? Reproducible, reliable ... evidence.
For now, I answered the same for your proxy Steven (Krab) on AVS Forum who claimed I couldn't pass 320 kbps MP3 against the original. So I just took the clips in play and ran that test too:
Not interested in mp3 ABX logs here, sorry. But shoot me a PM with links to the files if you are so interested in me trying to ABX them (for whatever reason?).
Of course not. In a hardware company like Meridian, audio engineers design audio equipment. They have nothing to do with critical listening.
From the summary section of Stuart paper:
Differences were demonstrated here in a double-blind test using non-expert listeners who received minimal training.
An electrical engineer designing circuits, a PCB layout person, an FPGA programmer/designer, etc. are all "audio engineers" but nothing in their training or work gives them expertise in listening test.
It is a lay assumption that an "audio engineer" is an expert listener. They are not as I explained to Arny earlier.
Please don't keep posting this stuff. Good grief...
Good grief indeed.
First of all you conveniently skipped this:
And yes, they additionally could prepare for each specific test! Audio engineers also have no "technology experience", right. What the actual f?
It seems that all you're interested in at this point is contradicting me, posting even more rubbish than usually.
So yeah, you just confirmed that last part.
Secondly, audio engineers usually deal with recording, manipulation, mixing and reproduction. Where does it say that the listeners were "only" audio equipment designers?
Also, another fallacy, I never said that they are expert listeners on the issue at hand.
And of course engineers that design high end audio components would not know anything about how e.g. tuning algorithms in their DSP components would lead to tiny audible improvements, which requires listening? No one working at such a company would be interested in critical listening, right? They must have used the "audio engineer interns", some of which were 65 years old, right?!
C'mon amirm, aren't you above such nonsense?
The key phrase that one finds in Amir's first post to that thread is:
"I did not level match anything. However, once I found one source was worse than the other, I would then turn up the volume to counter any effect there. Indeed, doing so would close the gap some but it never changed the outcome. Note that the elevated level clearly made that source sound louder than the other. So the advantage was put on the losing side."
Maybe he's so highly trained and skilled as a listener (his claim), he doesn't need to level match either. Or maybe that bizarre statement throws into grave doubt what he claimed to have learned at Microsoft about controlled listening tests?
Either way, I can see why anyone who read that thread and saw the "results" of that, ummm, "test", might have a wee bit of skepticism at these latest completely unsupervised results of Windows files testing on a Windows pc.
Not to mention that he either still doesn't know, or care, what a Red Herring is, despite being told at least 5yrs ago. Thus includes in nearly every post.
Maybe both Meyer and Moran could send him some nice smoked Red Herring for Christmas.
cheers,
AJ
I understand what he is posting, but half of the time I don't understand why he even would post the things he does.
Is that a red herring? Must be. AJ, help!!! (<- no, I'm not serious)
... Now, is the alternative of using sighted tests better? Not at all. The magnitude of differences is so small that placebo effect easily dwarfs it. So we are stuck between two non-ideal methods to get to the "truth." This is why I like mathematical analysis such as what Stuart does so beautifully. Or measurements. We take out the untrustworthy listener out of the picture and get truly objective data.
We then come up against those who will not accept that measurements can be more sensitive than our ears. "There are differences that we can hear but that cannot be measured." "You're not measuring the right things." And so on.
Your skepticism is due to lack of experience and knowledge. You can't turn that around and put the work on me to prove you wrong. We have peer reviewed tests in the form of Stuart's tests. We also have peer reviewed tests in the form of Meyer and Moran.
By AES.
Amir, if your online self administered Windows ABX tests showed here on numerous occasions to be flawed/gameable, are evidence for 16/44 audibility somewhat on par with the BS test results, when will we expect you to submit them as a paper to the AES?
Certainly you could use another award like they gave the BS paper and as an AES member, submit your setup, methods, listener training, results, etc. as a paper for peer review?
I know what to listen for.
Amir, wouldn't someone like JJ also know what to listen for?
Why does he settle for CD and reject the Hi-Rez scam products? Are you claiming to have better hear....excuse me, "listening" ability than your teacher?
cheers,
AJ
I want the bits prior to conversion to 16/44.1.
Why?
An award winning luminary of the industry had this to say:
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
Transparent processing. What more do you want?
Or don't you believe him?
cheers,
AJ
You know, AJ has posted that Meridian blurb strongly endorsing TPDF dither (as AES gods Lipshitz and Vanderkooy first did long before Meridian) at least a half dozen times now....admittedly only about half as often as you post your fabulous ABX results, but still, you can't have missed it -- and he's requested your thoughts on it every time, re: the disjunction between it and what was reported in Meridian's convention paper. Do you plan to ever engage?
Let me tell you a story first .
I am working for Sony and our boss Dr. Doi (http://en.wikipedia.org/wiki/Toshitada_Doi) is replaced by another executive. We get summoned to Tokyo and through back channel hear that the new guy is a pain in the you know what and is going to scream our heads off. We are talking about board member of Sony here. We go to the meeting and he shows up together a few other people. Immediately he starts to pound on us like you would never see in a corporate world. Very rude and obnoxious and angry. Japanese executives don't ever act this way but he thought he should act "western" to show us he knew our ways. After a few minutes he finishes and everyone looks at my boss and I expecting some explanation. Without looking at each other, both of us instinctively returned the favor by acting Japanese: saying absolutely nothing. A minute goes by. You could cut the tension with a knife. I mean who would dare not answer a Japanese senior executive? But we sat there. Another minute goes by and still silence. By then the low-level Japanese who usually say nothing start to get nervous and we are still holding our silence. What happened next was remarkable: the new Japanese executive in very clam and friendly tone said, "of course we know you are doing excellent work. Just work with our finance department to forecast your expenses better." And the meeting ended right then and then!
You should have seen the reaction from our counterparts when the exec left. They could not believe the outcome. Of course we were quite relieved and happy and were thankful that our years of working Japanese had paid off in turning the tables on them as the exec was trying to do with us.
Why am I telling you this? No particular reason. OK, I kid . I know how to respond to AJ. When it is your turn to deal with him, you can answer him the way you want. I have already answered his question in this thread but he keeps asking about it. It is who he is. When you learn that as well as I, you will understand.
Now, if you also missed the answer and are unwilling to look back for it, ask it in your own language and I will explain. Don't look for it as a response to AJ. I need him to stew just like the rude Japanese exec.
Why am I telling you this? No particular reason.
Like your constant references to what M&M did when the argument is the BS Paper results, or asking people about them taking tests (since at least '09) when discussing your particular test results, or.....haha, just kidding.
I know how to respond to AJ.
Great, Would you mind addressing the lack of any transparency data for the BS test system, the speakers, the switching software (especially given the recent online test fiascos with switching artifacts yielding false positives), why Meridian continues to make claims about 16/44 transparency, why no 16/44 TPDF version of the BS test track exists, when you will submit your ABX logs to AES, etc, etc?
Btw, you do realize you are sitting on a gold mine of valid data with those unsupervised online ABX results, yes? Submitted as an AES paper, along with the BS awards paper, would be a double whammy for all those skeptical deaf folks who settle for inferior 16/44 due to their inferior trained "listening" skills. Like JJ.
It is who he is.
I do like turntables. Even though I own no records.
Have a good night amir, you know I always have fun conversing with you.
cheers,
AJ
An electrical engineer designing circuits, a PCB layout person, an FPGA programmer/designer, etc. are all "audio engineers" but nothing in their training or work gives them expertise in listening test.
It is a lay assumption that an "audio engineer" is an expert listener. They are not as I explained to Arny earlier.
...
Please don't keep posting this stuff. Good grief...
Good grief indeed.
I can't recall seeing the term 'audio engineer' mean quite what Amir says....but, hey, 'what happens in private industry, stays in private industry', right? and I'm just a layman. I would suggest, though, as someone who has some scientific paper writing experience, that maybe Meridian should have anticipated this and been more explicit if they meant 'electrical engineers who design and build audio hardware and software but had no substantial experience whatsoever in discriminating subtle audio differences prior to this experiment'
...but had no substantial experience whatsoever in discriminating subtle audio differences prior to this experiment'
That is what "non-expert listeners who received minimal training." means in the industry/research community.
Let me tell you a story first .
No....I think not.
Rather, let me put you back on ignore here, too. I'll go back to enjoying the show. The choice bits of your clown show that others quote , will suffice to tell me whether you are trying to divert the thread.
Let me tell you a story first .
No....I think not.
Too late!
Rather, let me put you back on ignore here, too. I'll go back to enjoying the show. The choice bits of your clown show that others quote , will suffice to tell me whether you are trying to divert the thread.
You take care Steven. Sorry to have caused you stress on two forums.
So, for the love of the flying spaghetti monster, could you please finally answer these simple questions amirm?
The main points:
- audibility of filters operating at 21+ kHz with
your hearing (please don't give me that training shtick, tell us what you hear, how you detect these filters etc.)
- what system you are using (I've read something about a laptop and headphones, which? what was your DirectSound configuration in Windows, etc.? Include whatever details you think could even remotely cause false positives.)
Also:
Show me one of "those tests" that has properly processed files, but you still hear a difference between them.
Then tell me what difference you hear, and where if applicable, and with what equipment and settings (player, DirectSound playback format if that is your API of choice, ...).
If this checks out then I will take a closer look (I did take a look at some of the files mentioned in the train wreck thread before - they were faulty) and ABX them.
This is what people usually do in this forum, by themselves, because they're interested in the truth. Are you?
And:
Anyway, if you finally answer the questions further above I would be highly interested in preparing a 21+ kHz filter test for you where you cannot, let's say, produce false positives as easily. Are you up to putting your money where your mouth is?
Wow, so now audio engineers who work with audio daily - according to you - are not trained listeners?
The rather obvious issue here is that not all audio engineers do the same work, receive the same training, etc.
For example a person with no formal training at all ever can spend a bit of time behind a mixing console and call himself an audio engineer, especially if those hours net a top 40 hit.
So can a power switching circuit designer who designs a switch mode power amp. So can a guy who operates a diffusion oven for a chip factory that makes audio chips.
My friend Clark has made a nice life for himself and his family, a lot of which involved critically listening to audio systems and teaching "Audio Engineers" how to critically listen to audio systems.
I've been long lamenting over the fact that engineering degrees, even degrees with audio pretensions are granted to people with zero training in psychoacoustics.
The poster boys for this can be seen all over the audio profession - even people who make strong pretenses about their strong pyschoacoustics background who write dozens of AES and IEEE papers,
none of which are backed up by even jus tone proper psychoacosutical test. I'd make up a list by looking at an index of papers in the JAES but Amir would burst a blood vessel! ;-)
So, for the love of the flying spaghetti monster, could you please finally answer these simple questions amirm?
Day one of "Dealing with Amir" 101: The clearer, the more revealtory the question, the less likely you are to receive a clear answer. They don't call him "The Dancing Man" for nothing! ;-)
IME this sort of behavior can get you into surprisingly high places in business, but usually it doesn't last forever.
No, I started up pretty "deaf." I remember being shocked that I could not hear much difference between 128 kbps MP3 and the source. That bothered me . So I started on a path to become a trained listener and after literally hundreds of hours of testing, and tons of education learning about psychoacoustics and algorithms in audio processing, all of a sudden found myself in a super unique situation of outperforming everyone else around me in such tests. I have never liked the phrase but everyone would call me the "golden ear."
And let's distinguish between hearing and listening. My hearing is shot due to age. I am embarrassed to say that I can't even hear 12 Khz well. The reason I do well still I think is that I focus on what I think technically matter and combine that with my training to look for small differences. I suspect people with my training but intact frequency response will be able to do a lot better. I actually ran into one such person at Microsoft who was working with a partner of ours. He was hearing high frequency distortions I could not. We hired him immediately . And I moved off from doing a lot of the listening tests myself.
Let me confess. I had no idea I could do this after so many years after retirement. I just ran the tests because people like Arny kept egging me on so I gave it a try.
Good morning amir,
Something about this little story has been bugging me a bit, perhaps you could enlighten us.
First, let me state, that if I could outrun Usain Bolt, the entire world would know about it. For obvious reasons, if I claimed to be able to do so, unsupervised, only in my backyard, by myself, no witnesses or oversight, folks might rightly be a bit skeptical and want to see a demo, maybe at the Olympics or some other similar witnessed, supervised, documented event, on camera, with electronic timing, etc, etc, etc.
Ok, so you claim above to have become this elite aural athlete, a real Golden Ear (per your coworkers). Seemingly many, many years ago (prior to retirement), you could "listen" to 16/44 (which had been around for a lot longer) and hear artifacts.
This certainly would predate M&Ms 2007 myth-buster.
Is there any record anywhere, of you demonstrating this elite aural athleticism, at say, the fully documented listening Olympics, supervised, to your peers? Or anyone (Obviously there was no mention in the +/-10% volume debacle on AVS in 2009)?
Or is the very first demo to the world earlier this year (2014), when you the ex-MS exec, sat at your Windows pc (possibly with your occasionally barking dog as the sole witness) and "passed" Arnie's corrupt ABX windows computer file ABX online test? And all similar subsequent ones of course.
Just curious, is all. Thanks.
cheers,
AJ
OK, finally I get to this technical point:
[...]
Having read the Stuart paper and prior citations, I am starting to think this may be due to time domain impact. It certainly is not frequency domain since I can't hear the ultrasonics. Just as pre-echo is a very audible time domain artifact (although created in frequency domain), maybe these filters act the same way.
After reading this I get an even stronger sense that you must have been cheating in a lot of the logs you posted, or you have some seriously anomalous hearing (regarding detecting lossy compression).
The filters used have linear phase. The pre-echo is no echo but filter ringing at over 21 kHz that only acts on the energy that is up there.
If you can't hear well past 12 kHz then how would you detect the ringing of a filter at over 21 kHz? So please explain why this filter ringing is of concern for you.
The answer is right in what I wrote but since it was misunderstood, I am going to explain the basics.
What you describe is not "ringing" but passband ripple. Yes, it is sometimes called ringing but given the fact that I was talking about time domain (see the new highlight in red), you shouldn't have gone to frequency domain.
There is no disagreement there. The paper makes it clear and I have said the same that the filters are near perfect in
frequency domain:
For both FIR lters, the ripple depth over the passband was a maximum of 0.025 dB, and the stopband attenuation was 90 dB.Note the correct terminology of "ripple" not ringing. But yes, that is as ruler flat as we need it with just +-.025db variation and aliasing truncated by 90 db. I wouldn't be able to hear that 0.025 db variation if it were in the range that I can hear. And this as you say is in the range that I can't hear. So again, no disagreement that that the frequency domain analysis doesn't indicate why this could be audible.
What followed in my post was a hypothesis. It is a hypothesis that is stated in the paper as I indicated. As any hypothesis, it is being offered as a potential answer, not proof. There is foundation in it but not one that I have personally investigated and hence the phrase I used:
"I am starting to think..."What is that thinking? It is what I described. What is happening in time domain. I created two filters in Matlab using the same 0.025 db passband ripple and 90 db out of band rejection. The first one is very similar to what Stuart used in his study (not identical because I am not performing any optimization):
.
The way you refer to pre-echo as "not being an echo" is totally nonsensical. Of course it is not an echo. It is *pre*-echo as it happens prior to the signal itself. From our old friend the wiki: http://en.wikipedia.org/wiki/Pre-echo (http://en.wikipedia.org/wiki/Pre-echo)
Pre-echo (not to be confused with reverse echo) is a digital audio compression artifact where a sound is heard before it occurs (hence the name). It is most noticeable in impulsive sounds from percussion instruments such as castanets or cymbals.Hey, what do you know? It says the same thing I did .
The psychoacoustic component of the effect is that one hears only the echo preceding the transient, not the one following – because this latter is drowned out by the transient. Formally, forward temporal masking is much stronger than backwards temporal masking, hence one hears a pre-echo, but no post-echo.It did it again!
So as you see, the whole confusion is due to thinking we are talking about filter ripple and frequency domain, when in multiple references it was clear that I was talking about time domain. And hence, my hearing limitation in frequency domain is no barrier to hearing such a phenomena. Learning to hear pre-echo even in minute amounts is a skill I had to develop to hear compression artifacts. So if that is what is at play, then that is the reason I may be hearing the difference.
Stuart makes multiple references to the same thing:
These parameters were chosen to o er a reasonable
match to the downsampling lters used in good-
quality A/D converters or in the mastering process;
we wanted to minimise the ripple depth and max-
imise the stop band attenuation in order to reduce
audible ringing artefacts, as described by Lagadec
[31].Have you read Lagadec's paper? If not, I suggest you do so. It explains the above and includes with it (informal) listening test observations of the impact of such ringing.
When the analysis was restricted to just the high-
yield audio sections, performance was signi cantly
better for the 48-kHz lter than for the 44.1-kHz l-
ter. This is perhaps surprising given that the difference
in spectral content between the filtered signals
was between 22050 Hz and 24000 Hz. A time-domain
explanation could be that the length of the 44.1-
kHz lter was longer than the 48-kHz one: 4.25 ms
compared with 3.9 ms; this could have resulted in
350 s less pre- and post-ringing, potentially render-
ing the 44.1-kHz [sic I think he means 48 khz] filter less audible.
This explanation is consistent with the idea of time smearing of ne
temporal details mentioned earlier [1, 14].So now you see why I needed a bit of time to respond.
The above is also the reason I say that if CD had picked 48 Khz as the sampling, I would have no beef with it. We have plenty of room to implement our filter above 20 Khz. But by picking 44.1, it leaves us a small margin forcing sharper filters and more time domain ringing.
Applying the same analysis to the sharp cut-off filter that was tested in the listening test gives us 9.5-2.5 = 7 milliseconds. In other words the span of ringing is now about 5 times wider in time than it was in the short cut off.
Is this audible? I gave a reference in my post as to how it might be: pre-echo. In psychoacoustics there is a concept of temporal masking. If you hear a loud sound, what follows it very quickly may not be audible. So if we look at our impulse itself, it creates a pretty nice shadow over what is to follow it. But looking behind it, there is impulse (assuming silence for now) and backward temporal masking is quite limited. This could, could, make the pre-ringing audible.
Note that since this is in time domain, it has nothing to do with my ability to hear high frequencies.
That bit doesn't make any sense. The ringing is at the filter cut-off frequency: 24kHz. If you can't hear 24kHz, you can't hear 24kHz. The fast it's before an impulse, or fades in, doesn't help you.
All audio samples go through the above transformation, not just high frequency ones. My truncated frequency response has no bearing on this situation unless the test tones are only above my hearing which is not the case in this musical segment. I am pretty sure I can hear what music is playing .
Your truncated frequency response matters because the pre-echo is always 24kHz, and the frequencies in the musical segment matter, because unless they get near 24kHz (i.e. have signal components or transitions or tones or clicks that reach up that far) they don't excite it. You won't see it.
It's not to say I reject this theory completely. It's been advocated many times. If there's something a little wrong anywhere in the system that makes ultrasonic signals fall down into the audible band in some way, then less pre-ringing could be audibly better, even when it's originally only at 24kHz.
But you need something that makes the 24kHz signal drop into the audible range before you can start talking about temporal (un)masking. Most of the things that could do that are things you don't want in a decent audible system.
Cheers,
David.
P.S. You've got 3kHz between 19kHz and 22kHz with CD's 44.1kHz sampling. You can have your gentle filter. As I'm sure you know, Bob will sell you one on the playback side. It'll cut so early (18kHz IIRC) and gently that it will removing all that nasty 22kHz ringing, even if it's in the source.
As I'm sure you know, Bob will sell you one on the playback side. It'll cut so early (18kHz IIRC) and gently that it will removing all that nasty 22kHz ringing, even if it's in the source.
I have a much easier solution that doesn't require new hardware from anyone: get the high sampling rate version of the content. Problem solved. You all can do the same if you choose. Tell me again why you guys are arguing.
Tell me again why you guys are arguing.
Again?! Isn't 18 pages of it sufficient?
I have a much easier solution that doesn't require new hardware from anyone: get the high sampling rate version of the content. Problem solved. You all can do the same if you choose. Tell me again why you guys are arguing.
The false puffiness over the fear that SRCs will cause audible artifacts in 5 of 9 attempts under specific criteria, when far larger issues remain unresolved in the process as they relate to what is actually provided to the buying public?
P.S. You've got 3kHz between 19kHz and 22kHz with CD's 44.1kHz sampling. You can have your gentle filter. As I'm sure you know, Bob will sell you one on the playback side. It'll cut so early (18kHz IIRC) and gently that it will removing all that nasty 22kHz ringing, even if it's in the source.
It has already been pointed out on this thread that real world CD players using actual real world DAC chips have low pass filter transition bands that are 2-3 KHz wide, and not the 0.5 KHz transition bandwidth that Meridian is trying to tell us characterize a "Typical Digital Filter in a HiFi Playback system". You can count on amateur experimenters to not catch this detail or understand its significance.
Whatever wave files that amateur listeners are actually listening to and with what options is mystery meat since they will still use the old Foobar Release 1.x ABX comparator, prepare their own test files, and won't put them up for independent examination.
Amateurs also often fail to use standard terminology while characterizing ringing which generally involves 10% variations - 10% below peak (90%) to a 10% or less residual.
The above is also the reason I say that if CD had picked 48 Khz as the sampling, I would have no beef with it. We have plenty of room to implement our filter above 20 Khz. But by picking 44.1, it leaves us a small margin forcing sharper filters and more time domain ringing.
Glad to see we're back to filters
I finally got the (interesting) paper and am still in the process of digestion. Plenty of questions.
First of all I'd like to compliment the authors with an unprecedented promotion of Haydn string quartets!
The paper doesn't IMO really separate the issues of dither and filtering. Filtering alone would be Conditions 1 and 4, although I can't find info how the processed signal was converted to 24 bits to feed the monitors (truncation, dithering?).
Am I the only one who thinks that a 90 dB stopband attenuation isn't spectacular ? Perhaps it's good enough, but when testing for very low level artifacts almost anything might matter. What about the <15 dB noise of the speakers (spec from website)? It would have been nice to see its spectrum in Fig.3.
About pre-ringing: when comparing a hi-res signal and a filtered (SRC) version in a null-test, there is no significant difference (frequency and time domain) in the passband. I would like to see more evidence that the >20kHz ringing is audible. I remain skeptical.
In 4.4 the authors announce future tests which include "TPDF dither with and without noise-shaping filters such as those used to limit noise in DSD playback, minimum-phase filters and apodizing filters". Since the processing is done in Matlab, they could have done those tests right away, IMHO. I'd like to see tests that not only determine which formats are not good enough, but also investigate which format can be really transparent (excluding abuse by incompetent engineers). As a recording engineer I'd like to know if 24/96 is good enough, or if there is evidence that we need to go even higher.
Here is an older pic i used for something different but it may show where the terrific ringing happens with different filters.
(http://i59.tinypic.com/zim2ya.png)
Tell me again why you guys are arguing.
Again?! Isn't 18 pages of it sufficient?
I would think so and I am looking at ending my participation. It is a shame though that I can't get people to agree with the simple business principal that high resolution downloads are happening. And by downloading such bits I am free of worrying whether downgrading them is audible or not.
I have a much easier solution that doesn't require new hardware from anyone: get the high sampling rate version of the content. Problem solved. You all can do the same if you choose. Tell me again why you guys are arguing.
The false puffiness over the fear that SRCs will cause audible artifacts in 5 of 9 attempts under specific criteria, when far larger issues remain unresolved in the process as they relate to what is actually provided to the buying public?
I have no fear. I am simply avoiding the problem altogether. And I don't care about that misdirection to change the topic. How a piece of music is produced is invariant to this discussion. However it is created, a further conversion to 16/44.1 is not necessary. Why do you guys have such a hard time with this concept? For CD that conversion is mandatory since the format requires. With digital distribution there is no such restriction so we can and are getting the pre-converted bits.
I am simply avoiding the problem altogether.
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
What problem? 16/44 is transparent, per award winning digital luminary BS/Meridian.
You certainly can't demonstrate any ability to audibly detect this imaginary "problem", not supervised. Never have, never will.
cheers,
AJ
For CD that conversion is mandatory since the format requires.
Yep and audibly completely transparent. There is not a whit of evidence to suggest CDs aren't. So anything that cost more for the same 2 channels with zero audible benefit, is a $cam.
cheers,
AJ
I have a much easier solution that doesn't require new hardware from anyone: get the high sampling rate version of the content. Problem solved. You all can do the same if you choose. Tell me again why you guys are arguing.
The false puffiness over the fear that SRCs will cause audible artifacts in 5 of 9 attempts under specific criteria, when far larger issues remain unresolved in the process as they relate to what is actually provided to the buying public?
For reasons we can only speculate about, Stuart and the hi rez cheerleading squad (which maybe should be rebranded as the Redbook Fearmongering Corps) would rather focus on eliminating the ant in the room. It's odd because surely there is money in upgraded *mastering* and in *room treatment/correction*, things which would address the real problems with modern home audio.
With digital distribution there is no such restriction so we can and are getting the pre-converted bits.
Ok amir, you keep (Red Herring) challenging other folks to take audio tests, including myself. I'll bite. I'll do a Hi-Rez comparison just like in the BS paper.
Where on the 2L digital distribution site can I find a restricted RPDF downsampled CD 16/44 version of the pre-converted bits Haydn track?
There is such a thing being distributed to end users as the BS paper has concocted, yes? This non-transparency distributed music isn't just an audiophile imagination bogeyman, right?
I'll get my transparent processing Meridian 518 player warmed up while awaiting your answer.
cheers,
AJ
It is a shame though that I can't get people to agree with the simple business principal that high resolution downloads are happening.
Oh, don't worry, we agree with that part. But marketing and sales has nothing to do with real benefits, something an ex-Microsoft person should be very intimate with.
This part of the BS paper always struck me as odd when testing something which quite possibly is related to very high frequencies, since 65 year-olds are typically expected to have high frequency loss:
"Eight listeners took part in the test, seven of whom were male. Most of the listeners were audio engineers, and their ages ranged from 25 to 65. All reported normal hearing, although this was not tested formally."
But I realize now the 65 year old, at the time of the testing, was most likely Bob Stuart himself [b. 1948]. Assuming he was also responsible for preparing the individual song segments he'd be in a unique position to be better prepared to listen for specific things like level mismatch, timing misalignment, sub-optimal dither artifacts including noise modulation, etc. and could have designed the song segment timings to make these specific issues particularly easy to ID.
I have a much easier solution that doesn't require new hardware from anyone: get the high sampling rate version of the content. Problem solved. You all can do the same if you choose. Tell me again why you guys are arguing.
The false puffiness over the fear that SRCs will cause audible artifacts in 5 of 9 attempts under specific criteria, when far larger issues remain unresolved in the process as they relate to what is actually provided to the buying public?
For reasons we can only speculate about, Stuart and the hi rez cheerleading squad (which maybe should be rebranded as the Redbook Fearmongering Corps) would rather focus on eliminating the ant in the room. It's odd because surely there is money in upgraded *mastering* and in *room treatment/correction*, things which would address the real problems with modern home audio.
It appears there is no need to speculate because the AES is working with the industry to bring hi-res to market. Consider the following two links:
http://www.aes.org/events/137/press/?ID=265 (http://www.aes.org/events/137/press/?ID=265)
http://www.aes.org/events/137/specialevents/?ID=4222 (http://www.aes.org/events/137/specialevents/?ID=4222)
Then take a look how Sony is marketing it.
http://discover.store.sony.com/High-Resolution-Audio/ (http://discover.store.sony.com/High-Resolution-Audio/)
It appears there is no need to speculate because the AES is working with the industry to bring hi-res to market. Consider the following two links:
http://www.aes.org/events/137/press/?ID=265 (http://www.aes.org/events/137/press/?ID=265)
http://www.aes.org/events/137/specialevents/?ID=4222 (http://www.aes.org/events/137/specialevents/?ID=4222)
Yes, it makes perfect sense that DEG, a group that is almost exclusively
MCH (http://degonline.org/about-us/) would want to promote the benefits of "Hi-Rez" audio. Given that there is not a whit of evidence it can be heard in 2ch...and amirs other luminary, Dr Floyd Toole, found that audible issues (loudspeaker) discrimination went
down, as more channels were added to the soundfield. So much so, that to make things easier, they test in mono.
I wonder if they use "Hi-Rez" mono??
Someone we know says they've been there....
cheers,
AJ
The answer is right in what I wrote but since it was misunderstood, I am going to explain the basics.
You don't need to explain the basics to me.
I know signal processing basics, let alone science basics.
So let me correct where you get the basics wrong...
What you describe is not "ringing" but passband ripple.
No, I don't use technical terms ambiguously or even with the wrong meaning.
I described filter ringing at over 21 kHz and not passband ripple.Yes, it is sometimes called ringing but given the fact that I was talking about time domain (see the new highlight in red), you shouldn't have gone to frequency domain.
Given that fact it seems you don't know what you're talking about. Again, for the nth time, the filter shows
ringing at over 21 kHz as visible in the impulse response (although frequency domain representation is equivalent anyway).
There is no disagreement there. The paper makes it clear and I have said the same that the filters are near perfect in frequency domain:
For both FIR lters, the ripple depth over the passband was a maximum of 0.025 dB, and the stopband attenuation was 90 dB.
Note the correct terminology of "ripple" not ringing. But yes, that is as ruler flat as we need it with just +-.025db variation and aliasing truncated by 90 db. I wouldn't be able to hear that 0.025 db variation if it were in the range that I can hear. And this as you say is in the range that I can't hear. So again, no disagreement that that the frequency domain analysis doesn't indicate why this could be audible.
No, you are confused.
I'm not talking about ripple but ringing. I never even mentioned ripple.
Also, 0.025 dB ripple does not mean +/- 0.025 dB. Ripple already specifies the max. variation.
What followed in my post was a hypothesis. It is a hypothesis that is stated in the paper as I indicated. As any hypothesis, it is being offered as a potential answer, not proof. There is foundation in it but not one that I have personally investigated and hence the phrase I used: "I am starting to think..."
Yes, and I asked you
how you could hear 21+ kHz ringing of a linear phase filter, that operates at 21+ kHz (except for the ripple, which seems negligible so I didn't even mention it in the first place).
What is that thinking? It is what I described. What is happening in time domain. I created two filters in Matlab using the same 0.025 db passband ripple and 90 db out of band rejection. The first one is very similar to what Stuart used in his study (not identical because I am not performing any optimization):
{img}
This is for the 48k sampling case. We are filtering everything above 24 Khz and starting to do that at 23.5 Khz as the paper mentions. This is not the frequency domain response which would be boring in its flatness as described above and in your post. But rather the impulse response of the filter.
Now we see "ringing." The center is our signal that we fed to the filter (a sharp spike). What we have now is pre- and post-ringing. The impulse is replicated prior to its appearance (pre-ringing) and after its appearance. The math dictates that it behave this way. Indeed this is a computer simulation based on the mathematics of the signal processing.
Before I say more, let's look at the same filter, but this time we relax the constraints so that the filtering starts at 21 Khz instead of 23.5. In other words, instead of giving the filter just 500 Hz to go from max to -90 db, we give it 3,000 Hz to do the same. The picture changes dramatically:
{img}
Don't worry about the details of the graph. The scales are different so this one is much more magnified. Pay attention to red arrows here and in the previous graphs. In this instantiation, ringing substantially subsides 0.2 milliseconds and 1.7 milliseconds. So the difference is roughly 1.5 milliseconds. This is the duration of our "distortion" if you will.
Applying the same analysis to the sharp cut-off filter that was tested in the listening test gives us 9.5-2.5 = 7 milliseconds. In other words the span of ringing is now about 5 times wider in time than it was in the short cut off.
All of this
was already pointed out in #362 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881524), but with more accurate numbers.
And no, this is
no distortion. This is
21+ kHz ringing of a linear phase low pass filter.
The worst case (which Meridian claims to reflect real world filters - it doesn't seem to..) gives you 4.5ms for the total pre-ringing part. This ignores that the coefficients drop rapidly, for example in this worst case, to -20 dB within a small fraction of a ms.
Also, at the risk of repeating myself again, this ringing is at 21+ kHz. The filter operates at 21+ kHz and only on the energy that is left up there.
Is this audible? I gave a reference in my post as to how it might be: pre-echo. In psychoacoustics there is a concept of temporal masking. If you hear a loud sound, what follows it very quickly may not be audible. So if we look at our impulse itself, it creates a pretty nice shadow over what is to follow it. But looking behind it, there is impulse (assuming silence for now) and backward temporal masking is quite limited. This could, could, make the pre-ringing audible.
First of all, echoes are generally understood to be reflections that the listener can
perceive individually as distinct events and therefore require a
delay of several tens of ms to multiple seconds. That's why I dislike the term "pre-echo".
The filter above doesn't fit any of this, especially not the delay. Remember.. linear phase.
It may come as a surprise to you, but an ideal single tone rings
infinitely at its specific frequency. Why am I telling you this?
Well, if you sum up tones up to a specific point, let's say 22.05 kHz, you will get an impulse response (and therefore FIR lowpass filter) that also rings
infinitely. The
ringing is that of the sharp cutoff at 22.05 kHz. Everything below that is
passed perfectly without "distortion" or delay.
See #364 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881529).
Note that since this is in time domain, it has nothing to do with my ability to hear high frequencies. All audio samples go through the above transformation, not just high frequency ones. My truncated frequency response has no bearing on this situation unless the test tones are only above my hearing which is not the case in this musical segment. I am pretty sure I can hear what music is playing .
Ringing at 21+ kHz of a filter... How often do I need to repeat this?
Take a look at #364 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881529). It shows a sine sweep from 20 to 23 kHz.
As you subtract the filtered signal, nothing is left up to wherever the filter starts rolling off, which is above 21 kHz for the filters above.This is a
time domain waveform.
So how do you hear a difference? The way you refer to pre-echo as "not being an echo" is totally nonsensical. Of course it is not an echo. It is *pre*-echo as it happens prior to the signal itself.
Exactly, it is not an echo in any sense hence I prefer the (imho more correct) term
filter pre-ringing. Now if you weren't so confused about how filters operate...
The psychoacoustic component of the effect is that one hears only the echo preceding the transient, not the one following – because this latter is drowned out by the transient. Formally, forward temporal masking is much stronger than backwards temporal masking, hence one hears a pre-echo, but no post-echo.
It did it again!
You mean you
demonstrated your ignorance once again? The context of passages from the wiki you quoted talks specifically about lossy compression that makes use of
frequency domain transformations. Simplified (just for you amir!) this is like chopping the signal into many frequency domain chunks.
Guess what this means?
Many steep bandpass filters that cover the whole audible range from 20 Hz to 20 kHz. This is an
entirely different issue than a lowpass filter operating at 21+ kHz.
Do you understand the difference?
And by downloading such bits I am free of worrying whether downgrading them is audible or not.
There is some logic in there. There is also some logic in capturing the microphone output as losslessly as possible, even if we can't hear it. DXD looks like a safe format for that, since most microphones don't have a bandwidth of more than 100 kHz. Higher-res downloads are more expensive and the claim is that the higher the bitrate, the better the sound quality. From older papers I got the impression that Bob Stuart seems to think that audible differences disappear above 24/96. Besides the satisfaction of "having the studio quality", what would be the point of buying higher formats ?
Some prices for an album in euro, from 2L.no , the provider of the samples of this paper:
stereo:
mp3 09.00
16/44.1 15.00
24/96 21.00
24/192 23.00
DSD64 28.00
DSD128 33.00
DXD 37.00
multichannel:
24/96 25.00
DSD64 33.00
(sorry, can't get tabs to work)
It appears there is no need to speculate because the AES is working with the industry to bring hi-res to market. Consider the following two links:
http://www.aes.org/events/137/press/?ID=265 (http://www.aes.org/events/137/press/?ID=265)
http://www.aes.org/events/137/specialevents/?ID=4222 (http://www.aes.org/events/137/specialevents/?ID=4222)
'the full range of sound'...hmmmm. Do they mean *sound* as in *what can be heard*?
Then take a look how Sony is marketing it.
http://discover.store.sony.com/High-Resolution-Audio/ (http://discover.store.sony.com/High-Resolution-Audio/)
This is actually clever of them:
Discover subtle details and artistic nuances in your favorite
music that you’ve never heard before. Feel the power
and presence of a live performance in your living room.
Or experience what it’s like to sit in on a live studio recording.
It’s all possible with the superior quality of High-Resolution Audio.
With quality greatly surpassing that of MP3 and CD, the difference is clear.
So, to cover their asses, they could claim to mean the slightly more revealing noise floor in the most quiet parts of a recording, or the very loudest parts of a recording with a dynamic range >96dB (probably a *live* recording) or the slight effects of mediocre antialias or antiimaging filtering
The only part where they back into bad habits is implying that the 'quality greatly surpasses' CD and that it translates to 'audible performance greatly surpassing CD's' and that the difference is 'clear' -- like, no training required, no special effort at 'discovery'.
And by downloading such bits I am free of worrying whether downgrading them is audible or not.
There is some logic in there. There is also some logic in capturing the microphone output as losslessly as possible, even if we can't hear it. DXD looks like a safe format for that, since most microphones don't have a bandwidth of more than 100 kHz. Higher-res downloads are more expensive and the claim is that the higher the bitrate, the better the sound quality. From older papers I got the impression that Bob Stuart seems to think that audible differences disappear above 24/96. Besides the satisfaction of "having the studio quality", what would be the point of buying higher formats ?
IIRC Stuart thinks the 'problems' with Redbook disappear at ~55kHz, ~20 bits (not sure about the bitdepth, it's whatever Fielder claimed in his papers) So either 88.2/24 or 96/24 , being common formats, were the the logical choices.
I don't recall him ever advocating more than that, as being either necessarily or useful.
It could however be useful for something like Plangent Processing, which relies on recorded ultrasonic bias tones to correct tape flutter.
On page 2 I find two rather conflicting statements and I'm not sure what conclusion the authors draw:
It has further been suggested [22, 23, 24] that listeners can discriminate timing differences of the order of 5 µs and below, which, if correct, would require a Nyquist frequency of 32 kHz or higher...
And below that:
However, it can not be assumed that the auditory system cannot extract time differences that are shorter than the periods between successive samples, even when convolved with a sinc function in D/A conversion (assuming a sufficient signal-to-noise ratio).
It is my understanding (references to JJ, David Griesinger e.a.) that "In Physics, the accuracy of timing is not determined by the bandwidth, but roughly by the product of the bandwidth and the signal to noise ratio".
Am I correct that the authors say that the time resolution argument based on sample rate and excluding SNR is not valid ?
Stuart has written papers that show that 52 kHz / 11 bits are the absolute minimum PCM channel using noise shaping, "capable of replicating the information received by the ear". 14 bits "ought to be adequate" to offer enough headroom. That would be 18.2 bits for a rectangular channel (TPDF, no noise shaping).
Tell me again why you guys are arguing.
Again?! Isn't 18 pages of it sufficient?
I would think so and I am looking at ending my participation. It is a shame though that I can't get people to agree with the simple business principal that high resolution downloads are happening. And by downloading such bits I am free of worrying whether downgrading them is audible or not.
I have answered this to my satisfaction, if not yours, already.
There are any number of inaudible objective improvements one could make to the recording and reproduction process. Most could fall into the "they can't make it worse so why not" category. Add them together, and you have spent all your money and all your life chasing numerical improvements with little or no audible benefit. Meanwhile the real improvements have passed you by. That is why we have to let the inaudible improvements pass us by, and go and chase audible improvements instead.
(The "you" in that statement is "you" the hi-res audio proponent or "you" the hi-res audio industry.)
So, why should we care about THIS inaudible imrovement? On what basis do you spend money and time on this one, but not 100 others?
Cheers,
David.
And by downloading such bits I am free of worrying whether downgrading them is audible or not.
There is some logic in there. There is also some logic in capturing the microphone output as losslessly as possible, even if we can't hear it. DXD looks like a safe format for that, since most microphones don't have a bandwidth of more than 100 kHz. Higher-res downloads are more expensive and the claim is that the higher the bitrate, the better the sound quality. From older papers I got the impression that Bob Stuart seems to think that audible differences disappear above 24/96. Besides the satisfaction of "having the studio quality", what would be the point of buying higher formats ?
Some prices for an album in euro, from 2L.no , the provider of the samples of this paper:
stereo:
mp3 09.00
16/44.1 15.00
24/96 21.00
24/192 23.00
DSD64 28.00
DSD128 33.00
DXD 37.00
multichannel:
24/96 25.00
DSD64 33.00
(sorry, can't get tabs to work)
Noting, once again, that they won't sell you the DXD 5.1 "master".
I wouldn't be surprised if amir was running now, after being unable to answer simple questions, demonstrating a lack of knowledge in even the basics and me asking him if he would do a test where he couldn't cheat as easily (of course he never answered this either).
And by downloading such bits I am free of worrying whether downgrading them is audible or not.
There is some logic in there. There is also some logic in capturing the microphone output as losslessly as possible, even if we can't hear it. DXD looks like a safe format for that, since most microphones don't have a bandwidth of more than 100 kHz. Higher-res downloads are more expensive and the claim is that the higher the bitrate, the better the sound quality. From older papers I got the impression that Bob Stuart seems to think that audible differences disappear above 24/96.
Hi Kees. Give it to you to see the reasonable argument. As to Stuart, no. He actually has far lower limits. Without noise shaping, and a rectangular channel, he advocates 20 bits at 56 Khz. Since 56 Khz doesn't exist, he says pick the next step up which would be 88 or 96 Khz.
Besides the satisfaction of "having the studio quality", what would be the point of buying higher formats ?
If it is higher than how the music was recorded/produced, then I have no use for it. If higher is what is recorded and mixed, then I like to get my hands no those bits as I don't trust anyone to do the right things in the chain to bring it down to something lower.
Some prices for an album in euro, from 2L.no , the provider of the samples of this paper:
stereo:
mp3 09.00
16/44.1 15.00
24/96 21.00
24/192 23.00
DSD64 28.00
DSD128 33.00
DXD 37.00
multichannel:
24/96 25.00
DSD64 33.00
(sorry, can't get tabs to work)
Yup. Everyone has a choice of what bits to get.
Tell me again why you guys are arguing.
Again?! Isn't 18 pages of it sufficient?
I would think so and I am looking at ending my participation. It is a shame though that I can't get people to agree with the simple business principal that high resolution downloads are happening. And by downloading such bits I am free of worrying whether downgrading them is audible or not.
I have answered this to my satisfaction, if not yours, already.
There are any number of inaudible objective improvements one could make to the recording and reproduction process. Most could fall into the "they can't make it worse so why not" category. Add them together, and you have spent all your money and all your life chasing numerical improvements with little or no audible benefit. Meanwhile the real improvements have passed you by. That is why we have to let the inaudible improvements pass us by, and go and chase audible improvements instead.
You mean I can't do both? How come?
And who is to say it is inaudible? MP3 artifacts are inaudible to many but I assume not to you and certainly not to me. What standard do we use for that?
So, why should we care about THIS inaudible imrovement?
I am not asking at all that you care. I am asking why people here and other forums go crazy, absolutely crazy at the idea that someone goes and downloads higher resolution bits. Why? How is bothering them? Read Krab's posts. Or AJ's. Or Arny's. Or Xnor. Or even our moderator. These are not logical reactions to me going buying high resolution audio. I know all of their arguments yet they keep repeating them. As you do here . Do they think I or others haven't hear them? Why keep repeating it and of all people tell me about it?
Can you answer the psychology of this? I mean I have this new guy Xnor in my life who is determined to say I have cheated. Cheated! How on earth did he decide it is appropriate to wake up one morning and decide his mission in life is to demonstrate that. And to what effect? What is driving him? You know him better than I. Culturally it is pretty far over the line to go there. I would think he would do that if he thought I had done him some personal harm. But to advocate getting the bits prior to conversion to 16/44.1? That is all it took for him to write 12-quote replies now? How did that become a priority for him?
Have people lost all sensibility?
I mean isn't it stupid of all people to take me on? I take double blind tests. I have built a career based objective science. I am an engineer. I talk about science and technology of audio. And this is the person that Krab says is going to put on "ignore" list? What is left to read then if it is not what I just explained.
On what basis do you spend money and time on this one, but not 100 others?
You are but one of a few reasonable people I have seen here. Can you explain to me? If not them, can you explain how you stand next to them?
Now I enjoy the banter so not saying this stuff to have you help me in that regard. But the whole affair is the most illogical thing there is. Let folks consume high resolution audio in peace. No harm will come to any of you. I promise. You will suffer less in life waking up to read my posts .
From such price lists the hi-re$ agenda is painfully obvious.
Good morning amir,
Good evening AJ. Today is your birthday. You get to go and celebrate with a good dinner? What? It is not your birthday? Well, it must be some cause for celebration since I decided to answer one your posts!
Something about this little story has been bugging me a bit, perhaps you could enlighten us.
Something is bugging me too and is along the lines of what I just wrote David. Why do you come here AJ? What all these posts? Why the anger and frustrations over someone buying high resolution audio? Can you think through your motivations and answer in your reply?
The rest of your answer in part 2.
I am asking why people here and other forums go crazy, absolutely crazy at the idea that someone goes and downloads higher resolution bits. Why? How is bothering them? Read Krab's posts. Or AJ's. Or Arny's. Or Xnor. Or even our moderator. These are not logical reactions to me going buying high resolution audio. I know all of their arguments yet they keep repeating them. As you do here . Do they think I or others haven't hear them? Why keep repeating it and of all people tell me about it?
I am not going crazy, in fact, I couldn't care less what files you download. You are also in no position to judge what is logical, given your previous posts.
You know all arguments, yeah right, but are unable to answer even simple questions?
Can you answer the psychology of this? I mean I have this new guy Xnor in my life who is determined to say I have cheated. Cheated! How on earth did he decide it is appropriate to wake up one morning and decide his mission in life is to demonstrate that. And to what effect? What is driving him? You know him better than I.
amir, for once try to stop being stupid.
I said that your own posting made it even more likely that you cheated. And your constant evasion of simple questions related to those tests support that assumption.
What is driving me? I already told you amirm. ... something you don't seem to be interested in.
Culturally it is pretty far over the line to go there. I would think he would do that if he thought I had done him some personal harm. But to advocate getting the bits prior to conversion to 16/44.1? That is all it took for him to write 12-quote replies now? How did that become a priority for him?
What are you even talking about? You are talking absolute rubbish.
I asked simple questions. You demonstrably evaded and went at length in your replies just to contradict me. You admitted to stereotyping. What is wrong with you, amirm?
I mean isn't it stupid of all people to take me on? I take double blind tests. I have built a career based objective science. I am an engineer. I talk about science and technology of audio. And this is the person that Krab says is going to put on "ignore" list? What is left to read then if it is not what I just explained.
But you demonstrably lack knowledge about basic science, logic, statistics and fail to make replies that are not riddled with fallacies, or evasion maneuvers.
You talk down on me, but take a look at the mirror. As someone has PM'd me: "he has no interest in intellectual honesty or scientific rigor". Well, I couldn't agree more and everyone can see it. (Although, given previous posts, I tend to think it is ignorance and an agenda rather than simply "no interest".)
And by downloading such bits I am free of worrying whether downgrading them is audible or not.
There is some logic in there.
Sorry Kees, don't see it. Amirlogic perhaps, not "logic" logic.
It may be audiophile logic to buy Clarity Cap MR (http://www.hificollective.co.uk/pdf/claritycapmr.pdf) for "worry free", non-"downgraded" audio, whether non-boutique caps are audible or not (the drum up some doubt method). Why? Because Clarity fabricated a singing cap themselves, which was, surprise, surprise, audible using blind tests in a University study. IOW, contrive a scenario that does not exist in real life, sell "worry free" widget to audiophiles based on this. Method sound familiar?
Some prices for an album in euro, from 2L.no , the provider of the samples of this paper:
stereo:
16/44.1 15.00
24/96 21.00
Yep, with no evidence the 24/96k version can be distinguished, since the 16/44 version must be RPDF dither doctored, heard in an iso-ward, with questionable system transparency, etc, etc, etc....
Not seeing the logic there at all.
cheers,
AJ
Is there any record anywhere, of you demonstrating this elite aural athleticism, at say, the fully documented listening Olympics, supervised, to your peers?
I imagine you have done special things in your work that are not public and can be verified this way, right?
But yes, there are hundreds of people at Microsoft that know about my listening abilities many of them first hand. Vast majority of them are not active on forums though.
I didn't think there would have been a need years later to prove that on some forum. I would have gotten notarized statements from witnesses otherwise.
This being the Internet, I trust you accept Internet type evidence. I did a search on AVS and quickly landed on a post by Ben Waggoner. Ben is quite a famous compressionist. You can google his name to find more. Anyway, he worked in my group and said this on AVS unprompted: http://www.avsforum.com/forum/114-hdtv-sof...ml#post13633874 (http://www.avsforum.com/forum/114-hdtv-software-media-discussion/1017012-dolby-digital-640kbps-just-good-better-than-dts-755kbps-3.html#post13633874)
Even Amirm has said that he cannot make this claim:
Surely you wouldn't imply that 640k tracks are indistinguishable from TrueHD, would you?
Amirm: "No I wouldn't"
Amir is also a golden-ears audio compression expert.
Transparency can be achieved by any of the right compression, content, and audience .
Digital Media Technology Insider with Microsoft
My compression blog[/color]
Or is the very first demo to the world earlier this year (2014), when you the ex-MS exec, sat at your Windows pc (possibly with your occasionally barking dog as the sole witness) and "passed" Arnie's corrupt ABX windows computer file ABX online test? And all similar subsequent ones of course.
Just curious, is all. Thanks.
Arny's corrupt ABX? I will let him address you on that.
I tell you what AJ. I will make a deal with you. I will fly to Florida to see you. You pay half of my expenses and I pay the other half. I will bring just my laptop and headphones and repeat the same tests I have post here. If I pass, you pay the other half of my expenses plus another $1,000 I give to charity. If I lose, I will do the reverse and you can decide what to do with the extra $1,000. What say you?
Alternatively you fly to Seattle and I take the test here in front of you and we only wager the $1,000.
Amir's, position as I understand it in all of this is as follows.
When he wants to buy music, he wants to buy the masters of the recording. If they only exist in 44.1, then fine. If it's 96 or 112 then fine. He doesn't want anything that's been up or down converted. If you think about it, that's his preference and as far as preferences go, it's whatever floats your boat.
How we each justify our preferences is an individual matter. There are rational, irrational, and any number of reasons why we choose things. A problem generally arises when we vocalize our reasonings to others. After all, not everything we say is going to make sense to the other person or there may even be errors in our thinking.
Amir has demonstrated to his satisfaction that he can key in on certain aspects of the conversion process with great success. That may well play apart in his internal justification process. He's aware that hi-res, even compressed, takes up more room and it costs more. Significantly in my opinion but to him that's irrelevant so long as what he bought is the master. It may well be that the master still suffers from loudness issues or even poor recording or mastering. But if it's the original then that's what matters.
The rest of what transpires are matters of verbal jousting. For me, the pity is that the complaints the public has had with the recordings are not going to go away. Look at those links I posted earlier and IMO, it's not about doing a better job with recording. Studios are going to put out what sells and in whatever format, moving the sliders around until they get something the customer approves. A crappy picture at 800x600 is still going to be crappy at mega resolutions.
Why do you come here AJ?
You mean why have I been here since '09? To discuss audio, one of my interests. Like this BS test.
Why did I bait you into coming here? I thoroughly enjoy our conversations, because your lack of answers, two-stepping, etc. do provide answers.
What do I also enjoy? Turning the tables on the elitists, bullies, illogical condescending snobs and highly self deluded "audiophiles" that permeate my audio hobby . Giving them healthy doses of their own medicine.
Nothing you could be cognizant of, figure out, much less worry about.
Now, about this highly contrived BS paper....
Oh and BTW, the thread is about the BS test, not "What Amir Bolt wants with audio files".
cheers,
AJ
FYI, I passed all of my double blind ABX tests using my laptop and headphones.
You probably think you're getting a hard time here, but you should take it as a compliment that no one has challenged you on this. The assumption must be that, given your employment history, you must know what you're doing. Anyone else who arrived on HA and made this claim would get a really hard time, due to a long history of drivers, sound cards, and transducers that don't take kindly to ultrasonic content. ABXing hi-res vs 16/44.1 on such crappy equipment is easy, and it's not because hi-res is audibly better.
Cheers,
David.
Are there some details on the system amir used? Which operating system/version, which headphones, which soundcard, which sound API, which DirectSound common sampling rate was configured (if applicable) ... ?
You spoke too soon David. As you see Xnor and AJ are on full bore inquisition with no regards to my prior experience.
I didn't answer you xnor because all of this has been hashed out on AVS Forum. And mzil gave you part of it. I know you haven't seen them but I thought you would at least pay attention to what David said.
So that you know what he is referring to, my team at Microsoft was responsible for entire Audio/Videos tack in Windows including DS, mixing and resampling pipeline, etc. I championed ditching the horrible one in XP and creating a proper one. I hired JJ to head that activity and the results were a major step up from XP.
Most of my testing was done on my rather new HP Zbook 14. I take these tests casually in our family room with TV on and such (hence the dogs barking). I like to use my Etymotic headphones because they seal so well. I was challenged on that so I repeated the test with my Shure IEM. To really put that argument to bed, I also ran a test with my uncomfortable Paradigm IEM. I have a stock audio stack in my laptop and have Foobar running in default configuration. Since all of the tests have been at 24-bit/96 Khz, I have set that as the audio property of the sound card. There is still dither added on the way in/out of the kernel stack but seeing how I had positive results, I didn't need to go with WASAPI or ASIO to eliminate that.
I have run Arny's Ultrasonic IM test and passed that (I post that in one of the threads here).
New version of Foobar ABX spits out the audio stack selection so here is David's test I ran:
foo_abx 2.0 beta 4 report
foobar2000 v1.3.5
2014-11-13 09:16:06
File A: limehouse_maximum_phase_100.wav
SHA1: 722dc26db8d4ce666dc03875b2c8d4570d22b521
File B: limehouse_reference.wav
SHA1: e8ad96830d23cad4bba5bf822ce875ae452b9e7c
Output:
DS : Primary Sound Driver
09:16:06 : Test started.
09:16:48 : 01/01
09:16:56 : 02/02
09:17:04 : 03/03
09:17:14 : 04/04
09:17:21 : 05/05
09:17:29 : 06/06
09:17:38 : 07/07
09:17:45 : 08/08
09:17:52 : 09/09
09:18:02 : 10/10
09:18:08 : 11/11
09:18:14 : 12/12
09:18:20 : 13/13
09:18:28 : 14/14
09:18:36 : 15/15
09:18:42 : 16/16
09:18:42 : Test finished.
----------
Total: 16/16
Probability that you were guessing: 0.0%
-- signature --
5b42b06c414b6ba77a3998695bf119a2d57663c0
Edit: Forgot to mention the OS. It is Windows 7 Pro.
BTW, you asked me what test to run. Start with the above. It is in the parallel link. Please post your answer here. AJ, same to you.
Is there any record anywhere, of you demonstrating this elite aural athleticism, at say, the fully documented listening Olympics, supervised, to your peers?
I imagine you have done special things in your work that are not public and can be verified this way, right?
But yes, there are hundreds of people at Microsoft that know about my listening abilities many of them first hand. Vast majority of them are not active on forums though.
I didn't think there would have been a need years later to prove that on some forum. I would have gotten notarized statements from witnesses otherwise.
This being the Internet, I trust you accept Internet type evidence. I did a search on AVS and quickly landed on a post by Ben Waggoner. Ben is quite a famous compressionist. You can google his name to find more. Anyway.....
Unfortunately Amir, those aren't the droids we're looking for again.
So I'll answer for you: No, you have zero evidence of demonstrating any such ability. Ever. Particularly for 16/44.
I tell you what AJ. I will make a deal with you. I will fly to Florida to see you. You pay half of my expenses and I pay the other half. I will bring just my laptop and headphones and repeat the same tests I have post here. If I pass, you pay the other half of my expenses plus another $1,000 I give to charity. If I lose, I will do the reverse and you can decide what to do with the extra $1,000. What say you?
Alternatively you fly to Seattle and I take the test here in front of you and we only wager the $1,000.
Unfortunately Amir, those aren't the droids we're looking for either.
Certainly wouldn't pass ITU-R BS.1116-2, or be submittable to AES. We wouldn't want to fall into that Hobbyist trap, right?
However, you are still welcome to accept my multiple invites to attend an audio show I exhibit at, have a beer with some banter. With your laptop.
cheers,
AJ
I didn't answer you xnor because all of this has been hashed out on AVS Forum. And mzil gave you part of it.
mzil gave me part of what? Your system, system configuration, what you precisely hear and when ...?
Are you confused again? If all of this information is available then why didn't you post a simple link to that particular post instead of posting pages of evasion and nonsense?
So that you know what he is referring to, my team at Microsoft was responsible for entire Audio/Videos tack in Windows including DS, mixing and resampling pipeline, etc. I championed ditching the horrible one in XP and creating a proper one. I hired JJ to head that activity and the results were a major step up from XP.
I gotta laugh because the resampler of Vista, 7 had absolutely horrible performance, such as when recording approaching an ENOB of ~8 bits. Also there have been so many reports about problems with the whole new audio stack (I've had absurd bugs as well) that it's not even funny anymore.
It took Microsoft roughly ~5 years to fix the most annoying problems and bad performance.
Most of my testing was done on my rather new HP Zbook 14. I take these tests casually in our family room with TV on and such (hence the dogs barking). I like to use my Etymotic headphones because they seal so well. I was challenged on that so I repeated the test with my Shure IEM. To really put that argument to bed, I also ran a test with my uncomfortable Paradigm IEM. I have a stock audio stack in my laptop and have Foobar running in default configuration. Since all of the tests have been at 24-bit/96 Khz, I have set that as the audio property of the sound card. There is still dither added on the way in/out of the kernel stack but seeing how I had positive results, I didn't need to go with WASAPI or ASIO to eliminate that.
[...]
Edit: Forgot to mention the OS. It is Windows 7 Pro.
Thanks. You haven't specified the IEM models, but most of these models have rolled-off treble (which doesn't matter much anyway since you said your hearing rolls off above 12k and you claim that the filter ringing at 21+ kHz somehow distorts audio below <12 kHz).
Now there is one thing left:
what did you specifically hear in these tests?New version of Foobar ABX spits out the audio stack selection so here is David's test I ran:
"Primary Sound Driver" could be anything, but doesn't matter anyway since, I'm sorry, I'm not trusting these logs for (what should be by now) obvious reasons.
So, anyway, are you up for a test where e.g. a spectrum analyzer running in the background wouldn't help you?(I'm still waiting for you to finally explain
what you specifically hear in these tests. Would make it easier to design a test that specifically targets your super-trained abilities.)
What say you?
Contact this guy in WA (https://home.comcast.net/~retired_old_jj/).
Have him setup/have someone run a valid, robust test. Use his home brewed speakers (the big ones with the ribbon), they are capable of what we need. Save some travel $$/headaches.
Post on Youtube (won't be any need for submitting to AES with the results you get). Or here.
Spend the $1k on CDs or maybe some therapy for these bizarre cravings.
cheers,
AJ
No, you are confused. I'm not talking about ripple but ringing. I never even mentioned ripple.
Oh? Did you not immediately follow up with this post?
Adding to the ringing post above:
(http://xserv.compress.to/xnor/audio/images/ringingchirp1.png)
The background signal (blue) shows a sweep from 20 kHz to 23 kHz.
The rest show this signal filtered with the following pass band edges:
magenta = 20.5 kHz
cyan = 21.591 kHz
red = 22 kHz
Stop band starts at 22.05 kHz in all cases.
Here the same graph when subtracting the filtered signals from the original:
(http://xserv.compress.to/xnor/audio/images/ringingchirp2.png)
Here's what happens when you mess up the filtering (see cyan line which doesn't null anymore with the original signal): png (http://xserv.compress.to/xnor/audio/images/ringingchirp3.png)
Where would I see "ringing" in those? You clearly say they are frequency sweeps from 20 to 23 Khz. Is that how we show the Impulse response and ringing? With a spectrum display from 20 to 23 Khz? You are telling us that those visualizations are the same as these I post?
(http://amirviews.smugmug.com/photos/i-q35sw3V/0/X2/i-q35sw3V-X2.png)
My graph shows ringing in time domain. Exactly the topic David and I were discussion. You clearly confused passband ripple in frequency domain with time domain ringing. It doesn't get any simpler than this.
Why do you come here AJ?
Why did I bait you into coming here? I thoroughly enjoy our conversations, because your lack of answers, two-stepping, etc. do provide answers.
What do I also enjoy? Turning the tables on the elitists, bullies, illogical condescending snobs and highly self deluded "audiophiles" that permeate my audio hobby . Giving them healthy doses of their own medicine.
There we go. The truth comes out. No interest in having a technical discussion about the topic at hand. A thread is created to discuss an AES paper and your priorities are "condescending snobs and highly self deluded audiophiles?" Why not go and waste forum bandwidth somewhere else where that is the topic?
Nothing you could be cognizant of, figure out, much less worry about.
Good that you know why I don't answer most of your posts. I am not remotely worried about them. So rant away my good friend.
Oh and BTW, the thread is about the BS test, not "What Amir Bolt wants with audio files".
Oh? I thought it was about
turning the tables on the elitists, bullies, illogical condescending snobs and highly self deluded "audiophiles" that permeate my audio hobby? Not so smart as the Oracle said in movie Matrix.
What say you?
Contact this guy in WA (https://home.comcast.net/~retired_old_jj/).
I don't want to wind up on his list of idiots by doing so. You can contact him though since you have so much passion in this.
cheers,
AJ
AJ I have been meaning to say this but keep forgetting. Your picture in your avatar has the wrong aspect ratio. It is squashed horizontally. Are you not even aware of such obvious flaws? Good grief man. I have not even seen the original but I am pretty sure it is wrong.
I don't want to wind up on his list of idiots by doing so.
You're not on it already? Haha, just kidding
Well, he does seem rather qualified for such endeavors, don't you think? Thought you were friends too.
Ah well. Your side could have finally swooned in victory over the stone deaf party poopers...but alas.
Another null wouldn't have mattered, your side rejects those.
You can contact him though since you have so much passion in this.
Nope, I'll try this again, even though I know you could never get it, but the onus is squarely on you/your camp.
Guess we'll just have to stick with Amir Geller, rather than Amir Bolt.
AJ I have been meaning to say this but keep forgetting. Your picture in your avatar has the wrong aspect ratio. It is squashed horizontally. Are you not even aware of such obvious flaws? Good grief man. I have not even seen the original but I am pretty sure it is wrong.
Here's the original but I'm just too lazy to fix the forum one.
Have a nice weekend Amir, now I think I'll go listen to some nice 4ch 16/44 CD. Since PSR at 320 kbps never took off and the MCH Hi-Rez stuff still tends to suck spatially.
cheers,
AJ
On page 2 I find two rather conflicting statements and I'm not sure what conclusion the authors draw:
It has further been suggested [22, 23, 24] that listeners can discriminate timing differences of the order of 5 µs and below, which, if correct, would require a Nyquist frequency of 32 kHz or higher...
And below that:
However, it can not be assumed that the auditory system cannot extract time differences that are shorter than the periods between successive samples, even when convolved with a sinc function in D/A conversion (assuming a sufficient signal-to-noise ratio).
It is my understanding (references to JJ, David Griesinger e.a.) that "In Physics, the accuracy of timing is not determined by the bandwidth, but roughly by the product of the bandwidth and the signal to noise ratio".
Am I correct that the authors say that the time resolution argument based on sample rate and excluding SNR is not valid ?
A goodly number of the cited references lack reliable substantiation.
For example reference 23 is
M. N. Kunchur. Audibility of temporal smearing and time misalignment of acoustic signals. Technical Acoustics, 17, 2007. You can read it here: http://boson.physics.sc.edu/~kunchur/paper...s---Kunchur.pdf (http://boson.physics.sc.edu/~kunchur/papers/Audibility-of-time-misalignment-of-acoustic-signals---Kunchur.pdf) .
If I can remember the critiques of it when it first came out, it is actually implicitly criticized in the new Meridian paper by this statement:
"However, it can not be assumed that the auditory
system cannot extract time diff erences that are
shorter than the periods between successive samples,
even when convolved with a sinc function in D/A
conversion (assuming a sufficient signal-to-noise ratio)."
and this relates to the following false claim in Kunchur's paper:
"...use of digital carriers limits the shortest resolvable time interval to about half the sampling interval (which for CD would be 11 ?s);"
One of the problems with the new Meridian paper is that if it is considered to be an AES refereed paper, it adds a number of serious errors (such as overlooking ABX1982 which is after all a refereed JAES paper) and apparent folklore such as:
"Listeners were able to
listen to as many labelled pairs of extracts as they
liked before progressing to the test. The filter used
here was an FIR filter with a frequency transition
band spanning 8-10 Hz. This fi lter was chosen as it
would have been straightforward for most listeners
to identify differences introduced by its application."
On page 2 I find two rather conflicting statements and I'm not sure what conclusion the authors draw:
It has further been suggested [22, 23, 24] that listeners can discriminate timing differences of the order of 5 µs and below, which, if correct, would require a Nyquist frequency of 32 kHz or higher...
And below that:
However, it can not be assumed that the auditory system cannot extract time differences that are shorter than the periods between successive samples, even when convolved with a sinc function in D/A conversion (assuming a sufficient signal-to-noise ratio).
It is my understanding (references to JJ, David Griesinger e.a.) that "In Physics, the accuracy of timing is not determined by the bandwidth, but roughly by the product of the bandwidth and the signal to noise ratio".
Am I correct that the authors say that the time resolution argument based on sample rate and excluding SNR is not valid ?
You may want to read this thread (http://www.stereophile.com/content/interesting-papers-1), starting from around the 12th or 13th post or so.
He schools some of the fashion engineer luminaries from amirs camp rather nicely.
cheers,
AJ
You may want to read this thread (http://www.stereophile.com/content/interesting-papers-1), starting from around the 12th or 13th post or so.
Thanks for the link. It confirms what I've read elsewhere. It's just sad that it's not common knowledge. Just look at this article from Yamaha Professional Audio (http://www.yamahaproaudio.com/global/en/training_support/selftraining/audio_quality/chapter5/09_temporal_resolution/):
To also accurately reproduce changes in a signal’s frequency spectrum with a temporal resolution down to 6 microseconds, the sampling rate of a digital audio system must operate at a minimum of the reciprocal of 6 microseconds = 166 kHz.
...
table 504: Main decision parameters for the selection of a digital audio system’s sample rate
Audio quality issues: desired temporal resolution
48 kHz 20 ?S - high quality
96 kHz 10 ?S - very high quality
192 kHz 5 ?S - beyond human threshold
On the topic of filters, can someone explain to me the use of impulse response energy as displayed in Fig.2 ? I know how to interpret amplitude response, but have never seen an energy plot like that.
For those who have not read the whole of this thread, I'd note that the question of the use of no dither when quantising to 16-bits (test conditions 2 and 5), or rectangular dither when quantising to 16-bits (test conditions 3 and 6) could be regarded as a subsidiary matter. This is because the paper reports that there was a statistically significant correct identification of an audible difference for test condition 1, i.e. filtering to emulate a resampling to 44.1kHz, and
without any quantisation to 16 bits. (As for conditon 4, a filtering to emulate resampling to 48kHz, and without any quantization to 16 bits, "the t-test just failed to reach significance at the 5% level". )
If it is true that a mere filtering at 24-bit depth for the 22.05kHz Nyquisit limit of 44.1kHz sampling was of itself identifiable (in particular for certain "high yield" [easier to spot differences] sections of the music), it becomes a subsidiary matter what effect a subsequent quantisation to 16-bits might have had. The "damage" or "impairment" had already occurred, or so it would appear.
Amirm, a few posts above, raises as a possible explanation the matter of ringing, with emphasis on the time domain aspect (the timing and duration of the ringing), rather than the frequency domain aspect (primarily, the frequency of the oscillation):
OK, finally I get to this technical point:
...
The above is also the reason I say that if CD had picked 48 Khz as the sampling, I would have no beef with it. We have plenty of room to implement our filter above 20 Khz. But by picking 44.1, it leaves us a small margin forcing sharper filters and more time domain ringing.
I note that had the filtering been performed with non-linear phase, then indeed signal frequencies in the audible range could well have been shifted in phase, and particularly when listening with headphones, such phase changes might have become audible as a subtle change. However the paper indicates that
linear phase Finite Impulse Response filters were used. The filter for the notional 44.1kHz sample rate had a transition band from 23500 to 24000Hz. That suggests that audible frequencies at 20kHz and below would not only have been unaffected in amplitude, but unaffected in phase.
In a separate thread in this forum on the audibility of a 20kHz brickwall filter, the promising ABX results reported have been for maximum phase shift, not linear phase shift. See: Audibility of 20kHz brick wall filtering, samples provided for ABXing - 24/96 sound card required (http://www.hydrogenaud.io/forums/index.php?showtopic=68524).
No, you are confused. I'm not talking about ripple but ringing. I never even mentioned ripple.
Oh? Did you not immediately follow up with this post?
Yes, and for the nth time, #364 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881529) shows
only the time domain. x-axis shows time, y-axis the amplitude of the sine wave that increases in frequency over time.
I am not talking about ripple, never was. The filters visualized in #364 have flat passband,
otherwise the subtraction in the 2nd plot would show some remaining signal in the passband (below 20.5, 21.5, 22 kHz for each filter respectively).
You still seem mightily confused about all of this, especially how basic filtering works.
Where would I see "ringing" in those? You clearly say they are frequency sweeps from 20 to 23 Khz. Is that how we show the Impulse response and ringing? With a spectrum display from 20 to 23 Khz? You are telling us that those visualizations are the same as these I post?
I posted these sweeps
to show you what a linear phase lowpass filter does.
These filters operate
linearly. It doesn't add to the signal, it doesn't remove anything either, and since it is linear phase it doesn't change the timing either. All it does is attenuate above 20 or 21 kHz.
A
minimum phase filter, that only has post-ringing and no pre-ringing, works similarly except for that fact that it
delays parts of the signal, reaching the highest delay at the cutoff frequency (again above 20 or 21 kHz). This delay increases smoothly with frequency however, so it has an effect on frequencies far below 20 kHz as well.
[img showing impulse response]
My graph shows ringing in time domain. Exactly the topic David and I were discussion. You clearly confused passband ripple in frequency domain with time domain ringing. It doesn't get any simpler than this.
Yes,
ringing of a filter like any other filter has, but at 21+ kHz. Again, for the 100th time, I am not talking about passband ripple.
Yes, it couldn't be any simpler than this but you're still confused which is why I will post another example...
First we generate a test signal:
(http://xserv.compress.to/xnor/audio/images/lpfilter1.png)
This is a random signal that contains mostly high-frequency energy up to 21500 Hz, critically sampled at 44.1 kHz. (Excuse the visualization with matlab plot function..)
Then we generate a filter:
(http://xserv.compress.to/xnor/audio/images/lpfilter2.png)
(y-axis shows time in seconds)
This is an absolutely terrible filter, very long with 122ms total pre-ringing. Transition band is only 50 Hz wide, so extremely steep. Again, lots of filter ringing.
Now we filter the above signal with this terrible, bad ringing filter:
(http://xserv.compress.to/xnor/audio/images/lpfilter3.png)
(different visualization #1 (http://xserv.compress.to/xnor/audio/images/lpfilter4.png), #2 (http://xserv.compress.to/xnor/audio/images/lpfilter5.png))
Oh noes,
the filtered signal matches the original signal perfectly (also see #1 (http://xserv.compress.to/xnor/audio/images/lpfilter4.png), #2 (http://xserv.compress.to/xnor/audio/images/lpfilter5.png)). Why? Because .. see the explanations above that I tediously had to repeat many times.
And for the curious, here's what happens if the filter does not have any pre-ringing but is: min phase (http://xserv.compress.to/xnor/audio/images/lpfilter6.png).
On the topic of filters, can someone explain to me the use of impulse response energy as displayed in Fig.2 ? I know how to interpret amplitude response, but have never seen an energy plot like that.
What figure 2?
Amir's, position as I understand it in all of this is as follows.
When he wants to buy music, he wants to buy the masters of the recording. If they only exist in 44.1, then fine. If it's 96 or 112 then fine. He doesn't want anything that's been up or down converted. If you think about it, that's his preference and as far as preferences go, it's whatever floats your boat.
How we each justify our preferences is an individual matter. There are rational, irrational, and any number of reasons why we choose things. A problem generally arises when we vocalize our reasonings to others. After all, not everything we say is going to make sense to the other person or there may even be errors in our thinking.
Amir has demonstrated to his satisfaction that he can key in on certain aspects of the conversion process with great success. That may well play apart in his internal justification process. He's aware that hi-res, even compressed, takes up more room and it costs more. Significantly in my opinion but to him that's irrelevant so long as what he bought is the master. It may well be that the master still suffers from loudness issues or even poor recording or mastering. But if it's the original then that's what matters.
Well said but let me clarify a bit the last part. I do not go and buy high resolution content. Most of my purchases are CDs. But I like to see a thriving market for high resolution downloads. I want to have the option to get the high resolution version should it not suffer from loudness compression of CD. And importantly, in the final eventuality where CD distribution diminishes. I routinely run into albums I want to buy that are only available in MP3/AAC. CD distribution is antiquated and over time this trend will continue to decline. I foresee a world where there is compressed MP3/AAC for the general public and high resolution for distribution to audiophiles.
So unless you are a fan of capping your fidelity to 256 kbps MP3/AAC (in which case we won't be on speaking terms ), you ought to support this development. No harm will come to any of you due to expansion of high resolution content. None whatsoever. But if you keep pouring negative vibes it, and should it have a material effect, you are doing the rest of us disservice who care about fidelity.
The rest of what transpires are matters of verbal jousting. For me, the pity is that the complaints the public has had with the recordings are not going to go away. Look at those links I posted earlier and IMO, it's not about doing a better job with recording. Studios are going to put out what sells and in whatever format, moving the sliders around until they get something the customer approves. A crappy picture at 800x600 is still going to be crappy at mega resolutions.
And our best hope, and one that we have influence over, is support high resolution downloads. We have zero, absolutely zero power to influence how music is created for the mass market. They want loudness equalization in their playlists and labels/talent want to develop on that and then some. High resolution downloads however, are going under the radar. In many cases we have a direct link to people who master them (see Bruce Brown who masters a lot of content for HD Tracks: http://www.whatsbestforum.com/forumdisplay...By-Bruce-Brown) (http://www.whatsbestforum.com/forumdisplay.php?235-The-Pro-Audiophile-By-Bruce-Brown)). Not doing so will cause us to eventually face the "crappy 800x600" in the form of loudness compressed and lossy compressed MP3/AAC.
Rest of this is all emotional chatter. Thanks for bringing voice of reason to this forum Chu.
On the topic of filters, can someone explain to me the use of impulse response energy as displayed in Fig.2 ? I know how to interpret amplitude response, but have never seen an energy plot like that.
What figure 2?
(http://arnyk.smugmug.com/Illustrations/i-P9SwRBh/0/L/audibility%20of%20typical%20digital%20fig%202-L.png)
This is a random signal that contains mostly high-frequency energy up to 21500 Hz, critically sampled at 44.1 kHz. (Excuse the visualization with matlab plot function..)
Looks like the plot is lost. We are not talking about re-filtering a 44.1 Khz track. We are talking about original that is at 192 Khz and has content past 22.05 Khz. Go ahead and simulate a transient with full bandwidth of 96 Khz and demonstrate how your filter did nothing. You won't be able to of course as that would be audio alchemy. The result would be my matlab graph that shows clear pre-ringing. That pre-ringing will step on what came prior to that transient. Music has full spectrum of content, not just the ultrasonics.
I don't want to keep going and lose everyone else. So I refer you to my ex-architect and friend, JJ's presentation to AES since everyone accepts him as the authority: http://www.aes.org/sections/pnw/ppt/jj/adc.ppt (http://www.aes.org/sections/pnw/ppt/jj/adc.ppt)
(http://amirviews.smugmug.com/photos/i-QJsQxSL/0/S/i-QJsQxSL-S.png)
(http://amirviews.smugmug.com/photos/i-vh7XVjz/0/L/i-vh7XVjz-L.png)
Exactly what I said and statement from Stuart.
(http://amirviews.smugmug.com/photos/i-kbsQ4Q4/0/M/i-kbsQ4Q4-M.png)
Same simulation I showed. Then this:
(http://amirviews.smugmug.com/photos/i-r8cBMRt/0/M/i-r8cBMRt-M.png)
And the same thing I said and Stuart did. This is the plot. There may be audible problems here but hard to test.
That's it. I am not here to prove it is audible. I said it may be the cause since we know the testers like me could not hear the ultrasonics. You can stick to your theory of people cheating. I will stick to potential technical explanations like JJ.
AJ, I hope you didn't have a heart attack reading that JJ speculates about audibility based on signal processing concepts and that of psychoacoustics. It is how we do things in real life. We don't ignore problems that we can demonstrate on paper absence of listening tests.
[figure 2]
Looks like
20*log10(abs(imp)) ... where imp is the filters impulse response.
This is a random signal that contains mostly high-frequency energy up to 21500 Hz, critically sampled at 44.1 kHz. (Excuse the visualization with matlab plot function..)
Looks like the plot is lost. We are not talking about re-filtering a 44.1 Khz track. We are talking about original that is at 192 Khz and has content past 22.05 Khz. Go ahead and simulate a transient with full bandwidth of 96 Khz and demonstrate how your filter did nothing. You won't be able to of course as that would be audio alchemy. The result would be my matlab graph that shows clear pre-ringing. That pre-ringing will step on what came prior to that transient. Music has full spectrum of content, not just the ultrasonics.
I don't want to keep going and lose everyone else
As if that would happen.
So I refer you to my ex-architect and friend, JJ's presentation to AES since everyone accepts him as the authority:
http://www.aes.org/sections/pnw/ppt/jj/adc.ppt (http://www.aes.org/sections/pnw/ppt/jj/adc.ppt)
(http://amirviews.smugmug.com/photos/i-QJsQxSL/0/S/i-QJsQxSL-S.png)
(http://amirviews.smugmug.com/photos/i-vh7XVjz/0/L/i-vh7XVjz-L.png)
Exactly what I said and statement from Stuart.
(http://amirviews.smugmug.com/photos/i-kbsQ4Q4/0/M/i-kbsQ4Q4-M.png)
Same simulation I showed. Then this:
(http://amirviews.smugmug.com/photos/i-r8cBMRt/0/M/i-r8cBMRt-M.png)
And the same thing I said and Stuart did. This is the plot. There may be audible problems here but hard to test.
JJ says the following:
"This suggests that for higher sampling rates, we do not want the ‘fastest’ filter, rather a filter with a wider transition band, and narrower time response."
and then note who is using filters whose transition band is something like 4x+ narrower than commercial DACs.
IME when JJ says stuff like this:
How would I do that?
Get a DAC with a SLOW rolloff running at 4x (192K).
Make a DC to 20 K Gaussian pulse at 192kHz.
Downsample by zeroing 3 of every 4 samples and multiplying the others by 4.
Generate a third signal with a TIGHT filter.
Compare the three signals in a listening test
It is more like a rhetorical question. He knows what the answer should be if you do the experiment properly.
The other thing is that tests like the above are violently unrealistic. People listen to music, not worst case highly abstract test signals like unit impulses.
Note that not even the recent Meridian paper is based on listening to pink noise, white noise, or unit impulses. That would be too obvious. If you tried to present such a thing at a real professional conference, you'd be laughed out of the room. You might even get your paper rejected based on its abstract.
Note that not even the recent Meridian paper is based on listening to pink noise, white noise, or unit impulses. That would be too obvious. If you tried to present such a thing at a real professional conference, you'd be laughed out of the room. You might even get your paper rejected based on its abstract.
The goal of any distribution channel/format is transparency. That transparency better be there for all signals. Who is to say some music does not have noise like characteristics? The famous Chris Jones No Sanctuary track starts with a pure tone. Are you going to say that is not "music" because it appears to have a test tone in it?
Now, if this were an impossible thing to accomplish, we could listen to the argument. But if all we have to do is go to >= 48 Khz sampling rate why would we want to settle for anything but full transparency? We wouldn't.
So you see the problem with your position. This is why I said I like to get the high resolution bits and be done with it. The moment you go down to 16/441. you get cornered having to exclude test cases like you are doing. Get the high res bits and no issues like this. We can user a super gentle filter at 88.2 Khz. Keep the bits at 24 so we don't have to worry about what dither is or is not applied and we are golden. All doable. All feasible. And all free with respect to hardware to play it.
Looks like the plot is lost. We are not talking about re-filtering a 44.1 Khz track.
I used this to explain to you, for the 100th time, that
the ringing and the frequency the filters operate are at over 21 kHz. #255 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881162)?
This was in response to #468 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881885), because up until now it didn't seem you understood how the filters work.
We are talking about original that is at 192 Khz and has content past 22.05 Khz. Go ahead and simulate a transient with full bandwidth of 96 Khz and demonstrate how your filter did nothing. You won't be able to of course as that would be audio alchemy. The result would be my matlab graph that shows clear pre-ringing. That pre-ringing will step on what came prior to that transient.
What you claim is the "original" is actually not the original either. It is also filtered (had to be, greetings from Shannon), just with a different filter that rings at a higher frequency.
Also, the content up there has considerably less energy than the content at lower, audible frequencies. And you have to consider hearing loss and also the attenuation of high frequencies through air ...
And no, the pre-ringing would not "step on what came prior to that transient". You are again confusing this with echo. The filter would build up output at an extremely low level and rapidly rise in energy at the transient (since that is where the energy is concentrated in such a filter) and would only act on the energy that is left up there.
In short, again, ringing at 21+ kHz.
Music has full spectrum of content, not just the ultrasonics.
What? Ultrasound is only the stuff above 20 kHz. In music most energy is located far below that, in the acoustic range.
I don't want to keep going and lose everyone else. So I refer you to my ex-architect and friend, JJ's presentation to AES since everyone accepts him as the authority: http://www.aes.org/sections/pnw/ppt/jj/adc.ppt (http://www.aes.org/sections/pnw/ppt/jj/adc.ppt)
So, have you done the suggested test with a 20 kHz gaussian pulse and a steep filter at 21+ kHz? It looks like once again you have no idea what you're even quoting. (jj also makes a reference to this being a problem with lossy codecs, which I also explained why it's an audible problem there, see the last lines of #446 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881832).)
You think you're pulling a rabbit out of your magic hat, but its actually a gun shooting into your own foot.
That's it. I am not here to prove it is audible. I said it may be the cause since we know the testers like me could not hear the ultrasonics. You can stick to your theory of people cheating. I will stick to potential technical explanations like JJ.
Now you are contradicting yourself.
Not only did you claim to hear clear audible problems, but run around showing off ABX logs. You say all it comes down to, to hear this 21+ kHz ringing, is training. But you still
haven't even given a hint at what specifically you hear.
You claim to be able to hear steep filters at 21+ kHz. Okay,
if I sent you some files could you identify such filters? (I, of course, wouldn't send you files that allowed you to cheat using a spectrum analyzer, for example.)
I am not here to prove it is audible.
You are, have been trying and failing, because you can't.
That's what the relentless posting of useless gameable computer files ABX logs are about, desperately trying to prove audibility.
AJ, I hope you didn't have a heart attack reading that JJ speculates about audibility based on signal processing concepts and that of psychoacoustics.
It's precisely because I've been well versed with his writings for a while now, that I've been encouraging you to seek treatment from him. Have him set up the proper test, switching, timing, levels, etc, pick the best room in your house, bring over his speakers, use the 2L "Hi Rez" Haydn track to compare against his 16/44 downsampled TPDF "CD" version (since none exists on the website).
Post on Youtube.
It is how we do things in real life. We don't ignore problems that we can demonstrate on paper absence of listening tests.
Great, so get on with your Amir Bolt audible demonstration Youtube video and quit wasting your time posting nonsense on forums.
In the meanwhile, remember to regenerate your power, use high end USB and SPDIF cables, Bryston amps, Clarity MR caps and lift your speaker cables off the floor, etc, etc.
Just in case. Oh and make sure it all costs a nice profit premium. Just in case.
cheers,
AJ
Was my pic misleading or wrong?
http://www.hydrogenaud.io/forums/index.php...st&p=881802 (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&view=findpost&p=881802)
Above is a gentle filter with very low ringing, the mid shows the ringing going well down to audible frequencuies with a gentle non-linear filter but as post-ringing.
Below is a typical steep linear filter showing its ringing around the filters frequency.
The goal of any distribution channel/format is transparency.
Wrong. It's sales, just like you learned at Sony and MS.
That transparency better be there for all signals.
It is:
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
No listener has ever used available distribution channel/format to demonstrate otherwise. Quite the opposite, as M&M showed.
Keep the bits at 24 so we don't have to worry about what dither is or is not applied and we are golden.
Sure, at higher co$t and greater profit, with about 10% fidelity to the original soundfield....but in "higher resolution" stereo. I bet Neil Young will sound so much better in 24bits.
Sounds like a plan. What do you intend to call this 10% fidelity cottage industry, the 2ch "Worry free High End"?
When do the listener training sessions begin? Installation of the iso-ward rooms and new loudspeaker standards? When do Berkley DACs start featuring "loop" buttons, to "zone in" on segments during listener enjoyment sessions in the easy chair?
Inquiring minds want to know Amir, please enlighten us.
cheers,
AJ
Not only did you claim to hear clear audible problems, but run around showing off ABX logs. You say all it comes down to, to hear this 21+ kHz ringing, is training.
Oh? Again, the plot is lost. Here is my comment that you objected to and caused the 12-part replies:
Having read the Stuart paper and prior citations, I am starting to think this may be due to time domain impact. It certainly is not frequency domain since I can't hear the ultrasonics. Just as pre-echo is a very audible time domain artifact (although created in frequency domain), maybe these filters act the same way.
You say I have claimed to hear audible problems. I have not. I have shown that I can tell files apart. That is what is required of me: detect lack of transparency.
You say I am running around with ABX logs. That is what people like to see as evidence of passing such tests. So I have provided them. I ran a tool that I don't like because people asked and asked for me to run them that way. So I did. Now there is something wrong with that? Maybe I should have just declared things audible and left it at that just like you are doing in reverse.
You say it all comes to hear 21 Khz ringing. How on earth can you read that in my post? I was thinking out loud with someone who I respect to know the topic, David. I provided no proof or that it is that specific to 21 Khz ringing or whatever. I said the paper and its references put out a theory in that regard. That is, why the 48 Khz sampling results were superior to 44.1 Khz. Both filter out ultrasonics that we can't hear and the only difference left was their filter lengths. The paper doesn't make more out of it than I just did.
I asked you to read Stuart's reference to earlier work. You didn't, did you? Here it is the excerpt:
Dispersive Models for A-to-D and D-to-A Conversion SystemsRoger Lagadec and Thomas G. Stockham, Jr.
Roger Lagadec is an AES Fellow: http://www.aes.org/technical/heyser/aes107.cfm (http://www.aes.org/technical/heyser/aes107.cfm)
Known for his pioneering work in digital audio, Roger Lagadec now focuses on Internet commerce. He received his Ph.D. in digital signal processing for telecommunications. He has served as an AES governor.The late Thomas Stockham has no less credentials:
His awards and professional accomplishments are numerous. A few among them are: After receiving his Sc.D. degree at MIT he was appointed Assistant Professor of Electrical Engineering in 1959. In 1968 he left MIT and joined the faculty at the University of Utah to help create their Computer Science Department. In 1974 he served on the panel of experts assembled to examine the 18 minute gap in the Watergate Tapes....He received a 1988 Emmy, the first ever technical Grammy Award in 1994, and a 1999 Scientific /Engineering Academy Award (Oscar) for his contributions to the creation of digital sound. He is internationally recognized as the father of digital recording....
In 1998 he was elected to the National Academy of Engineering for his contributions in the field of digital audio recording. He also won a technical Oscar. The AES awarded him a fellowship in 1978, and the prestigious Gold Medal in 1987. He was president of the society from 1982-83 and served on the Board of Governors in
1988.This is work that you read first before posting and posting:
9. Some Conclusions
Using admittedly exaggerated filter designs in order to make
effects evident, it has been possible to show that the dispersive
character of filters must be taken into account. Such concepts as
ripple, bandwidth and phase response do not describe exhaustively
the acoustical behavior behavior of analog or digital filters,
especially when the criteria is not (or should not be) the audibility
for untrained listeners, but the perceptibility by professionals
and people used to the quality of digital audio.See? I stand on shoulders of giants when I talk about this topic. This is not some idle speculation.
But you still haven't even given a hint at what specifically you hear.
And I am not going to. Seems like you are unaware of what an ABX test means. It means differentiating two samples. That is it. It doesn't tell you what was different. It doesn't tell you which version was better. It just says I can differentiate the two files. Indeed, in no case did I attempt to identify which file is which. I put them in Foobar playlist and instantly ran the plug-in. Neither before or after the test did I attempt to see which file was which. I ran the ABX test, found the difference and reported it. To keep asking qualitative questions is wrong. It just is.
Now, I gave you a method for you to discover the same. You asked me to PM you with test files to run. I mentioned an example here and you were flippant about it. Here it is again: http://www.hydrogenaud.io/forums/index.php?showtopic=68524 (http://www.hydrogenaud.io/forums/index.php?showtopic=68524). You see foobar logs there of people who say they have passed them. I put in my results in one of these two threads.
Run the test and report back the log file. We won't accuse you of doctoring them. Let's see if you can hear a difference. If you can, then you answer your own question. If you cannot then it answers my question that you are not a critical listener.
You claim to be able to hear steep filters at 21+ kHz. Okay, if I sent you some files could you identify such filters? (I, of course, wouldn't send you files that allowed you to cheat using a spectrum analyzer, for example.)
See above. Let's see how many more back and forths it takes for you to run a double blind ABX test. The very test we keep asking others to run. But somehow, we are gun shy about running it ourselves.
And please state if you think the others who passed David's test are cheating.
Wrong. It's sales, just like you learned at Sony and MS.
Seeing how you are in the business of selling audio gear, I am unclear why you are so critical of your own selected profession Ammar. I read in all places, the Stereophile that this was your demo gear at AXPONA 2012 show:
Source was a laptop running Windows Media Player, feeding a NuForce uDAC-2, VT 01 preamp, and MB 200 monoblocks. (I can't find the last two on NuForce's website, but it's what company founder Ammar Jadusingh, shown in the photo, scribbled in my notepad.
Monoblocks? Over $6,000 in amplification alone? A near $5,000 tube pre-amp? Seems like you are exercising sales tactics there my friend. And contradicting your posts here.
At least you did well with respect to your player as WMP was developed in my team!
Sounds like a plan. What do you intend to call this 10% fidelity cottage industry, the 2ch "Worry free High End"? When do the listener training sessions begin? Installation of the iso-ward rooms and new loudspeaker standards?
Nah. I passed the tests on my laptop using stock DAC. No tube pre-amp. No monoblocks. You just have to not be a good listener.
When do Berkley DACs start featuring "loop" buttons, to "zone in" on segments during listener enjoyment sessions in the easy chair?
That is not what a DAC does Ammar. It has no control over the source. You want to send me your Nuforce DAC to compare to it?
Inquiring minds want to know Amir, please enlighten us.
That is inconsistent with what you said intentions were in these threads:
turning the tables on the elitists, bullies, illogical condescending snobs and highly self deluded "audiophiles" that permeate my audio hobby? No mention of enlightenment.
You say I have claimed to hear audible problems. I have not.
Thank goodness for this admission. The BS paper sure made it sound like there were audio problems with 16/44, not "transparency" problems, which would contradict the 518 user guide.
So 16/44 doesn't have audible problems.
It "lacks transparency", when "listening". So we don't need 24 bits because of audible issues, we need it for "worry free transparency when listening".
Thanks amir. Have you contacted JJ yet to begin the therapy?
cheers,
AJ
Seeing how you are in the business of selling audio gear, I am unclear why you are so critical of your own selected profession Ammar
How is it being "critical" to point out the obvious? "Hi-Rez" distribution is about sales, as any other commercial, non-charity venture. Meridian is also in the business of selling audio gear, yes?
Monoblocks? Over $6,000 in amplification alone? A near $5,000 tube pre-amp? Seems like you are exercising sales tactics there my friend.
Absolutely. Demoing my speakers at the shows is a sales tactic. I'm not the one in denial about the reality of commercial endeavors, like "Hi-Rez".
Sharing room costs with other manufacturers also makes sense, when one doesn't manufacture upstream components, just speakers. Why create fear, prejudice and worry unnecessarily? I want folks like you to hear my speakers, not peek in the room and start worrying.
Nah. I passed the tests on my laptop using stock DAC. No tube pre-amp. No monoblocks.
No witnesses either.
Sorry, the dog doesn't count.
xnor passed the test without hearing audible differences also. Did I mention the uselessness of online ABX tests and subsequent logs with regards to the BS paper?
That is not what a DAC does Ammar. It has no control over the source.
Ok, so when do the listener training sessions, etc, etc. begin for "Hi-Rez" buyers?
How is the person purchasing an SACD, "Hi-Rez" from the 2L store, or your store Madrona, suppose to get the benefits of this "transparency", when "listening" if it isn't audible vs 16/44? If there is no training classes, no loop functions on Berkleys etc, no requirements for iso-ward rooms?
The sole benefit is lack of "worry"?
cheers,
AJ
Thank goodness for this admission. The BS paper sure made it sound like there were audio problems with 16/44, not "transparency" problems, which would contradict the 518 user guide.
They did find audible problems because in their case, they outline them:
It was reported that filtering gave "softer edges" to
the instruments, and "softer leading edges" to musi-
cal features with abrupt onsets or changes. Echoes,
when audible, were identi ed as being affected the
most clearly by the ltering. It was felt that some
of the louder passages of the recording were less ag-
gressive after filtering, and that the inner voices (sec-
ond violin and viola) had "a nasal quality". Over-
all, the filtered recording gave a "smaller and
flatter auditory image", and speci cally the physical space
around the quartet seemed smaller.So definite problems and nothing to do with dither. Transparency was not achieved.
So 16/44 doesn't have audible problems.
So you are from Mars.
It "lacks transparency", when "listening". So we don't need 24 bits because of audible issues, we need it for "worry free transparency when listening".
Thanks amir.
Yeh, the problem goes away when I stop listening and just read the lyrics. But yes, ignorance is really bliss here. Trust me, you don't want to be in my shoes and that of others who are sensitive to these artifacts. It takes away from the enjoyment. Just like a musician hearing a wrong note and ruining the performance for him, while I may not even notice it not being a musician, we are different in that regard. Just be careful to not foist your understanding of audio devoid of critical listening skills onto others.
Have you contacted JJ yet to begin the therapy?
Nah. If you need therapy and you think he can provide it, please contact him yourself.
Edit: fixed OCR problems in the cut and past again .
They did find audible problems because in their case, they outline them:
Bingo! Yes!
So what the heck does your ABX logs have anything to do with the BS Paper, if you just admitted you heard no audible problems with them?
That your issue with 16/44 stems from some sort of "worry" when "listening" malady.
The title of the paper is "
Audibility of Typical Digital Filters in a Hi-Fi Playback System."
Please let's concentrate on that, not your non-audible ABX logs.
Trust me, you don't want to be in my shoes and that of others who are sensitive to these artifacts. It takes away from the enjoyment.
You're absolutely right I wouldn't want to be an audiophile....and stricken with all these "worry" maladies. I actually enjoy listening to music, live and reproduced. Even if reproduced by a clock or XM radio. If I remotely became cognizant of having these "worries", I'd seek treatment or have myself committed.
If you need therapy and you think he can provide it, please contact him yourself.
I get all the therapy I need at the Concert Halls like these (http://music.arts.usf.edu/) (check out those "Hi-Rez" prices Amir), or listening to standard "Low-Rez" FM (http://www.wusf.usf.edu/music/channel/classical_music) over factory speakers or streaming, even on a laptop. I'm not the one with the digital disorders. I certainly wouldn't mind hearing JJs system if it's set up for PSR. Which I'm sure even at 320Kbps, would blow away any "Hi-Rez" 2ch reproduction of the same event, including the 2L Hadyn type stuff.
You really do need to take a Logic 101 class Amir. Figure out why the burden of proof rest squarely on your side/camp, even with this BS paper, which is preliminary at best and leaves many unanswered questions.
cheers,
AJ
How is it being "critical" to point out the obvious? "Hi-Rez" distribution is about sales, as any other commercial, non-charity venture.
So low-res audio is not about sales? Why would you pay more for the CD vs MP3 Ammar? Did you play MP3s at the show?
Meridian is also in the business of selling audio gear, yes?
Yes, you are both in the business of selling audio gear. Big difference otherwise though. In the case of Bob Stuart, he comes with proper education and prestigious award of AES Fellow whose work is referenced and published. He has published a listening test/paper that garnered an award from his peers at Audion Engineering Society in the form of "Best Peer-Reviewed Paper."
You are a hobbyist who says one thing here, and then shows up at an audiophile show and does another. No educational experience. No professional experience. No record of taking any of these tests.
Monoblocks? Over $6,000 in amplification alone? A near $5,000 tube pre-amp? Seems like you are exercising sales tactics there my friend.
Absolutely. Demoing my speakers at the shows is a sales tactic. I'm not the one in denial about the reality of commercial endeavors, like "Hi-Rez".
Well you must be or you would post under your real name Ammar. Be proud of your position and let your customers see who you really are. Ditto for members here. Don't tell us you are a vegetarian and advocate the same, only to go home and have steak one night, and bacon the next morning.
Oh? Again, the plot is lost. Here is my comment that you objected to and caused the 12-part replies:
Having read the Stuart paper and prior citations, I am starting to think this may be due to time domain impact. It certainly is not frequency domain since I can't hear the ultrasonics. Just as pre-echo is a very audible time domain artifact (although created in frequency domain), maybe these filters act the same way.
You say I have claimed to hear audible problems. I have not. I have shown that I can tell files apart.
So you say you cannot hear audible problems but can tell the files apart (with a spectrum analyzer?), and show off with ABX logs. And you wonder why I said it's more likely that you cheated than hearing genuine differences? Some may even see this as admission to exactly that.
You say I am running around with ABX logs. That is what people like to see as evidence of passing such tests. So I have provided them. I ran a tool that I don't like because people asked and asked for me to run them that way. So I did. Now there is something wrong with that?
You are contradicting yourself again:
Our shtick in these discussions has always been a challenge for the other camp to run an ABX test to prove their subjective impressions of audio. Since I am not in the opposing camp, and see value in blind tests, I ran the tests. Not one but many. Not the first version but the follow ups. Not just on AVS but also here and on WBF Forum.
[...]
If our real intentions are to encourage others to believe and run such tests, we have failed and failed badly in that mission.
In the process of making excuses you're shooting yourself in the foot. Again.
Maybe I should have just declared things audible and left it at that just like you are doing in reverse.
Oh my ... I guess you will never learn.
Please finally read wiki/Philosophic_burden_of_proof (https://en.wikipedia.org/wiki/Philosophic_burden_of_proof) carefully. Additional links will also teach you about the
many fallacies that you still keep making happily after 10 pages.
You say it all comes to hear 21 Khz ringing. How on earth can you read that in my post?
I don't, but keep on making those straw man arguments.
In fact, I asked you countless times
what you hear. You evaded, and evaded, and evaded until above you say you don't hear but I guess magically tell the files apart (hence my exhaustive remarks regarding false positives, cheating).
9. Some Conclusions
Using admittedly exaggerated filter designs
Just stop there amirm. You don't see a problem here?
Also, the cited paper is largely outdated. Over 30 years ago the filters in A/D/A converters looked very different. You know that but you cherry-pick (another fallacy btw).
But please show us where it demonstrates the audibility of a linear-phase lowpass with a transition band of roughly 2 kHz (that's what we find in modern devices), with a cutoff frequency above ~21 kHz.
And I am not going to. Seems like you are unaware of what an ABX test means. It means differentiating two samples. That is it. It doesn't tell you what was different. It doesn't tell you which version was better. It just says I can differentiate the two files. Indeed, in no case did I attempt to identify which file is which. I put them in Foobar playlist and instantly ran the plug-in. Neither before or after the test did I attempt to see which file was which. I ran the ABX test, found the difference and reported it. To keep asking qualitative questions is wrong. It just is.
Honestly, I've never heard such lame excuse. Never.
Of the hundreds of people that posted ABX that I have seen, not one was unable to even hint at what differences they heard.
It seems that you are unaware of what "ABX test" means. It is a method
to identify audible differences between two tracks.
This also makes the test, that I would have prepared for you, irrelevant since you weasel out of even having the ability to hear differences. Yet you magically produce positive ABX logs.
Now, I gave you a method for you to discover the same. You asked me to PM you with test files to run. I mentioned an example here and you were flippant about it. Here it is again: http://www.hydrogenaud.io/forums/index.php?showtopic=68524 (http://www.hydrogenaud.io/forums/index.php?showtopic=68524). You see foobar logs there of people who say they have passed them. I put in my results in one of these two threads.
Run the test and report back the log file. We won't accuse you of doctoring them. Let's see if you can hear a difference. If you can, then you answer your own question. If you cannot then it answers my question that you are not a critical listener.
You claim to be able to hear steep filters at 21+ kHz. Okay, if I sent you some files could you identify such filters? (I, of course, wouldn't send you files that allowed you to cheat using a spectrum analyzer, for example.)
See above. Let's see how many more back and forths it takes for you to run a double blind ABX test. The very test we keep asking others to run. But somehow, we are gun shy about running it ourselves.
And please state if you think the others who passed David's test are cheating.
(http://i326.photobucket.com/albums/k426/snickasevans/dailysmiley/wallbash.gif)
I will simply ignore the lesser logical fallacies in this one. (see wiki link above for starters)
So, you set up a false dichotomy where either
you win or
you win (at least from what I can tell you seriously do think that?!) based on your
flawed premises. Jesus help us all.
I've also explained to you, multiple times now, that non-linear-phase will cause a distortion of phase in the audible range even if the filter operates at 20 kHz. But the test you point me to is one with a maximum-phase filter?! So even if I posted a successful log
and told you what differences I heard your flawed argument would still fail horribly.
Seriously ... I can take a lot, but this is getting much too painful. You are truly one of a kind, amirm.
So you say you cannot hear audible problems but can tell the files apart (with a spectrum analyzer?),
Plot is lost again. I was very clear that I never analyzed the files, before or after the test.
...And you wonder why I said it's more likely that you cheated than hearing genuine differences?
That is a reflection on you. That you think these online discussions are so important that one should resort to winning them by cheating. I don't go there. You can say or do whatever but I won't question your integrity. That you do so in post after post, means that you and I have very different thresholds here. You think winning an argument against a few Internet posters writing under an alias is justification for cheating. You need to re-examine that logic.
And please state if you think the others who passed David's test are cheating.
(http://i326.photobucket.com/albums/k426/snickasevans/dailysmiley/wallbash.gif)
I will simply ignore the lesser logical fallacies in this one. (see wiki link above for starters)
So you will say in each test I cheated in my logs yet when I ask you about others in this very forum, taking a test from one of our members, and you won't go there. What this means is that you have a personal beef with me. In other words, we are not having a discussion of science.
but this is getting much too painful. You are truly one of a kind, amirm.
Yes, it does get painful when you demand evidence and turn around and accuse the other person after delivering it of a cheat. Please allow me to post this forum's Terms of Service #8:
8. All members that put forth a statement concerning subjective sound quality, must -- to the best of their ability -- provide objective support for their claims.
Acceptable means of support are double blind listening tests (ABX or ABC/HR) demonstrating that the member can discern a difference perceptually, together with a test sample to allow others to reproduce their findings. Graphs, non-blind listening tests, waveform difference comparisons, and so on, are not acceptable means of providing support.
ABX double blind listening test results are provided.
Exact same content is provided per my last post to you.
But you refuse to take the same tests to back your claims of inaudibility. This must be a one-way street, right? We demand the tests but heaven forbid if someone asks us to run the same. All of a sudden the request becomes "painful." The person asking gets called names. They committed a crime. They asked for double blind tests. How dare they? We are beyond such requests. When we declare something to be inaudible, it is. We don't need to prove that against direct evidence that says otherwise. A published study in a conference? Means nothing. They must have made simple mistakes that we can point out. Except that we can't.
So we resort to personal remarks. That must be what "objective support" mean in the TOS #8. If the few of us talk loud enough, and rude enough, maybe that becomes objective.
You want to be so clearly on record as refusing to run double blind ABX test that I and a number of others on this and other forums have run, be my guest. The choice is yours.
Edit: fixed formatting.
So low-res audio is not about sales?
Sure it is. But it's not a $cam, like inaudible benefit "Hi-Rez" for more $$. I've never seen it advertised as "worry free" Hi-Rez. Have you?
Why would you pay more for the CD vs MP3 Ammar?
It's audibly better, I'm into home audio not ipods, storage is cheap, unlike in the era of mp3's birth.....and it isn't a $cam. Unfortunately for your amirlogic, the thread is Hi-Rez vs CD, zero to do with mp3.
Yes, you are both in the business of selling audio gear.
I sell speakers and you and Bob both sell audio gear, including "Hi-Rez" related products yes. The thread's about Hi-Rez and relates to the products you both promote and sell.
Big difference otherwise though. In the case of Bob Stuart, he comes with proper education and prestigious award of AES Fellow whose work is referenced and published. He has published a listening test/paper that garnered an award from his peers at Audion Engineering Society in the form of "Best Peer-Reviewed Paper."
You are a hobbyist who says one thing here, and then shows up at an audiophile show and does another. No educational experience. No professional experience. No record of taking any of these tests.
You're unraveling and letting me get in your head again as usual amir. The thread is about the BS paper, not our personalities. Please, "worry" about the BS paper, thanks. It's going to be judged on it's merit, not BS "credentials".
Well you must be or you would post under your real name Ammar. Be proud of your position and let your customers see who you really are.
Ummm, Amir, that's me in the avatar, me at all those very public audio shows, where people refer to me by my real initials "AJ", much like "JJ" or Amir"M".
Or "BS". Remember, one of the authors of the paper "Audibility of Typical Digital Filters in a Hi-Fi Playback System"? The thing you should be worried about.
If you care to address some of the issues, the dither doctoring, zero system transparency data, possibility of switching artifacts, the fact that no "CD"16/44 TPDF downsampled version of the 2L Hadyn file exists, etc., etc, please do so.
cheers,
AJ
So low-res audio is not about sales?
Sure it is. But it's not a $cam, like inaudible benefit "Hi-Rez" for more $$.
Who says? Can you prove you can hear the difference between CD and MP3 Ammar? This forum requires objective proofs as such. Otherwise, we have to declare it as not having a differential in quality and the extra price difference a $scam.
Remember, this forum requires such proofs of audible difference. See TOS #8 in my last post.
So you say you cannot hear audible problems but can tell the files apart (with a spectrum analyzer?),
Plot is lost again. I was very clear that I never analyzed the files, before or after the test.
But you admit to not hearing actually a difference? Okay, thanks that's all.
That you think these online discussions are so important that one should resort to winning them by cheating.
Sigh, it's not about the online discussion. You cannot tell anyone with a straight face that you have no affiliation with anyone selling hi-res "scam" (as AJ put it) so there seems to be an agenda.
But nothing of this really matters. You are simply derailing again.
I care about truth, but as soon as anyone inquires details you weasel out by posting mountains of nonsense, fallacies and excuses. Yes, that is painful. Everyone can see that.
You need to re-examine that logic.
We've already established several pages back that you better shouldn't talk about logic, like at all.
So you will say in each test I cheated in my logs yet when I ask you about others in this very forum, taking a test from one of our members, and you won't go there. What this means is that you have a personal beef with me. In other words, we are not having a discussion of science.
No. Learn to read. #495 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881978)
I'm not going to waste time explaining this AGAIN, like every other tiny thing I had to explain 10 times to you.
Yes, it does get painful when you demand evidence and turn around and accuse the other person after delivering it of a cheat. Please allow me to post this forum's Terms of Service #8:
8. All members that put forth a statement concerning subjective sound quality, must -- to the best of their ability -- provide objective support for their claims. Acceptable means of support are double blind listening tests (ABX or ABC/HR) demonstrating that the member can discern a difference perceptually, together with a test sample to allow others to reproduce their findings. Graphs, non-blind listening tests, waveform difference comparisons, and so on, are not acceptable means of providing support.
Jeez.. learn to read!
"Subjective sound quality" ... means statements like "the bass with device X sounds boosted" or "the cymbal crash in this mp3 sounds fuzzy".
"To the best of their ability" ... no comment
"the member can discern a difference" ... you admitted yourself you can't hear differences, you just produce ABX logs
"reproduce their findings" ... so far I haven't seen members here reproducing most of your claims
... and everyone here knows that ABX logs can be botched or false positives. The new ABX component, for example, has
known bugs that can lead to false positives. The person might not even realize that, hence my question
what you hear several pages back, which you evaded and weaseled out of painfully.
But you refuse to take the same tests to back your claims of inaudibility.
I stop here and skip the following garbage. This sentence alone is just
way too dumb.
It shall be my new forum signature.
I sell speakers...
You do more than that Ammar. This is what "parttimeaudiophile" web site says you showed at this year's AXPONA show:
A rack full of ModWright Instruments gear was driving the VSFT-3, including a $5,250 KWA 100SE amp and $3,750 LS100 tube preamp. For sources, AJ had a Modwright modded Sony HAP-Z1ES and Oppo BDP-105. I didn’t get a chance to explore the HAP-Z1ES all that much, but I’m pretty sure that was driving the bus while I was in the room. The mods are pretty similar to what I have in my own Modwright-ed Oppo, but include an external PSU, a tube stage and a completely reworked output stage. $3k gets the mod on your own unit.
MG Audio Design cables were used throughout, including the giant flat ribbon Planus III speaker cables and the silver Planus AG1 interconnects.
You have to be kidding me Ammar. MG AUdio Design Cables? Flat ribbon speaker cable? 10 foot version of that cable retails for $2,300 or $230/foot!
Modded Sony and Oppo players? Did you even get a measurement to see if these mods change the specs let alone have better sound? I have yet to run into one of these modder who have any instruments to verify their changes.
Oh well. Let me know when you actually believe in the stuff you are posting here.
Remember, this forum requires such proofs of audible difference. See TOS #8 in my last post.
Beware AJ, amirm has discovered the ToS. He doesn't understand it, but happy like a little child finding a dollar he spreads the word to everyone, just like he did after he found out that he can produce ABX logs.
So low-res audio is not about sales?
Sure it is. But it's not a $cam, like inaudible benefit "Hi-Rez" for more $$.
Who says?
You did:
You say I have claimed to hear audible problems. I have not.
You admitted you couldn't hear audio problems with the 16/44 version. So how isn't charging more and claiming
audible benefits a $cam?
Can you prove you can hear the difference between CD and MP3 Ammar?
I don't have to, as it's only purpose would be to Red Herring or Ad Hominem the arguments about the BS paper.
And no I won't meet you in the schoolyard for a fight after classes either Amir.
Pull yourself togther now, let's get back to the award winning BS paper.
Remember, this forum requires such proofs of audible difference. See TOS #8 in my last post.
If you feel I've violated forum rules, please report it. Otherwise, let's get back to the BS paper and you inability to defend it, the real source of your worries.
cheers,
AJ
I sell speakers...
You do more than that Ammar.
Wrong Amir. The only thing I sell is speakers. On my website, only speakers. Nothing else.
Now, we both know you do sell Hi-Rez products (Berkley) hardware. So I understand fully you've got some skin in the game and have a very strong desire for the BS paper to be legit.
So let's address some the issue around the BS test, like the dither doctoring, zero system transparency data, possibility of switching artifacts, the fact that no "CD"16/44 TPDF downsampled version of the 2L Hadyn file exists, etc., etc.
Are you a bit worried about some of these legitimate issues raised?
cheers,
AJ
Can you prove you can hear the difference between CD and MP3 Ammar?
I don't have to, as it's only purpose would be to Red Herring or Ad Hominem the arguments about the BS paper.
Ah, so the premium cost of CD over MP3 is justified. But past that, it is a sin.
The rules we invent. It is like saying you can go from 3 to 4 bedroom house. But oh, more than that? It is a sin.
And no I won't meet you in the schoolyard for a fight after classes either Amir.
Chicken.
If you feel I've violated forum rules, please report it. Otherwise, let's get back to the BS paper and you inability to defend it, the real source of your worries.
Seems like you missed what is on the title page:
). Award winning Ammar. Peer reviewed Ammar. Co-written by AES Fellow. With strong background in psychoacoustics.
Don't need no defending I say!
Tables are turned AJ. The impossible has happened. People have passed ABX DBT tests of high res vs CD's 16/44. Peer reviewed published tests have shown the same even down to the filtering alone.
And that is not all. We have just about all of you on record refusing to run double blind ABX tests. Maybe you could have lived with the above but with this? Not possible Ammar. You will be haunted by it and strongly so.
Me? I ran the tests. I held a position that has come true now. That quality engineering matters. That the good enough camp will have its hat handed to them one day. The day and year came .
Folks on AVS gave up on protests. Same will happen here. Just takes a bit of time for the reality to sink in. That by very standard we demand proof, proof has been provided. Attempts to change the rules now will just look silly as heck.
Grown up "objectivists" declaring ABX test logs subject to corruption! This whole forum is based on that bylaw. But we now say live witnesses are required. Don't come here with no DBT ABX results. That is what we are saying. Why? Because we can't allow ourselves to lose the "good enough" war.
Let me know when you understand truly what it means to be an objectivist because it is clear you are living same made up notion.
Edit: grammar and a couple of additions.
If you read the fine-print you'd have seen that the award was given on basis of the abstract and a 750-word summary, which has been peer-reviewed. They explicitly say that the paper itself has not been peer reviewed. How can a journal in good conscience hand out an award in that case?
And anyway, highlighting the names of the authors or the fact that this paper got an (dubious) award also means nothing at all (http://en.wikipedia.org/wiki/Argument_from_authority). What's more interesting is that all of the people work at a company with a strong interest to push "hi-res" mass-market adoption.
Are you ever going to stop trying to shift the burden of proof? Again, it makes you look retarded and impossible to take seriously. Your deluded statements do not help your case either. You may even be partially right, but no one here is going to take you seriously the way you state your case. I am not an engineer so I do not understand the technical details, but I do not need to if one side of this never ending debate is clearly irrational. You lose by default. It's getting old.
If you read the fine-print you'd have seen that the award was given on basis of the abstract and a 750-word summary, which has been peer-reviewed. They explicitly say that the paper itself has not been peer reviewed. How can a journal in good conscience hand out an award in that case?
They can because that is not the situation at hand as I explained to Arny earlier in this thread:
Winner of best *peer-reviewed* paper award. You say you have the paper but didn't see that? And positioned it as not being peer-reviewed?
The copy of the paper I just obtained from the AES web site starts out:
"This Convention paper was selected based on a submitted abstract and 750-word precis that have been peer reviewed
by at least two quali ed anonymous reviewers. The complete manuscript was not peer reviewed. This convention
paper has been reproduced from the author's advance manuscript without editing, corrections, or consideration by the
Review Board. The AES takes no responsibility for the contents."
Hi Arny. That is just a boilerplate AES puts on every conference paper these days. The Stuart paper is different and has this additional line I post:
Winner of the AES 137th Convention Best Peer-Reviewed Paper Award
That is the result of this (rather) new category of paper submission to the conference: http://www.aes.org/events/137/authors/137thCallForPapers.pdf (http://www.aes.org/events/137/authors/137thCallForPapers.pdf)
Authors may submit proposals in three categories:
1. Complete-manuscript peer-reviewed convention papers (submit at www.aes.org/137th_authors)
2. Abstract-precis-reviewed convention papers (submit at www.aes.org/137th_authors)
3. Synopsis-reviewed engineering briefs (submit at www.aes.org/137th_ebriefs)
For the complete-manuscript peer-reviewed convention papers (category 1), authors are asked to submit papers of 4–10 pages to the
submission site. Papers exceeding 10 pages run the risk of rejection without review. These complete-manuscript papers will be reviewed
by at least two experts in the field, and authors will be notified of acceptance by 2014 June 18. Final manuscript with revisions
requested by the reviewers have to be submitted before 2014 July 9. If rejected as a convention paper (Cat. 1), the proposal may still be
accepted for categories 2 or 3.
Award selection is then made from these complete, peer-reviewed papers. Winning the award opens the door to the paper being published in the journal.
I think we have slammed the door shut on any statement regarding standing of this paper in the eyes of professional AES community. It is excellent work. It is the most careful double blind test of small differences due to coding of audio that has been published. It is an inconvenient truth to be sure for many, but hopefully the search for knowledge and learning trumps that negative emotion.
As to your other comment,
And anyway, highlighting the names of the authors or the fact that this paper got an (dubious) award also means nothing at all (http://en.wikipedia.org/wiki/Argument_from_authority). What's more interesting is that all of the people work at a company with a strong interest to push "hi-res" mass-market adoption.
it seems the nature of Audio Engineering Society is not understood. This is an
industry organization. I would say some 90% of the papers are from companies describing the work they have done for their commercial enterprise. Check out Dr. Toole and Dr. Olive's work. Dr. Toole used to work for Harman and Dr. Olive still does. Dr. Olive was elected the president of AES. If you dispose of their work due to commercial interest, you will lose some of the best work done every about speakers and room acoustics. AES is where the industry gets together and shares knowledge.
Even if that were not the case, appearance of bias is not bias. Who do you think will go and rush and buy Meridian hardware because this paper came out? Doesn't your AVR player 24/192? How would you using that capability result in increased sales for Meridian? Yes, there is some side benefit in potentially justifying the apodizing filter and such in Meridian equipment but they have no lock on that and at any rate, that is a side comment.
All of this said, I praise the paper sarcastically because for weeks your members, mzil kept saying on AVS Forum thread that all sins of Meyer and Moran are to be forgiven because their work was "peer reviewed and published by AES." Here is what he said when like you, thought the paper was not peer reviewed:
In numbers, yes, but in terms of human hearing there is still, to this date, no peer reviewed, published study which has found even a single individual, anywhere, under any conditions, using either music or specially designed test tones (to make the challenge even easier), who can distinguish between CD quality sound and Hi-res even in an ideal, specially treated room, using $20K-$80K/pr speakers [$46k?], under controlled, blind, level matched conditions, sourced from the same master recording, at normal listening levels, with statistical significance.
As they say, the bed was made and now folks have to sleep in it .
As to appeal to authority, sad to see that argument used again when it has no relevance in real life. When our industry luminaries speak, we listen. They have knowledge, experience and data which we do not have. The notion that they are no better than AJ in florida, or xnor in Australia is absurd. Their work trumps any member posting under an alias with no industry or research experience. Please don't ask me to live in the fantasy we have created for ourselves otherwise .
Are you ever going to stop trying to shift the burden of proof? Again, it makes you look retarded and impossible to take seriously. Your deluded statements do not help your case either. You may even be partially right, but no one here is going to take you seriously the way you state your case. I am not an engineer so I do not understand the technical details, but I do not need to if one side of this never ending debate is clearly irrational. You lose by default. It's getting old.
Shift the burden of proof? Proof has been provided. Every DBT thrown at me I have taken and passed. We now are discussing double blind published study that has shown across 160 trials to better than 95% statistical confidence that listeners could hear difference between high resolution source and filtered one. The burden is all yours to prove otherwise.
And why do you think you were on the other side of the fence ever? I know we have convinced ourselves in forums that such proof can't exist. But that is symptom of living in a bubble, not reality. I can cite paper after paper that demonstrates CD can lack transparency. Here is professor Vanderkooy who despite his high qualification, is more on you all's side:
J. Vanderkooy. A digital-domain listening test
for high-resolution. AES 129th Convention,
San Fransisco, 2010.
At the outset, let me state my bias that CD-quality audio
(44.1 kHz, 16 bit) is essentially transparent. Under
pristine conditions it may just be possible to hear
residual channel noise.
[..]
My colleagues and I have done some fairly good tests
ourselves, and while we cannot say definitively that
good digital systems all sound alike, the significance of
any differences were minor in our assessment. Clearly
better tests are needed, and this proposal seeks to garner
broad support for a new test concept.
And so a better test was developed which is the topic of this thread. The above is reference [7] in Stuart's paper out of 35. Yes, 35 references. And you thought that work can be dismissed out of hand like that? Burden of proof? Really?
So please excuse me for once again being disappointed in another emotional outburst instead of a constructive technical contribution. Is there no hope that this forum acts professionally and the discussion remains on topic than personal?
Edit: typos.
Every DBT thrown at me I have taken and passed.
Amirm, I see you passed a comparison of non-linear filtering on this forum in the 20kHz brickwall filtering thread, and a particular comparison of non-aligned sources on another forum (AVS), in its "take 2" thread. I have not seen your attempt at ABXing linear filtering in the 20kHz brickwall thread, or time-aligned files in the AVS take 2 thread (my own samples X2 and Y2).
Given that your hearing is not as responsive to high frequencies as it used to be, it would be quite reasonable for you to decline. However if you wish to display a current prowess, then I would invite you specifically to try ABXing the linear 20kHz brickwall filter example on this forum,
1 and the time-aligned, level aligned, linear filtered for 44.1kHz, On_The_Street_Where_You_Live examples X2 and Y2 on AVS Forum.
2 If you try, and hear no differences, please advise.
_______________
The following footnotes identify where the comparison sources may be found.1. 20kHz reference and maximum phases files:I've filtered some 24/96 audio with the linear phase (050) and maximum phase (100) 20k impulses, and uploaded the results here:
...
The files I linked to in this thread have expired.
To make sure the files remain accessible, I've also put them online on my website.
Note that these are slightly modified files made on Jul 9 2010 by bandpass (http://www.hydrogenaudio.org/forums/index.php?showuser=56644).
The shell-script used to generate the latest samples is available as well.
...
limehouse_20k.zip (http://galaxyclassics.com/public/limehouse_20k.zip)
...
2. AVS forum time-aligned, level matched, linear filtered comparison files:{Link to AVS post = http://www.avsforum.com/forum/91-audio-the...ml#post29132777 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1598417-avs-aix-high-resolution-audio-test-take-2-a-20.html#post29132777) }
Amirm,
as you have indicated an ability to ABX test pairs provided by AVS for this thread, I wonder whether you would be good enough to try out the test pair below, On_The_Street_Where_You_Live_X2.flac and On_The_Street_Where_You_Live_Y2.flac, one of which is the first two minutes of a true 24/96 hi-res version as supplied by AVS Forum, and the other of which is the same file constricted to 16/44 and then resampled back to 24/96, using readily available software, the SoX utility.
I would be interested in:
- Whether you feel you hear differences.
- If so, whether you are able to establish that objectively using ABX software.
- How pronounced the subjective differences are for your ears, and what character the differences have for your ears.
I note that the dither used was a slightly modified version of TPDF dither. It is not the rectangular dither used for the AES 137th Convention Paper, The audibility of typical digital audio filters in a high-fidelity playback system.
I note also that the files are time aligned and level-matched. This can be verified by inverting one, playing both together, using an editor such as Audacity, and hearing silence.
{Link to AVS post = http://www.avsforum.com/forum/91-audio-the...ml#post29078954 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1598417-avs-aix-high-resolution-audio-test-take-2-a-19.html#post29078954) }
In this post, I provide an updated version of level-matched, time-aligned, files for the first two minutes of On_The_Street_Where_You_Live.
If anyone does try comparing X2 and Y2, at a normal, moderate listening volume, and succeeds with an ABX test, I'd be interested in any subjective comments as to how pronounced they found the difference(s) to be for their hearing.
...
The new files have been encoded as flac and can be downloaded from these links:
On_The_Street_Where_You_Live_X2.flac
On_The_Street_Where_You_Live_Y2.flac
If anyone wishes to verify the test files they are free to download SoX and then run the commands specified in the full wording of the above quoted post. (You may find it convenient to place the original 24/96 .wav file in the same directory as sox.exe and run the command prompt from that directory.)
Are you ever going to stop trying to shift the burden of proof? Again, it makes you look retarded and impossible to take seriously. Your deluded statements do not help your case either. You may even be partially right, but no one here is going to take you seriously the way you state your case. I am not an engineer so I do not understand the technical details, but I do not need to if one side of this never ending debate is clearly irrational. You lose by default. It's getting old.
Shift the burden of proof? Proof has been provided. Every DBT thrown at me I have taken and passed.
Which discredits the body of work. Nobody is perfect. Also the results could be confirmed to a higher degree with Foobar2000 R2, but very little such work has been provided. Even that which has been provided is incomplete - a complete set for a confirmable test includes the files that were compared.
And so a better test was developed which is the topic of this thread.
Except that neither the paper nor the proposed test are all that wonderful.
The paper is misnamed because in fact no actual "...Typical filters in a HiFi playback system..." were ever used. The Matlab simulations could have been reasonably exact, but several incorrect critical arameters were used.
Second naming issue is the fact that the paper purports to cover sample conversion from 24 bits to 16 bits but again there is debilitating asymmetry between the tests that were done and actual best practice. TPDF dither was not used. Spectrally shaped ditehr was not used. These technologies have been well known since the 1990s or earlier and one of the co-authors proudly designed and sold equipment featuring it. Note that typically the 24 -> conversion is not done in HiFi system by listeners at all, but is instead is done during production. So this part of the paper is completely off-topic as compared to its title, not to mention incorrectly performed to the point of incompetence if not malice.
The paper falls on its face right out the door by erroneously defining all ABX tests as being the 1950 test that everybody ignored for over 20 years after its publication for evaluating audio gear, because it was so inappropriate insiesitive and hard to use. A far better test was described in the JAES in 1982 and is widely used (and illustrated in the Meyers and Moran paper that this paper criticizes), but the improved test was ignored because long experience has taught us that ignoring ABX1982 is an established tradition in golden-eared circles. Of course Amir, you repeated the same mistake in your own work on this forum so I guess we know who you align yourself with.
The paper attempts a review of previous work but ignores the most serious problem with its immediate predecessor Meyer and Moran, which is that Meyer and Moran made the mistake of taking the music industry at their word and presumed that every recording that was labelled high resolution actually was high resolution. Glossed over was the fact that 1/3 to 1/2 of the entire universe of recordings they chose from were actually low resolution recordings. The paper cites other work that is so bad (Kunchur) that reasonableness forced them to criticize the laughable Kunchur work's major failing, but they did not identify which paper they were criticizing.
The paper bases itself on a goodly number of theories that have never been reliably tested and are therefore controversial. We have the proverbial house of cards built on shifting sands. For example there is a claim that it is reasonable to train listeners using recordings that were downsampled using filters with excessively narrow transition bands. but there is reasonable criticism of this theory on the grounds that the artifacts are ultrasonic. Examination of some of the program material that was used suggests that far poorer than best practices were used to produce it. For example it seems to contain passages with copious aliasing.
There appears to be no evidence that listener training involved anything but sighted evaluations, which probably partially explains the weak results. What's wrong with training people do do blind listening tests by letting them practice with blind listening tests? I think that part of the problem was the awkward nature of the blind tests themselves and their execution which seems to be largely manual.
Did I mention weak results? People, we got weak results. There is a tradition that requires an audible situation to be detectable 70% of the time to be worthy of serious consideration, but this paper only obtained 56% reliable detection. Even though a relatively large number of trials were run, only 95% confidence was achieved. The problem with this is that when there is 95% confidence we can expect about one chance in about 20 that the results were really due to random guessing. 17 different musical segments were tested, so the 1 in 20 chance may be a real possibility.
Furthermore the tests that form the core of the experimental data were also performed using filters with excessively narrow transition bands. If narrow transition bands are such a problem, why ruin your own work with them? Obviously, best current practices were not tested.
The alleged better test itself is not well described and seems to be restrictive and desensitizing to the listener.. The segments of music were rigidly defined and are too long to allow listeners to home in on critical passages and exploit the fact that memory for small audible details pretty well evaporates after 2 seconds or less. Ironically ABX1982 has always allowed the listener this flexibility and more. Perhaps the authors are behind on their reading. I don't see discussion of all relevant test parameters such as the listening room or critical details such as switch-over time.
There are many more points that should probably be criticized, but I don't want to beat an obviously dead horse. ;-)
Did I mention weak results? People, we got weak results. There is a tradition that requires an audible situation to be detectable 70% of the time to be worthy of serious consideration, but this paper only obtained 56% reliable detection. Even though a relatively large number of trials were run, only 95% confidence was achieved. The problem with this is that when there is 95% confidence we can expect about one chance in about 20 that the results were really due to random guessing. 17 different musical segments were tested, so the 1 in 20 chance may be a real possibility.
I found the paper a bit hard to follow in its presentation of the results in that the analysis was mainly based on lumping together results from the different test segments ("sections") of the music, where some sections yielded what were described as "high yield" results. The paper states (on page 9):
We analysed the percent-correct scores for each of sections 1-17 across all conditions, and found that some sections yielded a higher ratio of correct results than others. For example, sections 2, 6 and 17 gave correct-to-total ratios of 0.714, 0.710, and 0.760 respectively, whereas sections 1, 7 and 10 gave correct-to-total ratios of 0.541, 0.455 and 0.509 respectively; this is a mean difference of about 23%. [emphasis in bold added]I would have found it helpful if the results for condition 1
alone (the filtering for notional 44.1kHz sampling) had been provided on a section by section basis. The low pass processing for condition 1 used linear Finite Impulse Response filtering with a narrow transition band from 21591 to 22050 Hz.* If some sections were apparently more "revealing" of the effect of the filtering, it would have been of interest to see what the results were for condition 1 itself, without the co-mingling of the additional effects of reduction to 16 bits with no dither (condition 2) or with rectangular dither (condition 3).
Replicating the test filesHaving said the above, I note that attempts are being made in another thread to review the frequency distribution of the source material. See 15 Haydn_ String Quartet In D, Op_ 76, No_ 5 - Finale - Presto 3 best, 3 best music segments 9174 audibiliity of typical digital filters (http://www.hydrogenaud.io/forums/index.php?showtopic=107527&pid=881999&st=0&#entry881999).
Ideally at least some of the sections of the musical piece used for the paper would be available for download both in their reference form, and their filtered form, for HydrogenAudio members to listen to for themselves,
** and to form their own opinions on the apparent extent of any audible difference, and its subjective character. Also, it should be possible for someone to verify that the particular Matlab settings do in fact yield the filtered forms used. I presume the so-called "peer review" would merely have been of the wording in the paper. I presume it would
not have involved an audit inspection of the apparatus used to obtain the experimental data on which the paper was based.
_________
*
Not from 23500 to 24000 Hz as I misstated at my post #475 above. That transition band was for the filtering for notional 48 kHz sampling, condition 4. **
The files would be stereo 24/192 PCM.
Ah, so the premium cost of CD over MP3 is justified. But past that, it is a sin.
The rules we invent. It is like saying you can go from 3 to 4 bedroom house. But oh, more than that? It is a sin.
There's some Amirlogic in there somewhere. Unfortunately, only you know where/what it is.
+/- 10% volume method ring a bell?
Award winning peer-reviewed paper
Yes Amir, reviewed and awarded by 2 anonymous AES members. You and some other member with strong pecuniary interests in "Hi-Rez"?
Is that why the dither doctoring, zero system transparency data, possibility of loudspeaker and switching artifacts were all missed?
Stay tuned....
People have passed ABX DBT tests of high res vs CD's 16/44.
Yes, you and xnor have both showed passing gameable/cheatable online ABX tests is easy. But remember you have admitted to hearing no audible difference with these 16/44 ABX tests, just like xnor.
That your passing is due to other methods. Was it the +/- 10% volume method you invented many years ago for creating desired results (http://www.avsforum.com/forum/86-ultra-hi-end-ht-gear-20-000/1136745-establishing-differences-10-volume-method-14.html#post16385882)?
I think anyone here who has slogged even the page I linked will, realize that you've had a long history of making claims about digital maladies and creating results of your unsupervised "blind" tests that are, umm, slightly unorthodox.
It came as no surprise to anyone here then, that you admitted to not hearing audible differences in the samples and were differentiating them by alternate means.
But we now say live witnesses are required.
For your method of obtaining differences in the files when you admit to not hearing audible differences, yes.
But please also remember, the AES BS paper is based on supervised, formal blind tests, not online cheatable/gameable ABX computer games. Your luminary, BS, appears to dismiss ABX as a valid method anyway, so you can't have it both ways. Claim the results of your non-audible difference ABX tests are valid, when the BS paper says they are not.
cheers,
AJ
The paper is misnamed because in fact no actual "...Typical filters in a HiFi playback system..." were ever used.
Yes, I noted this (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=880436) a while back.
"Audibility of Atypical Digital Filters in a Hi-Fi Playback System"....
I probably should also have noted, a playback system with unknown switching, loudspeaker, etc, etc, transparency to the test.
It will be interesting if it's ever revealed who the 2 anonymous reviewers were. See if we know at least one.
cheers,
AJ
The paper is misnamed because in fact no actual "...Typical filters in a HiFi playback system..." were ever used.
Yes, I noted this (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=880436) a while back.
"Audibility of Atypical Digital Filters in a Hi-Fi Playback System"....
I probably should also have noted, a playback system with unknown switching, loudspeaker, etc, etc, transparency to the test.
It will be interesting if it's ever revealed who the 2 anonymous reviewers were. See if we know at least one.
IME the names of the people on the AES review board are in selected issues of the Journal. I've been reviewing the last 13 years worth of issue on my Kindle, and I know I've seen that list just lately. There are about 20 people on the list, several of whom I know personally. None were Amir.
The paper is misnamed because in fact no actual "...Typical filters in a HiFi playback system..." were ever used.
Yes, I noted this (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=880436) a while back.
"Audibility of Atypical Digital Filters in a Hi-Fi Playback System"....
I probably should also have noted, a playback system with unknown switching, loudspeaker, etc, etc, transparency to the test.
I think we know what speakers were ($80k/pr Meridians) , and they are active speakers so that covers the amplifier part. A Macbook's internal audio interface was the rest of the system.
What we don't know is what the room was like, listening distance, etc. I'll bet the speakers were boresighted on the listeners ears.
I think we know what speakers were ($80k/pr Meridians) , and they are active speakers so that covers the amplifier part. A Macbook's internal audio interface was the rest of the system.
Right, but my statement you quoted was that we don't know anything about the
transparency of the setup to the test. I've raised possible issues with the loudspeakers, the recent online ABX fiascoes show that the switching software needs to be transparent (and non-cheatable of course) also.
There is zero evidence of any of that for this test.
What we don't know is what the room was like, listening distance, etc. I'll bet the speakers were boresighted on the listeners ears.
Ye, I noticed that with amirs silence when I asked him if he could figure out what the speaker distance was based on the test description. The whole paper seems incredibly vague in many areas. Hopefully all will come to light at some point.
cheers,
AJ
This is clearly a hopeless case.
Let me close the chapter "amirm" (all based on evidence found in this thread (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881982)):
- stereotyping people, which he even admitted to
- keeps on making straw men and red herrings among countless other fallacies
- keeps shifting the burden of proof but doesn't even notice it
- sets up flawed arguments where he either wins or wins (that is all he seems to care about anyway)
- dares to speak about logic and science when he has demonstrated (dare I say it) willful ignorance
- thinks this is a war
- admitted to not hearing differences but still magically produces ABX logs
- rather posts multiple pages of nonsense than answering simple questions, one of which he finally explicitly evaded with a lame excuse
- poor reading comprehension (fun fact: I am from Australia according to him)
- even defends fallacies (!)
... and loads of other nonsense.
By now, and this is probably the worst, I think that most if not everyone here can see that "he has no interest in intellectual honesty or scientific rigor".
I take the paper for what it is. amir on the other hand blindly accepts whatever fits his agenda. amir, do you know that there are peer-reviewed papers for homeopathy, and actually quite a bit more than just 1 paper, but the scientific consensus is still: nonsense, quackery, a sham?
If you had any intellectual honesty, you would have to admit to believe in all kinds of nonsense based on some peer-reviewed papers.
This all boils down to: no credibility, no honesty, no interest in truth. I will simply link back to this post whenever it fits, and trust me, this will be every other post if he continues like that.
I am sorry that it came to that.
Ye, I noticed that with amirs silence when I asked him if he could figure out what the speaker distance was based on the test description.
You did? I don't read your posts all the time Ammar. But this one is pretty easy to answer. From the paper:
For a system gain of 75 dB, the loudest peak passage was measured as 102 dB SPL at the listening position, somewhat lower than the level we would expect from a live performance at a distance of 3 m.
So for us with the broken Imperial system, it was about 10 feet.
The whole paper seems incredibly vague in many areas. Hopefully all will come to light at some point.
To lay people, sure. Remember, AES presentations are not for random person on some audio forum. The audience is expected to be people not only in professional audio field, but with subject matter expertise. Remember that AES puts a burden on length of the paper:
For the complete-manuscript peer-reviewed convention papers (category 1), authors are asked to submit papers of 4–10 pages to the
submission site. Papers exceeding 10 pages run the risk of rejection without review. The current paper is 12 pages including its 35 references.
Compare this to 4 pages or so for Meyer and Moran where were note given the source frequency response, the system frequency response, the frequency response and crosstalk characteristics, frequency response of the restricted channel (i.e. 44.1 Khz), room description, etc., etc.
All of this said, sure, more detail is always better as it teaches us how these tests are conducted.
Every DBT thrown at me I have taken and passed.
Amirm, I see you passed a comparison of non-linear filtering on this forum in the 20kHz brickwall filtering thread, and a particular comparison of non-aligned sources on another forum (AVS), in its "take 2" thread. I have not seen your attempt at ABXing linear filtering in the 20kHz brickwall thread, or time-aligned files in the AVS take 2 thread (my own samples X2 and Y2).
Given that your hearing is not as responsive to high frequencies as it used to be, it would be quite reasonable for you to decline. However if you wish to display a current prowess, then I would invite you specifically to try ABXing the linear 20kHz brickwall filter example on this forum,1 and the time-aligned, level aligned, linear filtered for 44.1kHz, On_The_Street_Where_You_Live examples X2 and Y2 on AVS Forum.2 If you try, and hear no differences, please advise.
Hi ML. Good to see another familiar face here and a more reasonable one at that.
Turns out I did run one of the tests above. The results were posted in this thread though not that one. David was kind enough to link to it:
Another ABX report of these samples can be found here:
http://www.hydrogenaud.io/forums/index.php...st&p=880594 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=880594)
Here is the results once more:
----------
I gave it a try on the same clip MLXXX had done:
foo_abx 1.3.4 report
foobar2000 v1.3.5
2014/11/13 08:40:05
File A: C:\Users\Amir\Music\HA Forum Tests\limehouse\limehouse_maximum_phase_100.wav
File B: C:\Users\Amir\Music\HA Forum Tests\limehouse\limehouse_reference.wav
08:40:05 : Test started.
08:40:37 : 00/01 100.0%
08:42:19 : 00/02 100.0%
08:43:22 : 00/03 100.0%
08:44:21 : 01/04 93.8% <--- Difference found.
08:45:14 : 02/05 81.3%
08:45:21 : 03/06 65.6%
08:45:34 : 04/07 50.0%
08:45:43 : 05/08 36.3%
08:45:52 : 06/09 25.4%
08:46:00 : 07/10 17.2%
08:46:10 : 08/11 11.3%
08:46:20 : 09/12 7.3%
08:46:29 : 10/13 4.6%
08:46:39 : 11/14 2.9%
08:46:51 : 12/15 1.8%
08:47:00 : 13/16 1.1%
08:47:10 : 14/17 0.6%
08:47:18 : 15/18 0.4%
08:47:26 : 16/19 0.2%
08:47:34 : 17/20 0.1%
08:47:42 : 18/21 0.1%
08:47:49 : 19/22 0.0%
08:47:55 : Test finished.
----------
Total: 19/22 (0.0%)
=========
I also re-ran it with the latest ABX plug-in with signatures:
2014-11-13 09:16:06
File A: limehouse_maximum_phase_100.wav
SHA1: 722dc26db8d4ce666dc03875b2c8d4570d22b521
File B: limehouse_reference.wav
SHA1: e8ad96830d23cad4bba5bf822ce875ae452b9e7c
Output:
DS : Primary Sound Driver
09:16:06 : Test started.
09:16:48 : 01/01
09:16:56 : 02/02
09:17:04 : 03/03
09:17:14 : 04/04
09:17:21 : 05/05
09:17:29 : 06/06
09:17:38 : 07/07
09:17:45 : 08/08
09:17:52 : 09/09
09:18:02 : 10/10
09:18:08 : 11/11
09:18:14 : 12/12
09:18:20 : 13/13
09:18:28 : 14/14
09:18:36 : 15/15
09:18:42 : 16/16
09:18:42 : Test finished.
----------
Total: 16/16
Probability that you were guessing: 0.0% -- signature --
5b42b06c414b6ba77a3998695bf119a2d57663c0
===========
While I have you here, would you please represent if you cheated in any way in running the above test and getting positive outcome? I am asking because I post the above but I am accused of cheating by xnor and Ammar. More in the next post.
2014-11-13 09:16:06
File A: limehouse_maximum_phase_100.wav
SHA1: 722dc26db8d4ce666dc03875b2c8d4570d22b521
File B: limehouse_reference.wav
SHA1: e8ad96830d23cad4bba5bf822ce875ae452b9e7c
Output:
DS : Primary Sound Driver
09:16:06 : Test started.
09:16:48 : 01/01
[... snip ...]
09:18:42 : 16/16
09:18:42 : Test finished.
----------
Total: 16/16
Probability that you were guessing: 0.0%
-- signature --
5b42b06c414b6ba77a3998695bf119a2d57663c0
===========
While I have you here, would you please represent if you cheated in any way in running the above test and getting positive outcome? I am asking because I post the above but I am accused of cheating by xnor and Ammar. More in the next post.
>>>
0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029) <<<
Arny, I agree, especially your comments about the filter parameters.
Luckily, even if the effects of pre-ringing at 21+ kHz were audible to some special people with special signals, you can simply re-filter with a less steep filter to reduce the pre-ringing at 21+ kHz.
edit: Depending on what you want you can easily reduce pre-ringing of Meridian's filter down to ~1/6th, better than the ringing in 48 kHz converters if you want. That is without using any imaging tricks.
Ye, I noticed that with amirs silence when I asked him if he could figure out what the speaker distance was based on the test description.
You did? I don't read your posts all the time Ammar. But this one is pretty easy to answer. From the paper:
For a system gain of 75 dB, the loudest peak passage was measured as 102 dB SPL at the listening position, somewhat lower than the level we would expect from a live performance at a distance of 3 m.
So for us with the broken Imperial system, it was about 10 feet.
Right. So were the test subjects sitting 10' from the speakers, or is 102db somewhat lower than the level we would experience with a listener sitting 10' from a "live performance", like an orchestral stage?
You are saying it's absolutely 10' from speakers Amir? No clarification needed? Cause that pushes things up to (ballpark) around 108db @ 1m from loudspeaker, which is awful close to their specified 112db absolute maximum.
I wonder what sort of non-linearties might appear when a DR 1" Beryllium metal dome is driven that hard with and without 22k band limiting? Might there be some down in the audible bandwidth <20k? When 1' DR domes typically start exhibiting as much as 3db compression at a mere 95db? Hmmm...
Let's see from the info provided by the paper:
Compare this to 4 pages or so for Meyer and Moran....
Nope, just another Red Herring by the lay person who is devoid of logic as lay people often are, thus cannot comprehend it has no relevance to the validity of the BS paper results. The BS paper/results/validity is all that matters here.
cheers,
AJ
Given that your hearing is not as responsive to high frequencies as it used to be, it would be quite reasonable for you to decline. However if you wish to display a current prowess, then I would invite you specifically to try ABXing the linear 20kHz brickwall filter example on this forum,1 and the time-aligned, level aligned, linear filtered for 44.1kHz, On_The_Street_Where_You_Live examples X2 and Y2 on AVS Forum.2 If you try, and hear no differences, please advise.
I addressed your first request in my last post. On running your set of converted AVS files, as you know, I did give them a try by downloading the 150 (?) megabyte package:
First, thanks so much for creating and uploading these files. This is what I call constructive discussion .
Something is very puzzling or drastically wrong though if I am understanding you right. I compared X and Y against each other in foobar. I immediately heard a difference at the start of the tracks. One has tons more noise added to it. You can hear it both before the music starts and during. The noise levels are very high and not consistent with proper dither.
I pulled the files into Audition and played them with the same issue. I zoomed into the start of the "Y" file and it indeed has noise that is elevated to -75 dbFS (~12 bits of signal to noise ratio). The file shows the 22 Khz cut off so it is the Audacity converted one. This is really, really broken if that is the conversion that Audacity performed. Would you please verify your workflow and confirm what I am saying here? Because if correct, it means Audacity signal processing is worse than junk. And the generated noise will really screw up any "null" type tests as it will dwarf signal dependent differences.
Members here should notice that I did not provide any ABX results. Would have been an easy win. But not an ethical one. So I didn't and instead let you know what was broken in the test.
As to running them again, I am sorry but I won't do it. The mountain of ABX results I have provided has meant nothing to the objectors here. This is tedious work and something I don't enjoy doing. But let's ask other people to do it and post their results.
Paraphrasing/summarizing the Stuart paper, a copy which I have, in my very layman way, the abstract could be rewritten as,
We have determined that it is possible to create a coin that comes up heads about 56% of the time with high reliability.
I am asking because I post the above but I am accused of cheating by xnor and Ammar.
That is a lie. I said the
tests were cheatable.
I said you are determining the files apart by means other than
hearing audible differences. You have already admitted this:
You say I have claimed to hear audible problems. I have not. I have shown that I can tell files apart.
However, none of this is relevant to the BS paper and the blind tests they performed to obtain their results. As a matter of fact, the BS paper largely dismisses the results of ABX type tests, which would include your online cheatable games versions.
You can't have it both ways Amir.
cheers,
AJ
As to running them again, I am sorry but I won't do it.
I agree, as ABX (specifically unsupervised tests run on Windows computers) is generally poor for these small impairments, especially the psychogenic ones created by "worry"!!!
I'd suggest MUSHRA, but your friend/neighbor within state JJ would be ideal to select, set up and administer such a test/method for you Amir.
Then the whole world would know who Amir Bolt is.
The arguments that Hi-Rez 2ch is an utter $cam would be laid to rest, as would this doctored BS paper.
You hold the keys in your hands for your camp Amir. Victory is possible. What a shame the car won't start.
cheers,
AJ
I wonder what sort of non-linearties might appear when a DR 1" Beryllium metal dome is driven that hard with and without 22k band limiting? Might there be some down in the audible bandwidth <20k? When 1' DR domes typically start exhibiting as much as 3db compression at a mere 95db? Hmmm...
I already addressed this Ammar. Here is the explanation again.
The peak SPL measurements are performed using test signals, not the music that the listeners heard. We have the full spectrum of the music that was heard in the paper:
. The total energy hitting the tweeter is pretty small in the ultrasonic region.
I don't know the crossover frequency of the speakers. But taking a guess of a few Khz, the level there averages about 50 db with peaks of about 70 db. Far cry from 95 db.
Ooops. Oh, that's right, the paper has zero relevant info regarding the tweeters performance within the audible band, at 108db, or anything regarding speakers transparency to the test.
No one listened to full spectrum white noise at 108 db spl. They listened to music. The relevant information is right up there in the measurements.
Maybe the utter lack of switching software transparency info? The filter "emulations"?
Many other people know that answer here but none are replying to correct your misconception. Anyway, I will explain it to you.
The filters are created in Matlab. Matlab is a math and signal processing program and lets one create such transformations as filter with ease. They read in the source file at 192 Khz, create the filtered versions and then using a GUI front-end they wrote for Matlab, switch between the streams. Matlab is either playing unprocessed source file in memory, or the processed one. The software pipeline through the computer is identical otherwise.
There is no hardware switch box. Just as there is none in Foobar when we compare two files with it.
So the switching fixture is transparent by nature.
The lack of clarity that you speak of symptomatic of not knowing the basics of tools and techniques in this field. That is to be expected. The paper as I keep saying, is written for other industry experts, not lay hobbyists on forums with no signal processing background.
I am asking because I post the above but I am accused of cheating by xnor and Ammar.
That is a lie. I said the tests were cheatable.
Well, this is your chance to clarify your position.
I pulled in the files as presented to me for the tests into the Foobar playlist. I select them, right click and invoke ABX plug-in. I select a segment and listen to A and B to see if I can reliably tell them apart. If I think I can, then I attempt to identify X/Y as being one of them.
If I fail, I go back to pick a different segment. Here is example results:
foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/11 06:18:47
File A: C:\Users\Amir\Music\AIX AVS Test files\Mosaic_A2.wav
File B: C:\Users\Amir\Music\AIX AVS Test files\Mosaic_B2.wav
06:18:47 : Test started.
06:19:38 : 00/01 100.0%
06:20:15 : 00/02 100.0%
06:20:47 : 01/03 87.5%
06:21:01 : 01/04 93.8%
06:21:20 : 02/05 81.3%
06:21:32 : 03/06 65.6%
06:21:48 : 04/07 50.0%
06:22:01 : 04/08 63.7%
06:22:15 : 05/09 50.0%
06:22:24 : 05/10 62.3%
06:23:15 : 06/11 50.0% <---- difference found reliably. Note the 100% correct votes from here on.
06:23:27 : 07/12 38.7%
06:23:36 : 08/13 29.1%
06:23:49 : 09/14 21.2%
06:24:02 : 10/15 15.1%
06:24:10 : 11/16 10.5%
06:24:20 : 12/17 7.2%
06:24:27 : 13/18 4.8%
06:24:35 : 14/19 3.2%
06:24:40 : 15/20 2.1%
06:24:46 : 16/21 1.3%
06:24:56 : 17/22 0.8%
06:25:04 : 18/23 0.5%
06:25:13 : 19/24 0.3%
06:25:25 : 20/25 0.2%
06:25:32 : 21/26 0.1%
06:25:38 : 22/27 0.1%
06:25:45 : 23/28 0.0%
06:25:51 : 24/29 0.0%
06:25:58 : 25/30 0.0%
06:26:24 : Test finished.
----------
Total: 25/30 (0.0%)
You can see how I went from not being able to tell the difference, to being able to do so with 100% confidence. Not some total that statistically was valid. But with absolutely correctness in trial after trial.
I hereby represent that the above is all I did. I did not use a spectrum analyzer. I did not edit the log file. I listened until I could identify A/B in double blind ABX testing. That is it. There is no more to the story.
I like to bring all of this to a close Ammar. Please clearly state on the record that you do or do not trust above to be what I have done. If it is not, then please state on what basis/data you are claiming so. Otherwise I hope this is the last time you make an accusation of unethical conduct to me.
foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/11 06:18:47
File A: C:\Users\Amir\Music\AIX AVS Test files\Mosaic_A2.wav
File B: C:\Users\Amir\Music\AIX AVS Test files\Mosaic_B2.wav
06:18:47 : Test started.
06:19:38 : 00/01 100.0%
06:20:15 : 00/02 100.0%
06:20:47 : 01/03 87.5%
06:21:01 : 01/04 93.8%
06:21:20 : 02/05 81.3%
06:21:32 : 03/06 65.6%
06:21:48 : 04/07 50.0%
06:22:01 : 04/08 63.7%
06:22:15 : 05/09 50.0%
06:22:24 : 05/10 62.3%
06:23:15 : 06/11 50.0% <---- difference found reliably. Note the 100% correct votes from here on.
[... snip ...]
06:26:24 : Test finished.
----------
Total: 25/30 (0.0%)
You can see how I went from not being able to tell the difference, to being able to do so with 100% confidence. Not some total that statistically was valid. But with absolutely correctness in trial after trial.
I hereby represent that the above is all I did. I did not use a spectrum analyzer. I did not edit the log file. I listened until I could identify A/B in double blind ABX testing. That is it. There is no more to the story.
[/color][/b]
You also admit to not hearing an audible difference.
>>>
0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029) <<<
So that others, if they so choose, can replicate your work, Amir, would you please provide the approximate time for what you heard as well as describe the phenomena?
Every DBT thrown at me I have taken and passed.
Amirm, I see you passed a comparison of non-linear filtering on this forum in the 20kHz brickwall filtering thread, and a particular comparison of non-aligned sources on another forum (AVS), in its "take 2" thread. I have not seen your attempt at ABXing linear filtering in the 20kHz brickwall thread, or time-aligned files in the AVS take 2 thread (my own samples X2 and Y2).
Given that your hearing is not as responsive to high frequencies as it used to be, it would be quite reasonable for you to decline. However if you wish to display a current prowess, then I would invite you specifically to try ABXing the linear 20kHz brickwall filter example on this forum,1 and the time-aligned, level aligned, linear filtered for 44.1kHz, On_The_Street_Where_You_Live examples X2 and Y2 on AVS Forum.2 If you try, and hear no differences, please advise.
Hi ML. Good to see another familiar face here and a more reasonable one at that.
Turns out I did run one of the tests above. The results were posted in this thread though not that one. David was kind enough to link to it:
Another ABX report of these samples can be found here:
http://www.hydrogenaud.io/forums/index.php...st&p=880594 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=880594)
Here is the results once more:
----------
I gave it a try on the same clip MLXXX had done:
foo_abx 1.3.4 report
foobar2000 v1.3.5
2014/11/13 08:40:05
File A: C:\Users\Amir\Music\HA Forum Tests\limehouse\limehouse_maximum_phase_100.wav
File B: C:\Users\Amir\Music\HA Forum Tests\limehouse\limehouse_reference.wav
08:40:05 : Test started.
08:40:37 : 00/01 100.0%
08:42:19 : 00/02 100.0%
08:43:22 : 00/03 100.0%
08:44:21 : 01/04 93.8% <--- Difference found.
08:45:14 : 02/05 81.3%
08:45:21 : 03/06 65.6%
08:45:34 : 04/07 50.0%
08:45:43 : 05/08 36.3%
08:45:52 : 06/09 25.4%
08:46:00 : 07/10 17.2%
08:46:10 : 08/11 11.3%
08:46:20 : 09/12 7.3%
08:46:29 : 10/13 4.6%
08:46:39 : 11/14 2.9%
08:46:51 : 12/15 1.8%
08:47:00 : 13/16 1.1%
08:47:10 : 14/17 0.6%
08:47:18 : 15/18 0.4%
08:47:26 : 16/19 0.2%
08:47:34 : 17/20 0.1%
08:47:42 : 18/21 0.1%
08:47:49 : 19/22 0.0%
08:47:55 : Test finished.
----------
Total: 19/22 (0.0%)
=========
I also re-ran it with the latest ABX plug-in with signatures:
2014-11-13 09:16:06
File A: limehouse_maximum_phase_100.wav
SHA1: 722dc26db8d4ce666dc03875b2c8d4570d22b521
File B: limehouse_reference.wav
SHA1: e8ad96830d23cad4bba5bf822ce875ae452b9e7c
Output:
DS : Primary Sound Driver
09:16:06 : Test started.
09:16:48 : 01/01
09:16:56 : 02/02
09:17:04 : 03/03
09:17:14 : 04/04
09:17:21 : 05/05
09:17:29 : 06/06
09:17:38 : 07/07
09:17:45 : 08/08
09:17:52 : 09/09
09:18:02 : 10/10
09:18:08 : 11/11
09:18:14 : 12/12
09:18:20 : 13/13
09:18:28 : 14/14
09:18:36 : 15/15
09:18:42 : 16/16
09:18:42 : Test finished.
----------
Total: 16/16
Probability that you were guessing: 0.0%
-- signature --
5b42b06c414b6ba77a3998695bf119a2d57663c0
===========
While I have you here, would you please represent if you cheated in any way in running the above test and getting positive outcome? I am asking because I post the above but I am accused of cheating by xnor and Ammar. More in the next post.
W
You also admit to not hearing an audible difference.
No I have not. I said I heard no audible *problems*. That is *not* the same as audible differences. This is what you asked me:
Not only did you claim to hear clear audible problems, but run around showing off ABX logs. You say all it comes down to, to hear this 21+ kHz ringing, is training.
Oh? Again, the plot is lost. Here is my comment that you objected to and caused the 12-part replies:
Having read the Stuart paper and prior citations, I am starting to think this may be due to time domain impact. It certainly is not frequency domain since I can't hear the ultrasonics. Just as pre-echo is a very audible time domain artifact (although created in frequency domain), maybe these filters act the same way.
You say I have claimed to hear audible problems. I have not. I have shown that I can tell files apart. That is what is required of me: detect lack of transparency.
A problem denotes something specific. ABX testing does not diagnose problems. Its sole job is to see if the tester can tell the files apart by listening. That is what I have done. I listened to the files and found the difference. That is it. I explained every step in my previous post.
>>> 0 credibility <<<
It is not material what a random poster on a forum thinks of the data in front of them. No one learns anything from that unless you are telling us you are some kind of authority to be believed with no foundation to your accusations. Are you?
I have asked you to contribute constructively to the discussion by running the tests that you say we have must have cheated to pass. You won't. Instead you keep getting personal this way which does nothing but raise the noise floor of the forum.
Since you won't listen to my pleading to put aside your emotions and act objectively, please allow me to deal with this differently since this is what is dominating the discussion now:
Does the larger membership and forum moderators think this is what the discussion should be? That we ask for ABX DBT results as the forum TOS #8 requires but the moment it is present, we call the person a cheater? Is this how you like to be known? I personally am fine with folks going there. It shows total lack of objectivity and interest in the technical topic. It becomes a referenceable link in the future when we stand on high horse and claim we are all about double blind tests. We are that if it produces negative results. If it produces positive, oh, the person must have cheated. Sorry, no. That is not objectivity. That is biased thinking.
Edit: highlighted above in red.
So that others, if they so choose, can replicate your work, Amir, would you please provide the approximate time for what you heard as well as describe the phenomena?
No. As I have mentioned, foobar does not save the segments markers and I have no record of them as such.
I did provide some data though in one of the tests: http://www.whatsbestforum.com/showthread.p...ll=1#post277513 (http://www.whatsbestforum.com/showthread.php?15255-Conclusive-quot-Proof-quot-that-higher-resolution-audio-sounds-different&p=277513&viewfull=1#post277513)
Not quite the topic of the thread but nevertheless part of similar arguments, here is a comparison of Arny's 24/96 file downsampled to 16/44.1 compared to 320 kbps MP3 of the same:
=================
foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/19 19:45:33
File A: C:\Users\Amir\Music\Arnys Filter Test\keys jangling 16 44.wav
File B: C:\Users\Amir\Music\Arnys Filter Test\keys jangling 16 44_01.mp3
19:45:33 : Test started.
19:46:21 : 01/01 50.0%
19:46:35 : 02/02 25.0%
19:46:49 : 02/03 50.0% << dog barked in my ear wanting to go out
19:47:03 : 03/04 31.3%
19:47:13 : 04/05 18.8%
19:47:27 : 05/06 10.9%
19:47:38 : 06/07 6.3%
19:47:46 : 07/08 3.5%
19:48:01 : 08/09 2.0%
19:48:19 : 09/10 1.1%
19:48:31 : 10/11 0.6%
19:48:45 : 11/12 0.3%
19:48:58 : 12/13 0.2%
19:49:11 : 13/14 0.1%
19:49:28 : 14/15 0.0%
19:49:52 : 15/16 0.0%
19:49:56 : Test finished.
----------
Total: 15/16 (0.0%)
I just selected the beginning of the file and the difference was very clear to my ears.
So let's have you all run the same test and see what you hear. Please don't keep asking me to explain to you what a Quince tastes like. I can't. You need to eat one (preferably a ripe one!). Better yet, smell one.
Note that as I have quoted multiple times, we do have such data for Stuart's test:
It was reported that filtering gave "softer edges" to
the instruments, and "softer leading edges" to musi-
cal features with abrupt onsets or changes. Echoes,
when audible, were identi ed as being a ected the
most clearly by the ltering. It was felt that some
of the louder passages of the recording were less ag-
gressive after ltering, and that the inner voices (sec-
ond violin and viola) had "a nasal quality". Over-
all, the ltered recording gave a "smaller and
atter auditory image", and speci cally the physical space
around the quartet seemed smaller.
Listeners described that quantization gave a "rough-
ness" or "edginess" to the tone of the instruments,
and that quantization had a signi cant impact on
decay, particularly after homophonic chords, where
"decay was sustained louder for longer and then died
suddenly". This could be an e ect of quantization
distortion; it is interesting that this was audible even
in a 24-bit system, and is consistent with the hy-
potheses of Stuart [29] that 16 bits are not sucient
for inaudible quantization.Did this do anything for you? I suspect the answer is no .
Just take the test for heaven's sake. Why is there so much hesitation to do so? Take it. Listen carefully. Try hard. You know others did it. Use that as motivation to find the differences and become a more critical listener. This is what others who took the test have gained.
Start with the easier MP3 case above. If you can't tell the difference at 320, try it at 256. Keep going lower until you do. Then work backward to higher bit rates and focus on the same differences you heard before.
Here is the results once more:
----------
I gave it a try on the same clip MLXXX had done:
foo_abx 1.3.4 report
foobar2000 v1.3.5
2014/11/13 08:40:05
File A: C:\Users\Amir\Music\HA Forum Tests\limehouse\limehouse_maximum_phase_100.wav
File B: C:\Users\Amir\Music\HA Forum Tests\limehouse\limehouse_reference.wav
...
===========
While I have you here, would you please represent if you cheated in any way in running the above test and getting positive outcome? I am asking because I post the above but I am accused of cheating by xnor and Ammar.
Well with the maximum phase file compared with the reference file, I did hear a slight dullness, back in January 2009. At the time my equipment was fairly basic. I used pc motherboard high definition sound, and medium quality hi-fi speakers. (These days if attempting such an ABX I'd use my Xonar sound card, and Sennheiser HD 800 headphones.)
I certainly didn't cheat. I simply performed the ABX with foobar, once I felt confident that I could hear a difference.
A maximum phase filter can delay frequencies extending well down into the audible range. I have found that the timbre of a tone with harmonics can change if one of the harmonics is altered in phase relative to the fundamental. [As a simple practical test, if I use Audacity at a sample rate of 96kHz to create a track with a 3kHz sinewave at a level of 0.5, and another track with a 9kHz sinewave at a level of 0.1, the addition of the 9kHz track alters the character of the sound. If I then delay the 9kHz tone by 90 degrees and play both tracks again, I find the sound has a slightly different character again, compared with the 9kHz tone being in phase.]
I was really more interested in your trying 2Bdecided's
linear phase file _50.wav (as I mention below). I recall that I was unable to discern any audible difference with that file, in January 2009.
Given that your hearing is not as responsive to high frequencies as it used to be, it would be quite reasonable for you to decline. However if you wish to display a current prowess, then I would invite you specifically to try ABXing the linear 20kHz brickwall filter example on this forum,1 and the time-aligned, level aligned, linear filtered for 44.1kHz, On_The_Street_Where_You_Live examples X2 and Y2 on AVS Forum.2 If you try, and hear no differences, please advise.
I addressed your first request in my last post.
You may have misunderstood my first request. I was not asking you to reproduce an ABX log for 2Bdecided's
maximum phase file, as I had already seen an ABX log by you for that file. I was asking you to try his
linear ("50") phase file, a much more difficult challenge, given that a linear phase file for 22.05 kHz should not affect the phase of frequencies in the audible range. I felt this request quite relevant, in light of your suggestion that the [linear] filtering for the BS paper might have been audible, on account of its effects in the time domain.
On running your set of converted AVS files, as you know, I did give them a try by downloading the 150 (?) megabyte package:
First, thanks so much for creating and uploading these files. This is what I call constructive discussion .
Something is very puzzling or drastically wrong though if I am understanding you right. I compared X and Y against each other in foobar. I immediately heard a difference at the start of the tracks. [color="#FF0000"]One has tons more noise added to it. [/color] You can hear it both before the music starts and during. The noise levels are very high and not consistent with proper dither.
I pulled the files into Audition and played them with the same issue. I zoomed into the start of the "Y" file and it indeed has noise that is elevated to -75 dbFS (~12 bits of signal to noise ratio). The file shows the 22 Khz cut off so it is the Audacity converted one. This is really, really broken if that is the conversion that Audacity performed. Would you please verify your workflow and confirm what I am saying here? Because if correct, it means Audacity signal processing is worse than junk. And the generated noise will really screw up any "null" type tests as it will dwarf signal dependent differences.
Members here should notice that I did not provide any ABX results. Would have been an easy win. But not an ethical one. So I didn't and instead let you know what was broken in the test.
As to running them again, I am sorry but I won't do it. The mountain of ABX results I have provided has meant nothing to the objectors here. This is tedious work and something I don't enjoy doing. But let's ask other people to do it and post their results.
I didn't provide an extended response on AVS forum at the time when the criticism was first made, but after having had time to analyse the files I am sorry but I disagree. A null test with the files I initially provided is in fact quite successful
at a moderate listening volume. With moderate gain, the dither cannot be heard, not for my hearing anyway! The frequency components peaked at -104dB at around 9kHz, as can be seen in the first graph [after accounting for the 30dB amplification] in the attachment to AVS forum post #564 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1598417-avs-aix-high-resolution-audio-test-take-2-a-19.html#post29078954) of the AVS Forum thread
AVS/AIX High-Resolution Audio Test: Take 2. I do not see noise at -75dB in any frequency band in the first graph for the first second (after allowing for the 30dB amplification I performed before running the frequency analysis). I would concede that applying 30dB amplification to low level quantisation/dither noise for the purpose of a frequency analysis graph might create some degree of anomaly in accuracy of the dB levels, as the noise is of such a low level, spread over a very wide range of frequencies.
On reflection, I don't think the conversion was "broken", merely that the dither that the old version of Audacity had created was somewhat noisy. However, as it was noisier than usual practice, and in view of the complaints, by yourself, and others, I went to the trouble of creating replacement files X2 and Y2, using SoX, set for a slightly modified TPDF dither.
I note that one person on AVS (chip_mk) did comment on my files X2 and Y2. He stated that they were subjectively more similar to each other than the AVS Forum A2 and B2 files (for On_The_Street_Where_You_Live). Despite the lack of corroboration of the AVS forum member's claims by way of ABX software, I took this subjective comment as a vindication of my efforts. Certainly for my ears the AVS files had a difference in "tone" in addition to the misalignment of approximately 10mS. I think these factors were enough to invalidate the AVS Forum exercise, i.e. it was too easy to tell the files apart (as I had done with a formal ABX). Extraneous factors had crept in, firstly the lack of time alignment, and secondly some tonal difference. I am unable to explain how the tonal difference arose. Why AVS Forum could not simply have provided time-aligned files capable of being audibly nulled at a moderate listening gain (by inverting one of the files) is a little surprising. I guess it was due in part to lack of familiarity with this type of exercise, and the importance of matching the files. Not only were the files not time aligned, but even when time aligned and with one file inverted, they would not null to silence at a moderate gain setting.
I agree with your comment that ABX testing is tedious work. It can be even more tedious if differences are very subtle. I take on the challenge only very occasionally myself. I make a record of the character of the difference for my hearing.
Ooops. Oh, that's right, the paper has zero relevant info regarding the tweeters performance within the audible band, at 108db, or anything regarding [!--sizeo:3--][span style=\"font-size:12pt;line-height:100%\"][!--/sizeo--]speakers[/size] transparency to the test.
That's too bad.
The peak SPL measurements are performed using test signals, not the music that the listeners heard.
For a system gain of 75 dB, the loudest peak [!--sizeo:4--][span style=\"font-size:14pt;line-height:100%\"][!--/sizeo--]passage[/size] was measured as 102 dB SPL at the listening position, somewhat lower than the level we would expect from a live performance at a distance of 3 m.
The irrelevant information is right up there in the measurements.
1pas·sage
noun \ˈpa-sij\
b : a phrase or [!--sizeo:3--][span style=\"font-size:12pt;line-height:100%\"][!--/sizeo--]short section of a musical composition[/size]
Sorry Amir, this loudspeaker stuff is way over your head. Huge problem when hobbyists try to analyze deliberately cryptic professional work.
Maybe the utter lack of switching [!--sizeo:3--][span style=\"font-size:12pt;line-height:100%\"][!--/sizeo--]software[/size] transparency info?
There is no hardware switch box. Just as there is none in Foobar when we compare two files with it.
So the switching fixture is transparent "by nature".
"Worry" syndrome obviously creates serious reading comprehension problems.
The lack of clarity that you I speak of with is symptomatic of not knowing the basics of tools and techniques in this field. That is to be expected. The paper as I keep saying, is written for other industry experts, not lay hobbyists on forums with no signal processing background.
Agreed.
cheers,
AJ
I hereby represent that the above is all I did.
Uh huh.
No I have not. [...]
I've already dealt with your excuses. 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
It is not material what a random poster on a forum thinks of the data in front of them. No one learns anything from that unless you are telling us you are some kind of authority to be believed with no foundation to your accusations. Are you?
1) Exactly.
2) Another fallacy. See 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
I have asked you to contribute constructively to the discussion by running the tests that you say we have must have cheated to pass. You won't. Instead you keep getting personal this way which does nothing but raise the noise floor of the forum.
Stop posting noise and the noise floor will go down. See 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
Since you won't listen to my pleading to put aside your emotions and act objectively, please allow me to deal with this differently since this is what is dominating the discussion now:
This is based objectively on what you posted in the past: 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
Does the larger membership and forum moderators think this is what the discussion should be? That we ask for ABX DBT results as the forum TOS #8 requires but the moment it is present, we call the person a cheater? Is this how you like to be known?
I've already dealt with your fallacious arguments and deliberate obtuseness. See 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029) and posts before that.
I personally am fine with folks going there. It shows total lack of objectivity and interest in the technical topic. It becomes a referenceable link in the future when we stand on high horse and claim we are all about double blind tests. We are that if it produces negative results. If it produces positive, oh, the person must have cheated. Sorry, no. That is not objectivity. That is biased thinking.
I have a referenceable link for your nonsense: 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
Future posts of nonsense will be condensed into the most funny line to spare other readers of risking to read the same nonsense twice accidentally.
Adding to my previous post (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882035) about ringing.
Here's what happens if you
re-apply a less steep 44.1 kHz filter than the one used in the paper:
(http://xserv.compress.to/xnor/audio/images/lpfilter7.png)
red = 21.591 kHz passband edge, 22.05 kHz stopband
blue = red filtered with a less steep filter that has a cutoff point of roughly 21 kHz, same stopband
Visualization is done using a high oversampling factor and:
20*log10(abs(filter))
For all I care, those who produce the music may use as steep filters as they like, because I'm resampling during playback anyway.
Again, this is without imaging trickery that Meridian actually makes use of in their products... (which distances their claim of "real-world filters" even more from reality).
Since you won't listen to my pleading to put aside your emotions and act objectively, please allow me to deal with this differently since this is what is dominating the discussion now:
Does the larger membership and forum moderators think this is what the discussion should be? That we ask for ABX DBT results as the forum TOS #8 requires but the moment it is present, we call the person a cheater? Is this how you like to be known?
I am only an infrequent contributor to this forum, and lack familiarity with personalities behind a number of the member names, but I have found the level of vitriol in this thread unusually high.
My reaction to an unexpectedly good ABX result would be firstly to query the equipment and the experimental technique. And then if reasonably feasible, attempt to replicate the test result with help from those who claim to have heard a difference (see below).
So that others, if they so choose, can replicate your work, Amir, would you please provide the approximate time for what you heard as well as describe the phenomena?
Yes, if the subjective difference in sound is at its most obvious in only selected segments of the files under comparison, it can be very useful to identify where one or more of those segments are. It can help others do their "fishing" at a promising spot! (Sometimes, though, a difference is pretty much continual.)
In any case, it can be very useful to state the subjective character of the difference(s). This can helps others in
their attempts at hearing. Rarely, a difference could be so minor and fleeting as to defy characterisation, but even describing it in that vague way could conceivably be of some help.
MLXXX, let me save you some time:
After several pages of painful back and forth:
But you still haven't even given a hint at what specifically you hear.
And I am not going to.
9. Some Conclusions
Using admittedly exaggerated filter designs
Just stop there amirm. You don't see a problem here?
Also, the cited paper is largely outdated. Over 30 years ago the filters in A/D/A converters looked very different. You know that but you cherry-pick (another fallacy btw).
Oh the "it is too old argument." Arny does that once in a while.
It must be that human evolution in 30 years that make the results invalid. We must hear differently now than then.
Or maybe it is because someone discovered digital filtering in the ensuing 30 years. Digital audio must be a lot younger than we think.
Fortunately for us neither is the case. The paper is referenced in Stuart's. You have not read it. It is not a review of A/D and D/A converters as to be too old. No.
The paper explains how as we went from analog to digital, we introduced distortions that did not exist in analog domain. And that while the general public may not care, professional listeners should.
This is the opening salvo:
1.3
Simple experiments and perceptible effectsA good starting point is to base on experiments which, although
seemingly safe and unobjectionable, still lead to clearly perceptible
effects. They make it possible to identify some of the
reasons for the lack of fidelity in conversion, and to derive
design criteria from them. Fortunately, two such experiments are
known, one involving a seemingly ideal digital filter of absolutely
flat characteristics and linear phase, the other involving
very steep digital filters (again with linear phase). The next
chapter will describe the effects which were encountered with an
"ideal" filter, the explanations which were found for them, and
techniques for eliminating them. The second set of experiments,
which is based on the difference between "gentle" and "brickwall"
digital filters (all of them with linear phase) will be described
at the end of the paper.
This is the reference given right when the FIR filter discussion occurs in Stuart's paper. And you call that cherry picking? We are discussing Stuart paper. The references given there are 100% proper to cite. I didn't go rummage through some archive and pick some random blog to cite. This is work presented by luminaries of the industry and is completely relevant to the current discussion.
I also provided much more recent paper from JJ. Here is that again from his 2006 presentation:
. 30 years later the problem is still acknowledged. And with it being mentioned in Stuart's paper, it still won the peer review award.
Is it true? I don't know. We know it is something to think about. That is all that these references were meant to back. You don't want to think about it, don't.
I was really more interested in your trying 2Bdecided's linear phase file _50.wav (as I mention below). I recall that I was unable to discern any audible difference with that file, in January 2009.
Oh sorry. I was wondering why at the same time you were saying I took the test and I didn't .
I did very quick try and these are the results for linear phase:
=======================
foo_abx 2.0 beta 4 report
foobar2000 v1.3.5
2014-11-23 13:38:11
File A: limehouse_linear_phase_050.wav
SHA1: 661058f46dfb7de9fd2687344ece857f0ae1531a
File B: limehouse_reference.wav
SHA1: e8ad96830d23cad4bba5bf822ce875ae452b9e7c
Output:
DS : Primary Sound Driver
13:38:11 : Test started.
13:38:31 : 01/01
13:38:39 : 02/02
13:38:48 : 02/03
13:38:58 : 03/04
13:39:06 : 04/05
13:39:18 : 05/06
13:39:24 : 06/07
13:39:31 : 07/08
13:39:40 : 07/09
13:39:48 : 08/10
13:39:56 : 09/11
13:40:03 : 10/12
13:40:09 : 11/13
13:40:17 : 12/14
13:40:28 : 13/15
13:40:36 : 14/16
13:40:36 : Test finished.
----------
Total: 14/16
Probability that you were guessing: 0.2%
-- signature --
b28120a9d23f5f5e1706c1bdd66b655564163639
I distinctly remember losing concentration the couple of times above. But since the results show 99.8% non-chance outcome, I figured it is good enough to post and not suffer any more .
I didn't provide an extended response on AVS forum at the time when the criticism was first made, but after having had time to analyse the files I am sorry but I disagree. A null test with the files I initially provided is in fact quite successful at a moderate listening volume.
My impression of the noise is far higher than you and mzil. Maybe we are subjectively different when it comes to perception of it. At the start of the track it sounded like a white noise generator next to your ear while your stereo played .
On reflection, I don't think the conversion was "broken", merely that the dither that the old version of Audacity had created was somewhat noisy.
That was my point. Unskilled people go and look up signal processing algorithms and implement them. Without critical listening skills, they think the job is done and ship it. This yet again demonstrates that there are two types of listeners: those that can hear non-linear artifacts and those who can't.
Speaking of that, I downloaded your latest tracks. At 50.9 to 51.1 I thought the "s" in street sounded distinctly different. There was more lisping in one than the other. In the trial mode I was able to consistently tell the difference for a good sequence of trials. But when I ran with the test without feedback, I think I got down to 30% probably of chance or some such thing. I am just not motivated to try again and don't remember where the critical segment was in my original testing. So if you like to declare this a loss for me, you can . I am just too lazy to try harder and see if I can pass it.
Despite the lack of corroboration of the AVS forum member's claims by way of ABX software, I took this subjective comment as a vindication of my efforts. Certainly for my ears the AVS files had a difference in "tone" in addition to the misalignment of approximately 10mS. I think these factors were enough to invalidate the AVS Forum exercise, i.e. it was too easy to tell the files apart (as I had done with a formal ABX).
Yet, hardly anyone has managed to pass that test, with ease or not!
Ultimately what I like people to take to the bank is just that: that we don't hear the same. The data is 100% compelling in that regard. And very easy to verify. Anyone who disagrees should try to duplicate our results. They can try to find the flaws just the same. But I suspect with full knowledge of all of that, they don't have critical enough listening skills to find and hear the difference in DBT ABX tool.
And given that proof, you cannot extrapolate from your hearing to the rest of the people. Just because you don't hear the difference doesn't mean it is inaudible and sufficient justification for you to go around and accuse people are being wrong to observe otherwise.
I am unable to explain how the tonal difference arose. Why AVS Forum could not simply have provided time-aligned files capable of being audibly nulled at a moderate listening gain (by inverting one of the files) is a little surprising. I guess it was due in part to lack of familiarity with this type of exercise, and the importance of matching the files. Not only were the files not time aligned, but even when time aligned and with one file inverted, they would not null to silence at a moderate gain setting.
Actually they duplicated a real-life situation. Took a professional audio workstation tool, Sonic Solutions, and converted the files to 16/44.1. Precisely how real music is produced. That in double blind tests we could tell the difference it means that what people observe in the wild can very much be true. That transparency is not there.
It matters not that the timing difference may be the reason. What matters in this context is that we are wrong. We are wrong to say they are imagining things when they compare the high res to 16/44.1. We are making idealized assumptions that are not true in reality.
So I think it is good that we discovered how professional resamplers are not the animals we think they are. There is no visibility into their design as I have mentioned. I imagine hardly any music is produced using Sox resampler. Engineers use the professional tools. And if those tools produce non-transparent results, then we should get the masters and not be subject to this.
These are 100% logical and defensible conclusions we can draw. Anyone who doesn't want to go there has fingers in the ears and in denial.
Oh the "it is too old argument." Arny does that once in a while.
It must be that human evolution in 30 years that make the results invalid. We must hear differently now than then.
Or maybe it is because someone discovered digital filtering in the ensuing 30 years. Digital audio must be a lot younger than we think.
Amir, the master of straw man arguments and red herrings. See 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
I've asked you this:But please show us where it demonstrates the audibility of a linear-phase lowpass with a transition band of roughly 2 kHz (that's what we find in modern devices), with a cutoff frequency above ~21 kHz.
This would be a worst case scenario that still matches reality, while the paper uses a transition band of about 450 Hz. No, I'm not saying that is audible normally either, save your straw man arguments (at the risk of bursting).
But even if the files were filtered with ridiculously small bandwidth, we could still filter with a more sane filter and remove the characteristics of the bad filter.
But, yes, you absolutely do cherry-pick whatever fits your agenda, see 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
You know that you can make allpass filters, cables, ... or almost any other device or component exhibit
audible differences, you just need to ridiculously overdo it. And no honest person would answer a question what that audible difference is with: "I'm not going to tell you", see 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
But that is what your argument is now based on.
This is the reference given right when the FIR filter discussion occurs in Stuart's paper. And you call that cherry picking?
Oh amir, playing deliberately obtuse again, are we?
"admittedly exaggerated filter designs"Can you not even accept the words you quote? (Granted, they were not highlighted by that stupid annoying red color you seem to love to use to show us what you cherry-pick.) But then again, you don't even seem to accept that all the filter does is ring at over 21 kHz...
My reaction to an unexpectedly good ABX result would be firstly to query the equipment and the experimental technique.
My reaction would be, let's see you pass a supervised, administered listening test, not one sitting unsupervised with a Windows computer. A guy named JJ lives in WA...
30 years of zero audible evidence, now suddenly numerous "passes" of online ABX games on Windows computers, unsupervised in 2014....with a bit of history (http://www.avsforum.com/forum/86-ultra-hi-end-ht-gear-20-000/1136745-establishing-differences-10-volume-method-14.html) in between.
And now back to our scheduled programming: "Audibility? of Atypical Digital Filters in an unconfirmed transparency Hi-Fi Playback System, blaring near maximum volume in an iso-ward,
without using low specificity ABX".
cheers,
AJ
Amir, the master of straw man arguments and red herrings.
Straman this, red herring that. I have never, ever heard anyone utter those phrases in real life or technical conferences. You would be laughed at for using those high-school debating terms.
Through our bullying tactics, we have made up these rules of discussion. Oh that piece of data hurts? Let me just claim it is "straw man." If that doesn't work, then it must be red herring. If that doesn't work it is appeal to authority. If that doesn't work it is excluded middle or whatever that nonsense is called.
Leave those things to someone who won't call you on it as a substitute for proper technical answer. You clearly demonstrate your lack of knowledge when you use these phrases.
And no honest person would answer a question what that audible difference is with: "I'm not going to tell you"
Do people have a different definition of honesty in Australia? I have been to your beautiful country and don't recall people speaking differently.
"I am not going to tell you" is just that: a choice. Ton of data has been shared. Research report being discussed. Countless tests by me and others across wide range of test conditions. You have not accepted a single one. Mr. Xnor is righ, and the rest of the world include the experts who wrote the paper in question and others who reviewed it, wrong. Right. More demands for data is made. As if giving them more would make them more civil or interested in knowledge.
I ask you to repeat our listening tests. You know, double blind ABX tests. You refuse. Your refuse! If I were you I would go and change my alias and come back as a different person after this debacle. It is that embarrassing.
You guys talk big about blind tests. Demand it left and right. Put it in the forum TOS. But heaven forbid for you to be in the hot seat to run them. Oh no. The other party is accused of being a cheater, and demand neverending additional data. Anything but running the test yourself.
With straight face you walk around proud of this. Oh I am going to put it in my signature that I won't run any blind test to show whether I can or cannot tell the difference that others have reported beside Amir. Go ahead, I love to see my name in all of your posts. Someone will do a search and land on this thread .
You talk big about science. But you have no idea what the AES organization is, or the provenance of papers published within. On page 1 it is declared to not be peer reviewed. I correct that. It is repeated again by Arny. I correct it one more time. And have to do it a third time at the end of this thread. Is it not clear that the vocal few here don't really read or understand audio literature?
Buying an AES paper? Nah, not going to happen. That costs a few dollars. Stomping your feet in posts is free. Never mind that you learn nothing about audio by doing that. You are going to do it. If it doesn't work, there is always "straw man" this, and "red herring that."
Let me cut through this:
The conclusions are:
1. We don't all perceive non-linear distortions the same. This is proven beyond any doubt whatsoever. Anyone who wants to challenge this should show up passing the same tests. Therefor you cannot extrapolate your hearing as your belief as to what is or is not audible.
2. Our demand of a double blind ABX test is an empty bluf. Should the person pass it under all conditions we define, we can still call them a cheater and do away with the results. We refused to run them ourselves.
3. We now have award winning peer-reviewed paper showing that to better than 95% statistical confidence, filters have audible effects and so does dither/truncation. It is not 100% proof but it is heck of a lot better than anonymous members stomping their feet and singing "liar liar, pants on fire."
There. Live with these .
My reaction to an unexpectedly good ABX result would be firstly to query the equipment and the experimental technique.
My reaction would be, let's see you pass a supervised, administered listening test, not one sitting unsupervised with a Windows computer. A guy named JJ lives in WA...
Don't think MLXXX lives here but whatever.
This is the forum TOS:
8. All members that put forth a statement concerning subjective sound quality, must -- to the best of their ability -- provide objective support for their claims. Acceptable means of support are double blind listening tests (ABX or ABC/HR) demonstrating that the member can discern a difference perceptually, together with a test sample to allow others to reproduce their findings. Graphs, non-blind listening tests, waveform difference comparisons, and so on, are not acceptable means of providing support.
The only test of validity is providing samples for others to reproduce it. That is it. You guys are choosing to not try to do that. There is no other recourse in the forum TOS. Making up new rules about live witnesses and such are as silly as you using $230/foot speaker cable at high end audio shows Ammar.
30 years of zero audible evidence, now suddenly numerous "passes" of online ABX games on Windows computers, unsupervised in 2014....
The answer is simple: no one had presented a test to run until now. Scott/Mark did so on AVS as did Arny. Arny said no one had ever passed his test and made it into a challenge to do so. He even went as far as saying people could not tell the difference down to 32 Khz. So I gave it a try. And here we are.
Let me share you an interesting tidbit on AVS. Frank was one of my detractors here. He reported this on Arny's test, trying to demonstrate it could not be passed:
============
First ABX results with Jingling keys:
foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/09 14:17:03
File A: \\diskstationone\music\jitter\test\keys jangling band resolution limited 3216 2496.wav
File B: \\diskstationone\music\jitter\test\keys jangling full band 2496.wav
14:17:03 : Test started.
17:07:14 : 00/01 100.0%
17:07:41 : 00/02 100.0%
17:08:23 : 00/03 100.0%
17:09:27 : 00/04 100.0%
17:09:45 : Test finished.
----------
Total: 0/4 (100.0%)
My upper hearing range is about 14kHz. I doubt further training will do much good.
====
So total failure. As it happens, I too was running the test in parallel and reported my results:
32 Khz versus 96 Khz
=================================
foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/09 06:10:07
File A: C:\Users\Amir\Music\Arnys Filter Test\keys jangling band resolution limited 3216 2496.wav
File B: C:\Users\Amir\Music\Arnys Filter Test\keys jangling full band 2496.wav
06:10:07 : Test started.
06:10:38 : 01/01 50.0%
06:10:50 : 02/02 25.0%
06:11:07 : 03/03 12.5%
06:11:23 : 04/04 6.3%
06:11:36 : 05/05 3.1%
06:12:00 : 06/06 1.6%
06:12:14 : 07/07 0.8%
06:12:26 : 08/08 0.4%
06:12:38 : 09/09 0.2%
06:12:49 : 10/10 0.1%
06:13:00 : 11/11 0.0%
06:13:23 : 12/12 0.0%
06:13:42 : 13/13 0.0%
06:13:48 : Test finished.
----------
Total: 13/13 (0.0%)
A remarkable thing happened. Once he knew this was a test that could be passed, he tries again:
======
A and B along with X and Y I listen from slightly louder 'sparkle' in the hirez file. (Difficult to dicribe the sound)
foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/09 20:49:23
File A: \\diskstationone\music\jitter\test\keys jangling band resolution limited 3216 2496.wav
File B: \\diskstationone\music\jitter\test\keys jangling full band 2496.wav
20:49:23 : Test started.
20:51:10 : 01/01 50.0%
20:52:25 : 02/02 25.0%
20:52:46 : 03/03 12.5%
20:53:12 : 04/04 6.3%
20:53:58 : 05/05 3.1%
20:54:20 : 06/06 1.6%
20:54:47 : 07/07 0.8%
20:55:25 : 08/08 0.4%
20:55:49 : 09/09 0.2%
20:56:16 : 10/10 0.1%
20:56:44 : 11/11 0.0%
20:57:15 : 12/12 0.0%
20:57:23 : Test finished.
----------
Total: 12/12 (0.0%)
======
Completely positive identification! Here is a person who says he can't hear past 14 Khz yet he could tell a resample that limit the response to 16 Khz (32 Khz sampling). Xnor, there is your description of the difference. Happy now? I suspect not .
Straman this, red herring that. I have never, ever heard anyone utter those phrases in real life
Of course not, those are taught in Logic classes. Still not too late for you too take one.
Research report being discussed. Countless tests by me and others across wide range of test conditions.
Right. You have utterly subverted the discussion about the BS paper/results by conflating it with cheatable/flawed unsupervised online ABX games. Just as quoted above.
Your ABX results are absolutely unrelated to the BS paper/results.
You should start a separate thread on your new methods for passing unsupervised ABX tests on Windows computers. Make that the title if you wish, maybe add: "A follow up to the 2009 +/- 10% volume method for identifying DACs (http://www.avsforum.com/forum/86-ultra-hi-end-ht-gear-20-000/1136745-establishing-differences-10-volume-method.html#post16216826)"
It ought to be a doozy.
In the meanwhile, how about addressing the issues raised here
about the BS paper and stop polluting the thread with your ABX conflations.
cheers,
AJ
Making up new rules about live witnesses and such are as silly as you using $230/foot speaker cable at high end audio shows Ammar.
Supervision is to ease my type "worries", the cables, yours.
Take that Logic class Amir.
The answer is simple: no one had presented a test to run until now.
Baloney. You claim to have been running your own filter artifacts test regiment (http://www.avsforum.com/forum/86-ultra-hi-end-ht-gear-20-000/1136745-establishing-differences-10-volume-method.html#post16216826) since 2001
Originally Posted by amirm
Thanks for creating the thread. I must say though, I feel like I am on trial or something . We are all here for a fun exchange. If what I present doesn't meet some high bar, so be it....
The comparison I performed was using a Mark Levinson No360S against the on-board DACs in five to six DVD-A and SACD players, all playing the same time sync'ed CD. In other words, I would listen to the analog output of the player while its digital output would feed the ML DAC. All front panel lights were turned off in addition to video circuits (yes, all of that made a difference in fidelity).
The two sources were fed to the dual inputs of a Stax "earspeaker" electrostatic headphone amp. If you are not familiar with Stax, you can read learn more about them here: http://www.stax.co.jp/Export/ExportProducts.html (http://www.stax.co.jp/Export/ExportProducts.html). I have three of their units and results are consistent across the board although the highest end unit does make the job a bit easier. Using headphones allowed me to completely eliminate the room and take advantage of the amazing transparency of these headphones to listen for the slightest differences. To latter point, I would often listen to material at levels well above what I would use for listening to music, allowing me to hear detail that would otherwise be lost.
I then picked material that made it easier to detect differences between DACs. I am not going to disclose what constitutes such content. Without such material, the job can range from difficult to impossible. One has to know what could be damaged by a DAC and then use music that has such content. To give you an example, when you compress music, it is the transients that suffer. So something like guitar music is much more revealing than say, violin as the latter is much more harmonic than the sharp impulses of a guitar. Voices play the same role. None of these are useful for testing DACs though so don’t use that as a hint to the question posed . You can’t test the cornering of a car if you just drive it straight….
The comparison was then conducted without knowing which input is which, sitting in front of the headphone amp and toggling back and forth. When necessary, I would go back and re-listen. Once I found which one sounded worse, I would then repeat the exercise by randomizing the inputs and seeing if I could still identify which one was worse. My success rate was 100% in the second test (i.e. could always verify that the first result was not by chance). This testing was repeated a number of times comparing the different sources against each other and the ML.
I did not level match anything. However, once I found one source was worse than the other, I would then turn up the volume to counter any effect there. Indeed, doing so would close the gap some but it never changed the outcome. Note that the elevated level clearly made that source sound louder than the other. So the advantage was put on the losing side.
The results above were later objectively shown to be backed by some science in Stereophile magazine. In reviews of said players and Mark Levinson, it was shown that the former would only resolve to 14 or 15 bits of audio samples. Turning off the front panel pushed some up to 16 bits or so. The ML on the other hand, was tested to have equiv. of 19.5 bits. This is contrast to all the DACs being rated at "24 bits."
Now this testing is a few years old (probably circa 2001 to 2002). Maybe DACs have improved so much that the $20 part in the player is just as good as my then $8000 Mark Levinson DAC (which was hand tuned). If so, then I like to know who has tested the new ones and details of their methodology.
There you have it. Was it worth the wait?
Originally Posted by Chu Gai
1. Am I to understand that you placed a CD into either the SACD or DVD-A and then ran an analog connection to the headphone amp followed by a digital connection from either the SACD or DVD-A to the Mark Levinson No360S followed by another analog connection to the headphone amp?
Correct. That allowed me to have near zero delay in comparisons, something that is sadly missing from HDMI world today, making any DBL test of this kind invalid in my opinion.
2. You indicate that your test probes were music, yes? Does that mean you ran original CD's or were those mp3's? If the latter what details can you provide on the means of compression.
What kind of question is that for heaven's sake. Of course these were music titles. And no, they were not MP3s. I gave the example of music compression as to have people understand that you have to use the right material for the problem you want to investigate. As with audio compression, there is material that is more revealing of DAC issues than others. And the reason for that, just like audio compression, is firmly planted in science and objective evaluation.
3. Was the Mark Levinson No360S modified and if so were you able to test a unit that was unmodified?
Modified was the wrong term to use here. No360 came in two flavors. One that came from the assembly line and the other version (the "S" I think) that was hand calibrated at ML for better performance. I seem to recall this option cost me a few thousand dollars more.
4. If it was only you doing the connections and then placing a blanket or whatever over everything, how was it that it was blind?
Pretty simple. I took advantage of old age and what it does to memory . Seriously, I would grab the RCA cables and plug them into the back without paying attention to which input was which. After the test, I would then trace the cable to the source.
In the other thread, I made a point about being honest with yourself in these tests. And that is what is going on here. All of the equipment was free to me so I had no interest to defend one or the other. I would have been just as happy to see these $1000+ sources outperform the ML so that I didn't have to use that box for my testing all the time to rule out equipment differences.
Look at it this way. If I didn't care about a fair outcome, I wouldn't even bother to go through the lengths I did to test the equipment. Raise your hand if you have two copies of multiple identical CDs, DVD-As, and SACDs. Seems like my hand is the only one up! I even have the same title in SACD and DVD-A (from Chesky).
Now let me ask you this. Have you participated in double blind tests? If so, what tests were they?
This is an old topic for you Amir.
cheers,
AJ
Of course not, those are taught in Logic classes.
Your English is failing you Ammar. I said I have never seen anyone use these phrases at a technical conference. Not where they teach it.
You argue like these guys running for the office:
https://www.youtube.com/watch?feature=playe...p;v=kPwW8nBVc0g (https://www.youtube.com/watch?feature=player_embedded&v=kPwW8nBVc0g)
Funny as heck .
Straman this, red herring that. I have never, ever heard anyone utter those phrases in real life or technical conferences. You would be laughed at for using those high-school debating terms.
LOL you are serious, aren't you?
You don't even seem to know what these terms mean. And yeah, usually these words do not come up often in technical discussions because
smart people actually know what these terms mean and try to avoid such fallacies.
You on the other hand are like a fountain spewing them endlessly.
Through our bullying tactics, we have made up these rules of discussion. Oh that piece of data hurts? Let me just claim it is "straw man." If that doesn't work, then it must be red herring. If that doesn't work it is appeal to authority. If that doesn't work it is excluded middle or whatever that nonsense is called.
That is right,
these terms are nonsense to you because you don't even understand them. And no, you cannot call out fallacies whenever you feel like it, that just further demonstrates your
gross ignorance.
There is a basic level of understanding of logic, statistics, science ... that you should have when you participate in a technical discussion. But most importantly is intellectual honesty, all of which you demonstrated to lack over and over again (see 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029)).
Leave those things to someone who won't call you on it as a substitute for proper technical answer. You clearly demonstrate your lack of knowledge when you use these phrases.
The irony is breathtaking. It truly is.
Also, how about a technical answer to this:
>>>"
But please show us where it demonstrates the audibility of a linear-phase lowpass with a transition band of roughly 2 kHz (that's what we find in modern devices), with a cutoff frequency above ~21 kHz."
>>>"And no honest person would answer a question what that audible difference is with: "I'm not going to tell you""
Do people have a different definition of honesty in Australia? I have been to your beautiful country and don't recall people speaking differently.
Can everyone see that? That reading comprehension problem? He again says I'm from Australia.
I've got old news (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029) (<- 0 credibility post) for you:
I'm not from Australia. But hey, I've only referenced that post a trillion times.
It looks like I have to explain every word to you (quote from Wikipedia):
Intellectual honesty is an applied method of problem solving, characterized by an unbiased, honest attitude, which can be demonstrated in a number of different ways, including but not limited to:
One's personal beliefs do not interfere with the pursuit of truth;
Relevant facts and information are not purposefully omitted even when such things may contradict one's hypothesis;
Facts are presented in an unbiased manner, and not twisted to give misleading impressions or to support one view over another;
References, or earlier work, are acknowledged where possible, and plagiarism is avoided.
Intentionally committed fallacies in debates and reasoning are sometimes called intellectual dishonesty.
"I am not going to tell you" is just that: a choice. Ton of data has been shared. Research report being discussed. Countless tests by me and others across wide range of test conditions. You have not accepted a single one. Mr. Xnor is righ, and the rest of the world include the experts who wrote the paper in question and others who reviewed it, wrong. Right. More demands for data is made. As if giving them more would make them more civil or interested in knowledge.
If you keep on doing this (and with this I mean all of this) then I will have to put you on ignore. You will be the first there, heck, you are a first for a lot of things on this forum, probably on the Internet.
I've explained to you why I'm skeptical towards your logs. I wanted to check if you truly can hear differences. You came up with excuses and made any possibility to remotely verify your abilities impossible. That's a prime example of intellectual dishonesty.
I ask you to repeat our listening tests. You know, double blind ABX tests. You refuse. Your refuse! If I were you I would go and change my alias and come back as a different person after this debacle. It is that embarrassing.
That you have extremely poor reading comprehension and reasoning abilities is embarrassing. I
already explained this to you. Scroll back.
I am not going to play your backwards game where you either think you win or you think you win.
With straight face you walk around proud of this. Oh I am going to put it in my signature that I won't run any blind test to show whether I can or cannot tell the difference that others have reported beside Amir. Go ahead, I love to see my name in all of your posts. Someone will do a search and land on this thread .
Here's what you said:
>>>"
But you refuse to [..] back your claims of inaudibility."
I
neither claimed that filters are inaudible,
nor does it logically make any sense to provide evidence for inaudibility. This kind of nonsense is what I'd expect to hear from a grade schooler, but not from a grown man.
This is the bottom of the basics. But you don't even seem to get those.
Yes, for the love of god,
I hope that many people will find this thread. They won't believe this is real, until they hit page ~17+.
You talk big about science. But you have no idea what the AES organization is, or the provenance of papers published within. On page 1 it is declared to not be peer reviewed. I correct that. It is repeated again by Arny. I correct it one more time. And have to do it a third time at the end of this thread. Is it not clear that the vocal few here don't really read or understand audio literature?
Are
you seriously telling me what I know and what I don't?
Where did I say the paper is not peer-reviewed? Do you even know what that means?
Let me cut through this:
The conclusions are:
1. We don't all perceive non-linear distortions the same. This is proven beyond any doubt whatsoever. Anyone who wants to challenge this should show up passing the same tests. Therefor you cannot extrapolate your hearing as your belief as to what is or is not audible.
2. Our demand of a double blind ABX test is an empty bluf. Should the person pass it under all conditions we define, we can still call them a cheater and do away with the results. We refused to run them ourselves.
3. We now have award winning peer-reviewed paper showing that to better than 95% statistical confidence, filters have audible effects and so does dither/truncation. It is not 100% proof but it is heck of a lot better than anonymous members stomping their feet and singing "liar liar, pants on fire."
There. Live with these .
1. Where on earth do you see non-linear distortions? You don't even know what an LTI system is, but talk big about filters? What the..
2. No, I've already explained this to you. A log is only a starting point, not conclusive proof. Cheating, false positives are always a possibility. An intellectually honest person provides more than just a log, he/she will point out what was heard, where, with which system and setup readily. He/she will also will do everything to eliminate the chance of a false positive. (You on the other hand posted pages of noise to finally evade with an excuse. 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).)
Then the community will start trying to reproduce what was heard .... and so on and on. Already explained all this.
This is
basic science stuff that is taught in middle school.
3. I also already told you that I take the paper for what it is, and there are still unknowns. Your stereotyping of me, which you even admitted, prevents you from actually understanding my position. You don't even care about any of this. All this is for you is a "war" (your own word), that you apparently need to win at all cost, which sadly seems to include your sanity.
If you keep on doing this (and with this I mean all of this) then I will have to put you on ignore.
Why do you people say this? Isn't that an oxymoron? You want to ignore someone, ignore them. Why declare that you are ignoring them?
And why is it always said as a threat? I don't write my responses for you. This is a public forum not a private discussion. I have a point of view and you guys facilitate getting that out by the writing your posts. So unless you stop posting, then whatever else you are doing is of no consequence to me.
So please, by all means, don't wait for anything else. Put me on your ignore list. That should take the noise level of the thread down given how emotional your posts have been.
Why do you people say this? Isn't that an oxymoron? You want to ignore someone, ignore them. Why declare that you are ignoring them?
And why is it always said as a threat? I don't write my responses for you. This is a public forum not a private discussion. I have a point of view and you guys facilitate getting that out by the writing your posts. So unless you stop posting, then whatever else you are doing is of no consequence to me.
So please, by all means, don't wait for anything else. Put me on your ignore list. That should take the noise level of the thread down given how emotional your posts have been.
I say this because you've reached a point that is far beyond my usual trolling threshold, but you continue. And by the post of the day it seems to get worse, which I never even deemed possible from any non-troll.
No, it is not an oxymoron. In the next paragraph you say yourself it is a public forum. Now think... think hard...
No, I haven't said that you're on my ignore list now. Learn to read.
I know that whatever I write has no consequences in your brain. After several pages of literally painful responses that much is clear to me, and should be to anyone really.
So you actually want me to put you on ignore? Jeez... it just got worse.
So you actually want me to put you on ignore? Jeez... it just got worse.
Seems that I can't win. You say you will put me on ignore list and I say do it now there is another protest on that. So don't! I don't know what to tell you.
P.S. My deepest apologies to you for mistaking you from Australia rather than Austria. I have temporary bi-focal glasses on a very high DPI display and just didn't notice the difference. My favorite engineering blogger Dave Jones is from Australia and routinely chastises people for confusing the two countries and here I am walking right into that. Again I am sorry. It was not intentional.
Speaking of Dave Jones and to bring some levity, here he is on "Audiophile Audiophoolery"
https://www.youtube.com/watch?v=m7ERMu825m4 (https://www.youtube.com/watch?v=m7ERMu825m4)
I have never, ever heard anyone utter those phrases in real life or technical conferences.
Of course not, those are taught in Logic classes.
Your English is failing you Ammar. I said I have never seen anyone use these phrases at a technical conference.
Honesty, basic reading comprehension and logic are not your bag Amir.
Can we get back to the topic of the BS paper, or is that one of your "worries" too now?
Perhaps contact BS to ask why so much obfuscation where even Amir was confused by the level of passage playback, speaker distance, etc? Or why no data for any system, speaker, switching software, etc are shown?
Any chance the BS paper follow up might include the 2L Hi Rez track vs an actual, 16/44 TPDF downsampled customer version?
cheers,
AJ
You should start a separate thread on your new methods for passing unsupervised ABX tests on Windows computers.
Per our kind moderator, this is a combo thread for both: http://www.hydrogenaud.io/forums/index.php...st&p=880887 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=106442&view=findpost&p=880887)
Hey krab, since no one can seem to keep the two topics separate, what do you think of my merging your two discussions into one enormous train wreck?
In the meantime, this discussion will now close.
Anyone who wishes to discuss the topic defined by the original post, send me a PM and I will re-open it. I am OK if you want to petition him to re-open that thread and we continue the discussions there. For now, if you don't mind I leave you be the one arguing with his decision .
Besides the satisfaction of "having the studio quality", what would be the point of buying higher formats ?
If it is higher than how the music was recorded/produced, then I have no use for it. If higher is what is recorded and mixed, then I like to get my hands no those bits as I don't trust anyone to do the right things in the chain to bring it down to something lower.
(snipped list of higher prices for higher resolution)
Assuming you really want to listen to a particular recording of some music:
If someone records at 1MHz and charges you $100 per track, will you pay to get your hands on this?
If someone records at 10MHz and charges you $1000 per track, will you pay to get your hands on this?
If someone records at 1GHz and charges you $1,000,000 per track, will you pay to get your hands on this?
I'm just wondering what your limit is.
Your entire argument is that you have no limit ("If higher is what is recorded and mixed, then I like to get my hands on those bits"). You keep implying folks here are unreasonable for setting a limit.
For this discussion to make sense, I need to know whether you really mean what you say, or whether you do have a limit, but it's different from ours.
Cheers,
David.
Can you answer the psychology of this?
You know the psychology. We see folks pushing barely audible / inaudible improvements to the detriment of easily audible improvements. It impacts negatively on the equipment and recordings that are available. Why wouldn't people who care about audio be annoyed about this?
Also, I think you're taking the flack for all the wild and unsubstantiated claims for hi-res that are thrown around in other places, even though the main argument you've pushed here is merely "why not"?
Cheers,
David.
Since you won't listen to my pleading to put aside your emotions and act objectively, please allow me to deal with this differently since this is what is dominating the discussion now:
Does the larger membership and forum moderators think this is what the discussion should be? That we ask for ABX DBT results as the forum TOS #8 requires but the moment it is present, we call the person a cheater? Is this how you like to be known?
I am only an infrequent contributor to this forum, and lack familiarity with personalities behind a number of the member names, but I have found the level of vitriol in this thread unusually high.
My reaction to an unexpectedly good ABX result would be firstly to query the equipment and the experimental technique. And then if reasonably feasible, attempt to replicate the test result with help from those who claim to have heard a difference (see below).
I agree. You'd normally start with an analysis of the laptop's output with these files and test files (with and without the headphones connected), and then an analysis of the performance of the headphones themselves. Not everyone has the equipment to do the first test, and few have the capabilities to do the second.
You'd normally leave the vitriol for rec.audio.high-end or where ever it is people like to converse in this way.
It wouldn't surprise me that much to find a laptop+headphones where hi-res sounded different from CD. I would want to analyse and understand what was happening before concluding that the best response to this discovery was to buy music in hi-res. That might be the right conclusion, but it might not be.
Cheers,
David.
You know the psychology. We see folks pushing barely audible / inaudible improvements to the detriment of easily audible improvements. It impacts negatively on the equipment and recordings that are available. Why wouldn't people who care about audio be annoyed about this?
I wholeheartedly agree. Anyone who listens to a stereo recording and/or a modern pop/rock/.. song or even just gets blown out of the seat by a commercial break, wouldn't even dream of putting a 21+ kHz lowpass filter on a list of things to improve the audio.
Also, I think you're taking the flack for all the wild and unsubstantiated claims for hi-res that are thrown around in other places,
You'd probably include me here, but I have to disagree. He's getting
flak because he's in a "war" that he has to win at all cost. Assuming that ones self-made enemies will blindly accept logs that are thrown around, not really question or object to anything that is said, glossing over fallacies ... is a huge mistake that backfired. It may work when you're a senior that can talk down on people, but not in a public forum, where you're not more equal than others.
even though the main argument you've pushed here is merely "why not"?
With qualifiers, and this really needs to be stressed, I am not even inclined to disagree.
There are still massive problems with that if I think of the prices of >44.1/16 tracks (some would say it's a scam), the misrepresentation of scientific findings and science in general to fit the agenda of people selling equipment (which they will tell you that you have to have as an audiophile) ...
Ah, so the premium cost of CD over MP3 is justified. But past that, it is a sin.
The rules we invent. It is like saying you can go from 3 to 4 bedroom house. But oh, more than that? It is a sin.
I use lossless for several reasons:
1. So I can encode my music into whatever lossy format I need in the future without introducing transcoding artefacts
2. So I can check the files on my HDD against AccurateRip at any point
3. So I can play my music through a surround sound system without revealing artefacts that were inaudible in stereo
These are real demonstrable benefits. It's not just about getting that warm fuzzy feeling from inaudible perfection.
I'm quite pragmatic though. Where some decent master is only available to me in lossy, whereas what's available as a lossless version is audibly inferior, I'll take the lossy version.
I think some other justifications for lossless do have a correlation with your argument, e.g.
"I want a perfect copy of what's on the CD even if I can't hear the benefit" / "I want a perfect copy of the master even if I can't hear the benefit"
"I want to do my own lossy coding because someone else might mess it up" / "I want to do my own downsampling because someone else might mess it up"
You would argue it's just a question of degree. I would suggest the evidence that lossy audio coding introduces audible problems is strong enough to make it a scientific fact. I would suggest the evidence that downsampling audio to CD quality introduces audible problems is nowhere near as strong. It may be right, or wrong.
Cheers,
David.
Besides the satisfaction of "having the studio quality", what would be the point of buying higher formats ?
If it is higher than how the music was recorded/produced, then I have no use for it. If higher is what is recorded and mixed, then I like to get my hands no those bits as I don't trust anyone to do the right things in the chain to bring it down to something lower.
(snipped list of higher prices for higher resolution)
Assuming you really want to listen to a particular recording of some music:
If someone records at 1MHz and charges you $100 per track, will you pay to get your hands on this?
If someone records at 10MHz and charges you $1000 per track, will you pay to get your hands on this?
If someone records at 1GHz and charges you $1,000,000 per track, will you pay to get your hands on this?
I'm just wondering what your limit is.
Your entire argument is that you have no limit ("If higher is what is recorded and mixed, then I like to get my hands on those bits"). You keep implying folks here are unreasonable for setting a limit.
I have no limit and yes, people here are illogical to have one. There is some confusion in what you are assuming about me though. So let me tell you another story .
I am working for Sony and my boss and I are practically living on airplanes going on 10 hour flights to and from Tokyo from SF Bay Area. This is early 1990s without all the technology we have to entertain ourselves (you should have seen me with my portable CD player and stack of CDs). On one of the flights, United gives us this survey to fill out about our interest in supersonic jets that would cut the time to 3 hours or some such thing. I start to fill it out and I get to your type of question: would pay 2X, 3X, 10X or something like that. Before I had a chance to think about it, my boss says, "check the 10X." I say but I don't think I will want to pay that much just to save 6 hours. He said the smartest thing: "let's motivate them to build it; once there we can decide if it is worth it or not. If they don't build it, we will never have a choice!" From now on I always check the most expensive box in such surveys .
So back to your question, I don't care what the retail price is. I want to encourage a stream of releases that are before the final mastering of the CD where both the 16/44.1 conversion and loudness compression occur. Yes, not all of them are free of the latter but more and more they are.
Answering your question then, my motivation ha nothing to do with the sampling rate. I don't care if the pre-CD master is 48 Khz. I take the better sounding one! Take Adele Live at Albert Hall. I think the Blu-ray version is only 48 Khz sampling but it is world's better than the CD. The latter has surely put through the grinder as it is not listenable compared to the Blu-ray version. This is what getting the upstream bits enables: a shot a better mastering. And a good shot at that.
When I bought the album in question, the L2 Nordic Sampler, I opted for 96 Khz. There was 192 at higher price still but it didn't matter to me. I was happy to see that supersonic version but chose one notch lower. In this case I knew there was no loudness compression applied to the 96 Khz so I opted for that. But you know what? I regret not getting the 192 Khz now just for the extra $4. I might go back and get that one just to compare even though it would set me back another $28 (really crappy of them to re-charge you the full amount for another sampling).
For this discussion to make sense, I need to know whether you really mean what you say, or whether you do have a limit, but it's different from ours.
Let me address the "ours" part. I am going to make an assumption that neither one of us has any interest in half a million dollar supercar. Personally even if I had the money, I wouldn't buy it. I have no use for it in any form or fashion. I am assuming there are many people like me. Should we start a campaign to outlaw development and sales of these cars? Yes I know, you are going to say it goes faster so there is real value. But put that aside for a moment as I just said, knowing that I have no interest in half a million dollar in a car. You are in the same boat with me. Should we go and get them to not be legal to sell?
That is what you are saying in your post. That how much something is priced for a different audience should be cause for us to get together here, and call each other names if someone disagrees.
I founded a company called Madrona Digital. Our "poor" customers have net worths many multiples of me. And the higher ones can afford to buy small countries . No, we don't sell them any high-end audio gear. That is a stupid business for anyone to try to get into (AJ will figure this out one day). But we will install automated lighting and shades that combined would cost $200,000! I don't wake up in the morning having any negative feelings about that or the fact that high-end lighting systems exist. In my own house the lighting system cost about $25,000. Someone with a $2 light switch just blew a gasket . I used to feel the same way but then I started to think of all the ways it would make our life more enjoyable and put it in. Just this week my wife and I decided that if someone rings the doorbell on the Intercom, it would be nice to have the front lights come on for 10 minutes but only if it is after dark. 20 minutes later that is how it worked.
Is the above of value to mass majority of people? No. Just like a 4 bedroom apartment.
Where I think the vocal few go wrong is that they pick battles when no one is asking them. If someone tries to convince me to buy a supercar, the will get strong, strong arguments. I want a large comfortable car because I haul a ton of stuff and drive far distances. It is fast enough for me and handles well enough. Everything about a supercar would be a step back for me at higher cost. So you are going to hear these arguments. But folks getting together in another forum trying to decide which supercar to buy? More power to them. They are not trying to convince me to buy so why business or logic do I have for shedding a tear.
This is what is wrong with the argument you are making. What business of ours is there to set a limit on price or specs? I don't buy LPs but people make them and folks buy them. Good for them all. It creates jobs and improves the economy . Ask me to buy an LP and I tell you that the convenience of digital is so important to me that I don't even want to hear it.
I am writing this long post because this is the ultimate problem and the cause of all of these written battles. People think they have some duty to fight these battles. They even fight the battle when no one is here to represent the other side! The only person we have who is promoting "snake oil" is AJ with his $2,300 cables. I have said no such thing.
If I ask them why they say they want to save some individual from wasting their money. Fine. But at what cost are you doing that? Xnor is determined to get there at the expense of looking totally unprofessional and going after someone who is interested in discussing the topic using double blind tests and science/research of audio. That would be like two republican leaders tearing each other up with the goal of signing up more democrats!
Here is the reality of it: the above is not the reason Krab, Any, AJ, mzil, xnor, etc. post what they post. Nope. We gather in these forums because it feels good to be known as an authority. Many people pick what I call "good enough" as their platform. It seems safe. We can immediately wear the cloth of "science" and say nothing is better than another, listening test says this and that, some textbook that we have read says something else and we are golden. We become the hallway monitor in school. Walk around the forum and stomp on any kid we can find.
Losing that power is a big deal. This is why xnor posts and posts. And before him Krab, Arny, AJ, mzil, etc. Mzil for pete's sake used to sell audio gear. The very same gear he is here to say should be illegal to buy! Give me a break.
I am not here to join a crowd for that reason. And certainly won't join the camp and be requested to paper over faults in our arguments. I want 100% transparency. Let me repeat: I want 100% transparency. If there is something wrong with our argument, we better offer it before the other camp does. We raved and raved about Meyer and Moran proving that there is no value to high resolution audio. All the while we heard the arguments that said, "wait a second, these guys didn't know which end was up; they didn't even test to see if their content was high resolution!" No, carried their report under our arm and used it as our bible to stomp on the other camp.
Well, that era is over. Far more careful test has been run which is the topic of this thread. And what do you know? It has a different outcome! What to do now? What? Oh, let's attack the messenger. Let's say that Amir is a cheater and see if we can create some smoke. Let's keep saying that an authority like James Johnston, my former audio architect and AES Fellow, would never agree with any of this. Well, this is another page from his presentation:
amir, would you be so kind to just take a quick (it's only 4 seconds) listen to imp_urhp.wav (http://www.hydrogenaud.io/forums/index.php?showtopic=107588) with your laptop at a similar volume level you made all those other ABX logs with and tell us if you hear something?
These are real demonstrable benefits. It's not just about getting that warm fuzzy feeling from inaudible perfection.
As is the case with getting the high resolution master prior to loudness compression and remastering for CD/MP3/AAC release. We need no double blind test, no signal processing lessons, nothing. I like to have the option to buy the original before a photocopy was made.
Since you won't listen to my pleading to put aside your emotions and act objectively, please allow me to deal with this differently since this is what is dominating the discussion now:
Since logical reasons for so-called high resolution audio seem to be lacking, strong advocates of it must be ruled by their emotions and not acting objectively.
And who is doing that?
These are real demonstrable benefits. It's not just about getting that warm fuzzy feeling from inaudible perfection.
As is the case with getting the high resolution master prior to loudness compression and remastering for CD/MP3/AAC release. We need no double blind test, no signal processing lessons, nothing.
The original is the tracks prior to mix down, and this is almost never for sale.
If you have the tracks prior to mix down, it takes serious work to obtain something that most people want to listen to.
I like to have the option to buy the original before a photocopy was made.
Of all the things that happen to the original tracks during production we know for sure that any downsampling to 44/16 is surely the most subtle and comparing it to photocopying is a complete and total distortion of the truth.
Most naive audiophiles have been sold the lie that their recordings would sound significantly more lifelike if they were in a high resolution format, and we now know that is for sure a lie out of the pit of hell.
Can you answer the psychology of this?
You know the psychology. We see folks pushing barely audible / inaudible improvements to the detriment of easily audible improvements. It impacts negatively on the equipment and recordings that are available. Why wouldn't people who care about audio be annoyed about this?
They can be annoyed all they want. But they have no business taking it out me.
I know all the arguments. And I know which ones are right, which ones we have made up to win a battle.
The moment anyone, ever, puts forward Meyer and Moran as a proper test, we need to boot them out of the objectivity camp. And certainly out of the camp of knowing anything about proper listening tests.
The moment anyone starts to make technical arguments when they have no professional or educational background in audio against someone who does, we need to boot them out of the objectivity camp. Lest we want to endorse that by reading forum posts, you wind up more knowledgeable than your doctor about medical field.
The moment anyone gets personal and makes comments like this in a technical thread:
Is amir deliberately obtuse? Is he trolling? Or is that just the way he is?
We need to boot them out of objectivity camp. Lest we want to look like a bunch of immature 16 year old games on a forum arguing about which is better: xbox of Playstation.
The moment anyone uses debating phrases like "appeal to authority" when a research paper is referenced and the author's qualifications are put forward, we need to boot them out of the objectivity camp. Lest we want to be seen as totally confused about how these discussions work in real life.
The moment someone is more interested in arguing than lifting a finger to run some tests, or buying a paper and reading it prior to arguing it is wrong, we need to boot them out of the objectivity camp. Lest we want to endorse that running with tag lines and repeating what others say is something constructive.
The moment someone questions the integrity of someone as a way to dismiss their hard work in producing data for the discussion, we need to boot them out of the objectivity camp. Lest we want this discussion to go right down the gutter and be ignorant of what FUD is.
This is what separates us. I won't do or endorse such things.
Also, I think you're taking the flack for all the wild and unsubstantiated claims for hi-res that are thrown around in other places, even though the main argument you've pushed here is merely "why not"?
It is even stronger than that. It says that anyone arguing against high-res is wasting their breath and forum bandwidth. This movement unlike the past ones is not at all governed by these arguments. This is not 2007. This is not a format battle that has cost to bring a new physical format to the market.
For me personally, yes, I like to have the option available to me. I foresea a future where CD format will decline. Before that happens, I like to see a solid market for digital downloads at > 256 kbps MP3/AAC. My entire career has been built on understanding both the business and technical aspects together. And I am telling you, one day we will wake up and all we can find is lossy compressed stuff with clear step down in quality to say nothing of loudness compression. I run into this situation multiple times a month now.
So join me in dialing down the rhetoric. No one or animal is being harmed here. Let's be truly objective and see the benefit here.
My time is nearly up here. I hope everyone thinks through these things and become at least as professional as you David. It would be a huge step forward even if we don't fully agree.
These are real demonstrable benefits. It's not just about getting that warm fuzzy feeling from inaudible perfection.
As is the case with getting the high resolution master prior to loudness compression and remastering for CD/MP3/AAC release. We need no double blind test, no signal processing lessons, nothing.
The original is the tracks prior to mix down, and this is almost never for sale.
No, the step before CD mastering is the stereo mix at the recorded/mixed bit rate and sampling rate which is almost always higher than the CD. They don't jump from the multitrack recording directly to an MP3.
I like to have the option to buy the original before a photocopy was made.
Of all the things that happen to the original tracks during production we know for sure that any downsampling to 44/16 is surely the most subtle and comparing it to photocopying is a complete and total distortion of the truth.
Why? A photocopy can be excellent or crappy. The copy may have saturated colors or be true to the original. The former would be like loudness compression in music.
Most naive audiophiles have been sold the lie that their recordings would sound significantly more lifelike if they were in a high resolution format, and we now know that is for sure a lie out of the pit of hell.
What is a lie is saying that about "most" audiophiles with no data to back it. Or imply that audiophiles are naive. Go to a forum like our WBF and you see the reality is not remotely like that. Here is a random example on "bogus" high-res downloads: http://www.whatsbestforum.com/showthread.p...i-res-downloads (http://www.whatsbestforum.com/showthread.php?15753-What-s-with-all-the-quot-bogus-quot-hi-res-downloads)
For a while it seemed that there was a fuss about upsampled 16/44.1 files sold as "hi-res", either in SACD's or 24/96 PCM (or even 24/192) and HDTracks responded by more carefully vetting the albums they sold. Now more recently the trend has reversed and there are a lot of "new" titles coming out. Acoustic Sounds has Stevie Wonder "Hotter Than July" and "In Square Circle", both known to be (and obviously) from 16/44.1 masters, MFSL has SACD's of Dire Straits "Brothers in Arms" and Los Lobos "Kiko" (likewise), HDTracks is selling The Allman Brothers "Complete Fillmore East Concerts" as a 24/192 download even though it was a 24/96 transfer from the analog masters (identified as such at prostudiomasters.com even though they also sell a 24/192 version, and there is even a low-pass filter used at about 29 kHz), the Jerry Garcia store sells a 24/176.4 album made from a 16/48 DAT master, and the list goes on.
These people are not stupid Arny. And certainly not the stereotype you paint them to be. You just need to go to the right forum where proper discussion happens instead of immediate war because someone dared to talk about high-res files. There, you see real knowledge at levels that you and I don't possess. Did you know the stuff he is sharing above down to where the master came from? I didn't.
You ask and get educated on the proper stereo masters and then you can make an economical choice as to buy or not buy. Can't do it here because DBT is demanded and if provided, you are accused of cheating if the outcome is positive. So step out of this cocoon and you see that the reality is not what you imagine, or more likely, say to make an argument.
Let's keep saying that an authority like James Johnston, my former audio architect and AES Fellow, would never agree with any of this.
Let's start: "how do you even imagine that one can hear a difference between two systems, one with noise 98 dB down and the other 146dB down, when the level is set to peak at 96dB?"
How do you (Amir) explain how "obviously difference" fails to show up in even the worst kind of ABX test? (unless it is the unsupervised, Windows pc, online cheatable variety, which are worthless and unrelated to the anti-ABX BS paper)
I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener. Have you any, bearing in mind that citing non-blind-testing proves nothing but the incompetence, the complete and total incompetence, of the person citing it as evidence.
Bear in mind the hard evidence for the persistance of loudness memory while you're at it.
08-01-2014, 10:07 PMjj_0001 avsforum (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1532092-debate-thread-scott-s-hi-res-audio-test-95.html#post26230585)
I have my doubts that SACD or DVDA are much, if any, of an improvement, but the test is just blisteringly hard to run, and more likely to respond to artifacts, either positively or negatively, than it is to actual differences. Time alignment, level alignment, frequency response in-band can all throw it positive, lack of training, bad test environment, bad time alignment, etc, can also cause false negatives. Subject verification, likewise, is an important issue.
So, I remain undecided, but I note that I own a lot of CD's and not a single SACD or DVDA, except for some people have sent me.
Quit posting Ammar and addressing me. Once I saw you going to shows and demonstrating your gear with snake oil stuff like modded players, I wrote you off as having any opinion worth discussing. All it took for you to change your colors was the opportunity to make a buck.
Anyone who really believed in what you say would have proudly gone to the show, use a PC, AVR and garden variety cable. But no, Mr. Ammar Jadusingh goes there and sells us all with $2,300 speaker cables, modded players and expensive amplification. No principals. No ethics. "Do as I say and not what I do."
I write articles like this on speaker wire: http://www.whatsbestforum.com/showthread.p...is-not-12-Gauge (http://www.whatsbestforum.com/showthread.php?14134-When-12-Gauge-Wire-is-not-12-Gauge)! You go to shows and put in "flat" speaker cables. I bet that added air to the music, removed the harsh digital edges, etc.
Can you answer the psychology of this?
You know the psychology. We see folks pushing barely audible / inaudible improvements to the detriment of easily audible improvements. It impacts negatively on the equipment and recordings that are available. Why wouldn't people who care about audio be annoyed about this?
They can be annoyed all they want. But they have no business taking it out me.
I think that it is well known what people take out on you Amir.
I know all the arguments.
You fooled me.
And I know which ones are right, which ones we have made up to win a battle.
You fooled me.
]The moment anyone, ever, puts forward Meyer and Moran as a proper test, we need to boot them out of the objectivity camp.
Nope. I'll take what that to mean that you want to boot them out of the Amir camp, and of course that is your right.
However, the Amir camp is not the objectivity camp.
I've explained to you what M&M is and what it is not. You have not learned a thing!
And certainly out of the camp of knowing anything about proper listening tests.
Given that how you've apparently swallowed the recent Meridian work, hook line and sinker...
Given the kind of crap that we've found you bragging about on AVS...
The moment anyone starts to make technical arguments when they have no professional or educational background in audio against someone who does, we need to boot them out of the objectivity camp.
If memory serves Brad Meyer studied Engineering at a well known Ivy League school, I think Harvard.
Given the monumental gross errors of yours that I corrected almost daily over at AVS, who are you Amir?
L
I write articles like this on speaker wire: http://www.whatsbestforum.com/showthread.p...is-not-12-Gauge (http://www.whatsbestforum.com/showthread.php?14134-When-12-Gauge-Wire-is-not-12-Gauge)!
I was a bit surprised at how easily you dismissed the possibility that the story of the coat hanger speaker wire was in fact correct. Taking your own data, two coat hangers would have a resistance of 38.8 milliohms, which into a 4 ohm load results in 0.08 dB of loss in the coat hangers. Even you would have to admit that this would not be audible.
I was also surprised that you do not seem to know the proper way to do a four wire resistance measurement, as indicated by your taking measures to clamp down hard on the ends of the wire to make good contact, as well as zeroing out probe resistance. In a proper four wire measurement none of this is necessary, but it would take a few seconds longer to attach four wires to the cable instead of two.
All in all I am not surprised that you found what seems to be a large variation in cable resistance, given how suspect you measurement technique was.
Quit posting Ammar and addressing me.
Nope. You could try putting me on ignore, but you know that doesn't work for me.
Generally speaking of course, ignoring arrogant, elitist, condescending, "Hi end" audiophile bullies doesn't work in real life....or cyberspace.
But no, Mr. Ammar Jadusingh goes there and sells us all with $2,300 speaker cables, modded players and expensive amplification.
Nope, still sell only speakers, ranging from $1800-$8500/pr. The accoutrements are not for "us all", just folks who "worry" about things like that. Know any?
Makes no sense to exhibit my speakers to you Amir, if you're going to be stressed an "worried". If it would ease your mind further, I could use some of the
$50,000 (2012 pricing) ML amps you peddle[/i] (http://www.madronadigital.com/Library/MarkLevinsonNo53Amplifier.html).
In comparison testing I have done, switching amplifiers using the classic class D configuration always sport incredible low frequency control and power. They beat out linear class AB amplifiers almost regardless of price. What they give up though is high frequency fidelity which I find somewhat harsh. The distortion is highly non-linear and challenging to spot but it is there. The Mark Levinson No 53 is the first switching amplifier I have heard which does not have this compromise. Its bass is amazingly authoritative: tight and powerful. Yet the rest of the response is absolutely neutral and pleasant.
If you have not heard these unique amplifiers, I highly encourage you to come into our showroom for a listen. We have a pair on hand driving our Revel speakers. I am confident that they will improve the sound of your current speakers given the ease with which they can drive any load regardless of how difficult they might be (and many high-end speakers are difficult to drive). We are happy to let you evaluate them with your own system to see the benefits of this technology. Hearing this amplifier was an eye-opener for me. I think it will be for you too.
My, my, that sounds mighty subjectivist..ish...despite you being in "our" camp...per self analysis.
No principals. No ethics. "Do as I say and not what I do."
You were saying (http://www.avsforum.com/forum/86-ultra-hi-end-ht-gear-20-000/1136745-establishing-differences-10-volume-method-14.html#post16386041)?
C'mon Amir, I know your savior BS paper has more holes than swiss cheese, but please don't get angry at me/play victim like you always do. Let's concentrate on why strong pecuniary interests may have lead to the BS paper and some rather cryptic and bizarre choices (some might even say doctoring).
Have you considered submitting your own AES paper to bolster the BS one, with all the evidence you've gathered? If not, why? What is it about that evidence, that would make it utterly useless for AES submittal/review?
cheers,
AJ
I liked your anecdote.
So back to your question, I don't care what the retail price is. I want to encourage a stream of releases that are before the final mastering of the CD where both the 16/44.1 conversion and loudness compression occur. Yes, not all of them are free of the latter but more and more they are.
That's a laudable goal, but the loudness compression doesn't usually occur there. Ironically (and this is part of the point I keep making) the more that people believe hi-res = better sound, the more likely the hi-res version is to suffer loudness compression. There are already hi-res releases that have suffered this fate. That's what happens when you pretend that "more bits than on a CD" is what is important. The really important things get overlooked. I think your hope that both improvements will go hand-in-hand is a little naive.
As sure as SD quality was messed up to introduce HD, and HD quality will be reduced to introduce UHD - that's how sure I am that hi-res releases will suffer loudness compression, unless people tell the truth: it's the loudness compression that really damages the music.
When I bought the album in question, the L2 Nordic Sampler, I opted for 96 Khz. There was 192 at higher price still but it didn't matter to me. I was happy to see that supersonic version but chose one notch lower. In this case I knew there was no loudness compression applied to the 96 Khz so I opted for that. But you know what? I regret not getting the 192 Khz now just for the extra $4. I might go back and get that one just to compare even though it would set me back another $28 (really crappy of them to re-charge you the full amount for another sampling).
Have you noticed the BluRay is cheaper, and supposedly includes 192kHz 5.1? And a hybrid SACD/CD.
If I ask them why they say they want to save some individual from wasting their money.
Do they? I think they want people to be informed. If people choose to "waste" their money, that's not so bad. It's mis-selling that's the real problem.
Here is the reality of it: the above is not the reason Krab, Any, AJ, mzil, xnor, etc. post what they post. Nope. We gather in these forums because it feels good to be known as an authority. Many people pick what I call "good enough" as their platform. It seems safe. We can immediately wear the cloth of "science" and say nothing is better than another, listening test says this and that, some textbook that we have read says something else and we are golden. We become the hallway monitor in school. Walk around the forum and stomp on any kid we can find.
Losing that power is a big deal. This is why xnor posts and posts. And before him Krab, Arny, AJ, mzil, etc. Mzil for pete's sake used to sell audio gear. The very same gear he is here to say should be illegal to buy! Give me a break.
I am not here to join a crowd for that reason. And certainly won't join the camp and be requested to paper over faults in our arguments. I want 100% transparency. Let me repeat: I want 100% transparency. If there is something wrong with our argument, we better offer it before the other camp does.
You see, that's exactly what I see some people here doing. Krab has been clear at least twice: Maybe the latest paper (if all the unclear parts are answered properly) shows what kind of audible difference hi-res brings. That would mean that it exists, but it's minuscule compared with what hi-res proponents are claiming.
Some people might be nearly as bad as you are caricaturing. It's not fair for me to say as they're not arguing with me. But you shouldn't tar everyone with the same brush. Some of the criticism is grounded in exactly the same objective reality you seek to inhabit.
Cheers,
David.
I have no limit and yes, people here are illogical to have one.
Oh, the
master of logic has spoken! See 0 credibility (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=500&p=882029&#entry882029).
Answering your question then, my motivation ha nothing to do with the sampling rate. I don't care if the pre-CD master is 48 Khz. I take the better sounding one!
Another assault on logic. Haven't you claimed that 48 sounds better than 44.1 (let's ignore your lame excuses about not telling us what you heard)?
And this topic is about filtering (which happens during sample rate conversion), yes?
Where I think the vocal few go wrong is that they pick battles when no one is asking them.
And you are the most vocal here, posting stories and pages upon pages of
noise and fallacious nonsense. And
you've said this is a "war". See 0 credibility (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=500&p=882029&#entry882029).
Your personal stories and net worth and preferences are of
no relevance to this discussion.
None. If you need someone to talk to then I'm sure you can find competent help elsewhere.
I am writing this long post because this is the ultimate problem and the cause of all of these written battles. People think they have some duty to fight these battles.
They only one here that even thinks this is a "battle" or "war" is
you. And you've made it abundantly clear that all
you're interesting in is winning this "war", not what is true.See 0 credibility (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=500&p=882029&#entry882029).
If I ask them why they say they want to save some individual from wasting their money. Fine. But at what cost are you doing that? Xnor is determined to get there at the expense of looking totally unprofessional and going after someone who is interested in discussing the topic using double blind tests and science/research of audio.
Putting words in other people's mouth
again. You are
demonstrably and objectively intellectually dishonest,
amirm.
I said before that I don't care at all what you or anyone else spends/wastes their money on. No interest. None.
After all that you've posted (see 0 credibility (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=500&p=882029&#entry882029)), you even dare to call me looking unprofessional? Ahaha. What a sense of humor.
Here is the reality of it: the above is not the reason Krab, Any, AJ, mzil, xnor, etc. post what they post. Nope. We gather in these forums because it feels good to be known as an authority. Many people pick what I call "good enough" as their platform. It seems safe. We can immediately wear the cloth of "science" and say nothing is better than another, listening test says this and that, some textbook that we have read says something else and we are golden. We become the hallway monitor in school. Walk around the forum and stomp on any kid we can find.
Losing that power is a big deal. This is why xnor posts and posts.
Thank the universe that there are not more people living in your
special reality. Thank you, thank you, thank you!
I am no authority. It seems you still even have bothered to look up "argument from authority". Again, a demonstration of willful ignorance (see 0 credibility (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=500&p=882029&#entry882029)).
This is like talking to a brick...
OK, I feel better now.
That is the most surprising of it all. I would feel so ashamed I wish the earth had opened and swallowed me up if I were in your skin ... several pages back.
The wish alone that "many people should read this thread", which you posted a page back, leaves me
dumbfounded. What is going on in your head, armirm?
And finally (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882145):
> imp_urhp.wav (http://www.hydrogenaud.io/forums/index.php?showtopic=107588) (4 seconds file)
You seem to have already downloaded it and listened to it. So how about you tell us what you hear?
Can you answer the psychology of this?
You know the psychology. We see folks pushing barely audible / inaudible improvements to the detriment of easily audible improvements. It impacts negatively on the equipment and recordings that are available. Why wouldn't people who care about audio be annoyed about this?
Indeed. And we can and should interrogate the 'psychology- or rather, the purposes -- of those push barely audible/inaudible improvements so aggressively.
Modern recording/mastering practice, problems with rooms and loudspeakers, the basic deficits of 2channel -- these are perceptually*huge* and prevalent factors in sound quality at home. Yet the hi rez faction is consumed by the 'danger' of mediocre filtering for a 22kHz bandwidth and the 'limits' of 96dB dynamic range.
Also, I think you're taking the flack for all the wild and unsubstantiated claims for hi-res that are thrown around in other places, even though the main argument you've pushed here is merely "why not"?
Apparently it's not incumbent on those who *sincerely* believe there is some (small) freedom in having 'hi rez' choices, to tamp down the wild claims made about hi rez elsewhere, or by people like Neil Young or the Stereophile/TAS crowd.
They seem very reluctant to take up that task.
We could interrogate the 'psychology' of that reluctance , too.
If memory serves Brad Meyer studied Engineering at a well known Ivy League school, I think Harvard.
Your recollection is faulty Arny. From the "crap" I have post on AVS Forum is his bio at the end of their paper:
. http://www.seas.harvard.edu/programs/engineering-sciences (http://www.seas.harvard.edu/programs/engineering-sciences)
The degree requirements for each of the three tracks for the Bachelor of Science in Engineering Sciences (S.B.) are included below.So no, he doesn't have an engineering degree from Harvard.
Funny little story with respect to his work at BBN. I worked for a computer company and we sold them one that had Unix operating system that I had worked on. They filed some bug about corruption of a mail inbox file that was large (large in that era). I answered them in email. Someone took that email from early 1980s and put it online for all eternity to be read! I can't find it now but it used to be that if you searched for me, that would be one of the earliest record of me existing online .
Given the monumental gross errors of yours that I corrected almost daily over at AVS, who are you Amir?
You said it Arny. Oh wait!
These are real demonstrable benefits. It's not just about getting that warm fuzzy feeling from inaudible perfection.
As is the case with getting the high resolution master prior to loudness compression and remastering for CD/MP3/AAC release. We need no double blind test, no signal processing lessons, nothing.
The original is the tracks prior to mix down, and this is almost never for sale.
Whoa. WHOA. Is *this* the line Amir is pushing now? We should favor hi rez because it will give us the master 'prior to loudness compression'?
First of all, who says loudness wars compression can't be baked in at the *mixing* stage? This has been going on since the days when Oasis was topping the charts.
Second, who says a mixed-down source with wide dynamic range, can't be mastered loud and compressed for 'hi rez' release? THEY HAVE BEEN AND STILL ARE.
Third, why does it have to be hi rez?
What is really being asked for here, then?
For old analog recordings it would seem, the request is: let's have digital release that doesn't reduce dynamic range to less than what analog tape would offer. CD can do that handily. In fact *that is what CD was supposed to offer in the first place*.
For digital recordings -- let's presume they are 'pure' digital original masters, and were recorded and produced at rates > 16/44 -- the request is, let's have a digital release that doesn't significantly reduce dynamic range of the master. CD can do that too.
Really, shouldn't we be agitating strongly for the release of *well-mastered* commercial releases? Hi rez is just a sideshow to that. It is a *distraction*.
So no, he doesn't have an engineering degree from Harvard.
Amir, can you explain how that Harvard fact prevented worried audiophiles like yourself, from hearing the benefits of Hi-Rez when "listening" in the large M&M test?
Why JJ says there's not a whit of evidence to support Hi-Rez?
Also, can you explain why your vast body of online cheatable ABX Windows files, taken unsupervised on a Windows computer, can't be submitted to the AES for review?
Especially if it supports the thread title BS paper. Thanks.
Apparently it's not incumbent on those who *sincerely* believe there is some (small) freedom in having 'hi rez' choices, to tamp down the wild claims made about hi rez elsewhere, or by people like Neil Young or the Stereophile/TAS crowd.
You mean other than this comment when discussing this very topic on AVS? http://www.avsforum.com/forum/91-audio-the...ml#post26578505 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1585994-avs-aix-high-resolution-audio-test-ready-set-go-6.html#post26578505)
It's like the Pono Kickstarter video where Neil Young had all those celebrities talking about how they never heard music sound that great outside of a studio. Wow, so awesome.
Agree that is a bunch of nonsense.
As to stereophile, I find them the best source of objective data in the form of their measurements and some of their technical tutorials. I ignore the rest and have said so clearly.
They seem very reluctant to take up that task.
We could interrogate the 'psychology' of that reluctance , too.
It is called objectivity .
No, the step before CD mastering is the stereo mix at the recorded/mixed bit rate and sampling rate which is almost always higher than the CD.
Obviously, the above is an artificial choice that ignores many relevant facts and suits the egregiously flawed agenda that many mislead people seem to want to follow.
But, let's pretend none of don't know any better and mistakenly agree that the final mix (one of many terms that are used to describe what you are actually describing) is the original and agree against reason and relevant facts that it is always at higher sample rate and longer data word length than a CD.
We know based even on just the highly flawed and biased recent Meridian AES paper that whether it sounds different than the CD version of it is difficult or impossible to determine, particularly if we wish for that determination to be adequately reliable. The proof of that is the manifold asymmetries and errors that it took to produce data to that end that supported the false conclusion even 56% of the time.
They don't jump from the multitrack recording directly to an MP3.
Irrelevant. This discussion is about Audibility of Typical Digital Filters in a Hi-Fi Playback System,
not the audibility of MP3 coding. Mention of this highly irrelevant topic is so far off topic that it raises serious questions about the mental acuity of any person who would try to bring this irrelevant issue into the discussion.
Haven't you claimed that 48 sounds better than 44.1 (let's ignore your lame excuses about not telling us what you heard)?
Show me the post.
Amir, please help us out here.
You claim that the inaudible HF artifacts created by downsampling the 24 bit master creates "worry" for you in the audio bandwidth 20-20Khz. Hence your reason for wanting "Hi-Rez", supporting this BS paper, etc.
Can you tell us if you find this HF performance (http://www.stereophile.com/content/mark-levinson-no53-reference-monoblock-power-amplifier-measurements) worrying?:
(http://www.stereophile.com/images/1212ML53fig01.jpg)
The reason I ask, is because here (http://www.madronadigital.com/Library/MarkLevinsonNo53Amplifier.html), you say:
Without going through the circuit design details, let's say it is challenging to create a passive filter that filters out the switching frequency but leaves the audio band perfectly flat. You see the ramification of this in response anomalies of some class D amplifiers that “ringing” at higher frequencies (frequency response has oscillations to it).
In comparison testing I have done, switching amplifiers using the classic class D configuration always sport incredible low frequency control and power. They beat out linear class AB amplifiers almost regardless of price. What they give up though is high frequency fidelity which I find somewhat harsh. The distortion is highly non-linear and challenging to spot but it is there. The Mark Levinson No 53 is the first switching amplifier I have heard which does not have this compromise. Its bass is amazingly authoritative: tight and powerful. Yet the rest of the response is absolutely neutral and pleasant.
If you have not heard these unique amplifiers, I highly encourage you to come into our showroom for a listen.
I'm just curious about this HF "worry" thing with 16/44 filtering, but not with the $50k (2012 pricing) ML amps you peddle. Hopefully you can help shed some light for us. Thanks as always.
cheers,
AJ
We know based even on just the highly flawed and biased recent Meridian AES paper that whether it sounds different than the CD version of it is difficult or impossible to determine, particularly if we wish for that determination to be adequately reliable.
Putting aside the mischaracterization of Stuart paper, no we don't know that Arny. High resolution masters not suffering from loudness compression have clearly audible fidelity difference that you have to be blind in addition to deaf to not hear it . From the bible of good enough camp, here comes a quote from Meyer and Moran:
Though our tests failed to substantiate the claimed advantages of high-resolution encoding for two-channel audio, one trend became obvious very quickly and held up throughout our testing: virtually all of the SACD and DVD-A recordings sounded better than most CDs—sometimes much better. Had we not “degraded” the sound to CD quality and blind-tested for audible differences, we would have been tempted to ascribe this sonic superiority to the recording processes used to make them.
Plausible reasons for the remarkable sound quality of these recordings emerged in discussions with some of the engineers currently working on such projects. This portion of the business is a niche market in which the end users are preselected, both for their aural acuity and for their willingness to buy expensive equipment, set it up correctly, and listen carefully in a low-noise environment.
Partly because these recordings have not captured a large portion of the consumer market for music, engineers and producers are being given the freedom to produce recordings that sound as good as they can make them, without having to compress or equalize the signal to suit lesser systems and casual listening conditions. These recordings seem to have been made with great care and manifest affection, by engineers trying to please themselves and their peers. They sound like it, label after label. High-resolution audio discs do not have the overwhelming majority of the program material crammed into the top 20 (or even 10) dB of the available dynamic range, as so many CDs today do."Missing that is missing the forest from the tree.
The proof of that is the manifold asymmetries and errors that it took to produce data to that end that supported the false conclusion even 56% of the time.
Sorry, no. How many times must we correct this misconception?
(http://amirviews.smugmug.com/photos/i-FKcswgv/0/XL/i-FKcswgv-XL.png)
56% is the threshold line. The medians for all tests are higher than that. So is the standard deviation with the exception of one.
There is also statistical analysis in this regard. Here is the paper itself:
One-sided t-tests were performed for each condition to
test the null hypothesis that the mean score was not
significantly di different from 56.25% correct: the out-
come of these is shown in Table 2. All means were
significantly different from chance other than that
for condition 4, although this t-test just failed to
reach signi cance at the 5% level (t=1.58; p=0.067).Do members have this much difficulty understanding such language?
We know based even on just the highly flawed and biased recent Meridian AES paper that whether it sounds different than the CD version of it is difficult or impossible to determine, particularly if we wish for that determination to be adequately reliable.
Putting aside the mischaracterization of Stuart paper, no we don't know that Arny. High resolution masters not suffering from loudness compression have clearly audible fidelity difference that you have to be blind in addition to deaf to not hear it .
Irrelevant. This discussion is about Audibility of Typical Digital Filters in a Hi-Fi Playback System, not the audibility of loudness compression. Mention of this highly irrelevant topic is so far off topic that it raises serious questions about the mental acuity of any person who would try to bring this irrelevant issue into the discussion.
Not only irrelevant but disingenuous. SD mastering need not suffer from loudness compression. So :
Masters not suffering from loudness compression have clearly audible fidelity difference from masters that do.'nuff said!
No need to lead with the modifier 'high resolution' there, as if a high resolution delivery format was something *required* in order deliver non-compressed audio. That would be a *LIE*.
If you were the AES, an industry organization, or for that matter all the many companies that make up the industry, what would be in your best financial interest? Would it be to manufactures best interests and promote hi-res because you can just sell a ton of sh!t or would you make a stand and actively look to reverse the loudness trend or whatever else is needed to improve recordings? In the former case, the chips fall where they may while in the second you run up against that people want to accommodate their bosses. Actually you have the same issues with the former. More cowbell.
If you were the AES, an industry organization, or for that matter all the many companies that make up the industry, what would be in your best financial interest? Would it be to manufactures best interests and promote hi-res because you can just sell a ton of sh!t or would you make a stand and actively look to reverse the loudness trend or whatever else is needed to improve recordings? In the former case, the chips fall where they may while in the second you run up against that people want to accommodate their bosses. Actually you have the same issues with the former. More cowbell.
It is not either or Chu. The recording engineers are fighting that fight. Psychoacoustics and signal processing people are fighting this one.
Here are some examples:
http://www.aes.org/e-lib/browse.cfm?elib=15934: (http://www.aes.org/e-lib/browse.cfm?elib=15934:) The Loudness War: Do Louder, Hypercompressed Recordings Sell Better?
https://secure.aes.org/forum/pubs/journal/?ID=350: (https://secure.aes.org/forum/pubs/journal/?ID=350:) Perceptual Effects of Dynamic Range Compression in Popular Music Recordings
http://www.aes.org/e-lib/browse.cfm?elib=15598: (http://www.aes.org/e-lib/browse.cfm?elib=15598:) The Loudness War: Background, Speculation, and Recommendations
http://www.aes.org/events/135/workshops/?ID=3712: (http://www.aes.org/events/135/workshops/?ID=3712:) AES New York 2013
Workshop W22
Saturday, October 19, 5:00 pm — 7:00 pm (Room 1E14)
W22 - Loudness Wars: Leave Those Peaks Alone
Panelists:
John Atkinson
Florian Camerer, ORF - Austrian TV - Vienna, Austria; EBU - European Broadcasting Union
Bob Ludwig, Gateway Mastering Studios, Inc. - Portland, ME, USA
George Massenburg, Schulich School of Music, McGill University - Montreal, Quebec, Canada
Susan RogersNote JA from stereophile in red. And we go cursing him left and right.
http://www.aes.org/e-lib/browse.cfm?elib=17085: (http://www.aes.org/e-lib/browse.cfm?elib=17085:) About Dynamic Processing in Mainstream Music
On and on. The problem with these papers is that they are preaching to the choir. The business decision makers from labels and the talent don't go to AES.
And you are way, way wrong on "sell a ton of sh!t" because of high-res. Playing > 16/44.1 has been a feature of home equipment now for decades. I am sure you can get a $200 AVR that claims to play 24/192. So high res does not drive sales. We need to do away with these talking points that don't pass the smell test.
Yes, there is a tiny market but growing one for DACs and such that connect to PC/Mac music servers. Nobody there though was waiting for an AES study to catapult their business.
Edit: the usual typos.
Amir, please help us out here.
[...]
The reason I ask, is because here (http://www.madronadigital.com/Library/MarkLevinsonNo53Amplifier.html), you say:
Without going through the circuit design details, let's say it is challenging to create a passive filter that filters out the switching frequency but leaves the audio band perfectly flat. You see the ramification of this in response anomalies of some class D amplifiers that “ringing” at higher frequencies (frequency response has oscillations to it).
Wow. Ripple, ringing, oscillations? It's all the same. (<- I'm not serious.)
Another explanation for the confusion of ripple vs. ringing at #428 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881786) and the following pages.
I was really more interested in your trying 2Bdecided's linear phase file _50.wav (as I mention below). I recall that I was unable to discern any audible difference with that file, in January 2009.
Oh sorry. I was wondering why at the same time you were saying I took the test and I didn't .
I did very quick try and these are the results for linear phase:
=======================
[color="#0000FF"]foo_abx 2.0 beta 4 report
foobar2000 v1.3.5
2014-11-23 13:38:11
File A: limehouse_linear_phase_050.wav
SHA1: 661058f46dfb7de9fd2687344ece857f0ae1531a
File B: limehouse_reference.wav
SHA1: e8ad96830d23cad4bba5bf822ce875ae452b9e7c
...
Total: 14/16
Probability that you were guessing: 0.2%
...
I distinctly remember losing concentration the couple of times above. But since the results show 99.8% non-chance outcome, I figured it is good enough to post and not suffer any more .
Thanks very much for trying this more challenging ABX exercise of comparing the linear phase file with the reference file.
I informally tried the maximum phase file again tonight and it still sounded duller than the reference file, as per my ABX done in January 2009. With the linear phase file, I sometimes thought I heard a difference but I couldn't get correct results even on a trial basis so there was no point in my proceeding with a formal ABX attempt.
..., I downloaded your latest tracks. At 50.9 to 51.1 I thought the "s" in street sounded distinctly different. There was more lisping in one than the other. In the trial mode I was able to consistently tell the difference for a good sequence of trials. But when I ran with the test without feedback, I think I got down to 30% probably of chance or some such thing. I am just not motivated to try again and don't remember where the critical segment was in my original testing. So if you like to declare this a loss for me, you can . I am just too lazy to try harder and see if I can pass it.
Amirm, many thanks for attempting this. Most appreciated. It provides indirect evidence for my contention that the AVS/AIX conversion was not particularly transparent (and that my own, simply performed, conversion using SoX, was closer to transparency).
Yes I found the "s" sound in "street" telling in the AVS/AIX conversion. In my own conversion with SoX software I sometimes thought I could hear something and I did get three correct results as a trial, but quickly fell off the rails in a formal ABX attempt, and did not proceed further.
If you were the AES, an industry organization, or for that matter all the many companies that make up the industry, what would be in your best financial interest? Would it be to manufactures best interests and promote hi-res because you can just sell a ton of sh!t or would you make a stand and actively look to reverse the loudness trend or whatever else is needed to improve recordings? In the former case, the chips fall where they may while in the second you run up against that people want to accommodate their bosses. Actually you have the same issues with the former. More cowbell.
AES is doing both; *but* I find it saddening that they are giving so much oxygen to 'high resolution' as if it were a game-changing problem-solver in consumer audio. I'm sure's there's at least a few notable AES members who feel the same.
'High resolution' certainly does
not solve the problem of loudness wars mastering; that trend has *nothing to do* with the comparative performance of standard and high resolution formats.
Oh, looks like I missed a few posts full of nonsense:
Unskilled people go and look up signal processing algorithms and implement them. Without critical listening skills, they think the job is done and ship it. This yet again demonstrates that there are two types of listeners: those that can hear non-linear artifacts and those who can't.
Wow, immediately attacking again.
There are clearly two types of people: those who have a clue what they are talking about, and those who merely think they are experts
and embarrassingly fail to even understand that the filters involved are linear, also see #547 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882095) where you
repeat the same nonsense of non-linearity.
Anyone who disagrees should try to duplicate our results. They can try to find the flaws just the same.
And with flaws you mean those producing false positives and flaws like spectrum analyzers running in the back?
You must have downloaded imp_urhp.wav (http://www.hydrogenaud.io/forums/index.php?showtopic=107588) (4 seconds file) by now. Can you tell us what you hear?
And given that proof, you cannot extrapolate from your hearing to the rest of the people. Just because you don't hear the difference doesn't mean it is inaudible and sufficient justification for you to go around and accuse people are being wrong to observe otherwise.
Oh wow, there is this shimmer of an "I get it" moment. You don't prove inaudibility (see my sig, it's funny). You prove that there's something to that claim of audibility by providing evidence that is gathered in independent experiments that could actually be reproduced. You know, that basic science stuff. And no, I'm not talking about your tinkerer ABX logs here.
And even then you cannot extrapolate that it is audible to everyone. As a random example, a placebo may actually cause a positive effect in 30% or 60% of the cases on the patients.
Statistically significant.Does that mean everyone profits from taking these placebos? No.
Does that mean that placebos actually contain active substances? No.
Should doctors therefore give placebos to everyone? No, because that would be
unethical.
There are definitely some connections to draw here.
Actually they duplicated a real-life situation. Took a professional audio workstation tool, Sonic Solutions, and converted the files to 16/44.1. Precisely how real music is produced. That in double blind tests we could tell the difference it means that what people observe in the wild can very much be true. That transparency is not there.
Actually, that is complete nonsense again.
Something like an extra time delay will destroy the ability to produce non-false positive ABX logs, but doesn't mean that the file itself sounds any less transparent. Anyone with a shred of
intellectual honesty (https://en.wikipedia.org/wiki/Intellectual_honesty) (see 0 credibility (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=500&p=882029&#entry882029)) would immediately see this and invalidate the test, because the test setup is flawed.
It matters not that the timing difference may be the reason. What matters in this context is that we are wrong. We are wrong to say they are imagining things when they compare the high res to 16/44.1. We are making idealized assumptions that are not true in reality.
What? It is absolutely crucial that such problems are eliminated in an ABX test. You are again sweeping big problems under the rug and assert that you are right. See 0 credibility (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=500&p=882029&#entry882029).
So I think it is good that we discovered how professional resamplers are not the animals we think they are. There is no visibility into their design as I have mentioned. I imagine hardly any music is produced using Sox resampler. Engineers use the professional tools. And if those tools produce non-transparent results, then we should get the masters and not be subject to this.
These are 100% logical and defensible conclusions we can draw. Anyone who doesn't want to go there has fingers in the ears and in denial.
Oh man, the pain! So you
didn't test the format but actually some resampler implementation and draw the conclusion that the format is the problem? 100% logical conclusions, from the master of logic (see 0 credibility (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=500&p=882029&#entry882029))? I somehow doubt it...
@MLXXX: The moral of the story is that if you provide a negative log then amirm will insult that you are either deaf (but his hearing extends to a whopping ~12 kHz) or that you are an unskilled buffoon. So he thinks he wins his war.
Or you provide a positive log, which he will accept blindly and to him is enough proof. So he thinks he wins his war.
Do you see the problem?
If you were the AES, an industry organization, or for that matter all the many companies that make up the industry, what would be in your best financial interest? Would it be to manufactures best interests and promote hi-res because you can just sell a ton of sh!t or would you make a stand and actively look to reverse the loudness trend or whatever else is needed to improve recordings? In the former case, the chips fall where they may while in the second you run up against that people want to accommodate their bosses. Actually you have the same issues with the former. More cowbell.
AES is doing both; *but* I find it saddening that they are giving so much oxygen to 'high resolution' as if it were a game-changing problem-solver in consumer audio. I'm sure's there's at least a few notable AES members who feel the same.
'High resolution' certainly does not solve the problem of loudness wars mastering; that trend has *nothing to do* with the comparative performance of standard and high resolution formats.
If stuff is already being recorded in hi-res then later down sampled to CD with subsequent tunes in various lossy formats, then I would guess its because you can charge more for hi-res. I saw where Sony was saying they hoped that hi-res would comprise something like 20% of downloaded material. From my perspective, the industry is seeing a steady increase in streaming music. Spotify, Pandora, Sirius/XM and many other outfits are supplanting traditional AM/FM and this is shifting the balance away from downloads. If lossless CD and its lossy offspring are going to comprise fewer downloads, then the problem comes how to recoup the profits that are and will be lost? Hi-res provides an answer or at least part of one.
The way we listen is changing rapidly and many simply no longer prop down in a chair listening to the exclusion of everything else. They're streaming what they want from their devices while checking email, tweeting, got the sports update going on and so forth.
Well, that era is over. Far more careful test has been run which is the topic of this thread. And what do you know? It has a different outcome! What to do now? What? Oh, let's attack the messenger. Let's say that Amir is a cheater and see if we can create some smoke. Let's keep saying that an authority like James Johnston, my former audio architect and AES Fellow, would never agree with any of this. Well, this is another page from his presentation:
(http://amirviews.smugmug.com/photos/i-LdhQVT6/0/O/i-LdhQVT6.png)
This is from JJ's presentation to the local chapter of AES on basics of digital audio: sampling and quantization. This doesn't paint a picture of transparency of 16/44.1 in any music you buy. If JJ volunteers it, we better darn well do the same. Anything else is a political move to censor data in order to achieve self-importance on some forum.
(1) Am I the only one who sees that a very well known best practice, the use of perceptually shaped dither, was specifically excluded?
(2) The presentation http://www.aes.org/sections/pnw/ppt/jj/fund_of_hearing.ppt (http://www.aes.org/sections/pnw/ppt/jj/fund_of_hearing.ppt) also says:
"The range of 20Hz to 20kHz is a reasonable choice for almost all human subjects."
"20 bits are probably sufficient for safe reproduction at home, or in most any venue." (note the absence of the word necessary)
"Given the noise in the modern world, 16 bits is probably sufficient in most places."
Thanks for the reference Amir - plenty of ammunition against your posts!
Am I the only one who sees that a very well known best practice, the use of perceptually shaped dither, was specifically excluded?
JJ has stated repeatedly he doesn't buy into the 'Hi-Rez" scams being peddled. Note that he also used "Pro-sumer" Hafler amps for his demos, not 50k $cam-amps with erratic HF performance, that those who lack critical listening skills might fall for.
Really puzzled why Amir would bring him up, but alas...
He also mentions JJ is an AES fellow, but refuses to say why he won't submit his Windows pc generated ABX file logs as part of a valid AES paper, supporting the BS paper under discussion. I wonder why that is?
cheers,
AJ
Notice the last sentence of the BS paper's opening abstract parrots the same old story we've heard for years, while indirectly plugging Meridian brand gear, YET THE PAPER LACKS ANY BACKING EVIDENCE TO SUPPORT THE SECOND CONCLUSION:
"Two main conclusions are offered...
... and second, an audio chain used for such experiments must be capable of high-fidelity reproduction."
There was no evidence presented showing that a lesser playback chain using speakers which weren't $46K Meridians wouldn't have achieved the exact same 56.25% successful Hi-res ID rate. For all we know a cheap mini system from a department store, with no response above 18kHz, might have revealed the noise modulation of the rectangular dither, or whatever it really was, JUST as well.
In other words, he's implying: "If you personally can't hear it, your system must not be good enough...Might I recommend..."
[Meridian DSP7200 SE speaker retail price based on this]: http://www.theabsolutesound.com/articles/m...lable-to-order/ (http://www.theabsolutesound.com/articles/meridian-special-edition-loudspeakers-now-available-to-order/)
If stuff is already being recorded in hi-res then later down sampled to CD with subsequent tunes in various lossy formats, then I would guess its because you can charge more for hi-res. I saw where Sony was saying they hoped that hi-res would comprise something like 20% of downloaded material. From my perspective, the industry is seeing a steady increase in streaming music. Spotify, Pandora, Sirius/XM and many other outfits are supplanting traditional AM/FM and this is shifting the balance away from downloads. If lossless CD and its lossy offspring are going to comprise fewer downloads, then the problem comes how to recoup the profits that are and will be lost? Hi-res provides an answer or at least part of one.
The way we listen is changing rapidly and many simply no longer prop down in a chair listening to the exclusion of everything else. They're streaming what they want from their devices while checking email, tweeting, got the sports update going on and so forth.
Recording and
production is 'high rez' , at least in the bitdepth dimension, for solid reasons -- >16bits allow more leeway for loud peaks and multiple digital transformations of the data, without fear of introducing audible artifacts. High sample rates allow less steep filtering in the ADC process, again lessening the change of introducing an audible artifact, though this has to be pretty bad to be routinely audible.
For a *delivery* format, this is all much overkill *and* takes up much more drive space/data bandwidth. Downconverting to redbook rates can and should be done with minimal 'hit' to sound quality, and considerable space saving. (It's not *lossy* in a practical or original sense of the word either, no matter what bullshit the hi rez cheerleaders spout.)
If the companies want to offer the non-downcoverted audio, fine, but they shouldn't charge *more* for it. The audio benefit is *in fact* miniscule at best if not inaudible in most situations, while the physical deficit -- taking up more space -- is still significant and easily detectable to anyone who needs to copy a collection of files from one drive to another.
And of course charging more for a *file* versus a physical disc of the same data -- that's just
Justifying price hikes on the bases of promised audible benefits of 'high rez' and on downloadibilty would be pretty blatantly a shuck. Ah, capitalism.
There is also statistical analysis in this regard. Here is the paper itself:
One-sided t-tests were performed for each condition to
test the null hypothesis that the mean score was not
significantly different from 56.25% correct: the out-
come of these is shown in Table 2. All means were
significantly different from chance other than that
for condition 4, although this t-test just failed to
reach significance at the 5% level (t=1.58; p=0.067).
Do members have this much difficulty understanding such language?
You obviously do have difficulties.The null hypothesis was: "the mean score is not significantly different from 56.25% correct". That is a curious null hypothesis, don't you think?
Then they used a one-sided t-test. So after doing their analysis they came to the result that allows rejection of this null hypothesis.
So the mean score was significantly different from 56.25% correct. Wow.
Condition 4 was barely different from that. Wow.
You know that condition 4 was of the highest quality, right? No extra quantization, 48 kHz, despite the arguably unrealistically steep filter.
I'm tired and my statistics is rusty, but doing a couple of 10 trial ABX tests with 70% to 50% pure guessing probability as result should be similarly statistically significant.
I've said it before and I will repeat it again: I take the paper for what it is. It's
no definite proof of the audibility of
either 16-bit quantization, even with botched dither, nor for questionably steep 44.1 kHz brickwall filters and
not even the combination of both. Not even close.
Also, I'm getting tired to remind you about imp_urhp.wav (http://www.hydrogenaud.io/forums/index.php?showtopic=107588) (
4 seconds file). Can you tell us what you hear?
^ And this is without a detailed analysis of the test setup, which includes the filtering and switching software/algorithms, hardware, randomization of trials ... and results and statistical analysis.
We know based even on just the highly flawed and biased recent Meridian AES paper that whether it sounds different than the CD version of it is difficult or impossible to determine, particularly if we wish for that determination to be adequately reliable.
Putting aside the mischaracterization of Stuart paper, no we don't know that Arny. High resolution masters not suffering from loudness compression have clearly audible fidelity difference that you have to be blind in addition to deaf to not hear it .
Irrelevant. This discussion is about Audibility of Typical Digital Filters in a Hi-Fi Playback System, not the audibility of loudness compression. Mention of this highly irrelevant topic is so far off topic that it raises serious questions about the mental acuity of any person who would try to bring this irrelevant issue into the discussion.
Not only irrelevant but disingenuous. SD mastering need not suffer from loudness compression. So :
Masters not suffering from loudness compression have clearly audible fidelity difference from masters that do.
'nuff said!
No need to lead with the modifier 'high resolution' there, as if a high resolution delivery format was something *required* in order deliver non-compressed audio. That would be a *LIE*.
Is it a lie, a false claim, an attempt to distract discussion from points he's in deep trouble over, or an inability to keep the relevant technology sorted out?
Whatever it is, it is repeated here: http://www.hydrogenaud.io/forums/index.php...st&p=882197 (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&view=findpost&p=882197)
Amir, Amir, Amir, dynamics compression and potential losses due to lossy compression are orthogonal with (do you know what that means?) with the audible properties of the digital filters in a typical Hi Fi system.
Notice the last sentence of the BS paper's opening abstract parrots the same old story we've heard for years, while indirectly plugging Meridian brand gear, YET THE PAPER LACKS ANY BACKING EVIDENCE TO SUPPORT THE SECOND CONCLUSION:
"Two main conclusions are offered...
... and second, an audio chain used for such experiments must be capable of high-fidelity reproduction."
[...]
In other words, he's implying: "If you personally can't hear it, your system must not be good enough...Might I recommend..."
[Meridian DSP7200 SE speaker retail price based on this]: http://www.theabsolutesound.com/articles/m...lable-to-order/ (http://www.theabsolutesound.com/articles/meridian-special-edition-loudspeakers-now-available-to-order/)
Since amirm asserts that ringing at 21+ kHz is audible to someone with a hearing rolled-off at 12 kHz and that linear filters suddenly magically are non-linear, all bets are off anyway.
$2 headphones could be enough ... certainly enough to produce positive ABX logs anyway, that are of course undeniable proof!!!
High-re$ gets a totally new meaning. You can use almost any crap system as long as you are an authority golden ear trained expert listener expert!!!
There was no evidence presented showing that a lesser playback chain using speakers which weren't $46K Meridians wouldn't have achieved the exact same 56.25% successful Hi-res ID rate. For all we know a cheap mini system from a department store, with no response above 18kHz, might have revealed the noise modulation of the rectangular dither, or whatever it really was, JUST as well.
Remember that the best condition (#4, no extra bit depth reduction, 48 kHz) was barely better than the 56.25% "
success" rate. So that does hint at some bias in their methodology.
While this would have been too much for the conference paper I agree that given their methodology, the same success or failure rate could have been achieved without the Meridians. amir has demonstrated undeniably that some $100-200 earphone will result in 100% success rate (slightly less if the dog is barking in the background) vs. really expensive Meridians!!!
--
I'm sorry, that was unnecessarily sarcastic.
For digital recordings -- let's presume they are 'pure' digital original masters, and were recorded and produced at rates > 16/44 -- the request is, let's have a digital release that doesn't significantly reduce dynamic range of the master. CD can do that too.
Not interested in a format that does not "significantly reduce dynamic range." Want a transparent format. Such a format not only should support the fully dynamic range as JJ, Stuart, and Fielder say (19 to 20 bits), but also some headroom. Then we can be sure that across all content and all listeners, we have delivered the bits as created.
We can do that trivially today. No need to say with a 30+ year format. No need at all.
No justification has been provided in this thread or anywhere else that some down conversion must be performed before we get the bits. No justification. You are constantly trying to say that we must settle for a photocopy when the original is readily available for us to get. And seemingly advocate that the original should be outlawed. Why? Because otherwise it would mean some sort of defeat relative to high-end audio where in reality that is not the case as even dollar store AVRs play "high res" content.
Really, shouldn't we be agitating strongly for the release of *well-mastered* commercial releases? Hi rez is just a sideshow to that. It is a *distraction*.
We are. We can bypass the commercialization of the format by the label and talent by going to the upstream high resolution release. The business rules are different. High res is assumed to be going to an audiophile who detests value loudness compensation whereas the mass market requires it.
Really, there is no there there. You guys have no argument whatsoever why we should stay with old standards pulled out of thin air to make a piece of plastic hold X minutes of content. There were no listening tests. No psychoacoustics analysis. Nothing.
It is wrong, wrong, wrong to keep advocating down conversion to CD's 16/44.1. Did I say it is wrong? I thought I did .
You cannot have the interest of audiophiles in mind by taking these positions much less having a heart attack over them any time the discussion goes in this direction. This is not snake oil. This is not fancy cables. No, there is science and business reasoning why we should support the move toward high resolution audio distribution. You don't have to buy that content. You just need to stop laundering the stale SACD/DVD-A arguments when everything about the situation is different now.
Not interested in a format that does not "significantly reduce dynamic range."
You're also not interested in intellectual honesty or scientific rigor. So not interested in what you want at all. Sorry.
Oh and now it's suddenly dynamic range that bothers you, while a couple of pages back you argued that it was far from audible even for the "excellent" recordings used in the paper? Oh yeah, that is arguing amirm-style.
No justification has been provided in this thread or anywhere else that some down conversion must be performed before we get the bits.
Again, buy whatever files you want. Nobody cares.
This is about the claim that digital filtering is audible and the evidence to support that claim.
Also, no justification has been provided for small cookies when you could have really big ones.
You are constantly trying to say that we must settle for a photocopy when the original is readily available for us to get.
No, that is just in your head. Give anyone the format of their choice, and let them pay extra and waste as much money as they want. (Although I would say that
if you buy a track, you buy the track, not some format. So any format of the track should be downloadable once you paid for a track.)
Really, there is no there there. You guys have no argument whatsoever why we should stay with old standards pulled out of thin air to make a piece of plastic hold X minutes of content. There were no listening tests. No psychoacoustics analysis. Nothing.
Oh amirm, I actually feel pity.
After 25 pages and countless pointers you still are completely lost.
You just need to stop laundering the stale SACD/DVD-A arguments when everything about the situation is different now.
What is different regarding the audibility?
You know, armirm, telling yourself something really really often still doesn't make it true. At least not in reality.
As for "old standards pulled out of thin air", 44.1/16:
a) Some people with still good hearing can hear up to about 19, maybe 20 kHz at unrealistic levels, but the vast majority of people have even trouble with >16 kHz.
b) The noise density of 16 bit is -137 dB. High fidelity playback levels shouldn't exceed an average SPL of 85 dB. You can do noise shaping. Rooms are noisy. Recordings have a noise floor.
You can figure out the rest.
amirm, have you ever even heard a 18 kHz tone? A 19 kHz tone? Let me tell you one thing: it's annoying, really really annoying.
Have you ever heard a jet engine with enabled afterburner in close flyby? Why would you even want anything close to such a dynamic range? Are you a masochist?
I've heard both and I'd categorize both as potential torture devices.
edit:
imp_urhp.wav (http://www.hydrogenaud.io/forums/index.php?showtopic=107588) (4 seconds file) is waiting. Tell us what you hear in your system!
Oh amirm, I actually feel pity.
After 25 pages and countless pointers you still are completely lost.
He's not lost, he's just struggling furiously to control the narrative. He gets away with that elsewhere, but he can't do it here.
High rez cheerleaders need to stop 1) bamboozling the public about the supposedly obvious audible benefits of high rez -- instead, tell them that for analog sources the benefits
should be essentially *nil*, and for digital sources, they
should be *miniscule at best*, which also means 2) stop lying/fearmongering about 'standard' rez delivery formats (as well as lossy compressed audio).
Tell them where the *real* problems are.
It
sure ain't with the digital filters.
It is wrong, wrong, wrong to keep advocating down conversion to CD's 16/44.1.
BS/Meridian:
.[/quote]
Did I say it is a $cam, $cam, $cam? Promoted and peddled by $cammers? (http://discover.store.sony.com/High-Resolution-Audio/)
The difference between
hearing and listening (tell me if you've seen this obfuscation weasel words before)
Discover subtle details and artistic nuances in your favorite
music that you’ve never heard before. Feel the power
and presence of a live performance in your living room.
Or experience what it’s like to sit in on a live studio recording.
It’s all possible with the superior quality of High-Resolution Audio.
With quality greatly surpassing that of MP3 and CD, the difference is clear.
You cannot have the interest of audiophiles in mind by taking these positions
(http://www.stereophile.com/images/1212ML53fig01.jpg)
In comparison testing I have done, switching amplifiers using the classic class D configuration always sport incredible low frequency control and power. They beat out linear class AB amplifiers almost regardless of price. What they give up though is high frequency fidelity which I find somewhat harsh. The distortion is highly non-linear and challenging to spot but it is there. The Mark Levinson No 53 is the first switching amplifier I have heard which does not have this compromise. Its bass is amazingly authoritative: tight and powerful. Yet the rest of the response is absolutely neutral and pleasant.
If you have not heard these unique amplifiers, I highly encourage you to come into our showroom for a listen.
Once again, the arguments about the BS paper completely subverted into "what Amir the 'worried' audiophile wants", rather than anything to do with the BS paper/results actually supporting the Hi-Rez $cam.
cheers,
AJ
Remember that the best condition (#4, no extra bit depth reduction, 48 kHz) failed to achieve the 56.25% "success" rate. So that does hint at some bias in their methodology.
Not at all. The success rate was above threshold of chance for the more critical segments. Per ITU BS1116:
Programme material
Only critical material is to be used in order to reveal differences among systems under test. Critical material is that which
stresses the systems under test. From Stuart's paper:
When the analysis was restricted to just the high-
yield audio sections, performance was signi cantly
better for the 48-kHz filter than for the 44.1-kHz filter.It is critical that in such tests we only use the data from content that reveals differences and testers who have the right acuity. Hobbyist tests routinely violate these rules and just throw whatever at the test and see what happens. This, from the response to the letters to the Journal of AES when the Meyer and Moran report came out, complaining about its poor statistical analysis:
Authors’ Reply2
Dr. Dranove has set requirements for our engineering
report that were not part of our plan, and then dismissed it
for failing to meet them. In hindsight it probably would
have been better for us not to cite the total number of trials
as there are issues with their statistical independence, as
well as other problems with the data. We did not set out to
do a rigorous statistical study, nor did we claim to have
done so. Accordingly it may not mean much to do a more
detailed data analysis, though we have done further work
on it that we will discuss later.
[...]
We did not know in advance what source material, what type
of system, or which subjects would be the most likely to
reveal an audible difference.
Didn't know what material, system or subjects likely to reveal an audible difference. In other words, shooting in the dark. Results is that they walked right into Simpson's Paradox. This is what passed for "scientific proof" that high res has no value. Right....
So no, the only bias I see are the vocal few whose audio ideology is now seriously questioned by Stuart's test/paper. And ad-hoc tests we have run. Nothing new there. Folks are being human. Just not objective.
While this would have been too much for the conference paper I agree that given their methodology, the same success or failure rate could have been achieved without the Meridians. amir has demonstrated undeniably that some $100-200 earphone will result in 100% success rate (slightly less if the dog is barking in the background) vs. really expensive Meridians!!!
I have no issue with this statement. Good ears trump good gear any day of the week and twice on Sunday .
I'm sorry, that was unnecessarily sarcastic.
No worries. Your posts make for great fodder to demonstrate our substantial bias in how we examine scientific data.
You're also not interested in intellectual honesty or scientific rigor. So not interested in what you want at all. Sorry.
Thanks for sharing your sentiments but I didn't write that for you Xnor. Remember, I don't write my posts for the few of you with such extreme prejudice that you would question the Pope on his Christianity should he say something unthwarted about audio.
This is about the claim that digital filtering is audible and the evidence to support that claim.
That evidence is in Stuart's paper. Other than jumping up and down, you have put forward nothing else to dispute it. Thanks to mzil, I get to remind you that the paper has won an award for best peer-reviewed paper. So you excuse me if I don't take opinion of it from anonymous posters seriously.
Really, there is no there there. You guys have no argument whatsoever why we should stay with old standards pulled out of thin air to make a piece of plastic hold X minutes of content. There were no listening tests. No psychoacoustics analysis. Nothing.
Oh amirm, I actually feel pity.
After 25 pages and countless pointers you still are completely lost.
Your lack of emotional maturity is apparent in every post. Instead of being focused on the technical discussion every sentence oozes with anger, emotional outburst, frustration, etc. You have no self-awareness of what these forums do to you. And importantly how you have been recruited as yet another martyr in this war. Folks are sitting back enjoying you spending all of your free time writing this stuff and fighting the good fight. I am happy that you are doing it because you show how unprofessional and unscientific we are about our approach to audio. So don’t stop on my account. But remember, there are folks who are egging you on and you are blindingly do so.
You just need to stop laundering the stale SACD/DVD-A arguments when everything about the situation is different now.
What is different regarding the audibility?
For one thing we can test them a hell of a lot better. We can test the files instead of two real-time audio streams that make it extremely difficult to focus on the critical segments. For another, it is very easy to perform mechanical analysis as everyone seems to be doing these days. The truth may have always been there. But we now have better tools to get it out.
Difficulty of running such tests with real-time sources resulted in sharply raised probability of negative outcome in double blind tests.
This is why you are seeing a new chapter in this book. Don’t dwell on the first as if that is the entire book.
You know, armirm, telling yourself something really really often still doesn't make it true. At least not in reality.
I worked in this field professionally. We are discussing a paper whose core author I know personally. I have cited critical reference from JJ who used to work for me and is a friend. You can declare what I say however you want but can’t change the facts I have stipulated:
1.High resolution master will completely, completely eliminate the risk of downgrading to 16/44.1 being audible. The entire argument becomes moot. Mess with the bits and you buy yourself a world of grief trying to prove transparency.
2.High resolution masters are become available in volume and no one is waiting for Mr. Xnor or any other forum to give them the OK to do so.
3.High resolution masters can and do come with better mastering than their counterpart CDs. Their business rules are sharply different where the content owner and distributor know that the customer is audio conscious as opposed to mass market. You can spit in the wind but the mass market products will not optimize for fidelity at the expense of other goals.
4.CDs will be on their way out. If not a couple of years, five or six years from now. On that day, we could face the highest fidelity being MP3/AAC or better than the CD in the form of high resolution stereo masters. Anyone who argues now, is by design in favor of the former and earns zero respect from me. Let me know when you become an audiophile and we can talk.
5.We hear differently. Some of us are able to be critical listeners and outperform others in detecting small differences. This is supported by considerable amount of research beyond personal data I have shared.
These are the facts and you can take them to the bank. Everything else is forum bickering substituting for the same.
Oh amirm, I actually feel pity.
After 25 pages and countless pointers you still are completely lost.
He's not lost, he's just struggling furiously to control the narrative.
That narrative has sunk your boat Steven. We are discussing peer reviewed, award winning paper/listening test that has disputed literally thousands of posts by you all to the contrary. You said this would never happen. You milked lack of data as data. Now it is time of reckoning. You are hoisted with your own petard. You demanded personal ABX DBT test. You have it. You demanded published report. You have it. You didn't ask for it but said report is written by an AES Fellow, not a hobby group.
These are the facts and are undisputable. They really are.
He gets away with that elsewhere, but he can't do it here.
It is not me that has brought your world to the end Steven: it is the sheer weight of evidence that you said would never exist. But exist it has. We are at the tail end of this process where folks are resorting to emotional comments, fallacious summaries like yours, etc. It is that against published AES report.
High rez cheerleaders need to stop 1) bamboozling the public about the supposedly obvious audible benefits of high rez
Nothing supposed about it. Using the freedom to use alternative masters, there are obvious audible benefits putting aside its better technical capabilities.
-- instead, tell them that for analog sources the benefits should be essentially *nil*
"Essentially?" You don't seem sure. Come back when you are and can bring some data with you.
, and for digital sources, they should be *miniscule at best*,
Not yours to declare their subjective value.
which also means 2) stop lying/fearmongering about 'standard' rez delivery formats (as well as lossy compressed audio).
There is the emotional outburst. If you have won, there is no need to talk like this.
Tell them where the *real* problems are.
It is you all who portray a mean, stubborn, biased image of objectivity. This is the problem. No problem otherwise. Folks are buying high res files and couldn't care less what you think.
It sure ain't with the digital filters.
Now you are sure? Above you weren't? Can't stay consistent even in the same post.
1.High resolution master will completely, completely eliminate the risk of downgrading to 16/44.1 being audible.
Your own authority Jim Johnson in a document you cited says that this risk does not exist in usual circumstances.
2.High resolution masters are become available in volume and no one is waiting for Mr. Xnor or any other forum to give them the OK to do so.
The actual original master tape is the original tracks that were mixed to produce the artistic work. This is never distributed and in general the high resolution descendants of it are not bit-perfect copies of it.
3.High resolution masters can and do come with better mastering than their counterpart CDs.
That is often a false claim because the CD versions often embody the same mastering as the high resolution versions.
I hesitate to post any further, as this thread is already very long, and has perhaps achieved as much as it's likely to achieve [unless perhaps revelations were to come to light of deficiencies in the BS paper experimental setup].
I am responding to a comment directed at me.
...
So I think it is good that we discovered how professional resamplers are not the animals we think they are. There is no visibility into their design as I have mentioned. I imagine hardly any music is produced using Sox resampler. Engineers use the professional tools. And if those tools produce non-transparent results, then we should get the masters and not be subject to this.
These are 100% logical and defensible conclusions we can draw. Anyone who doesn't want to go there has fingers in the ears and in denial.
Oh man, the pain! So you didn't test the format but actually some resampler implementation and draw the conclusion that the format is the problem? 100% logical conclusions, from the master of logic (see 0 credibility (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=500&p=882029&#entry882029))? I somehow doubt it...
@MLXXX: The moral of the story is that if you provide a negative log then amirm will insult that you are either deaf (but his hearing extends to a whopping ~12 kHz) or that you are an unskilled buffoon. So he thinks he wins his war.
Or you provide a positive log, which he will accept blindly and to him is enough proof. So he thinks he wins his war.
Do you see the problem?
Amirm makes the point that in practice professional sample rate conversions can introduce changes because of vagaries of the software. (Some people doing such conversions might be incapable of hearing such vagaries;
or might not even try to hear them in a careful A B comparison, relying on a [misplaced?] faith in the software.)
In relation to the recent test files over at AVS Forum being distinguishable after being "professionally" converted from 24/96 to 16/44, and then back to 24/96, I think two of the vagaries were irrelevant to sound quality (my points 1 and 2 below):
- In the first batch of test files, there was a very slight reduction in level attributable to a feature to prevent clipping for certain source files mastered right up to the maximum sample level of 0dB in the high resolution format. This slight level reduction would go undetected if listening to the converted files alone rather than doing a painstaking A B comparison. Importantly, this effect wouldn't of itself diminish the sound quality.
- In the second [level matched] batch of test files, a timing delay of around 10mS existed (for at least two of the converted files). This would go undetected if not doing a precision A B comparison. Again, of itself, this effect wouldn't diminish the sound quality of the converted files. Recording company conversions are not normally done for the purpose of creating comparison files for ABX testing!
- There was for my ears still some very subtle change in the tone with the batch 2 version of On The Street Where You Live. And I was able to ABX it. One of the members of AVS Forum thought the conversion was more euphonious (though he lacked the hardware setup to corroborate his [and his nephew's] ability to distinguish the second batch of AVS/AIX files. To my mind it is only this slight difference that was significant about the professional conversion. And it was still a very minor difference detectable by a certain percentage of people in an A B comparison, but without an A B comparison being available quite difficult to detect, I would think.
In relation to my own initial conversion for AVS Forum (resulting in comparison files X and Y for
On The Street Where You Live) being 'noisy' as regards the dither, I think amirm may have made a lot out of the slightly elevated noise. It wasn't something noticeable to me at a moderate playback level. I see that, over at AVS Forum, Kees has asked amirm for more detail about his listening setup, at post #650 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1598417-avs-aix-high-resolution-audio-test-take-2-a-22.html#post29292866) of the "take 2" thread.
In relation to my SoX conversion for AVS Forum (resulting in comparison files X2 and Y2 for
On The Street Where You Live), amirm acknowledged that the differences were difficult to hear for the purposes of a formal ABX test, though informally he had obtained some correct answers when preparing to do a formal ABX test.
So, yes, the particular conversion software used in a professional conversion could introduce a
slight effect noticeable under conditions of careful scrutiny in an A B comparison. [Actually I myself didn't have any preconceived notion that professional conversions would all necessarily be completely transparent, if A B compared under clinical conditions. I had not interpreted the Meyer and Moran study to suggest that. That study in my recollection was about listening broadly and not being told when a change was made to bottleneck the sound. That loose type of setting is less likely to reveal a hair-splitting difference that might be just detectable with clinical repetition of two short sections of a recording that have been subject to different processing.]
...1.High resolution master will completely, completely eliminate the risk of downgrading to 16/44.1 being audible...
...2.High resolution masters are become available in volume and no one is waiting...
People are entitled to buy whatever audio file format they wish. However sellers should be obligated to display a disclaimer in their download sites. Similar to what is done in cigarrete packages. Something like this:
"WARNING! If you think you can hear the difference between the hi rez file you are downloading and 16bit/44.1kHz CD standard you are seriously delusional. You should look for professional help immediately."...4.CDs will be on their way out...
...These are the facts...
Why do you say that?! Just wait until the people that are now "rediscovering" LPs start to "rediscover" the CD. They will be amazed with the great sound improvement!
...1.High resolution master will completely, completely eliminate the risk of downgrading to 16/44.1 being audible...
...2.High resolution masters are become available in volume and no one is waiting...
People are entitled to buy whatever audio file format they wish. However sellers should be obligated to display a disclaimer in their download sites. Similar to what is done in cigarrete packages. Something like this:
"WARNING! If you think you can hear the difference between the hi rez file you are downloading and 16bit/44.1kHz CD standard you are seriously delusional. You should look for professional help immediately."
From Meyer and Moran once more:
Though our tests failed to substantiate the claimed advantages of high-resolution encoding for two-channel audio, one trend became obvious very quickly and held up throughout our testing: virtually all of the SACD and DVD-A recordings sounded better than most CDs—sometimes much better. Had we not “degraded” the sound to CD quality and blind-tested for audible differences, we would have been tempted to ascribe this sonic superiority to the recording processes used to make them.
Plausible reasons for the remarkable sound quality of these recordings emerged in discussions with some of the engineers currently working on such projects. This portion of the business is a niche market in which the end users are preselected, both for their aural acuity and for their willingness to buy expensive equipment, set it up correctly, and listen carefully in a low-noise environment.
Partly because these recordings have not captured a large portion of the consumer market for music, engineers and producers are being given the freedom to produce recordings that sound as good as they can make them, without having to compress or equalize the signal to suit lesser systems and casual listening conditions. These recordings seem to have been made with great care and manifest affection, by engineers trying to please themselves and their peers. They sound like it, label after label. High-resolution audio discs do not have the overwhelming majority of the program material crammed into the top 20 (or even 10) dB of the available dynamic range, as so many CDs today do."You think Meyer and Moran were delusional of the differences they observe there?
Where do you guys get such a myopic view of this topic? Get out in the real world. It is not the distorted picture you see on forums like this.
Why do you say that?! Just wait until the people that are now "rediscovering" LPs start to "rediscover" the CD. They will be amazed with the great sound improvement!
The fact is that the market for 'hi rez' in and of itself -- as for all 'audiophile' niches -- is tiny relative to all audio content sales. And sales arent always about the 'resolution'. I've bought quite a few 'high rez' releases myself in the past few years....because they included 5.1 mixes and bonus tracks and 'flat transfers' of original masters, which are what I *actually* cared about and *obviously* make a difference. I had to buy the 'high rez' edition to get them.
Even so, if that changes -- if hi rez becomes the norm, if the industry somehow decides to replace all the most popular forms of profitable audio delivery content (currently CD and lossy compressed downloads and streaming) with 'hi rez' as the default (and that's the only way it will ever become even as popular as, say, LP is now).... I say *so what?*, from an audibility perspective.
Unless they commit at the same time to 'old school' mastering, it won't matter to those who actually understand the differences. The average consumer might 'believe' they got 'better sound' that 'even your wife could hear', but sonic difference, where it exists, will overwhelmingly come from different mastering (as Meyer and Moran correctly deduced). So why will Joe Listener believe it's due to the 'high rez' sauce? Because the assurances will flow like honey, the truth will be downplayed, and the noisy tap-dancing will be fast and furious and constant, as we've seen here performed by the industry's avatar from Madrona.
IOW *nothing will
substantially change* unless *mastering practices* change. And if they do, high rez still *won't be the reason for better sound* and won't be *necessary* to deliver that sound at home. Hi rez delivery format is at best an *appendage* , a *marketing tool*, a *comfort to the paranoid* ; and at worst a *diversion* of technical and educational resources better directed elsewhere.
Tap dancing CD Fearmongers need to STOP saying we need hi rez to get good sound. That's a lie. We don't. We need good recording and mastering, and good listening setups
From Meyer and Moran once more:...
In the most characteristic "subjective audiophile" fashion I am having a strong audio gut feeling that you purpously misquoted Meyer and Moran. They mean exactly the oposite of what you are saying since 2007.
...of the SACD and DVD-A recordings sounded better than most CDs...
That is a very illogical affirmation. It should have said it sounded diferent (not better). it s that from a scientific paper??!!
...High-resolution audio discs do not have the overwhelming majority of the program material crammed into the top 20 (or even 10) dB of the available dynamic range, as so many CDs today do.
Dynamic range compression does not depend on audio format. It is a choice made by the recording technician. Didn't Meyer and Moran know that?! I don't think they said that. I don't think you can be trusted as a source of information. Can you show some evidence?
Where do you guys get such a myopic view of this topic?
The only myopic views are coming from your posts. If a difference can't be heard in a properly set double blind test (to get rid of bias) than that difference is irrelevant. I think you know that but you choose to believe otherwise.
You think Meyer and Moran were delusional of the differences they observe there?
There is
no technical reason not to put a noise-shaped dithered and resampled "hi-res" master on a CD, and no test has shown that such a recording will be perceptually audibly different from the "hi-res" master. Any "better sounding" DVD/SACD recordings are due to engineers intentionally or ignorantly putting a worse master on CD.
My assertion is as follows:
The sound quality deficiencies of contemporary CD recordings are due to human error (intention or ignorance) and have nothing to do with the technological means available. Any perceived improvement in "hi-res" audio formats are due to different mastering decisions and practices. Therefore, the null hypothesis is that if properly downconverted to redbook audio standard, there will be no perceivable differences of "hi-res" and redbook recordings.This null-hypothesis can be (in theory easily) be rejected by down converting a "hi-res" release to redbook 16bit/44.1kHz using proper dither and resampling, converting it back up to the "hi-res" specs, and then succeeding in an double-blind ABX test against the original "hi-res" master.
The burden of proof is on the proponents of "hi-res" delivery formats to show there is an actual difference. Any other convoluted schemes or red-herrings do nothing towards to satisfactorily or scientifically show advantages of "hi-res" delivery formats. On the contrary, anyone engaging in convoluted nonsense and sleight-of-hands tricks to push his "hi-res" agenda disqualifies himself from this discussion.
Any "better sounding" DVD/SACD recordings are due to engineers intentionally or ignorantly putting a worse master on CD.
If you read M&M yourself directly, instead of sections taken out of context, you'll see that's EXACTLY their point:
"
Our test results indicate that all of these [SACD/DVD-A] recordings could be released on conventional CDs with no audible
difference."
http://www.drewdaniels.com/audible.pdf (http://www.drewdaniels.com/audible.pdf)
Any "better sounding" DVD/SACD recordings are due to engineers intentionally or ignorantly putting a worse master on CD.
If you read M&M directly, instead of sections taken out of context, including parts which they attribute to STUART*, you'll see that's EXACTLY their point.
After re-reading my post I changed the wording. I must admit that I only read these few lines and not the full paper. Just shows you how ill-intended amir is going into this discussion.
I think you've got to be careful. A lot of excellent hi-res recordings sound equally gorgeous on CD. The CD releases aren't crippled. It's more that the kind of recording engineers and labels who can be bothered with hi-res (especially early on) are the ones more likely to make great sounding recordings. Hence the average sound quality of all hi-res releases in the world is probably higher than the average sound quality of all CD release in the world - because a lot of rubbish has never been released in hi-res.
I think there was a point in time when you could say the same about CD. I think there is a time when you can say this about almost any new format. There's a point in the uptake when only the people who really give a damn are doing anything; before the mass market has jumped in and churned out rubbish in the new format, just like it did in the previous one
You will know that there is some mass market content released as hi-res, and some of it isn't very good at all. The best CDs sound better than the worst hi-res releases - and I predict the more hi-res releases there are, the more overlap there will be as more mediocre recordings are released in hi-res.
Cheers,
David.
P.S. forgive the gross generalisations. I know there are exceptions to all of this.
You think Meyer and Moran were delusional of the differences they observe there?
There are right (IOW Scientific) ways to do preference tests, and neither of the methodologies that M&M appear to have used to reach this conclusion being ABX1982 and Sighted Evaluation are among them.
Where do you guys get such a myopic view of this topic? Get out in the real world. It is not the distorted picture you see on forums like this.
My real world has a fairly important place in it for Science. Some people don't. They sell Truth and Science down the river when ever their ego gets hurt. They misquote their own sources and make their science up as they go along. In history some of these people were Stalin and Hitler. Not my favorite role models... ;-)
Wow, I don't think I've been in a Godwin'd thread before.
Any "better sounding" DVD/SACD recordings are due to engineers intentionally or ignorantly putting a worse master on CD.
If you read M&M yourself directly, instead of sections taken out of context, you'll see that's EXACTLY their point:
"Our test results indicate that all of these [SACD/DVD-A] recordings could be released on conventional CDs with no audible
difference."
http://www.drewdaniels.com/audible.pdf (http://www.drewdaniels.com/audible.pdf)
I have a number of Enhanced SACDs with both PCM and DSD versions of identically mastered recordings on different layers of the same disc. Both sound great. So M&M's assertion here has real world meaning to me. From this you may discern that I have a SACD player. I also have a DVD-A player. They sound great, but I attribute that to the source recordings and production.
Wow, I don't think I've been in a Godwin'd thread before.
My bad. Mentioning Stalin is in the same category but I feel the need to give credit where credit is due.
Well, it has already been Amired down. ;-)
Stick a fork in it...
Not at all.
I shouldn't have written "failed to", but "was barely different" from achieving a better than 56.25% success rate - all of which sounds like guesswork to me. Well, maybe with some extra bias in their methodology.
The success rate was above threshold of chance for the more critical segments. Per ITU BS1116:
Programme material
Only critical material is to be used in order to reveal differences among systems under test. Critical material is that which
stresses the systems under test.
From Stuart's paper:
When the analysis was restricted to just the high-
yield audio sections, performance was significantly
better for the 48-kHz filter than for the 44.1-kHz filter.
Yes, and if they had set their bar to 55% guessing instead of 56% guessing it would
not be significantly different. And the paper in other words: "when we cherry-pick the data, we can show significant differences."
That is easily within the margin of error of their experiment itself, because condition 6 (48 kHz, RPDF) had better detection rate than the worst condition. Not only in mean, but ~2% better in terms of lower standard error as well.
Those 2% would make every other condition statistically insignificant or condition 4 statistically significant. Of course then they'd have had to move the bar..
This, from the response to the letters to the Journal of AES when the Meyer and Moran report came out, complaining about its poor statistical analysis:
Oh, so you're doing that again.
Evading. Mudslinging. (see 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029))
I don't care about M&M and certainly not about your quote mining techniques.
This is what passed for "scientific proof" that high res has no value. Right....
Honestly, what you keep doing amirm is starting to
disgust me and very likely other readers as well.
So no, the only bias I see are the vocal few whose audio ideology is now seriously questioned by Stuart's test/paper.
Biased to reality, as opposed to some fantasy world where logic is an alien concept, fallacies are welcome and intellectual honesty is frowned upon (see 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029)).
I have no issue with this statement. Good ears trump good gear any day of the week and twice on Sunday .
Wow.
Directly contradicting your authority Stuart on that. I guess that makes their whole paper invalid according to you? (Not that anyone would care.)
You are spending hours crafting up these replies that are just noise, but cannot tell us what you hear in a 4 second file?
imp_urhp.wav (http://www.hydrogenaud.io/forums/index.php?showtopic=107588)
No worries. Your posts make for great fodder to demonstrate our substantial bias in how we examine scientific data.
But I'm greatly worried, about Microsoft and humanity in general when I read these posts.
And yes, my posts are a great (albeit very painful) demonstration of your 0 credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
Is there evidence that digital filters like those used in digital speakers and room correction devices are free of "deleterious" effects on audio signals ?
Is there evidence that digital filters like those used in digital speakers and room correction devices are free of "deleterious" effects on audio signals ?
It is hard to prove a negative. Is there any evidence that digital filters cause "deleterious" effects on audio signals?
It is hard to prove a negative. Is there any evidence that digital filters cause "deleterious" effects on audio signals?
Yes. In the Best Peer-Reviewed Paper under discussion conclusion no.1 is:
FIR filters that emulate downsampling for sample rates of 44.1 kHz and 48 kHz can have a deleterious effect on the listening experience in a wideband playback system.
I was just wondering if it was a good thing to test the effect of (steep) digital filters by using a playback system that apparently contains several digital filters. The specifications of the playback system are:
192 kHz D/A convers, wide frequency response (up to 40 kHz), very low differential group delay and compact impulse response with insignificant ringing.
There is quite some anecdotal evidence about audible effects of digital filters, so to me an all analogue path after the DAC would make more sense. Perhaps a reason could be that Meridian doesn't have any non-digital speakers ?
Is there evidence that digital filters like those used in digital speakers and room correction devices are free of "deleterious" effects on audio signals ?
Since a lot of them run at 44 or 48 KHz, we're back to the same old controversy. Lets put it this way, if set for flat frequency response they seem to hold no more potential for audible terror then any other reconstruction filters with the same corner frequencies.
Not all room correction devices are the same - the two main competing technologies involve on the one hand digital filters that emulate traditional minimum phase paramatric and graphic equalizers and crossovers, while on the other hand we have those that are FFT-based equalizers. Just to keep us on our toes, the FFT-based equalizers often seem to actually be hybrids, and mix both technologies because of the computational load of real-time high resolution FFTs, particularly at low frequencies. We are in the land of proprietary technologies, and these details are often far from being forthcoming.
Even if you could show "deleterious" effects of real-world filters, it doesn't seem to matter, because as
Meridian writes:
Early digital recordings suffer from harshness too, due to primitive filters. After painstaking research, Meridian developed a special filter [...] cleaning up the effects of early filters.
(Let's ignore the fact that this is
partly false, because aliasing is virtually impossible to remove from any sampled piece of inadequately filtered music. In other words if the A/D conversion is crap, you cannot un-crap it.)
Also see #539 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882067) for a demonstration.
It is hard to prove a negative. Is there any evidence that digital filters cause "deleterious" effects on audio signals?
Yes. In the Best Peer-Reviewed Paper under discussion conclusion no.1 is:
FIR filters that emulate downsampling for sample rates of 44.1 kHz and 48 kHz can have a deleterious effect on the listening experience in a wideband playback system.
I was just wondering if it was a good thing to test the effect of (steep) digital filters by using a playback system that apparently contains several digital filters. The specifications of the playback system are:
192 kHz D/A convers, wide frequency response (up to 40 kHz), very low differential group delay and compact impulse response with insignificant ringing.
There is quite some anecdotal evidence about audible effects of digital filters, so to me an all analogue path after the DAC would make more sense. Perhaps a reason could be that Meridian doesn't have any non-digital speakers ?
I don't know all the details about the Meridian speakers, but the fact that the Meridian supplied frequency response curve only goes up to about 40 KHz suggests that they may be based on 96 KHz sampling.
Is there evidence that digital filters like those used in digital speakers and room correction devices are free of "deleterious" effects on audio signals ?
According to BS/Meridian regarding the stuff they sell, no, they are transparent:
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
According to world renown expert JJ, no.
Let's start: "how do you even imagine that one can hear a difference between two systems, one with noise 98 dB down and the other 146dB down, when the level is set to peak at 96dB?"
How do you (Amir) explain how "obviously difference" fails to show up in even the worst kind of ABX test? (unless it is the unsupervised, Windows pc, online cheatable variety, which are worthless and unrelated to the anti-ABX BS paper)
I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener. Have you any, bearing in mind that citing non-blind-testing proves nothing but the incompetence, the complete and total incompetence, of the person citing it as evidence.
Bear in mind the hard evidence for the persistance of loudness memory while you're at it.
08-01-2014, 10:07 PMjj_0001 avsforum (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1532092-debate-thread-scott-s-hi-res-audio-test-95.html#post26230585)
I have my doubts that SACD or DVDA are much, if any, of an improvement, but the test is just blisteringly hard to run, and more likely to respond to artifacts, either positively or negatively, than it is to actual differences. Time alignment, level alignment, frequency response in-band can all throw it positive, lack of training, bad test environment, bad time alignment, etc, can also cause false negatives. Subject verification, likewise, is an important issue.
So, I remain undecided, but I note that I own a lot of CD's and not a single SACD or DVDA, except for some people have sent me.
According to subjectivist Amir, no, no deleterious effects of HF filtering, not with this $50k amp he sells and waxes about subjectively:
(http://www.stereophile.com/images/1212ML53fig01.jpg)
In comparison testing I have done, switching amplifiers using the classic class D configuration always sport incredible low frequency control and power. They beat out linear class AB amplifiers almost regardless of price. What they give up though is high frequency fidelity which I find somewhat harsh. The distortion is highly non-linear and challenging to spot but it is there. The Mark Levinson No 53 is the first switching amplifier I have heard which does not have this compromise. Its bass is amazingly authoritative: tight and powerful. Yet the rest of the response is absolutely neutral and pleasant.
If you have not heard these unique amplifiers, I highly encourage you to come into our showroom for a listen.
He even admitted to not hearing audible issues with 16/44 earlier in the thread, caught himself, slid, two stepped and then played his usual word games with, what he really meant was "problems".
IOW, there is zero reason to $pend more on "Higher Resolution" of 2channels, 10% of soundfield recordings - which no audiophile on earth, can demonstrate hearing on their systems, unless they create pathological scenarios, like cranking them up to dangerous levels with cherry picked material, in an iso-ward.
cheers,
AJ
Even if you could show "deleterious" effects of real-world filters, it doesn't seem to matter, because as Meridian writes:
Early digital recordings suffer from harshness too, due to primitive filters. After painstaking research, Meridian developed a special filter [...] cleaning up the effects of early filters.
(Let's ignore the fact that this is partly false, because aliasing is virtually impossible to remove from any sampled piece of inadequately filtered music. In other words if the A/D conversion is crap, you cannot un-crap it.)
Also see #539 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882067) for a demonstration.
There is a lot of vagueness in the statement "Early digital recordings suffer from harshness too..." What is an Early Digital Recording? Do they all sound the same? Is something created with the early one-off recorders made in several Japanese labs? Is it the first 3M digital mastering system? Is it a PCM-1 or a PCM-F1? Is it the first or second generation Soundstream recorder?
Despite all the Golden-Ears wailing about Bop 'Till You Drop, there are many who think it still sounds great. I don't recall anybody complaining too hard about the Telarc Soundstream based recordings whether distributed on digital or vinyl. So, where's the beef?
On balance digital mastering and production bypassed the slings and arrows of analog tape self-erasure, azimuth and saturation losses, and they can be audible.
There is a lot of vagueness in the statement "Early digital recordings suffer from harshness too..."
I have found more. In other places they talk specifically about "conventional 44.1/48 kHz sampled CDs and DVDs".
And:
This ‘apodising’ filter is so effective that it can correct problems further up the chain – even including fixing faults in the original recording.
What does their special filter look like? In SoX check allow aliasing, increase the bandwidth and set to min phase. That's it.
The apodizing filter starts to roll off at a lower frequency, and more gently, than a typical brick wall linear phase filter. It is asymmetric in the time domain, with minimal/no pre-ringing. It's not linear phase (though it's linear up to about 18kHz).
It will remove ~22kHz ringing in recordings. It will also reduce/remove aliasing above 20kHz (as well as real content of course).
If you count these things as faults, and have a recording that contains them, then it's true to say that this filter removes faults from the original recording.
There might be some combination of listener, recording and speaker where the slightly earlier attenuation of ultra high frequencies is audible.
Cheers,
David.
In the most characteristic "subjective audiophile" fashion I am having a strong audio gut feeling that you purpously misquoted Meyer and Moran. They mean exactly the oposite of what you are saying since 2007.
...
That is a very illogical affirmation. It should have said it sounded diferent (not better). it s that from a scientific paper??!!
...
Dynamic range compression does not depend on audio format. It is a choice made by the recording technician. Didn't Meyer and Moran know that?! I don't think they said that. I don't think you can be trusted as a source of information. Can you show some evidence?
Meyer and Moran 2007, presented evidence that a simple downconversion of 'hi rez' commercial releases to Redbook rates was not detectable under blind, level-matched conditions by a variety of listeners using several setups (including a professional recording room, and including gear and discs chosen by the listeners to be revealing). The paper ends with a 'Note on High Resolution Recordings' that explores why such recordings are reported to 'sound better' compared to CDs (a report they affirm). Here is that section
Please read it
in full and ponder the implications of their thesis:
A Note On High Resolution Recordings
Though our tests failed to substantiate the claimed advantages
of high-resolution encoding for two-channel audio,
one trend became obvious very quickly and held up
throughout our testing: virtually all of the SACD and
DVD-A recordings sounded better than most CDs—
sometimes much better. Had we not “degraded” the sound
to CD quality and blind-tested for audible differences, we
would have been tempted to ascribe this sonic superiority
to the recording processes used to make them.
Plausible reasons for the remarkable sound quality of
these recordings emerged in discussions with some of the
engineers currently working on such projects. This portion
of the business is a niche market in which the end users are
preselected, both for their aural acuity and for their willingness
to buy expensive equipment, set it up correctly,
and listen carefully in a low-noise environment.
Partly because these recordings have not captured a
large portion of the consumer market for music, engineers
and producers are being given the freedom to produce
recordings that sound as good as they can make them,
without having to compress or equalize the signal to suit
lesser systems and casual listening conditions. These recordings
seem to have been made with great care and
manifest affection, by engineers trying to please themselves
and their peers. They sound like it, label after label.
High-resolution audio discs do not have the overwhelming
majority of the program material crammed into the top 20
(or even 10) dB of the available dynamic range, as so
many CDs today do.
Our test results indicate that all of these recordings
could be released on conventional CDs with no audible
difference. They would not, however, find such a reliable
conduit to the homes of those with the systems and listening
habits to appreciate them. The secret, for two-channel
recordings at least, seems to lie not in the high-bit recording
but in the high-bit market.
The fact is that the market for 'hi rez' in and of itself -- as for all 'audiophile' niches -- is tiny relative to all audio content sales.
That's right. So any argument that this is all about making a lot of money is without merit.
And sales arent always about the 'resolution'. I've bought quite a few 'high rez' releases myself in the past few years....because they included 5.1 mixes and bonus tracks and 'flat transfers' of original masters, which are what I *actually* cared about and *obviously* make a difference. I had to buy the 'high rez' edition to get them.
So for those reasons you don't want to bad mouth these new distribution channels. If you enjoy non-copy protected multi-channel content, your only path is to get it from the same outfits that sell stereo high resolution. To get more variety of such content, you need to be supportive of what they are doing, not spreading so much negative vibe. Be positive. Be constructive. Let the market develop. Get what you need out of it, i.e. multi-channel releases without copy protection, and others can get that or stereo. More choices is always good for consumers.
Even so, if that changes -- if hi rez becomes the norm, if the industry somehow decides to replace all the most popular forms of profitable audio delivery content (currently CD and lossy compressed downloads and streaming) with 'hi rez' as the default ...
Not going to happen in a million years. What is going to happen is CD will get squeezed out. It is a physical format and requires inventory. It lacks convenience and the mass market could not care less about its incremental fidelity. And the people who care about fidelity will accept with open arms digital downloads at > CD fidelity or higher number of channels as you prefer. CD becomes a no-man land with high cost to keep in market.
Unless they commit at the same time to 'old school' mastering, it won't matter to those who actually understand the differences.
How would you "understand" the difference if all you have in your hand is a CD and compressed MP3/AAC? The only way you would know there is or is not something better is to get the upstream stereo mix. One of two things will then happen:
1. It is the same master in which case it may or may not demonstrate a small incremental fidelity.
2. It is a better master in which case the merit is obvious and something we all violently agree we want.
By making a case against high resolution you are throwing the baby with the bath water. You are biting your nose despite your face. You are forestalling option #2.
The average consumer might 'believe' they got 'better sound' that 'even your wife could hear',
Never seen any such "average consumer." Average consumer thinks all of us are crazy to talk about this stuff and has no understand or care about any of this. They are listening to compressed music and it is all they want. Convenience and adequate fidelity. Their needs are being met by the industry.
The problem for us, the audiophiles, is that after nearly 40 year introduction of the CD, and for the first time ever, the fidelity bar is going backward. Once CD gets diminished, that would be the day of mourning for anyone caring about fidelity. We allowed technological advancement to take us backward from what we want as audiophiles.
The high resolution/multi-channel distribution is a gift to us all. It really is. As you correctly said, the market is tiny so it has no business existing. But exist it does and it is growing due to fast conversion from physical media to media servers among audiophiles. Audiophiles like myself enjoy the convenience of digital distribution and for the right content, I don't mind the premium price.
Tap dancing CD Fearmongers need to STOP saying we need hi rez to get good sound. That's a lie. We don't. We need good recording and mastering, and good listening setups
As I explained above, your assumptions about this market are totally wrong. Music business has always been about combination of technology and business. You must know and appreciate both components. Average Joe caring about high res? Did you really think that? Do you really think you will get your multi-channel releases if you convinced people there is no value to high resolution music?
Lack of experience in the field is what is steering you wrong Steven. You can't just operate from the point of view of what you have read on argumentative forums. You will continue to lack a real feel for the levers that drive music market and consumer acceptance.
Your arguments have all been rendered moot. They have. We are getting high resolution releases and that is that. No amount of jumping up and down is going to change things. The barrier to adoption is far lower than it was for physical media in the form of DVD-A/SACD. It costs next to nothing to have digital bits spinning in the cloud for whoever wants to come and consume it. And consume they are. In 2007, that battle was fought in stereo magazines and enthusiast circles. This is not 2007. Consumers can share knowledge at speed of light and should a high res release come out with better mastering it will be snapped up and snapped fast. No one will come and check to see what you,xnor, arny, etc. think. You have been rendered irrelevant I am afraid.
How would you "understand" the difference if all you have in your hand is a CD and compressed MP3/AAC?
The mention of MP3 and AAC is just another off-topic red herring from a seemingly habitual creator of them - obviously the work of a very desperate poster who knows that he has long ago lost the argument for lack of relevant facts and reason.
The only relevant discussion relates to the CD.
It was simple science.
(1) I created a suite of recordings of music and high bandwidth natural sounds @ 24/96 using microphones with 40+ KHz bandpass.
(2) Using myself and friends as guinea pigs I ABXed them with downsampled/upsampled versions of themselves.
None of us could hear a difference. This was > 14 years ago when my hearing was much more sensitive to subtle differences.
The only way you would know there is or is not something better is to get the upstream stereo mix.
Which is/was not exactly rocket science. The name of the group of people who did this was Legion. The results were pretty consistent. Nobody heard nuttin'
One of two things will then happen:
1. It is the same master in which case it may or may not demonstrate a small incremental fidelity.
Easy enough - been there, done that.
2. It is a better master in which case the merit is obvious and something we all violently agree we want.
The implicit false claim here is that the only way to obtain a better master is to up the sample rate and/or word length. That has long been disproven by means practical examples.
The core weakness is that reliable unbiased information about better sound is like hen's teeth. As I keep pointing out even such information that was published in the JAES as part of a peer reviewed article seems like just anecdotes.
By making a case against high resolution you are throwing the baby with the bath water.
The problem with this false claim is that after a long and diligent search, there was no baby.
"
Our test results indicate that all of these recordings
could be released on conventional CDs with no audible
difference. They would not, however, find such a reliable
conduit to the homes of those with the systems and listening
habits to appreciate them. The secret, for two-channel
recordings at least, seems to lie not in the high-bit recording
but in the high-bit market.
"
So what do the second and third sentences mean?
That audiophiles are less likely to purchase a well mastered recording if only available on a CD or CD format (44/16) download?
Is the CD well that poisoned by unnecessarily bad product?
I think Amir you're a little guilty of wishful thinking.
Audiophile record labels create great sounding hi-res releases. They create great sounding CDs too.
Meanwhile mainstream-ish content, when it makes it out in hi-res, is a real mixed bag. I don't have a feeling for the proportion of good vs disappointing releases out there, but of the few I've tried, most have been disappointing. In a few cases a great SACD from an audiophile company has eclipsed an old and ropey mainstream CD. Without SACD, the same company would have put out an audiophile CD.
I'll happily give credit where a new master, released in hi-res, beats all available CD issues. I just can't join in with this fiction that the improvement in sound quality is due to the number of bits.
As for the price...
http://www.amazon.co.uk/After-Midnight-Nat...e/dp/B004C6QCXM (http://www.amazon.co.uk/After-Midnight-Nat-King-Cole/dp/B004C6QCXM)
Cheers,
David.
I don't know all the details about the Meridian speakers, but the fact that the Meridian supplied frequency response curve only goes up to about 40 KHz suggests that they may be based on 96 KHz sampling.
Meridian's website specifies for the speaker model DSP7200SE input:
1x balanced digital (AES/EBU) input on RJ45; 32kHz–96kHz sampling rates at up to 24 bit; MHR support. RJ45 input also carries RS232 and trigger.
[/size]
The main reason for my question was that where ringing artifacts of a >20kHz filter are expected to be above the audible range, filters in a speaker can be expected to be in the audible range. We don't know how steep these filters are and if they might influence the test of filters above 20 kHz.
The speaker's pulse response in Fig.1 (http://www.hydrogenaud.io/forums/index.php?act=attach&type=post&id=8072) doesn't show any pre-echo. Should that be sufficient evidence for a pre-ringing-free speaker ? The amplitude scale isn't very precise.
In general my objection to hi-res tests is that the higher format is automatically assumed to be the better quality. I would prefer a suitable acoustical source as reference, although reproducibility might be difficult in practice. Especially transducers (microphones and speakers) introduce noise and distortion, so eliminating them from the reference sound seems like a good idea.
I think Amir you're a little guilty of wishful thinking.
No. I know David Chesky, the founder of HD Tracks. I know Bruce Brown. He remasters a ton of high-resolution content for release in outlets like HD Tracks. And I have a decade of experience working with music labels, getting their buy-in, working with digital music distributors, etc. I go to shows and spend a lot of time talking to manufacturers of equipment. And do the same when they come to visit us at Madrona. So while anyone's crystal ball may be faulty, I am not talking about wishes. What I said is the consensus of the industry as of this moment.
Here is a recent reference showing the same:
Subjective Evaluation of High Resolution Recordings in PCM and DSD Audio Formats
Authors: Marui, Atsushi; Kamekawa, Toru; Endo, Kazuhiko; Sato, Erisa
Affiliations: Tokyo University of the Arts, Adachi-ku, Tokyo, Japan; TEAC Corporation, Tokyo, Japan
AES Convention:136 (April 2014)This is how the paper starts:
While the music industry actively releasing perceptually
coded (lossy) versions for almost all new music releases,
high-resolution audio production and consumption
also is increasing attraction supported by
broader bandwidth of the Internet realizing the music
distribution over the Internet and production of
relatively affordable high-resolution capable sound
recorders from several manufacturers. This is the trend in the industry and there is no denying it. High resolution audio distribution is becoming more popular and thank heavens for that .
Audiophile record labels create great sounding hi-res releases. They create great sounding CDs too.
They do. Doesn't change what I said about the CD as a physical format being squeezed out.
Meanwhile mainstream-ish content, when it makes it out in hi-res, is a real mixed bag. I don't have a feeling for the proportion of good vs disappointing releases out there, but of the few I've tried, most have been disappointing. In a few cases a great SACD from an audiophile company has eclipsed an old and ropey mainstream CD. Without SACD, the same company would have put out an audiophile CD.
Not in the future. Small audiophile companies have a severe need for move to digital distribution. Their volumes are low and producing physical goods very onerous. This is why David Chesky, one of those boutique labels was the one to spearhead HD Tracks. Their future most definitely is online distribution and at better than CD specs.
I'll happily give credit where a new master, released in hi-res, beats all available CD issues. I just can't join in with this fiction that the improvement in sound quality is due to the number of bits.
Then don’t. My argument is the former. That is why I say all of this technical discussion, listening test, etc. are moot. We have a path to get better masters. We better support that as to have a lifeline for high quality releases in the future, lest we want to get stuck with MP3/AAC. As I mentioned, I routinely run into music that I want to buy that has gone directly to MP3/AAC and is not available on CD. IF we don’t help grow the high-fidelity distribution of content, that will be our future. It is not a future I like.
As for the price...
http://www.amazon.co.uk/After-Midnight-Nat...e/dp/B004C6QCXM (http://www.amazon.co.uk/After-Midnight-Nat-King-Cole/dp/B004C6QCXM)
I don’t get your point. The SACD is £50.16 and the CD, £43.78. So neither is cheap for this special edition. HD tracks doesn’t have that title but it sells another in high res for $40 US or £25. So there, it is cheaper than the CD.
Then don’t. My argument is the former. That is why I say all of this technical discussion, listening test, etc. are moot.
So the 25 pages of mostly nonsense you posted here are all moot, just as the logs you throw around.
Great music doesn't come from the format.
Btw, you don't even need to produce a log for imp_urhp.wav (http://www.hydrogenaud.io/forums/index.php?showtopic=107588). Just listen for 4 seconds and tell us what you hear in your system.
If you can remember, you even
specifically asked me to create such a file. What do you fear, losing 4 seconds of your life while wasting thousands of seconds on these replies?
Let's see, argument ad populum, red herring, fear mongering.....
Bieber being released straight to MP3, oh the horror, the horror.
Now, about this BS paper....
cheers,
AJ
That is why I say all of this technical discussion, listening test, etc. are moot. We have a path to get better masters.
Good sounding recordings are the products of human skill and initiative. Now that good ADCs, microphones and other production equipment has become so highly affordable, human skill has become the universal weakest link.
Trouble is, you can't package up human skill and initiative into a bunch of numbers.
The great loss to the world of audio related to the belief in magic sample rates and word length numbers is that it distracts from the most relevant issues which are human skill and initiative.
2. It is a better master in which case the merit is obvious and something we all violently agree we want.
The implicit false claim here is that the only way to obtain a better master is to up the sample rate and/or word length. That has long been disproven by means practical examples.
Again, I couldn't care less about offering high res delivery, I'll take it, or standard, or even lossy. I have heard *every one of them* produce results that sound excellent to me. ( I *prefer* a lossless version so that I have more choice in making my own lossy versions. ) Those factors are NOT where the current problems with home audio lie.
I *do* care about the *deliberate* confusion of cause and effect that high rez cheerleaders are fomenting, for the same reason I care about any consumer bamboozling.
Even if they hew to the line 'buy this new high rez version because the mastering is better', the High Rez marketers are obscuring the truth.
The *transparent* line would be
that audiophile sound quality you are noticing right away isn't *due to* the high rez format, it's *due to* the mastering Since the informed consumer might then ask some embarrassing questions, which the Amirs of the industry would consider a *dangerous* line of inquiry, we won't see that line. It's too honest.
Let's see, argument ad populum, red herring, fear mongering.....
Bieber being released straight to MP3, oh the horror, the horror.
Now, about this BS paper....
The flood of off-topic posts should be interpreted as about as close to an admission of defeat as you'll get from a certain person.
Off-topic post = troll
"
Our test results indicate that all of these recordings
could be released on conventional CDs with no audible
difference. They would not, however, find such a reliable
conduit to the homes of those with the systems and listening
habits to appreciate them. The secret, for two-channel
recordings at least, seems to lie not in the high-bit recording
but in the high-bit market.
"
So what do the second and third sentences mean?
That audiophiles are less likely to purchase a well mastered recording if only available on a CD or CD format (44/16) download?
Is the CD well that poisoned by unnecessarily bad product?
I think those are perhaps the most fascinating lines. They echo Amir's hand-wringing, but from a different perspective and with a different thrust , coming after the paragraphs before them, where M&M report anecdotes from audio engineers indicating that the industry has purposely created a two-tiered mastering system. This, in turn, fosters a 'high bit market'. IOW, we get better mastering because of a market that has been persuaded that the improvement comes from the *format*.
(M&M seem unaware of objective evidence for this in 'hybrid' high rez releases like the
Dark Side of the Moon) SACD)
The only relevant discussion relates to the CD.
It was simple science.
(1) I created a suite of recordings of music and high bandwidth natural sounds @ 24/96 using microphones with 40+ KHz bandpass.
(2) Using myself and friends as guinea pigs I ABXed them with downsampled/upsampled versions of themselves.
None of us could hear a difference. This was > 14 years ago when my hearing was much more sensitive to subtle differences.
Hi Arny. Perhaps we should fill out more of this story as you told it on AVS.
It went like this:
Fortunately, the above speculations are not true. With complex music signals the limit of audiblity is more like 16 KHz due to masking, and that is more than plenty far enough away from 44.1 KHz with modern digital filters.
Ah yes. I recall you saying this before:
I've done experiements with sliding brick wall filters down to lower and lower frequencies. Usually, they start being barely audible around 16 KHz.
Can you please outline those tests? Were they double blind or sighted? And was the person performing the tests had his hearing tested to make sure they could hear above 16 Khz?
And a reference to masking of everything above 16 KHz would also make for great bedtime reading for me .
You kindly replied with this:
I did what I said in the previous post. I made recordings of live musicans in an exceedingly quiet and non-reverberent room @24/96 using 1/4" measurement mics that had strong content > 20 KHz, even 30 KHz.
I set up an ABX between the 24/96 files with a 16 KHz brickwall filter, and with full bandpass. I used speakers and amps with strong response > 30 KHz and put the listeners on axis of their supertweeters.
Were they double blind or sighted?
both
The training sequence was files brick walled at lower frequencies such as 9 KHz, and working up in logical steps.
And was the person performing the tests had his hearing tested to make sure they could hear above 16 Khz?
Yes.
So this testing was done at 32 Khz sampling. As you know I and a number of other people passed this test. So clearly you hearing was not that sensitive even then .
Kind of like the ballparks that were selling small and large beers but the amount in each was about the same.
The only relevant discussion relates to the CD.
It was simple science.
(1) I created a suite of recordings of music and high bandwidth natural sounds @ 24/96 using microphones with 40+ KHz bandpass.
(2) Using myself and friends as guinea pigs I ABXed them with downsampled/upsampled versions of themselves.
None of us could hear a difference. This was > 14 years ago when my hearing was much more sensitive to subtle differences.
Hi Arny. Perhaps we should fill out more of this story as you told it on AVS.
It went like this:
Fortunately, the above speculations are not true. With complex music signals the limit of audiblity is more like 16 KHz due to masking, and that is more than plenty far enough away from 44.1 KHz with modern digital filters.
Ah yes. I recall you saying this before:
I've done experiements with sliding brick wall filters down to lower and lower frequencies. Usually, they start being barely audible around 16 KHz.
Can you please outline those tests? Were they double blind or sighted? And was the person performing the tests had his hearing tested to make sure they could hear above 16 Khz?
And a reference to masking of everything above 16 KHz would also make for great bedtime reading for me .
You kindly replied with this:
I did what I said in the previous post. I made recordings of live musicans in an exceedingly quiet and non-reverberent room @24/96 using 1/4" measurement mics that had strong content > 20 KHz, even 30 KHz.
I set up an ABX between the 24/96 files with a 16 KHz brickwall filter, and with full bandpass. I used speakers and amps with strong response > 30 KHz and put the listeners on axis of their supertweeters.
Were they double blind or sighted?
both
The training sequence was files brick walled at lower frequencies such as 9 KHz, and working up in logical steps.
And was the person performing the tests had his hearing tested to make sure they could hear above 16 Khz?
Yes.
So this testing was done at 32 Khz sampling.
I'm talking about different tests that were as I said, they involved 44.1 KHz sampling.
These other 32 KHz tests were also done.
As you know I and a number of other people passed this test. So clearly you hearing was not that sensitive even then .
I know no such thing. Neither I nor any of my designated representatives were present at the AVS tests you seem to be describing, so I don't know for sure if the tests were performed properly. I was present at the tests I described so I have some confidence about them.
The other implication that I was referring to in the final section of M&M 2007 is that as a format becomes more popular, the mastering for it gets worse.
We *already* see 'loudness' mastering in 'high rez' releases...it's been evident for years now (http://www.avsforum.com/forum/112-surround-music-formats/1009384-hi-rez-dvd-sometimes.html). *
On 'audiophile' forums I routinely see high rez releases (*and downloads* ) critiqued for the sorts of 'bad' mastering moves long the bane of CD remasters (e.g. , extreme compression, 'smiley face' EQ, resulting in lack of 'crankability').
This sort of mastering happens because, over and over, since at least the 1960s, someone, somewhere in the production chain, decides that these mastering moves 'sound better ' -- e.g., to focus groups comparing short samples of audio that are not level-matched. There's solid psychoacoustic foundation for *that* being a predictable outcome.
So, if the industry pushes high rez downloads, what's to prevent that mindset from proliferating more? Won't that bring us back to where we started? Popular music genres will be given 'modern' mastering, classical and jazz, left pretty much alone?
This is what happens when the consumer is led to believe that a 'format' necessarily gives you great sound.
The focus of the consumer (especially the 'audiophiles' to whom this shilling is primarily aimed) should not be on 'high rez' but on MASTERING CHOICES.
*Amusingly , Steely Dan's Gaucho is one of the hi rez releases on M&M's sample list. In terms of dynamic range compression, does this look way better than CD (http://www.avsforum.com/forum/112-surround-music-formats/1009384-hi-rez-dvd-sometimes-2.html#post13464629)? Contrary to Tap Dancers who claim that their opponents have no criticisms for M&M, I'd say (and have said) that M&M were too naively accepting of the generalization that high rez releases have superior mastering. (That said, Gaucho sounds pretty 'audiophile' in any incarnation. The tale of its production is hilarious as a fable of late-1970s coke-fueled obsessiveness)
I know no such thing. Neither I nor any of my designated representatives were present at the AVS tests you seem to be describing, so I don't know for sure if the tests were performed properly. I was present at the tests I described so I have some confidence about them.
Well, neither I nor any of my designated representatives were present at the tests you are describing so I don't know for sure if the tests were performed properly! But I was present at my own tests and I have a lot of confidence about them.
You also said this on AVS:
When I served in the US Army (drafted) in the 1960s they had no clue about hearing protection. I qualified with 3 different firearms, worked on firing ranges, and worked routinely for about 30 months in a very noisy environment. While any damage that may have related to those experiences did not seem to hurt my hearing acuity that much when I was younger, these days things are far worse.
I now struggle to hear the effects of an 8 KHz brick wall filter at normal listening levels. So while Floyd Tooles comments may relate to Amir and you John, some of us did not have such protected lives, no fault of our own.
14 years ago would have been year 2000, some 40 years after the firearm damage to your ears. So surely your hearing was shot by the time you performed the 44 Khz tests, pun intended . I hope you agree that given what you have told us, your listening test results are not reliable measure of what anyone else would hear.
Arny's tests may indeed not be indicative of what others may hear but by the same token, neither are yours especially since you've been so coy about disclosing what you heard.
...
So, if the industry pushes high rez downloads, what's to prevent that mindset from proliferating more? Won't that bring us back to where we started? Popular music genres will be given 'modern' mastering, classical and jazz, left pretty much alone?
...
The focus of the consumer (especially the 'audiophiles' to whom this shilling is primarily aimed) should not be on 'high rez' but on MASTERING CHOICES.
...
Something I said to Amir a month ago:
"I share your hope for better sound, but also your fear that someone will screw it up. Apple already want higher than CD resolution as input to produce AAC files. Record labels want those tracks to stand out. How long will it be before CD-style compression/limiting of high resolution files becomes common? "
Arny's tests may indeed not be indicative of what others may hear but by the same token, neither are yours especially since you've been so coy about disclosing what you heard.
My answer to you has been to try the same test. But you don't. Yet keep asking me to do more. Here is a person who is one of detractors on AVS as you know, and kept ridiculing the idea of anyone passing these tests, including posting a number of failed ABX tests. He listened to me and gave it a serious try and reported as such: http://www.avsforum.com/forum/91-audio-the...ml#post25871786 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1532092-debate-thread-scott-s-hi-res-audio-test-71.html#post25871786)
Laptop? Practice?
Well, I decided to give my laptop a try since Amir did so well using his. Lo and behold, I had little difficulty with the 16/32 key jangling test. Not quite perfect, but I suspect a bit more practice would get me up to perfect.
My laptop is a Sony Vaio PCG-41412L with the HD upgraded to a SSD. All audio enhancements are off. I used a pair of Sony MDR-1R headphones.
The results speak for themselves; I found a critical segment that revealed an audible difference. I've had some practice, which helped—just as Amir suggested. Now, I can pass an ABX test I previously failed. I'll tackle the 16/44 test next. Oh, and it was a piece of cake to pick out the differences in the 16/16 and 22/16 tests.
foo_abx 1.3.4 report
foobar2000 v1.3.3
2014/07/19 11:26:49
File A: C:\Users\mark_000\Downloads\keys jangling band resolution limited 3216 2496.wav
File B: C:\Users\mark_000\Downloads\keys jangling full band 2496.wav
11:26:49 : Test started.
11:27:29 : 00/01 100.0%
11:28:58 : 00/02 100.0%
11:29:46 : 00/03 100.0%
11:29:59 : 01/04 93.8%
11:30:06 : 01/05 96.9%
11:30:16 : 02/06 89.1%
11:30:26 : 03/07 77.3%
11:30:34 : 04/08 63.7%
11:30:45 : 05/09 50.0%
11:31:00 : 06/10 37.7%
11:31:10 : 07/11 27.4%
11:31:29 : 08/12 19.4%
11:31:41 : 09/13 13.3%
11:32:05 : 10/14 9.0%
11:32:20 : 10/15 15.1%
11:32:30 : 11/16 10.5%
11:32:41 : 12/17 7.2%
11:32:52 : 13/18 4.8%
11:33:07 : 13/19 8.4%
11:33:16 : 14/20 5.8%
11:33:28 : 15/21 3.9%
11:33:40 : 16/22 2.6%
11:33:58 : 17/23 1.7%
11:34:12 : 18/24 1.1%
11:34:25 : Test finished.
----------
Total: 18/24 (1.1%)
Give the above test a try. It is 32 Khz sampling so there should be no doubt whatsoever that we have truncated useful bits. Hopefully you hear the difference and the material acts like training. Run the higher sample rate ones and listen to the exact same segment. If you can't hear the difference go back and re-listen to the 32 Khz. If nothing else, you will get a first hand feel for what we are talking about.
If you recall, I also shared with you the detailed subjective remarks from Stuart's paper which is the subject of this thread. What came out of that? Nothing, right? So this is yet another request to be ignored or written off as cheating, etc. This is not a constructive discussion as to have proper give and take. It is all take and complaining because the answer is not what you want to hear. And that will continue to be the case no matter what you ask me. I aspire for excellence and as long as folks are not, they are not going to hear from me what they want to hear.
...
So, if the industry pushes high rez downloads, what's to prevent that mindset from proliferating more? Won't that bring us back to where we started? Popular music genres will be given 'modern' mastering, classical and jazz, left pretty much alone?
...
The focus of the consumer (especially the 'audiophiles' to whom this shilling is primarily aimed) should not be on 'high rez' but on MASTERING CHOICES.
...
Something I said to Amir a month ago:
"I share your hope for better sound, but also your fear that someone will screw it up. Apple already want higher than CD resolution as input to produce AAC files. Record labels want those tracks to stand out. How long will it be before CD-style compression/limiting of high resolution files becomes common? "
Fear is the first letter in FUD . There is no reason to speculate negatively here. Become a customer and advocate of better mastering and we will get in this narrow distribution channel (stupid to even talk about it for mass market releases).
I keep repeating myself: this is not 2007. This is not the same old business. The world has changed around us. See this from many such examples, for one of the award winning mastering engineers, Doug Sax: http://mixonline.com/news/profiles/masteri...h.cHY5BQ9r.dpuf (http://mixonline.com/news/profiles/mastering-cutting-edge-quality-age-low-res-low-budgets-and-multiple-formats/366464#sthash.cHY5BQ9r.dpuf)
When he was invited to sit on Ludwig’s Platinum Mastering Panel this year, Doug Sax suggested they discuss mastering for multiple formats. “Ten years ago, you made a CD master and you were done,” Sax says. “Now you do a CD master, a master for iTunes, a high-resolution master for HDtracks, and a cutting master for vinyl.” See? Separate mastering. You think he is going to do the same thing he does for iTunes? or CD? He doesn't or he wouldn't have specifically named that as another branch.
You guys need to put aside negatively and learn the new world order. Things are moving in the right direction but only if you believe in best fidelity, not good enough.
So, if the industry pushes high rez downloads, what's to prevent that mindset from proliferating more? Won't that bring us back to where we started? Popular music genres will be given 'modern' mastering, classical and jazz, left pretty much alone?
This is what happens when the consumer is led to believe that a 'format' necessarily gives you great sound.
This could indeed result in a huge fail for the hi-res peddlers.
By suggesting to the masses that it is the
format that makes the music sound better
instead of the recording, mixing and mastering "choices", the people responsible for the horrible sound quality would get a free pass for what, another decade, two, until ultra-hi-res comes along?
The rationale behind this is that by making hi-res a commodity just like the CD, the uninformed average listener will just happily buy a crappy hi-res product as he/she is doing today with crappy CDs. CD audio would become obsolete, hi-res crap the norm rather than the exception.
I can already imagine amirm in 10 years, still wasting Internet traffic, but this time about ultra-hi-res, and how bigger is always better (except if it doesn't fit his momentary argument).
When he was invited to sit on Ludwig’s Platinum Mastering Panel this year, Doug Sax suggested they discuss mastering for multiple formats. “Ten years ago, you made a CD master and you were done,” Sax says. “Now you do a CD master, a master for iTunes, a high-resolution master for HDtracks, and a cutting master for vinyl.” [/color][/i]
See? Separate mastering...
Since it is already stablished that there is no audible difference between the hi rez formats and CD format, and vinyl is obsolete. It looks like their intention is to scam the final user (consumer).
When he was invited to sit on Ludwig’s Platinum Mastering Panel this year, Doug Sax suggested they discuss mastering for multiple formats. “Ten years ago, you made a CD master and you were done,” Sax says. “Now you do a CD master, a master for iTunes, a high-resolution master for HDtracks, and a cutting master for vinyl.” [/color][/i]
See? Separate mastering...
Since it is already stablished that there is no audible difference between the hi rez formats and CD format, and vinyl is obsolete. It looks like their intention is to scam the final user (consumer).
You must have no understanding of mastering to say that.
In the most characteristic "subjective audiophile" fashion I am having a strong audio gut feeling that you purpously misquoted Meyer and Moran. They mean exactly the oposite of what you are saying since 2007.
I am not a subjectivists nor are you an objectivists if you don't even read the published literature on this topic.
You said there should be a warning that says anyone who buys a high resolution title is delusional. I showed you evidence from prosecution's witness that high resolution releases, you know the real product, absolutely sound superior to their CD releases due to better mastering. As such you blanket assertion that hearing better fidelity is imaginary is wrong.
...of the SACD and DVD-A recordings sounded better than most CDs...
That is a very illogical affirmation. It should have said it sounded diferent (not better). it s that from a scientific paper??!!
No, it is not a scientific paper. It is an engineering report published in the Journal of Audio Engineering Society.
Dynamic range compression does not depend on audio format.
Technically no, but from business point of view, it absolutely does. The labels and talent consider CD and MP3/AAC formats as mass market so impose their wants in the form of loudness compression. There is no such business rule for high resolution releases.
It is a choice made by the recording technician. Didn't Meyer and Moran know that?! I don't think they said that. I don't think you can be trusted as a source of information. Can you show some evidence?
Who the heck is a "recording technician?"
The only myopic views are coming from your posts. If a difference can't be heard in a properly set double blind test (to get rid of bias) than that difference is irrelevant. I think you know that but you choose to believe otherwise.
You are in a thread where such a difference was heard in double blind tests and reported in award winning peer reviewed paper in Audio Engineering Society. Please do a bit of education before randomizing the thread.
And that was not even the point of my reply to you. It was the fact that by getting the upstream stereo master, it could be created to a different set of business rules than mass market releases.
Amir, I've already explained to you why I haven't done the tests and have never stated that you cheated. It may well be that due to various issues, be they time offsets or level differences, that the test can be considered cheatable but that is a far cry from accusingly you. A full disclosure, and really something more than what you posted on WBF, would be illuminating and could help to advance discussions in that area.
You must have no understanding of mastering to say that.
Yeah because
less extra compression, less extra EQ, ... in short less messing up the sound quality deliberately, that does
cost extra $$$! Audiophile logic. (I guess that's why audiophiles are the ones willing to pay for the extra cost.)
Technically no, but from business point of view, it absolutely does. The labels and talent consider CD and MP3/AAC formats as mass market so impose their wants in the form of loudness compression. There is no such business rule for high resolution releases.
But you want to make CD obsolete, right? It will happen eventually, and depending on the success of the hi-res peddler probably faster than we think.
So where is your niche market, where people have to pay more money for less destruction, when hi-res is the "mass market"?
You are in a thread where such a difference was heard in double blind tests and reported in award winning peer reviewed paper in Audio Engineering Society. Please do a bit of education before randomizing the thread.
No, you are in a thread where in the best condition the score was very close to the worst condition, and that was being better than about 56% correct.
Oh and "randomizing the thread"? See amir credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
A full disclosure, and really something more than what you posted on WBF, would be illuminating and could help to advance discussions in that area.
He's not interested in being honest (see amir credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029)).
He's not even willing to listen to this 4 second short test file: imp_urhp.wav (http://www.hydrogenaud.io/forums/index.php?showtopic=107588), and tell us what he hears in his system. Well I think he already did download and listen, but he's evading. Cognitive dissonance at its best.
amir, stop moving the goalposts. This discussion is (in part) about technical benefits of "hi-res" formats, not about de-facto benefits of "hi-res" releases due to bad mastering practices for CD releases. In any case, an easy, sensible, scientific and honest way to study the benefits of "hi-res" formats is the experiment I (and many others before me) described earlier (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882271).
I also urge you to study what "scientific" means (http://www.arachnoid.com/what_is_science/index.html).
If you keep posting the same nonsense over and over this discussion will close.
I am not a subjectivist
Reinventing the Audio Power Amplifier: Mark Levinson No 53
By Amir Majidimehr
But How Does it Sound?
In comparison testing I have done, switching amplifiers using the classic class D configuration always sport incredible low frequency control and power. They beat out linear class AB amplifiers almost regardless of price. What they give up though is high frequency fidelity which I find somewhat harsh. The distortion is highly non-linear and challenging to spot but it is there. The Mark Levinson No 53 is the first switching amplifier I have heard which does not have this compromise. Its bass is amazingly authoritative: tight and powerful. Yet the rest of the response is absolutely neutral and pleasant.
If you have not heard these unique amplifiers, I highly encourage you to come into our showroom for a listen. We have a pair on hand driving our Revel speakers. I am confident that they will improve the sound of your current speakers given the ease with which they can drive any load regardless of how difficult they might be (and many high-end speakers are difficult to drive). We are happy to let you evaluate them with your own system to see the benefits of this technology. Hearing this amplifier was an eye-opener for me. I think it will be for you too.
Would love to see you start a thread about "Audibility of Typical Filters in a Class D amp" Amir. You've mentioned TOS #8 a few times, so we know you are aware of it.
cheers,
AJ
But you want to make CD obsolete, right?
I have no power to do so. But I absolutely believe CD needs to be shot in the head and put out of its misery. In this day and age creating plastic waste, and spinning a motor to read 650 megabytes? Are you kidding?
It will happen eventually, and depending on the success of the hi-res peddler probably faster than we think.
No, hi-res has no influence on CD going away. It is the mass market that pays the bills and will cause its gradual demise.
So where is your niche market, where people have to pay more money for less destruction, when hi-res is the "mass market"?
No, there is no mass market for high res. There will be a market much larger than today. But still roundoff error in the grand scheme of things. Just like the unit volume for Porsche vs Ford.
Our role in all of this is to create a meaningful market where some segment of the content producers service us. There is already a good start so the chicken and egg problem has been solved. We just need to let it thrive.
If that causes you, Krab, AJ, etc. angst, that much the better.
When he was invited to sit on Ludwig’s Platinum Mastering Panel this year, Doug Sax suggested they discuss mastering for multiple formats. “Ten years ago, you made a CD master and you were done,” Sax says. “Now you do a CD master, a master for iTunes, a high-resolution master for HDtracks, and a cutting master for vinyl.” [/color][/i]
See? Separate mastering...
Since it is already established that there is no audible difference between the hi rez formats and CD format, and vinyl is obsolete. It looks like their intention is to scam the final user (consumer).
There is a format on the list that has the potential to add audible differences - iTunes, which I believe means AAC a well known perceptual encoder.
It is generally recognized that the transparency of modern perceptual coders may be less than perfect with certain musical sources.
It may be possible to produce a given musical work in such a way that it is encoded more accurately by perceptual coders. Seems like it could be a worthy goal for recordings that are going to widely distributed as perceptually encoded files.
BTW this post is off-topic in this thread and any further discussion of it should be moved to an appropriate thread.
amir, stop moving the goalposts. This discussion is (in part) about technical benefits of "hi-res" formats, not about de-facto benefits of "hi-res" releases due to bad mastering practices for CD releases.
I am not doing that. Others are doing so to randomize the topic. Here is one of many posts like that:
1. We "improve" the wrong thing, to the detriment of other things that would introduce a real/greater audible improvement.
Bingo. While Amirm and Stuart and Neil Young are flogging this technology, we're still getting 'loudness wars' mastering. And we are still stuck ina 2-channel paradigm. And listening in rooms that are often horrendous, acoustically.
What the hi rez cheerleading squad is doing is akin to trying to focus all attention on the font of a document that is too often contains gibberish.
The question is why...why would this be the focus, when *all of them* know where the actual 'obvious and audible and known about for years' problems lie?
I am also constantly being asked to provide my personal opinion so I do.
In any case, an easy, sensible, scientific and honest way to study the benefits of "hi-res" formats is the experiment I (and many others before me) described earlier (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882271).
Your recommendation there is very sound and one that I am 100% on board on. And to the dismay of few here, is one that Stuart recommended in his JAES paper, Coding for high resolution.
The problem with it is one of execution. There is no mandate, broad education or certification that anyone includes noise shaped dither. The default in my Adobe Audition CC for sample rate is white TPDF. Noise shaping is off by default. There are some 8 options for noise shaping. I challenge you to find any recording engineer that can explain what all of them mean.
And then there is the "marketing of it." We call dither noise. Imagine the reaction of someone producing music to "noise." They naturally have a suspicious opinion of adding dither.
Go to any pro audio forum and search for dither. YOu will find opinions all over the place. Here I did a quick search on Gearslutz where I hang out from time to time: https://www.gearslutz.com/board/mastering-f...n-any-type.html (https://www.gearslutz.com/board/mastering-forum/800645-dither-specifically-reaper-but-ill-take-conversation-any-type.html)
So, I just started using Reaper this year. It is the first time I have found myself in a situation where I can apply dither as I listen and toggle it on and off.
The dither sounds like poo.
The sound stage just gets so much smaller. So my questions are:
Is all dither obviously audible?
Does Reaper just ship with a bad dither plug?
Am I doing something wrong with a perfectly fine plug?
Is it all totally in my head and there is in fact no difference to hear because dither noise is just too low?
On a side note, this dither is going on post limiter, and then only on personal reference mixes. The final product will never see dither applied by me.
Instead of relying on all of these people understanding proper dither and its use, I say let's bypass the whole thing. Let's get the bits prior to down conversion. If you want to still have 16/44.1, you can convert it yourself with dither of your choice. There is no technological reason anymore to have these guys make random selections as to stuff the bits on CD.
I also urge you to study what "scientific" means (http://www.arachnoid.com/what_is_science/index.html).
If you keep posting the same nonsense over and over this discussion will close.
I have no emotional attachment to the thread or this forum. Feel free to close the thread any time you like.
On what is science, feel free to open another thread and we can have a good discussion. I assure you in five minutes I can show that none of you are following the first thing about that.
14 years ago would have been year 2000, some 40 years after the firearm damage to your ears. So surely your hearing was shot by the time you performed the 44 Khz tests, pun intended . I hope you agree that given what you have told us, your listening test results are not reliable measure of what anyone else would hear.
Ignores common knowledge about hearing damage:
http://www.worker-health.org/noisehearingloss.html (http://www.worker-health.org/noisehearingloss.html)
"It is important to recognize that noise-induced hearing loss is cumulative. This means that exposures to loud levels of noise off the job – gunfire, power tools, and loud music - can also contribute to a worker’s overall hearing loss."
I am fully aware that I'm responding to a poster who appears to be desperate to distract this discussion from its topic because the on-topic discussion leaves him little rational to say. However technical errors such as this one need to be addressed. Any further discussion of the topic of cumulative hearing damage should be moved to an appropriate thread.
I have no power to do so. But I absolutely believe CD needs to be shot in the head and put out of its misery. In this day and age creating plastic waste, and spinning a motor to read 650 megabytes? Are you kidding?
I was talking about CD audio, as in the format 44.1/16. That's one of the main parts the thread is about, not physical media.
No, hi-res has no influence on CD going away. It is the mass market that pays the bills and will cause its gradual demise.
No, there is no mass market for high res. There will be a market much larger than today. But still roundoff error in the grand scheme of things. Just like the unit volume for Porsche vs Ford.
What? That doesn't even make any sense.
If hi-res gains mass market traction then CD audio will eventually go away. Hi-res won't be that market where you can scam people to pay more money for less deliberate destruction of sound quality anymore, but the new mass market.
How can you fail to even understand these simple (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882377) postings (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882382)?
If that causes you, Krab, AJ, etc. angst, that much the better.
Despicable. But nothing else to expect, see amir integrity (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
imp_urhp.wav (http://www.hydrogenaud.io/forums/index.php?showtopic=107588) is still waiting.
Posts nonsense in over 27 pages, repeats fallacy after fallacy even after they are pointed out, posts multiple logical abominations ... and finally, posts a random ancedote to support another one of his off-topic journies in the same post where he finally posts this:
On what is science, feel free to open another thread and we can have a good discussion. I assure you in five minutes I can show that none of you are following the first thing about that.
Q.E.D. amir has disqualified himself from any rational or honest discourse. All people can objectively understand this by googling this thread, for a summary see amir integrity (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
Our role in all of this is to create a meaningful market where some segment of the content producers service us.
Wrong. Your (subjectivist) sides role, was to subvert the idea of fidelity to original soundfields and come up with ways to sell $50k "sound" amps and justify $50 2ch recordings because they possess "more resolution" and other such excrement.
This was long ago recognized by someone who might be considered the patriarch of the whole mess (http://www.stereophile.com/asweseeit/1107awsi/), despite him distancing himself from the idiocy at the end:
JA - Do you still feel the high-end audio industry has lost its way in the manner you described 15 years ago?
JGH - Not in the same manner; there's no hope now. Audio actually used to have a goal: perfect reproduction of the sound of real music performed in a real space. That was found difficult to achieve, and it was abandoned when most music lovers, who almost never heard anything except amplified music anyway, forgot what "the real thing" had sounded like. Today, "good" sound is whatever one likes. As Art Dudley so succinctly said [in his January 2004 "Listening," see "Letters," p.9], fidelity is irrelevant to music.
Since the only measure of sound quality is that the listener likes it, that has pretty well put an end to audio advancement, because different people rarely agree about sound quality. Abandoning the acoustical-instrument standard, and the mindless acceptance of voodoo science, were not parts of my original vision.
JA - I remember you strongly feeling back in 1992 that multichannel/surround reproduction was the only chance the industry had for getting back on course.
JGH - With fidelity in stagnation, spatiality was the only area of improvement left.
That's all this horse and pony show about 2ch "Hi-Rez" and what the crap that the BS paper peddles is all about. Squeezing more $$ out of the few idiots left, who long ago ceased to want anything remotely resembling fidelity to original soundfields.
Yes, "Higher Resolution" of Neil Youngs guitar amp studio construct. Whatever the heck that might be. I'd love to see someone identify that 24bit vs the 16/44 TPDF version under any circumstances.
Except an unsupervised online ABX Windows pc "test" of course.
cheers,
AJ
Addressing this bit of pseudo science:
1. Where on earth do you see non-linear distortions? You don't even know what an LTI system is, but talk big about filters? What the..
Down conversion involves both filtering and quantization. Quantization error is anything but a linear distortion. And likewise accuracy/quantization error in the filter math also creates non-linear distortion. Do you really think what they heard in Stuart's listening test was linear distortion? Linear distortion garners such adjectives? "Echoes, when audible, were identi ed as being a ected the most clearly by the ltering. It was felt that some of the louder passages of the recording were less aggressive after ltering, and that the inner voices (second violin and viola) had "a nasal quality". I don't think so. Not at all.
But let's suspend reality and follow your TLA buzzword. The cognitive part of the brain does not follow the rules of signal processing. Let's look at the impulse response again:
(http://amirviews.smugmug.com/photos/i-q35sw3V/0/L/i-q35sw3V-L.png)
From math/signal processing point of view we can say all of this ringing is benign. The theory says this is what must happen in time domain to implement the low pass function. But that is not how the psychoacoustics works. We hear each one of those samples as they come. The brain doesn't collect them all and then say, "ah, that was a FIR filter." It hears the individual samples as they get processed by the transfer function of the low pass filter. This is why you can't use your logic of "it is a linear function." Those sample transforms taken individually do not represent such.
To understand this topic fully, you must have broad understanding of multiple disciplines. Assuming it just about college textbook signal processing will surely cause you to fall into a ditch thinking the results created by Stuart's research is impossible. Only to see the opposite come true.
Ditto for assuming we are all sensitive to distortions the same way which was my point. We are not and that is amply proven in detection of lossy compression artifacts where expert listeners can hear distortions that masses cannot perceive at all.
3. I also already told you that I take the paper for what it is, and there are still unknowns. Your stereotyping of me, which you even admitted, prevents you from actually understanding my position. You don't even care about any of this. All this is for you is a "war" (your own word), that you apparently need to win at all cost, which sadly seems to include your sanity.
No, the war has been lost. It was lost before this thread even started. The discussion is months old. On WBF Forum the thread is 1,500 posts long and 41,000 views. On AVS it is on the order of 3,000 posts. Krab, Arny, mzil, etc. have made all of their arguments elsewhere and lost them. Specific to this discussion of my hearing ability with respect to distortions, Arny had this to say: http://www.avsforum.com/forum/91-audio-the...ml#post25344962 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1532092-debate-thread-scott-s-hi-res-audio-test-41.html#post25344962)
Amir, I reject the implication that you and I are in the same camp, audio-wise. IMO, you are a Golden Ear, pure and simple.
It is not easy to get him to volunteer such things as I am sure you know. But volunteer he had after he saw the few of us accomplish what most could not.
Amir, I reject the implication that you and I are in the same camp, audio-wise. IMO, you are a Golden Ear, pure and simple.
It is not easy to get him to volunteer such things as I am sure you know. But volunteer he had after he saw the few of us accomplish what most could not.
Only a seriously self deluded individual with overactive imagination, like for example, a "subjectivist" audiophile...could read that Arny quote in anything other than a pejorative sense.
No, the war has been lost.
No Amir, there is no "war", except within your mind of "worry" and other subjectivist audiophile issues. This is about distributed (to be) reproduced music, purportedly for musical enjoyment (which we all know isn't on the audiophile need list).
The BS paper is not an "attack" on rational, objective folks. It's an attempt by those with pecuniary interests in "Hi-Rez" to concoct some form of counter to the M&M 2007 "Hi-Rez" myth buster. It has more holes than swiss cheese, but we understand if it's all you got.
And no, 13yrs of unsupervised, unrecorded and in at least one case fabricated, test results, don't count.
I'm sure this isn't the last we are going to hear about this BS paper. So your psychogenic "war" isn't over quite yet.
cheers,
AJ
This could indeed result in a huge fail for the hi-res peddlers.
By suggesting to the masses that it is the format that makes the music sound better instead of the recording, mixing and mastering "choices", the people responsible for the horrible sound quality would get a free pass for what, another decade, two, until ultra-hi-res comes along?
The rationale behind this is that by making hi-res a commodity just like the CD, the uninformed average listener will just happily buy a crappy hi-res product as he/she is doing today with crappy CDs. CD audio would become obsolete, hi-res crap the norm rather than the exception.
I can already imagine amirm in 10 years, still wasting Internet traffic, but this time about ultra-hi-res, and how bigger is always better (except if it doesn't fit his momentary argument).
We already have over a decade of anecdotal cases where the mere label "high resolution" (sacd, dvd-a, bluray, HDtracks) is enough to send listeners into raptures. In not a few of those cases, the source was analog tapes, or SD PCM, or the mastering bears evidence of the 'loudness wars'.
We also have cases where some recognize that they've been shucked -- the mastering
doesn't live up to the high rez hype.
You get it. 2bdecided gets it. We, generally, (at HA) get it:
'It's the mastering, stupid.' The Dancing Man from Madrona *knows* it, but keeps dancing around it. He has a commercial -- and hugely ego-based -- disincentive to acknowledge 'getting it'.
Is it perhaps time to close the thread down? Is there anything more to be expected from Amir than propaganda, diversion, obfuscation, nit-picking, and Arny-baiting? Is there anyone besides him who will take up the 'cause' of 'high rez' here?
I wonder.
But you want to make CD obsolete, right?
He wants to make 'standard resolution' obsolete. Because, well, "high rez"!
Now, a true 'audiophile' would want to make crappy recording and mastering and playback conditions obsolete. They'd be pushing for mastering standards and for moving beyond the 2channel bottleneck. "Hi rez" delivery formats, which cannot *necessitate* or *guarantee* any of these things, and which provide miniscule audible improvements over SD only when listened with certain abnormal preconditions, would be a minor little side-skirmish in a much larger battle to render crap sound obsolete.
But instead.....
Addressing this bit of pseudo science:
1. Where on earth do you see non-linear distortions? You don't even know what an LTI system is, but talk big about filters? What the..
Down conversion involves both filtering and quantization. Quantization error is anything but a linear distortion.
But quantization error is a non-problem given that the down conversion is properly dithered. The means for effecting proper dither (TPDF dither of sufficient amplitude best practice being perceptually shaped PSD) has been a solved problem for everybody with a brain (except the perpetrators of the recent Meridian fiasco who used RPDF dither - oops!) for decades.
Sorry Amir, but this is just a diversion off-topic that could be legitimately discussed in another thread, but not here.
When he was invited to sit on Ludwig’s Platinum Mastering Panel this year, Doug Sax suggested they discuss mastering for multiple formats. “Ten years ago, you made a CD master and you were done,” Sax says. “Now you do a CD master, a master for iTunes, a high-resolution master for HDtracks, and a cutting master for vinyl.”
See? Separate mastering. You think he is going to do the same thing he does for iTunes? or CD? He doesn't or he wouldn't have specifically named that as another branch.
It isn't clear if Sax meant individual "mastering" or creation of delivery formats.
Mastering usually consists of two steps: a creative one where the rough audio material is analyzed and (hopefully) improved by applying EQ, dynamic range compression and any other techniques to make it sound better for the intended purpose. IME the final result is captured in a hi-res format and downconverted by using trusted tools. Ideally every non-losslessly converted format should be auditioned by a qualified engineer.
The second step is the creation of delivery formats, which is the non-creative part. Entering meta-data like PQ, ISRC, EAN, cd-text etc. This is important, time consuming but less critical work and is often done by assistant engineers.
Having done quite some mastering I can attest that multiple delivery formats require more work and consequently increase the invoice for the client.
When he was invited to sit on Ludwig’s Platinum Mastering Panel this year, Doug Sax suggested they discuss mastering for multiple formats. “Ten years ago, you made a CD master and you were done,” Sax says. “Now you do a CD master, a master for iTunes, a high-resolution master for HDtracks, and a cutting master for vinyl.”
Doug Sax endorsed Shakti Stones.
I would classify him with the cohort of mastering 'engineers' who, having learned how to make great recordings, think that everything they hear is real.
We hear each one of those samples as they come. The brain doesn't collect them all and then say, "ah, that was a FIR filter."
The purpose of the BS paper was (or at least should have been) to find out IF those samples make it at all to the brain. In theory those samples are supposed to be inaudible. That's why listening tests which indicate otherwise should be analyzed very carefully.
(http://amirviews.smugmug.com/photos/i-q35sw3V/0/L/i-q35sw3V-L.png)
From math/signal processing point of view we can say all of this ringing is benign. The theory says this is what must hear the in time domain to implement the low pass function. But that is not how the psychoacoustics works. We hear each one of those samples as they come. The brain doesn't collect them all and then say, "ah, that was a FIR filter." It hears the individual samples as they get processed by the transfer function of the low pass filter. This is why you can't use your logic of "it is a linear function." Those sample transforms taken individually do not represent such.
Sorry, but as written this is very misleading. The response shown is that of a digital FIR filter. In whole or in part, it is a linear transform, down to the arithmetical limits of the DSP. Such limits were 120dB down a couple of decades ago. You talk about hearing this. I am not convinced.
Cheers,
David.
Addressing this bit of pseudo science:
Let me refute your own words by using your own words:
---
Down conversion involves both filtering and quantization.
First:
Let's start at the top. This is the title of the paper:The audibility of typical digital audio filters in a high-fidelity playback system
No, I am not being pedantic . Please pay attention to what I have highlighted: digital audio filters. That is the mission of the paper. They start with 192 Khz music samples, then apply two filters to it. One is to represent CD sampling rate of 44.1 Khz and the other, 48 Khz. Here is the paper itself in technicolor to make sure these key points are not lost:
2.3. Signal processing and test conditions
Two kinds of linear-phase FIR (fnite impulse re-
sponse) filter were used, both of which operated at
192 kHz and both of which were implemented us-
ing TPDF (triangular probability density function)
dither at the 24th bit.
Secondly:
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
"prefect linearity"Oh no, it looks like your authorities have betrayed you.
As I said, you don't understand what you're talking about.
---
Do you really think what they heard in Stuart's listening test was linear distortion? Linear distortion garners such adjectives? "Echoes, when audible, were identi ed as being a ected the most clearly by the ltering. It was felt that some of the louder passages of the recording were less aggressive after ltering, and that the inner voices (second violin and viola) had "a nasal quality". I don't think so. Not at all.
Here's what you wrote to the objection of worse RPDF before:
But go ahead and move the green line up 6 db if it makes you feel better. It is still below the threshold of hearing.
Remember, you will lose a much larger battle if you stay on this path Arny. That such noise levels is audible in a 105 db SPL playback. And that they are so audible as to be an obvious flaw in the test.
So keep going at your peril.
Here you first say that even the worse 16-bit RPDF is not audible, but a few pages later you seriously claim that 24-bit TPDF causes
audible non-linear distortion.
---
And finally, a quote for the LOLs that doesn't need further comment:
Let's look at the impulse response again:
[...]
We hear each one of those samples as they come. The brain doesn't collect them all and then say, "ah, that was a FIR filter." It hears the individual samples as they get processed by the transfer function of the low pass filter. This is why you can't use your logic of "it is a linear function." Those sample transforms taken individually do not represent such.
And to make this even more funny:
To understand this topic fully, you must have broad understanding of multiple disciplines.
Listen to this 4 second short test file: imp_urhp.wav (http://www.hydrogenaud.io/forums/index.php?showtopic=107588), and tell us what you hear in your system.
No, the war has been lost.
Take your meds amir, there is no war. And stop refuting and contradicting and embarrassing yourself ... even more.
(http://amirviews.smugmug.com/photos/i-q35sw3V/0/L/i-q35sw3V-L.png)
From math/signal processing point of view we can say all of this ringing is benign. The theory says this is what must happen in time domain to implement the low pass function.
False claim. The above appears to be the response of a linear phase low pass filter to an impulse. However all low pass filters are not linear phase. For example a low pass filter can be minimum phase in which case there is no pre-ringing.
But that is not how the psychoacoustics works. We hear each one of those samples as they come.
Not on the planet earth, not if the listener is a human being, and not if the sample rate is much higher than a few 100 Hz at the most.
A good reference in this matter is found here: http://www.aes.org/sections/pnw/ppt/jj/fund_of_hearing.ppt (http://www.aes.org/sections/pnw/ppt/jj/fund_of_hearing.ppt)
"The inner ear is a mechanical filter bank, implementing a filter whose center-frequency tuning goes from high to low as one goes farther into the cochlea"
in short the stimulus that the brain responds to is the spectral content of the sound, not any imagined sample-by-sample analysis of digital data clocked in the 10s of KHz.
The above are two unbelievably poorly informed comments, especially given the alleged credentials of its author, and the fact that he has quoted the document I just referenced in this thread very recently.
False claim. The above appears to be the response of a linear phase low pass filter to an impulse. However all low pass filters are not linear phase. For example a low pass filter can be minimum phase in which case there is no pre-ringing.
That's not even the problem of his momentary claim (he changes them whenever he sees fit). To justify his repeated false comment about non-linear distortion in filters (that's what we were talking about) he had to go back to quantization, which he himself discarded before in this thread as being an audible problem - and that was with noise-modulating 16-bit RPDF.
(https://i.imgur.com/wJYjGZ4.gif)
False claim. The above appears to be the response of a linear phase low pass filter to an impulse. However all low pass filters are not linear phase. For example a low pass filter can be minimum phase in which case there is no pre-ringing.
That's not even the problem of his momentary claim (he changes them whenever he sees fit). To justify his repeated false comment about non-linear distortion in filters (that's what we were talking about) he had to go back to quantization, which he himself discarded before in this thread as being an audible problem - and that was with noise-modulating 16-bit RPDF.
(https://i.imgur.com/wJYjGZ4.gif)
I think that cat is who one becomes if one gives too much credibility is given to the wrong people.
There is an old saying: Engage brain,
then open mouth.
Truth be known everybody fails to follow this procedure every once in a while. The procedure seems to be absent or suspended permanently in some cases, eh? ;-)
Why aren't we discussing the BS paper and what they claim to have heard, instead of some subjectivist audiophile with a long history of absurd GE claims about hearing (via "listening") linear vs SMPS, SPDIF cables, DACs with +/-10% volume methodology, class d filtering, power regenerators, etc, etc, etc?
Last I checked, Amir was not present for the BS tests and was not part of that supervised listening panel (or any other, recorded, ever). He is not mentioned anywhere in the BS paper, which is purportedly the thread topic.
cheers,
AJ
In comparison testing I have done, switching amplifiers using the classic class D configuration always sport incredible low frequency control and power. They beat out linear class AB amplifiers almost regardless of price. What they give up though is high frequency fidelity which I find somewhat harsh. The distortion is highly non-linear and challenging to spot but it is there. The Mark Levinson No 53 is the first switching amplifier I have heard which does not have this compromise. Its bass is amazingly authoritative: tight and powerful. Yet the rest of the response is absolutely neutral and pleasant.
If you have not heard these unique amplifiers, I highly encourage you to come into our showroom for a listen. We have a pair on hand driving our Revel speakers. I am confident that they will improve the sound of your current speakers given the ease with which they can drive any load regardless of how difficult they might be (and many high-end speakers are difficult to drive). We are happy to let you evaluate them with your own system to see the benefits of this technology. Hearing this amplifier was an eye-opener for me. I think it will be for you too.
Can anybody explain to me how to distinguish the above 2 paragraphs from the usual subjectivist false claims that we all know and find to be unfounded, senseless and reprehensible?
But that is not how the psychoacoustics works. We hear each one of those samples as they come.
Not on the planet earth, not if the listener is a human being, and not if the sample rate is much higher than a few 100 Hz at the most.
A good reference in this matter is found here: http://www.aes.org/sections/pnw/ppt/jj/fund_of_hearing.ppt (http://www.aes.org/sections/pnw/ppt/jj/fund_of_hearing.ppt)
"The inner ear is a mechanical filter bank, implementing a filter whose center-frequency tuning goes from high to low as one goes farther into the cochlea"
in short the stimulus that the brain responds to is the spectral content of the sound, not any imagined sample-by-sample analysis of digital data clocked in the 10s of KHz.
Really? Here is the other JJ ES presentation I referenced earlier which you said you had read and were quoting sections from it:
.
This is why I say this discussion is not about science. Science would require someone to well, study science . When Arny says he has read about a topic and the very same presenter has said the opposite of what he is saying, and said reference was available to him just yesterday, then this is not a scientific discussion. It is the Xbox vs Playstation argument by two 16 year old that don't know a GPU from an FPU.
The notion that people with no educational or professional experience in the field waking up in the morning and post technical statements on a complex topic like this being a scientific approach requires suspension of disbelief. Somehow we have fooled ourselves into thinking getting our knowledge from forum posts and googling, and being loud means we are discussing "science." That is not remotely a possibility in real life.
That is on top of scientific method requiring unbiased approach to all data which is anything but what we are seeing here.
The above are two unbelievably poorly informed comments, especially given the alleged credentials of its author, and the fact that he has quoted the document I just referenced in this thread very recently.
Right... The very powerpoint I just quoted above which 100% disputes your read of the topic.
But that is not how the psychoacoustics works. We hear each one of those samples as they come.
Not on the planet earth, not if the listener is a human being, and not if the sample rate is much higher than a few 100 Hz at the most.
A good reference in this matter is found here: http://www.aes.org/sections/pnw/ppt/jj/fund_of_hearing.ppt (http://www.aes.org/sections/pnw/ppt/jj/fund_of_hearing.ppt)
"The inner ear is a mechanical filter bank, implementing a filter whose center-frequency tuning goes from high to low as one goes farther into the cochlea"
in short the stimulus that the brain responds to is the spectral content of the sound, not any imagined sample-by-sample analysis of digital data clocked in the 10s of KHz.
Really? Here is the other JJ ES presentation I referenced earlier which you said you had read and were quoting sections from it:
(http://amirviews.smugmug.com/photos/i-25ddQHJ/1/O/i-25ddQHJ.png)
It is exactly what I said. That the ear analyzes samples in time.
Amir, it would be nice if your revisionist history were accurate. However you didn't say what you are now claiming. You said:
"We hear each one of those samples as they come."
And speaking of revisionism, JJ actually said:
"It (the ear) does not wait for the whole filter to arrive."
So, how many samples long is a 44 KHz reconstruction filter?
For your claim: ""We hear each one of those samples as they come." to be relevant, every 44 KHz reconstruction filter would have to be just one sample long.
Reality is that these filters are from dozens to 100's of samples long. All JJ said was that our ears don't have to wait for the whole lot of from dozens to 100s of samples to arrive in order to start perceiving how the audio signal sounds.
There's a built in common sense reality check here. What kind of discrimination does the ear have over music samples that are one or two samples long? Can you name that tune in 25-50 microseconds? Of course you can't!
So its fail, fail, fail. First we have the pair of ludicrous errors, and then we have the purported support for the ludicrous errors based on grotesque misunderstandings and utter distortion of the meaning and sometimes even the actual contents of the quoted or paraphrased portions of the references.
If this were a high school debate what would the judges do if a high school student resorted to these tactics? I don't know because I've never seen a high schooler screw up this badly.
Why aren't we discussing the BS paper and what they claim to have heard, instead of some subjectivist audiophile with a long history of absurd GE claims about hearing (via "listening") linear vs SMPS, SPDIF cables, DACs with +/-10% volume methodology, class d filtering, power regenerators, etc, etc, etc?
Last I checked, Amir was not present for the BS tests and was not part of that supervised listening panel (or any other, recorded, ever). He is not mentioned anywhere in the BS paper, which is purportedly the thread topic.
cheers,
AJ
In the context of the paper, apparently the actual snippets from the musical selection are not going to be made available due to copyright concerns. Further the 192 kHz would simply be too large to email. This means that in the context of the BS paper there is no way to examine/evaluate/whatever the actual selections used. However the good news is that headphones, even if they don't extend to 40kHz should be adequate but then you'd have to create your own snippets.
Sorry, but as written this is very misleading. The response shown is that of a digital FIR filter. In whole or in part, it is a linear transform, down to the arithmetical limits of the DSP. Such limits were 120dB down a couple of decades ago. You talk about hearing this. I am not convinced.
I didn't talk about "hearing this." As in "this" being a specific thing. As I keep saying, the nature of ABX tests is not qualitative. And at any rate, the tests I performed were combination of filtering and quantization which is not a linear transform. Distinguishing files or even observing subjective difference doesn't spit out the cause.
Stuart's listening tests however does examine "this" to fair extent. They isolated just the filter and tested its audibility and got positive outcome to statistical confidence. And they were not searching down in 120 db SPL. They compared filter lengths between 44 and 48 Khz and hypothesized the difference in audibility to filter length. Clearly none of their subjects was hearing the extended ultrasonic response in frequency domain between 22 and 24 Khz.
As I just correct Arny, JJ's slide very clearly confirms the same that the brain is not a DSP machine where it captures a bunch of samples and then decides what it means. Note how JJ wrote that in capital letters. And this was my key point to xnor. That he can't use DSP explanations to rule out audibility. The brain's model of perception does not match the mathematical model of filter transforms. At the risk of stating the obvious, it can't hear the future! It has to judge the samples as they arrive and modified by the filter coefficients.
So sure, we can have our doubts but we can't rule it out as impossibility based on combined science of signal processing and psychoacoustics as xnor attempted.
Edit: usual typos.
I guess when jj talks about problems of ringing it is a problem especialy for lossy because the filters work near the audible band when cutting out masked content.
btw. i once saw nice pictures done with iZotope RX that shows the ringing together with its spectral distribution. Can anyone create one so that we don't have to look at this silly ringing picture the marketing uses since years?
I guess when jj talks about problems of ringing it is a problem especialy for lossy because the filters work near the audible band when cutting out masked content.
No, the presentation is not about lossy compression. He is using what he is hoping to be a familiar analogous problem in lossy compression as I did. Here is the title of the presentation:
(http://amirviews.smugmug.com/photos/i-QJsQxSL/0/XL/i-QJsQxSL-XL.png)
As you see, it is completely about the topic at hand. I also showed the follow up slide where he directly got into its relevance to anti-aliasing filters.
btw. i once saw nice pictures done with iZotope RX that shows the ringing together with its spectral distribution. Can anyone create one so that we don't have to look at this silly ringing picture the marketing uses since years?
Not sure what is wrong with the picture but here is JJ's:
(http://amirviews.smugmug.com/photos/i-kbsQ4Q4/0/XL/i-kbsQ4Q4-XL.png)
As I just correct Arny, JJ's slide very clearly confirms the same that the brain is not a DSP machine where it captures a bunch of samples and then decides what it means. Note how JJ wrote that in capital letters. And this was my key point to xnor. That he can't use DSP explanations to rule out audibility. The brain's model of perception does not match the mathematical model of filter transforms. At the risk of stating the obvious, it can't hear the future! It has to judge the samples as they arrive and modified by the filter coefficients.
So sure, we can have our doubts but we can't rule it out as impossibility based on combined science of signal processing and psychoacoustics as xnor attempted.
Oh boy, it hurts so much.
After contradicting and refuting yourself on the non-linear distortion part, of course you have to switch topics again and in the process put words in my mouth, fighting a straw man in your imaginary war.
No samples arrive at the ear, as you said. The ear receives a continuous mechanical wave that oscillates. Of course the ear doesn't wait for specific filters to arrive, it would have to wait a few seconds if we use a long filter.
As you've quoted many times now, the problem with pre-ringing jj mentions primarily are filter banks in codecs that
operate across the whole audible range. He even specifically gives the example of cutoff frequencies of 2 and 4 kHz.
As you send a short signal that contains energy from 1 to 3 kHz into a filter that cuts off at 2 kHz, the pre-ringing
at 2 kHz will build up slowly until it reaches the point where most energy is concentrated in the filter. This 2 kHz ringing can be audible.
I do not see how this is so had to get. Even a child should understand that ringing (which can be reduced or even eliminated, see previous posts)
outside the hearing range is different from clearly within the hearing range.
I also asked you to listen to a 4 second short test file (http://www.hydrogenaud.io/forums/index.php?showtopic=107588) for like ~20 times now, to shed some light on what you hear in your system.
Are you afraid to tell us? What's wrong amir?
These are pictures of impulse and frequency response. Nothing to wurry about.
Stuart's listening test showed audibility of such
Bzzzt,
wrong. Impossible for you to be cognizant of this, but this is both a causal and wishful thinking fallacy.
You have no clue what the cause is for what the BS test claims, whether it was system artifacts - speakers, switching software, time alignment, level alignment, frequency response in-band, etc, etc. or just the "emulated" filter/doctored dither.
cheers,
AJ
No, the presentation is not about lossy compression. He is using what he is hoping to be a familiar analogous problem in lossy compression as I did. Here is the title of the presentation:
You seriously cherry-pick the title of a presentation over the actual content of the actual slide you were just talking about a moment ago? You even highlighted
CODECS yourself.
Wow... I'm speechless.
amir credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029) = 0
No, the presentation is not about lossy compression. He is using what he is hoping to be a familiar analogous problem in lossy compression as I did. Here is the title of the presentation:
You seriously cherry-pick the title of a presentation over the actual content of the actual slide you were just talking about a moment ago? You even highlighted CODECS yourself.
Wow... I'm speechless.
amir credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029) = 0
This is the slide again:
. No on this planet as Arny would say. And the "two examples" that followed? They were two antialiasing filters, one with gentle and the other with sharp response. As I explained to Wombat, he is using pre-echo in codecs as a similar problem that is "hard to solve."
He also specifically talks about "converters" having pre-echo. Converters have nothing to do with codecs. That is the last letter in DAC/ADC.
You clearly can't follow technical explanations. Even a layman would see that you read the slide and presentation topic wrong.
btw. i once saw nice pictures done with iZotope RX that shows the ringing together with its spectral distribution.
That image was used in this thread (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=67619&view=findpost&p=604927) from 2008.
(http://www.galaxyclassics.com/public/4kHzLowPassFilters.jpg)
@amir
So what? We neither argue about codecs nor older rate converters.
btw. i once saw nice pictures done with iZotope RX that shows the ringing together with its spectral distribution.
That image was used in this thread (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=67619&view=findpost&p=604927) from 2008.
(http://www.galaxyclassics.com/public/4kHzLowPassFilters.jpg)
Thanks, this must have been it. Now we need this for a 21kHz filter to show where this ringing happens. Hopefully this will help to understand.
Amir can't defend the frivolous BS paper, thus has been reduced to one dive down the rabbit hole after another and quoting papers from a guy who says this:
Let's start: "how do you even imagine that one can hear a difference between two systems, one with noise 98 dB down and the other 146dB down, when the level is set to peak at 96dB?"
How do you explain how "obviously difference" fails to show up in even the worst kind of ABX test?
I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener.
Bear in mind the hard evidence for the persistance of loudness memory while you're at it.
I have my doubts that SACD or DVDA are much, if any, of an improvement, but the test is just blisteringly hard to run, and more likely to respond to artifacts, either positively or negatively, than it is to actual differences. Time alignment, level alignment, frequency response in-band can all throw it positive, lack of training, bad test environment, bad time alignment, etc, can also cause false negatives. Subject verification, likewise, is an important issue.
So, I remain undecided, but I note that I own a lot of CD's and not a single SACD or DVDA, except for some people have sent me.
And
In the usual stereo audio presentation, a partial sound stage consisting primarily of the front elements of the sound stage is created by two channels, either sampled from several microphones set in the original sound field or more often by a mixdown of many microphones placed both in proximity to the performers and out in the hall to capture the ambience. The information presented by the two channels, in either case, is a small fraction of the information in the original sound field. Additionally, this fraction is presented to the front of the listener. The presentation does not create an envelopment experience, where one is immersed in the original sound field, as the information is not present.
I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener.
Amir is implying JJ doesn't understand his own presentations. He's got it reversed, as usual.
cheers,
AJ
@amir
So what? We neither argue about codecs nor older rate converters.
??? Neither is the topic of JJ's presentation. Once more those are side remarks. The paper is about anti-aliasing filters. I will explain more but it is strange that I can't get this simple message across.
In comparison testing I have done, switching amplifiers using the classic class D configuration always sport incredible low frequency control and power. They beat out linear class AB amplifiers almost regardless of price. What they give up though is high frequency fidelity which I find somewhat harsh. The distortion is highly non-linear and challenging to spot but it is there. The Mark Levinson No 53 is the first switching amplifier I have heard which does not have this compromise. Its bass is amazingly authoritative: tight and powerful. Yet the rest of the response is absolutely neutral and pleasant.
If you have not heard these unique amplifiers, I highly encourage you to come into our showroom for a listen. We have a pair on hand driving our Revel speakers. I am confident that they will improve the sound of your current speakers given the ease with which they can drive any load regardless of how difficult they might be (and many high-end speakers are difficult to drive). We are happy to let you evaluate them with your own system to see the benefits of this technology. Hearing this amplifier was an eye-opener for me. I think it will be for you too.
Can anybody explain to me how to distinguish the above 2 paragraphs from the usual subjectivist false claims that we all know and find to be unfounded, senseless and reprehensible?
This is not a topic about amplifiers. I am unclear why there is no moderation to stop this off-topic branch.
But since there is so much interest about it, I created one on audibility of amplifier distortions: http://www.hydrogenaud.io/forums/index.php?showtopic=107604 (http://www.hydrogenaud.io/forums/index.php?showtopic=107604)
Please take your amplifier arguments there.
On a sidenote i remember how benchmark media in an official forum post declared ringing in their hardware as non-issue.
Meridian has a patended solution against it with their apodizing filter so interest in it to be an issue.
As I just correct Arny, JJ's slide very clearly confirms the same that the brain is not a DSP machine where it captures a bunch of samples and then decides what it means.
Never said it does. Amir you can argue with yourself as much as you wish, at least as long as the moderators pass on the Off-topic nonsense.
Just to review, Amir claimed: "We hear each one of those samples as they come."
On his first attempt to spin out this hole, he try to change what he said. He also tried to change what JJ said.
On his second attempt to spin his way out of this hole, he's trying to change what I said.
So here is the question - who else is he going to misquote? ;-)
Amir can't defend the frivolous BS paper, thus has been reduced to one dive down the rabbit hole after another and quoting papers from a guy who says this:
The paper doesn't need me to defend it. It has won an award for the best peer reviewed paper at this year's AES conference. Having you dispute it is like a fly hitting the windshield of the car slowing it down.
I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener.
You quote his forum posts, I quote his presentation to Audio Engineering Society:
[/quote]
The above slide could not be more clear. Don't confuse the JJ that is giving a formal presentation with one that is in forums smacking people like you around for fun . And of course your link is from a couple of years ago, prior to the publication of the listening test results we are discussing.
I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener.
You quote his forum posts, I quote his presentation to Audio Engineering Society:
Where does it mention hi-res? It doesn't mention it anywhere in your own quote. It just says it may affect your choice of sampling frequency. That could be 20.000001Khz or 14KHz or 19KHz or 42,000,000MHz.
Amir can't defend the frivolous BS paper, thus has been reduced to one dive down the rabbit hole after another and quoting papers from a guy who says this:
The paper doesn't need me to defend it. It has won an award for the best peer reviewed paper at this year's AES conference. Having you dispute it is like a fly hitting the windshield of the car slowing it down.
I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener.
You quote his forum posts, I quote his presentation to Audio Engineering Society:
The above slide could not be more clear. Don't confuse the JJ that is giving a formal presentation with one that is in forums smacking people like you around for fun . And of course your link is from a couple of years ago, prior to the publication of the listening test results we are discussing.
[/quote]
What's clear is that correcting severe and obvious errors and misrepresentations in certain people's posts doesn't change their behavior. The above quote is out of context, and its meaning is further clarified by other slides in JJ's presentation.
Please see this post made to this thread in the past day for a clarification of the misrepresentations made above:
http://www.hydrogenaud.io/forums/index.php...st&p=882429 (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&view=findpost&p=882429)
Let me highlight a part of JJ's slide that Amir curiously neglects .
A SPL range of 6dB -120dB being 'more than sufficient' for presentation, suggests that JJ is considering corner cases (i.e., rare cases) where 'presentation' on CD would not suffice.
This is pretty much the sort of case the whole hi rez-for-the-home edifice is built on. The vast majority of recordings ever made and offered for home listening would not require such a range. Certainly no *analog tape* recordings! And emphatically never any presentation in vinyl format, despite the beloved status of vinyl amongst 'golden ears'.
It accords with JJ also saying "I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener."
It would be rare indeed to find a case where the extra DR and bandwidth that high rez offers 'mattered' to a listener at home the way, say, different mastering EQ and compression 'matter', room correction 'matters', or more delivery channels 'matter'.
And of course your link is from a couple of years ago, prior to the publication of the listening test results we are discussing.
Liar. Part of that quote is from
Aug 2014 and I have linked it directly
multiple times in this thread, so there is no ambiguity and to stave off typical Amirspin.
There is absolutely nothing in this doctored BS paper that would conflict with anything I've read JJ state over multiple fora/presentations. There is probably very good reason why he wanted to offer PSR at 320kbps. Knowing fully well it would blow away your sides 2ch 24bit $cam.
Oh, he used Hafler AB amps too, for all the bass needed, not blingy 50k Class D $cam-amps.
His silence here is to save his old boss from further humiliation.
Or maybe he enjoys it. Who knows?
cheers,
AJ
Let me highlight a part of JJ's slide that Amir curiously neglects .
A SPL range of 6dB -120dB being 'more than sufficient' for presentation, suggests that JJ is considering corner cases (i.e., rare cases) where 'presentation' on CD would not suffice.
Sorry no. His data is not based what is rare and what is not. But rather, through a number of slides demonstrating the required dynamic range that would capture what we can hear. This is the slide on the top side:
(http://amirviews.smugmug.com/photos/i-KsdVqf7/0/L/i-KsdVqf7-L.png)
The hearing system uses a electromechanical gain stage to slide the limited dynamic range up and down through stiffening of OHC. See my post on AVS which may be easier to understand than JJ's: http://www.avsforum.com/forum/91-audio-the...ml#post24254752 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1512853-monitor-audio-rx6-distortion-2.html#post24254752)
The lower limit is described such by JJ:
(http://amirviews.smugmug.com/photos/i-x3MCbPf/0/L/i-x3MCbPf-L.png)
Now you see the 6 db reference. Again, my link above has a simpler explanation involving the brownian motion that sets the lower limit. JJ makes a good point about mapping this to ERB and hence getting 6 db unlike 0 db I assumed there. That gets you 114 db or 19 bits.
Any format that cannot present this dynamic range is missing what the hearing system can capture in a live situation.
Speaking of that topic which is not covered by JJ, you can read Fielder's JAES presentation on research he did by measuring the peak dynamic range of concert halls and their noise floor (accounting for conversion of noise to ERB as JJ suggests). He arrives at a requirement of 120 db there just the same. See my version of that write-up in the article I wrote for WSR Magazine: http://www.madronadigital.com/Library/RoomDynamicRange.html (http://www.madronadigital.com/Library/RoomDynamicRange.html)
CD as JJ clearly mentioned is a compromise here. Since we can adopt higher dynamic range formats with zero playback equipment cost to us, or any break in compatibility, there is little justification to stick to that spec when the physical format itself is becoming irrelevant over time.
I have yet to see a whit of evidence that "high-rez" matters for final presentation to a listener.
You quote his forum posts, I quote his presentation to Audio Engineering Society:
Where does it mention hi-res?
Who is buried in Elvis' grave?
It doesn't mention it anywhere in your own quote. It just says it may affect your choice of sampling frequency. That could be 20.000001Khz or 14KHz or 19KHz or 42,000,000MHz.
Which one of those is supported by your audio hardware?
Which one of those is supported by your audio hardware?
What has that got to do with the quote you posted? Which again doesn't mention hi-res, just different sample rates.
You really are the master of avoiding answering questions. Are you a politician?
In 29 pages, has anyone mentioned the other Bob Stuart co-authored paper from the same AES conference session? Number 9178?
http://www.aes.org/e-lib/browse.cfm?elib=17501 (http://www.aes.org/e-lib/browse.cfm?elib=17501)
It's not off topic. It includes very similar discussions. It includes very different ideas. It's classic Peter Craven territory. In one element, I thought "they're doing lossyWAV with the blocksize equal to one track!"
Cheers,
David.
Speaking of that topic which is not covered by JJ, you can read Fielder's JAES presentation on research he did by measuring the peak dynamic range of concert halls and their noise floor (accounting for conversion of noise to ERB as JJ suggests). He arrives at a requirement of 120 db there just the same. See my version of that write-up in the article I wrote for WSR Magazine: http://www.madronadigital.com/Library/RoomDynamicRange.html (http://www.madronadigital.com/Library/RoomDynamicRange.html)
As I've mentioned before I've loaded my Kindle with every issue of the JAES since 2001 and am working my way through them. In one article I noticed a comment that while it seems like he never published it, Fileder has mentioned in public that he did some research that showed that the average person listens to A/V at 70 dB SPL.
Which one of those is supported by your audio hardware?
What has that got to do with the quote you posted? Which again doesn't mention hi-res, just different sample rates.
You really are the master of avoiding answering questions. Are you a politician?
You were asking questions whose answer was in the very slide you were referencing. So I told you that with the reference to Elvis. The answer was given albeit in a light-hearted manner.
Since it is still not clear, I will go ahead and provide the detail. The phrase "hi-res" is a generic term and one coined in forums. Its absence in a presentation has no meaning for you to note. You need to read the content of the presentation to get the answer you are seeking.
CD's spec is 16 bits. JJ's presentation says you need 19 bits to cover the sensitivity range of the ear. And what gives you that in the marketplace is 24 bits which is the next jump from 16. That makes his requirement "hi-res" as in higher resolution than the CD.
Likewise, when someone says you probably need to worry about sampling rate to produce 20 Khz, they mean something better than 44.1. The next incremental value that is standardized in the industry is 48 Khz. And you go up from there to 88.2, 96, etc. There is no such thing as 22.00001 Khz and other random values you threw out.
These are the requirements for transparency. You are welcome to keep dialing them down as you prefer. But please don't mess with my dials .
JJ's presentation says you need 19 bits to cover the sensitivity range of the ear.
Two words: Flat PSD
The same presentation says:
"Given the noise in the modern world, 16 bits is probably sufficient in most places."
Say Amir, have you ever heard of perceptually shaped dither? If you have heard of it, do you have problems with its use?
JJ's presentation says you need 19 bits to cover the sensitivity range of the ear.
Two words: Flat PSD
The same presentation says:
"Given the noise in the modern world, 16 bits is probably sufficient in most places."
Not my copy of the presentation. That aside, not interested in "probably sufficient." That means transparency is not achieved for all content and all people. We have a format that does that and it is the source file prior to down conversion.
Say Amir, have you ever heard of perceptually shaped dither? If you have heard of it, do you have problems with its use?
Where I buy my music, they don't list what type of dither was used to produce the 16/44.1 from its higher resolution stereo master.
Remember, the best use of noise shaping is when your sampling rate is higher so that you can park the noise in the ultrasonic range where we know it is inaudible. Trying to stuff it in the audible band below the threshold of hearing is a tougher matter as you have to make sure my hearing is not better than that curve. Whereas I know I can't hear ultrasonics, noise or otherwise.
This is why I say if the CD had picked 48 Khz like the professional field had done, we would be far better off.
JJ's presentation says you need 19 bits to cover the sensitivity range of the ear.
Two words: Flat PSD
The same presentation says:
"Given the noise in the modern world, 16 bits is probably sufficient in most places."
Not my copy of the presentation. That aside, not interested in "probably sufficient." That means transparency is not achieved for all content and all people. We have a format that does that and it is the source file prior to down conversion.
The false claim here is the presumption that content with higher sample rate and/or word length exploits it. With 24/96 there is no content in the real world that exploits it. The musical selection used in the Meridian tests was a good example - the noise floor was only about 70 dB below peak levels and thus easily handled with 16 bits and best practices. Actually best practices would not be required, merely good practices such as avoiding RPDF dither should suffice.
Say Amir, have you ever heard of perceptually shaped dither? If you have heard of it, do you have problems with its use?
Where I buy my music, they don't list what type of dither was used to produce the 16/44.1 from its higher resolution stereo master.
JJ and I agree, as it was produced and as it will be most likely used, won't matter.
Remember, the best use of noise shaping is when your sampling rate is higher so that you can park the noise in the ultrasonic range where we know it is inaudible.
[/quite]
Ignores fact that noise shaping generally puts the noise below the threshold of hearing, and in the case of any real world recording the noise is also well below ambient noise.
More blustering and misdirection from Amir. Does he ever cease?
JJ notes that as we approach 120dB SPL, listening becomes *dangerous*. (I note Amir didn't highlight that part). JJ also knows what the absolute lower SPL limit of detectability is limited to the sound of atoms impinging on our ears, audible under conditions of extreme quietitude.
Someone please point me to recordings of real-world events, and playback in real-world rooms, where this
matters. We're talking corner cases again.
Do you really think all the audiophile ranting and raving about how bad CD sound is, has to do in reality with *any* of that?
re: attaining 'high fidelity' at home:
ELEPHANT IN THE ROOM:
bad mastering + bad in-room response at the listening position + too few delivery channels to provide spatial realism
FLEA IN THE ROOM:
0Hz-22kHz band inadequate to pass signals > 22kHz + 96dB dynamic range* inadequate for signals ranging 120dB
And which one do the Amirs of the world push as
the problem that needs addressing?
_________________
*more than that , with noise-shaped dither...up to 120dB 'effective' DR. As Monty (https://www.xiph.org/~xiphmont/demo/neil-young.html)notes:
Our -96dB noise floor figure is effectively wrong; we're using an inappropriate definition of dynamic range. (6*bits)dB gives us the RMS noise of the entire broadband signal, but each hair cell in the ear is sensitive to only a narrow fraction of the total bandwidth. As each hair cell hears only a fraction of the total noise floor energy, the noise floor at that hair cell will be much lower than the broadband figure of -96dB.
Thus, 16 bit audio can go considerably deeper than 96dB. With use of shaped dither, which moves quantization noise energy into frequencies where it's harder to hear, the effective dynamic range of 16 bit audio reaches 120dB in practice [13], more than fifteen times deeper than the 96dB claim.
120dB is greater than the difference between a mosquito somewhere in the same room and a jackhammer a foot away.... or the difference between a deserted 'soundproof' room and a sound loud enough to cause hearing damage in seconds.
Agrees with what J. Robert Stuart himself (http://file:///C:/Users/Steven/Downloads/coding2%20(1).pdf) wrote:
It is possible to exploit the frequency-dependent human hearing threshold by shaping the quantisation
and dither so that the resulting noise floor [of 16-bit audio] is less audible.
Figure 19 shows how the Meridian 518 (an in-band shaper) can allow a 16-bit transmission channel to
have a subjective noise floor more equivalent to a 20-bit ‘simple’ channel. If such a channel is to be
useful, the resolution of the links in the chain before and after the noise-shaped channel must be
adequate. In simple terms, this means mastering and playing back using well-designed converters
offering at least 20-bit resolution.
So, again, if you do Redbook 'right', you perceive 'hi rez' sound. But let's not tell people that when CD playback 'bad' it's because something's been done wrong...let'$ tell them that CD $ound i$ bad because CD i$ bad.
The false claim here is the presumption that content with higher sample rate and/or word length exploits it. With 24/96 there is no content in the real world that exploits it.
Anyone can declare anything. You need to back such statements Arny. Not just declare it.
Here is me backing the need for 120 db/20 bits to accomdate all cases: http://www.madronadigital.com/Library/RoomDynamicRange.html (http://www.madronadigital.com/Library/RoomDynamicRange.html). There are two journal of AES references at the end from two AES Fellows. You have something like that which says what you just said?
The musical selection used in the Meridian tests was a good example - the noise floor was only about 70 dB below peak levels and thus easily handled with 16 bits and best practices.
Here is the measurements from Stuart's article:
(http://amirviews.smugmug.com/photos/i-2qR2C5k/0/O/i-2qR2C5k.png)
At 3.5 Khz and eyeballing it, the noise floor of the content is around -25 db spl and the peak 90 db. So the total range is 90+25=115. How did you get the 70 number?
(http://amirviews.smugmug.com/photos/i-2qR2C5k/0/O/i-2qR2C5k.png)
At 3.5 Khz and eyeballing it, the noise floor of the content is around -25 db spl and the peak 90 db. So the total range is 90+25=115. How did you get the 70 number?
Nice job of cherry-picking from misleading data. The noise spectral density is said to have been measured using a 1 Hz bandwidth, but in fact the ear hears in critical bands which are about 1000 times wider around 3.5 KHz. The well known fallacy of measuring the perceptual qualities of music or ambient noise in narrow, constant frequency bands is being exploited.
Thanks for pointing this misleading data out, its another serious problem with the article. If you are going to do comparisons like this the frequency bands have to be perceptually relevant.
If you check the scientific literature, you'll find that a lot of noise measurements are done in octave bands. Pros know this, amateurs generally don't and make the mistake shown above.
The cherry picking part is the fact that the signal level shown (in a mistaken way) varies over a 40 dB range right around 3.5 KHz, and you picked your number from among the higher numbers in this range.
Hi Amir,
In another thread here, you linked a page with this article (http://www.madronadigital.com/Library/AudibilityofSmallDistortions.html) on your sales store website, titled "Audibility of Small Distortions" By Amir Majidimehr, the self assessed objectivist/non-hobbyist.
The “Q” indicates how steep the resonance is in the frequency domain. In the time domain (not shown), the higher the Q, the more “ringing” the system has. Ringing means that a transient signal (think of a spike) will create ripples that go on for some time after they disappear. An ideal system would reproduce that transient with zero ringing. The higher the Q of a resonance, the more ringing the system has. Reading what I just wrote, if I asked which one of the resonances on the right is more audible, you will likely say High Q. It seems natural that it has the highest amplitude change and more time domain impact per my explanation. Yet listening tests show the opposite to be true! The Low Q is more audible.
(http://www.madronadigital.com/Library/Resonances.png)
Here is the measurements from Stuart's article:
(http://i216.photobucket.com/albums/cc73/AJinFLA/DSP7200SErez.jpg)
What do you make of this and what would the implications be for audibility with and without typical and atypical filters?
Is it possible for out of band resonances to create harmonics in band?
cheers,
AJ
Looking at some of my past posts. I'm posting the entire abstract for the ABX1950 paper to correct any possible errors and/or omissions:
http://scitation.aip.org/content/asa/journ....1121/1.1917190 (http://scitation.aip.org/content/asa/journal/jasa/22/5/10.1121/1.1917190)
"
An understanding of the over‐all process of hearing depends upon proper interpretation of the results of many individual experiments. In the field of subjective experimentation the problem has been complicated by the wide variety of test procedures that characterize available data. If a common technique could be applied to the many different types of auditory tests, such as thresholds of acuity, masking tests, difference limens, etc., the organization of these data would be facilitated. The purpose of the present paper is to describe a test procedure which has shown promise in this direction and to give descriptions of equipment which have been found helpful in minimizing the variability of the test results. The procedure, which we have called the “ABX” test, is a modification of the method of paired comparisons. An observer is presented with a time sequence of three signals for each judgment he is asked to make. During the first time interval he hears signal A, during the second, signal B, and finally signal X. His task is to indicate whether the sound heard during the X interval was more like that during the A interval or more like that during the B interval. For a threshold test, the A interval is quiet, the B interval is signal, and the X interval is either quiet or signal. For a masking test, A is the masking signal, B is the masking signal plus the signal being masked, and X is either A or B repeated. The apparatus for the ABX test is mechanized so all details of the method can be duplicated for each observer, and the variability of manual operation eliminated. The entire test is coded on teletype tape to reduce the time and effort of collecting large quantities of data.
"
For example the last sentence points out that an automated procedure was used, and it could have and probably did include a list of pre-generated unknowns that were secret to everybody involved in the test, unless someone who could see the tape (likely) got clever and started reading the punched tape which is not impossible.
Thanks for pointing this misleading data out, its another serious problem with the article. If you are going to do comparisons like this the frequency bands have to be perceptually relevant.
Can someone explain to me what MAF and UEN thresholds (in Fig.3) are and how they differ from the well known Fletcher–Munson curves ?
MAF reflects sound pressure levels in the outside world, MAP reflects sound pressure levels near the eardrum. They are both testing loudness of sinusoids. Fletcher-Munson curves are also for sinusoids, but presented via headphones, i.e. not properly accounting for filtering of the head and pinna. Also, Fletcher-Munson curves are old and less accurate than modern measurements.
UEN shows the sound pressure level of noise (not sinudoids) that appears to be uniformly loud at all frequencies (no word on where it's measured: in the free field or near the eardrum).
Good morning Arny .
At 3.5 Khz and eyeballing it, the noise floor of the content is around -25 db spl and the peak 90 db. So the total range is 90+25=115. How did you get the 70 number?
Nice job of cherry-picking from misleading data.
Boy that is a grumpy way of starting a reply . I have now seen this phrasing used multiple times to dismiss out of hand various data point. It must be part of the parlance of this forum. It is another form of FUD so I hope we don't keep using the tactic when we are having an educated and deep technical discussion. More below.
The noise spectral density is said to have been measured using a 1 Hz bandwidth, but in fact the ear hears in critical bands which are about 1000 times wider around 3.5 KHz. The well known fallacy of measuring the perceptual qualities of music or ambient noise in narrow, constant frequency bands is being exploited.
Hmmm. Very, very odd comment Arny. How much a rock weighs has nothing to do with how much weight you can carry. The signal to noise ratio of a music track is a measurement. It has nothing to do with what part of it we can hear. Indeed you said it properly yourself:
"The musical selection used in the Meridian tests was a good example - the noise floor was only about 70 dB below peak levels and thus easily handled with 16 bits and best practices."See? You made no reference to what we can hear. Of course you didn't read the graph right so let's look at a marked up version to make it clear:
.
Your comment regarding critical band is just voodoo psychoacoustics in this context triggered by not understanding the method by which we get like data. Stuart graphs have been normalized to make it possible to compare a peak tone to a noise level. They need no further conversion to be used for computation of the signal to noise ratio. Indeed the very reason for existence of that graph is to make this very point regarding test music track.
Let's put all of that aside. We just went through JJ's slides that stipulated the ear's dynamic range at 114 db. So we need no further mistranslation of that. Our track has 115 db of dynamic range which means if we heard it live, we would be able to appreciate it all. A 24-bit recording system with real 20-bit response gives us that. A 16 bit system does not.
And just to bring some variety into the discussion, here is another trusted source on this topic:
From the Professor Vanderkooy's paper which we discussed earlier on his proposal to perform this test correctly:
A Digital-Domain Listening Test for High-Resolution
John Vanderkooy
Department of Physics and Astronomy, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1
Using non-subtractive flat triangular probability-function (TPDF) dither will add ?2/6 of noise power, so that the theoretical S/N ratio of 98 dB for 16-bit audio becomes about 93 dB.
While this may be audible by some listeners at elevated listening levels in a very quiet environment, especially at the beginning of a musical selection or at its final decay, its significance for audio quality is not as critical as other aspects.This is referenced in Stuart's paper. As you recall from our previous discussions, it outlines how one would create a much more reliable test of high resolution against down converted 16/44.1.
So I hope are done with the constant declarations that "music" is always 70 db or so and hence you don't need more than 12 to 13 bits. That is junk audio science if you forgive me for saying so.
Good morning Arny .
At 3.5 Khz and eyeballing it, the noise floor of the content is around -25 db spl and the peak 90 db. So the total range is 90+25=115. How did you get the 70 number?
Nice job of cherry-picking from misleading data.
Boy that is a grumpy way of starting a reply . I have now seen this phrasing used multiple times to dismiss out of hand various data point. It must be part of the parlance of this forum. It is another form of FUD so I hope we don't keep using the tactic when we are having an educated and deep technical discussion. More below.
The noise spectral density is said to have been measured using a 1 Hz bandwidth, but in fact the ear hears in critical bands which are about 1000 times wider around 3.5 KHz. The well known fallacy of measuring the perceptual qualities of music or ambient noise in narrow, constant frequency bands is being exploited.
Hmmm. Very, very odd comment Arny. How much a rock weighs has nothing to do with how much weight you can carry. The signal to noise ratio of a music track is a measurement. It has nothing to do with what part of it we can hear.
That's right Amir, in your radical subjectivist world it matters not in the least how well measurements correlate with perception or audibility.
Amir THD and jitter with any number of leading zeroes after the decimal point is what you seem to want to sell and the records of AVS have recorded this any number of times. How many times has this played out in your posts?
That's not the intent for the breed of objectivists of the kind I hang with.
If numbers are irrelevant to audibility, why these papers?
http://www.gedlee.com/distortion_perception.htm (http://www.gedlee.com/distortion_perception.htm)
I have a very slight concern about the filters they used, beyond the fact that they are unnecessarily steep. As far as I can tell, they didn't window them. I could be wrong on this, but let's assume for a moment I'm right. While the intended pre-/post-echos are at ~22kHz and ~24kHz, truncating the filter (rather than windowing it) can cause full-band echoes at the ends of the filter (i.e. +/-4ms). There's not enough information to know for sure, but best guess is they could be -76dB down. It could be a lot lower. It's probably not spectrally flat.
I'm not saying I think that it should be, or is, audible. I'm saying it's possibly something wrong in-band, albeit at a very low level. Temporal masking data suggests that it should be masked, though one big caveat is that temporal pre-masking declines dramatically with listener training.
It's a shame there's not that more information about the generation of the filters. They mention which MATLAB toolbox they used, but not all the parameters. A sentence more information, and those of us with MATLAB and that toolbox could re-create the exact filters ourselves.
Cheers,
David.
It is another form of FUD so I hope we don't keep using the tactic when we are having an educated and deep technical discussion.
Wow, the irony is stunning.
The noise spectral density is said to have been measured using a 1 Hz bandwidth, but in fact the ear hears in critical bands which are about 1000 times wider around 3.5 KHz. The well known fallacy of measuring the perceptual qualities of music or ambient noise in narrow, constant frequency bands is being exploited.
Hmmm. Very, very odd comment Arny. How much a rock weighs has nothing to do with how much weight you can carry. The signal to noise ratio of a music track is a measurement. It has nothing to do with what part of it we can hear. Indeed you said it properly yourself: "The musical selection used in the Meridian tests was a good example - the noise floor was only about 70 dB below peak levels and thus easily handled with 16 bits and best practices."
Wow, you're obviously absolutely clueless. Do you even know what signal-to-noise ratio means?
Of course you didn't read the graph right so let's look at a marked up version to make it clear:
The embarrassing part here is that I already explained spectral density ~20 pages back. How resistant can a human be to learning anything?
If you thought logically, for just one minute, you'd see that by your wrong method of graphically determining "SNR" you would have to say that 16-bit RPDF results in an "SNR" of over 130 dB given that track and a "noise floor" of about -145 dB.
Anyone can spot this as patently absurd, amirm. You are also misusing the term peak SNR, like all the other "deep technical" () terms which you apparently have no clue what they actually mean.
... I can't even be bothered to decipher the rest of your nonsense.
Since BS is armirs fellow he simply can ask him for the Matlab data.
That's right Amir, in your radical subjectivist world it matters not in the least how well measurements correlate with perception or audibility.
To do that, you need to read the measurements right. You were off by what, 45 db? Not a small error.
And I quoted JJ for the audibility part.
Amir THD and jitter with any number of leading zeroes after the decimal point is what you seem to want to sell and the records of AVS have recorded this any number of times. How many times has this played out in your posts?
THD? The only interactions I recall are of this nature posted just last year:
:
THD figure is an improper metric for audibility of distortion. It does not at all follow the rules of psychoacoustics. If I have a 5 Khz signal with a second harmonic at 10 Khz vs a 1 Khz signal with second harmonic at 2 Khz at the same level, the THD will be the same (with respect to the second harmonic). Audibility of these two situations however is very different due to frequency masking and resolution of auditory filters. One must know the spectrum of the harmonic distortions to determine the audibility which of course is never given (or often not measured individually).
We use THD because it used to be easy to measure and has become so ingrained in the industry that it continues to be used. But other than a gross measure, it has little value.
Arny answers thusly: http://www.avsforum.com/forum/91-audio-the...ml#post23952991 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1498174-amplifiers-affecting-speaker-frequency-response-8.html#post23952991)
THD is a reasonable metric for electronics and other situations where it can easily be reduced to the point where it is orders of magnitude below inaudible, psychoacoustics notwithstanding.
One leading critic of THD as a metric are Geddes and Lee who are good friends and whose papers I have actually read and recall a few relevant details from. Their criticism of THD is in the context of loudspeakers and other components that still may have audible nonlinear distortion.
Therefore the claim above that "...THD figure is an improper metric for audibility of distortion. " is inconsistent with science, reason and accepted recent research in the field of audio.
Clearly you are ignoring psychoacoustics and the work of Earl below in demonstrating the same in stating that.
That's not the intent for the breed of objectivists of the kind I hang with.
If numbers are irrelevant to audibility, why these papers?
http://www.gedlee.com/distortion_perception.htm (http://www.gedlee.com/distortion_perception.htm)
Let's quote Earl, shall we?
This is precisely where the signal-based distortion
metrics fail. In our next paper we will show that .01%
THD of one type of nonlinear system can be
perceived as unacceptable while 10% THD in another
example is perceived as inaudible. Even one of these
simple examples is sufficient to invalidate THD as a
viable metric for discussion of the perception of
distortion. Furthermore, 1% THD is not at all the
same as 1% IM, but we will show that neither
correlates with subjective perception. While some of
the signal-based metrics may be “better” than others,
it is our opinion they all fall short of what we are
seeking. I suggest instead of just having drinks with Earl, you actually listen to him and learn from him why THD is anything but "
consistent with science, reason and accepted recent research in the field of audio" that you wrote on AVS.
As JJ is fond of saying, almost any measurement that lacks a spectrum analysis, is useless in determining audibility.
Hi Amir,
Hey buster.
In another thread here, you linked a page with this article (http://www.madronadigital.com/Library/AudibilityofSmallDistortions.html) on your sales store website, titled "Audibility of Small Distortions" By Amir Majidimehr, the self assessed objectivist/non-hobbyist.
You exhibit a lot of frustration there partner. Any symptoms of dry mouth and restless sleep go with that?
The “Q” indicates how steep the resonance is in the frequency domain. In the time domain (not shown), the higher the Q, the more “ringing” the system has. Ringing means that a transient signal (think of a spike) will create ripples that go on for some time after they disappear. An ideal system would reproduce that transient with zero ringing. The higher the Q of a resonance, the more ringing the system has. Reading what I just wrote, if I asked which one of the resonances on the right is more audible, you will likely say High Q. It seems natural that it has the highest amplitude change and more time domain impact per my explanation. Yet listening tests show the opposite to be true! The Low Q is more audible.
(http://www.madronadigital.com/Library/Resonances.png)
Here is the measurements from Stuart's article:
(http://i216.photobucket.com/albums/cc73/AJinFLA/DSP7200SErez.jpg)
What do you make of this and what would the implications be for audibility with and without typical and atypical filters?
The audibility of resonances is based on research from Dr. Toole and Dr. Olive. It is based on listening tests of simulated resonances. It was made at low frequencies. You can't apply it to the above graph.
For one thing, the one you have circled on the right is in the ultrasonic/inaudible region. How the heck would you run a test and confirm Dr. Toole/Olive's results when the subjects can't hear the main tone let alone the variations from flat response?
On the prior one, the bandwidth of auditory filters are quite large by the time you get there. See my article on that here: http://www.madronadigital.com/Library/RoomReflections.html (http://www.madronadigital.com/Library/RoomReflections.html)
And this graph:
(http://www.madronadigital.com/Library/ERB.png)
You are way at the end of that graph. To determine audibility, you need a frequency response measurement that is psychoacoustically filtered per above. Such a filtering would smooth out a lot of ripples there.
Is it possible for out of band resonances to create harmonics in band?
How did you jump from linear resonances to non-linear distortions?
Wow, more quoting posts from other forums in an attempt at making others look as stupid as yourself. That's so pathetic.
amirm, how about you first learn what SNR means and come back then with a correction to your ridiculous graphical attempt at determining "SNR", as we can all see above?
Wow, more quoting posts from other forums in an attempt at making others look as stupid as yourself. That's so pathetic.
??? He asked me specifically about posts on AVS:
"Amir THD and jitter with any number of leading zeroes after the decimal point is what you seem to want to sell and the records of AVS have recorded this any number of times."What is pathetic about answering his question as he asked?
That aside, do you agree with my position or his?
Wasn't Earl's paper about speakers?
No samples arrive at the ear, as you said. The ear receives a continuous mechanical wave that oscillates. Of course the ear doesn't wait for specific filters to arrive, it would have to wait a few seconds if we use a long filter.
You don't say!
As you've quoted many times now, the problem with pre-ringing jj mentions primarily are filter banks in codecs that operate across the whole audible range. He even specifically gives the example of cutoff frequencies of 2 and 4 kHz.
2 to 4 Khz? Filter Banks? Sorry no. Not even close. The slides could not be clearer. Let's review them again:
.
Shouldn't have post just now. Reminded me to correct your post above .
Good morning Arny .
At 3.5 Khz and eyeballing it, the noise floor of the content is around -25 db spl and the peak 90 db. So the total range is 90+25=115. How did you get the 70 number?
Nice job of cherry-picking from misleading data.
Boy that is a grumpy way of starting a reply .
Just following your lead, Amir.
The noise spectral density is said to have been measured using a 1 Hz bandwidth, but in fact the ear hears in critical bands which are about 1000 times wider around 3.5 KHz. The well known fallacy of measuring the perceptual qualities of music or ambient noise in narrow, constant frequency bands is being exploited.
Hmmm. Very, very odd comment Arny. How much a rock weighs has nothing to do with how much weight you can carry.
A very odd and often very incorrect claim, especially if someone specifies their weight carrying ability in rocks or as is more common and historic: Stone.
The signal to noise ratio of a music track is a measurement. It has nothing to do with what part of it we can hear.
I think that the above almost unbelievable claim is actually something that you believe, Amir. It shows zero insight into the fact that the purpose of audio measurements is to provide some kind of representation of the subjective experience of listening to music through the equipment with a set of relevant objective measurements. Believing otherwise is common among subjectivists and other poorly informed audiophiles. In fact the subjective ragazines noften teach people that measurements are always meaningless.
Indeed you said it properly yourself: "The musical selection used in the Meridian tests was a good example - the noise floor was only about 70 dB below peak levels and thus easily handled with 16 bits and best practices."
I didn't reference that to a cherry picked measurement, or a measurement made over a 1 Hz band. You never asked and I suspect you never knew to ask. I've been trying to explain this and related subtleties to you on numerous occasions over many months on several forums and your recent post reflects poorly on my success in that effort.
One of the key concepts of that I've never been able to get over to you Amir is the concept of comparing comparable items. IOW, compare apples to apples.
Music and background noise are inherently non-comparable. Music is deterministic and composed of a collection of pure tones. Random noise is not. The energy in random noise is distributed over an infinite number of different frequencies. Accurate measurements of music only require that you measure over the range of frequencies that the music occupies which is often fairly limited. Background noise measurements require measuring over far wider ranges of frequencies. Usually noise measruements are made over the full audio band or over octaves or larger fractions of an octave. Noise measurements over 1 Hz bands are generally useless.
This is what he says about the impact of time domain response of those filters:
"If the filter has substantial energy that leads the main peak, this may be able to affect the auditory system."
So the fact that they are ultrasonic in that regard is not material because of the time domain effect. I am afraid that torpedoes your arguments throughout this thread.
The conclusion is not adequately supported by the out-of-context quote.
The logical error that is illustrated is that of interpreting a conditional and vague statement as if it is a general statement.
The more obvious conditional words in the quoted text are:
substantial
may
Substantial and May are not only conditional, they are also vague.
So the statement that is being used as a proof text for a global generality is both conditional (whether or not it is true in a specific case depends on unstated conditions) but it is also vague which can be illustrated by asking the questions:
How many percent or dB is Substantial?
What is the probability that is associated with May?
Wasn't Earl's paper about speakers?
No.
I linked the full text - please read it or at least the brief abstracts and if you have any further questions I would be happy to explain.
You don't say!
But that's exactly what you said. You said individual samples arrive at the ear. Quite funny indeed. Even funnier than not even knowing what SNR is.
As you've quoted many times now, the problem with pre-ringing jj mentions primarily are filter banks in codecs that operate across the whole audible range. He even specifically gives the example of cutoff frequencies of 2 and 4 kHz.
2 to 4 Khz? Filter Banks? Sorry no. Not even close. The slides could not be clearer. Let's review them again:
Are you blind? Are you trolling again (or should I say still)?
fund_of_hearing.ppt (http://www.aes.org/sections/pnw/ppt/jj/fund_of_hearing.ppt):
Slide 35:
“Linear Phase” (constant delay) filters do not have this phase shift, however they have a pre-ringing. In extreme cases (some older rate convertors, audio codecs) this pre-ringing is clearly audible. Not all pre-echo is audible.Slide 36:
- In Codecs this is a known, classic problem, and one that is hard to solve.
- In some older rate convertors, the pre-echo was quite audible.Slide 37 shows a
random filter, it says absolutely
nothing about audibility. It only demonstrated the
relationship between time- and frequency-domain.
adc.ppt (http://www.aes.org/sections/pnw/ppt/jj/adc.ppt):
Slide 63:
- The main lobe of a filter cutting off in 2.05 kHz must necessarily have a wider main lobe than the narrowest (in time) cochlear filter. df * dt >=1.
- The main lobe of a filter cutting off over 4kHz will have a main lobe a bit smaller than the narrowest cochlear filter.Slide 64:
reuses the same random filter as above to demonstrate the relationships between domains, not audibility.
It is absolutely ludicrous to use these examples to support audibility of pre-ringing at ~40 kHz. But we know why you do this, to win a war. See amir credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029) = 0.
"If the filter has substantial energy that leads the main peak, this may be able to affect the auditory system."
Yes, deliberately cherry-picking again. Yuck! And notice the
may be.
Here's what is says in the next line, again:
- In Codecs this is a known, classic problem, and one that is hard to solve.So the fact that they are ultrasonic in that regard is not material because of the time domain effect. I am afraid that torpedoes your arguments throughout this thread.
Since you've disqualified yourself a long time ago from any reasonable and rational discourse, I enjoy your posts for their comedic value, even if they are intellectually quite dishonest. Luckily people can google this thread and see your lack of honesty (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
You've just demonstrated your dishonesty again. Good job amirm.
Even a child should understand that ringing (which can be reduced or even eliminated, see previous posts) outside the hearing range is different from clearly within the hearing range.
Hopefully you see that you are completely mistaken here and you have to explain why it is hard to get, not me .
As I suspected, you do not understand what even a child would get.
I've also asked you to listen to a 4 second short test file (http://www.hydrogenaud.io/forums/index.php?showtopic=107588) for like ~20 times now, to shed some light on what you hear in your system.
Are you afraid to tell us?
Let's quote Earl, shall we?
This is precisely where the signal-based distortion
metrics fail. In our next paper we will show that .01%
THD of one type of nonlinear system can be
perceived as unacceptable while 10% THD in another
example is perceived as inaudible. Even one of these
simple examples is sufficient to invalidate THD as a
viable metric for discussion of the perception of
distortion. Furthermore, 1% THD is not at all the
same as 1% IM, but we will show that neither
correlates with subjective perception. While some of
the signal-based metrics may be “better” than others,
it is our opinion they all fall short of what we are
seeking.
Actually reading the paper being cited is informative. Please find me where in part 2 the .01%
THD of one type of nonlinear system was shown to be perceived as unacceptable by the average listener.
Secondly its quite clear that part 2 is partially based on crossover distortion which is well known to be audible at very low percentages because of asymmetries between the measurement environment and the listening environment. It is also so rare as to be irrelevant to modern reasonable quality audio gear.
Any symptoms of dry mouth and restless sleep go with that?
Nope. No Subjectivist audiophile BS claims about anything on my website or forum escapades, about amp$, wire$, Hi-Re$, etc.
If I like how something looks, feels, etc subjectively. I make no attempt to justify it objectively, be it a Ferrari or an ML amp. I sleep quite well because of it.
The audibility of resonances is based on research from Dr. Toole and Dr. Olive. It is based on listening tests of simulated resonances. It was made at low frequencies.
Well, you're either misinformed, lying, both, or...?
(http://i216.photobucket.com/albums/cc73/AJinFLA/TooleResonances.jpg)
That 12k (linear distortion) hump is certainly within the audible bandwidth, even to someone with as limited hearing as yours.
For one thing, the one you have circled on the right is in the ultrasonic/inaudible region.
Right. That's where one would expect to find Beryllium tweeter breakup resonances. Now the question is do the harmonics make it down into the audio band when excited, especially when driven to 108db @ speaker, near damage territory for a 1" DR dome.
DG's tests (http://www.davidgriesinger.com/intermod.ppt) seemed to indicate it might not be with his 2 tweeter sample. But it's not up to me or my side to do anything. It's up to your Hi-Re$ peddler camp to show that the system used in the BS test is transparent to the test and not generating false positives...and there is zero data regarding this.
How the heck would you run a test..
I wouldn't. You and the BS crew would, for any sort of honesty. Including the switching software and file alignment and....
cheers,
AJ
I think that the above almost unbelievable claim is actually something that you believe, Amir. It shows zero insight into the fact that the purpose of audio measurements is to provide some kind of representation of the subjective experience of listening to music through the equipment with a set of relevant objective measurements.
The purpose of my car is to be driven by me. But my car is an object independent of me. Measurements are measurements. How we interpret them is different. If you get the measurement wrong as you did, then the interpretation becomes completely false. Which is what happened when you used the 70 db number to proceed to tell us there is 12 to 13 bits of data in there. If you start with the right measurement value of 115 db, then you get the right bit depth which is 19.
One of the key concepts of that I've never been able to get over to you Amir is the concept of comparing comparable items. IOW, compare apples to apples.
And that is what Stuart has done. You see how there is only one set of axis there? Everything is normalized to the same two axis.
Music and background noise are inherently non-comparable. Music is deterministic and composed of a collection of pure tones. Random noise is not. The energy in random noise is distributed over an infinite number of different frequencies. Accurate measurements of music only require that you measure over the range of frequencies that the music occupies which is often fairly limited. Background noise measurements require measuring over far wider ranges of frequencies. Usually noise measruements are made over the full audio band or over octaves or larger fractions of an octave. Noise measurements over 1 Hz bands are generally useless.
Your starting point is correct. Where you end up is completely wrong. Stuart has literally written the book on topic of mapping noise to tone so that it can be compared to other things such as threshold of hearing that is based on test tones as was nicely explained by Alexey. I provided my article to you on this topic explaining the same: http://www.madronadigital.com/Library/RoomDynamicRange.html (http://www.madronadigital.com/Library/RoomDynamicRange.html). And noted the reference at the end:
“Noise: Methods for Estimating Detectability and Threshold,” Stuart, J. Robert, JAES Volume 42 Issue 3 pp. 124-140; March 1994Determining the "detectability and threshold" of noise requires conversion of noise power to music tones. The paper explains how this should be done to be perceptually correct (based no ERB) and Stuart has done the same in this paper. He provides the same reference in his most recent paper:
[30] J. R. Stuart. Noise: methods for estimating detectability and threshold. AES 93rd Convention, Berlin, 1993. And this is where it is used:
This [playback] level was chosen for comfort, and because it was high enough for details to be audible but also low enough that 16-bit RPDF dither would be inaudible at the listening position [30].RPDF dither is noise. And for it to be plotted correctly to show inaudibility requires that it be converted to the threshold of hearing base on tones. By making sure 1 Hz bins are used for the FFT of the music, we know there is no process gain and hence the numbers can be compared.
Any symptoms of dry mouth and restless sleep go with that?
Nope. No Subjectivist audiophile BS claims about anything on my website or forum escapades, about amp$, wire$, Hi-Re$, etc.
If I like how something looks, feels, etc subjectively. I make no attempt to justify it objectively, be it a Ferrari or an ML amp. I sleep quite well because of it.
Seeing how you go to shows and exhibit all the symptoms of subjectivism with modded players, ribbon speaker cables, etc, I say your are in denial and hence the reason I asked about other symptoms.
The audibility of resonances is based on research from Dr. Toole and Dr. Olive. It is based on listening tests of simulated resonances. It was made at low frequencies.
Well, you're either misinformed, lying, both, or...?
Here is my graph in my article you asked about:
(http://www.madronadigital.com/Library/Resonances.png)
And here is the reference in Dr. Toole's Book:
(http://amirviews.smugmug.com/photos/i-5t8PsXr/0/O/i-5t8PsXr.png)
This is the text for it:
Figure 19.9 shows examples of deviations from fl at for high- (50), medium-
(10), and low- (1) Q resonances at three frequencies when they were adjusted
to the audible threshold levels using pink noise in an anechoic chamber and
for the 200 Hz resonances detected when listening to typical close-miked,
low-reverberation pop and jazz.
So the only data for music is at 200 Hz which is what I referenced. The rest is for pink noise but even that was tested at max of 5 Khz, not the frequencies you asked me about.
The graph you are showing is a threshold shift between anechoic and reverberant space. That is a differential score for a different type of test.
That 12k (linear distortion) hump is certainly within the audible bandwidth, even to someone with as limited hearing as yours.
I don't know Ammar. Maybe I need the snake oil flat ribbon speaker cable you use with your speakers to hear them.
That aside, the hump in there playing the exact same thing in all the tests at 12 Khz. In other words it is invariant to the test. We are not performing a speaker review here Ammar.
For one thing, the one you have circled on the right is in the ultrasonic/inaudible region.
Right. That's where one would expect to find Beryllium tweeter breakup resonances.
Expecting and finding are two different things. Here is the explanation again:
(http://amirviews.smugmug.com/photos/i-2qR2C5k/0/O/i-2qR2C5k.png)
The *peak* level there is about 0 dbfs (relative to 105 db fs). The floor is about -30 db fs. You are saying that a Beryllium tweeter breaks up playing such low levels?
But let's say it does. Go ahead and tell us what the level of the IM sidebands would be.
DG's tests (http://www.davidgriesinger.com/intermod.ppt) seemed to indicate it might not be with his 2 tweeter sample. But it's not up to me or my side to do anything. It's up to your Hi-Re$ peddler camp to show that the system used in the BS test is transparent to the test and not generating false positives...and there is zero data regarding this.
And show we have. The Stuart paper has won an award for best peer reviewed paper at this year's conference. No one is waiting for lay people with no industry experience to approve anything. No medical research waits for approval of patients. Somehow you are confused thinking audio field is an exception because you have two ears. You need to fill the space between them with audio knowledge which doesn't happen by reading forum posts.
That is easy to show with your own reference. This is what David Griesinger, the author in that powerpoint says about Kiryu and Ashihara’s listening tests:
Their choice of source signal MAXIMIZES the (possible) audibility of an ultrasonic signal.
The sound pressure of the ultrasonic harmonics are equal to the sound pressure of the harmonics below 20kHz.
Now look at the ultrasonic components in Stuart's test above. There is no way any harmonic distortion in the ultrasonic range in stuart's test would create power that is equiv. to in-band frequencies.
If you get the measurement wrong as you did, then the interpretation becomes completely false. Which is what happened when you used the 70 db number to proceed to tell us there is 12 to 13 bits of data in there. If you start with the right measurement value of 115 db, then you get the right bit depth which is 19.
Words from a man that has posted this image:
(http://oi61.tinypic.com/2ajxc11.jpg)
This shows that you not only got the measurement wrong, you don't even know what the measurement is let alone what you measured.
I just analyzed the whole Haydn track. RMS amplitude of the noise floor at the end of the track is -58 dB +/-10 dB of that (min/max using a 50ms window).
With a max RMS amplitude of the total track reaching -10 dB, the dynamic range is about 60 dB, about 70 dB if you want to use 0 dBFS as peak.
If we filter out the noise below ~300 Hz we still get a -70 dB RMS noise floor.
(This is all relative to 0 dBFS, obviously.)
And that is what Stuart has done. You see how there is only one set of axis there? Everything is normalized to the same two axis.
No, you still don't understand.
You cannot subtract some musical peak amplitude from the noise spectral density, which is precisely what you did to calculate "SNR" (well, you thought you did, but you were not even close).
"This [playback] level was chosen for comfort, and because it was high enough for details to be audible but also low enough that 16-bit RPDF dither would be inaudible at the listening position [30]."
RPDF dither is noise. And for it to be plotted correctly to show inaudibility requires that it be converted to the threshold of hearing base on tones. By making sure 1 Hz bins are used for the FFT of the music, we know there is no process gain and hence the numbers can be compared.
It seems that you have no clue what you're talking about.
Yes, dither is noise.
This noise was however not converted in any way in the plot. Threshold of audibility of noise is shown as another line, which itself is unrelated to the noise floors.
1 Hz bins in the FFT does not ensure that there is no "process gain" (whatever that is in your mind) and the peak levels can actually be plotted accurately with a much higher bin width.
Let me repeat from my previous post:
If you had thought logically for just one minute you'd have seen that by your wrong method of graphically determining "SNR" you would have to say that 16-bit RPDF results in an "SNR" of over 130 dB given that track and a "noise floor" of about -145 dB.You really should look up some basic literature about signal analysis.
The *peak* level there is about 0 dbfs (relative to 105 db fs). The floor is about -30 db fs. You are saying that a Beryllium tweeter breaks up playing such low levels?
You seem confused. 105 dB
SPL not FS. Also, the noise floor is not at -30 dBFS, not even close.
The interesting questions are:
What is the
crossover frequency for the tweeter?
How much
power does it receive during loud passages?
Where are the measurements for this tweeter?
And show we have. The Stuart paper has won an award for best peer reviewed paper at this year's conference. No one is waiting for lay people with no industry experience to approve anything. No medical research waits for approval of patients. Somehow you are confused thinking audio field is an exception because you have two ears. You need to fill the space between them with audio knowledge which doesn't happen by reading forum posts.
Search this post (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029) for "homeopathy".
A paper with 60% positive outcome, peer-reviewed or not, doesn't make something a fact.
Peer-reviewed papers are not guaranteed to be correct either. You'd have to accept all kinds of nonsense if this was your "standard" (well, if it fits your agenda there doesn't seem to be much of a standard).
(Btw, I am not asserting that the tweeter caused audible artifacts, in case you want to attack some straw man.)
Seeing how you go to shows and exhibit all the symptoms of subjectivism with modded players, ribbon speaker cables, etc, I say your are in denial and hence the reason I asked about other symptoms.
Nope, nothing audiophile subjectivist shyster about that at all, like making objective claims about the "sound" of $cam amps one peddles for $50k (http://www.madronadigital.com/Library/MarkLevinsonNo53Amplifier.html), plying Hi-Re$, "Power Regenerators", etc, etc. $cams.
Not to mention outright fabricating listening tests, using +/-10% volume etc.
Those are definitely all subjectivist audiophile symptoms.
And here is the reference in Dr. Toole's Book
Which is where mine came from, i.e., Toole does not limit audibility of resonances <200hz (his resonance audibility data goes to 10K) to deflect from the fact that there is zero transparency info on the BS paper setup, including the speakers.
Naturally, since he's neither a $cam artist or a Hi Re$ peddler.
I don't know Ammar. Maybe I need the snake oil flat ribbon speaker cable you use with your speakers to hear them.
What about the $50k $cam-amp you peddle and claim to have done listening tests on? Would those $cam-amps with this audibility HF filters, be transparent (as you claim) to the BS paper doctored dither HF filtering?
(http://www.stereophile.com/images/1212ML53fig01.jpg)
cheers,
AJ
I provided my article to you on this topic explaining the same: http://www.madronadigital.com/Library/RoomDynamicRange.html (http://www.madronadigital.com/Library/RoomDynamicRange.html). And noted the reference at the end:
“Noise: Methods for Estimating Detectability and Threshold,” Stuart, J. Robert, JAES Volume 42 Issue 3 pp. 124-140; March 1994
Unfortunately Amir it appears that you did not read the middle of
your paper. More likely, it was quoted without being understood. I direct your attention to Figure 5, taken from a paper by Fielder.
(http://www.madronadigital.com/Library/FielderRecordingSpace.gif)
Now, for the first time in your life Amir you will hear an explanation a brief phrase taken from of the caption of this graph from
your paper which is:
"One Third Octave Comparison..."
This is how one compares noise to music, Amir. You don't use 1 Hz bands or any other constant frequency bands, you use bands composed of an third octave, or an octave or the same number of octaves or the same fraction of an octave.
Unfortunately I'm currently in Connecticut visiting my oldest son and his wonderful family for the Thanksgiving holiday having completed a similar visit with my youngest son and his family in Massachusetts. In my haste to leave home I left part of my JAES library back in Grosse Pointe. This laptop's SSD could hold it as it is only about 18 CDs' worth of data. I will have more fun with you and Mr. Stuart next week when I return to Grosse Pointe, God and Southwest Airlines willing. ;-)
I think there's some confusion because the UEN (red curve in the graph quoted many times in this thread) is quoted for 1Hz wide bands, and the noise floor in the same graph is quote for 1Hz wide bands. That's fine. The noise will cross the UEN at the same real-world signal level that a one-third octave measure of the same noise would cross the regular (tone referenced) threshold of hearing curve. The 1Hz wide measures are obviously very different from the one-third octave measures, but as long as you compare like with like, you get the same threshold for noise in the range where one-third of an octave is a reasonable approximation to the width of a critical band (i.e. the filter in the ear). Where it's not, it doesn't work, and an accurate UEN is a better way of doing it than one-third octave measures.
If you don't know what you're doing, it goes "wrong" for tones, or very shaped noise. The fact the curve for tones ends up being 20dB+ higher than the curve for noise shows part of the problem. The fact that a 1Hz bin implies an analysis time of one second is the other problem: any music which changes spectrum within a second isn't being analysed properly.
I'm not suggesting any mistake is being made in the paper (though as ever, I'd like more details). I do think people are getting a bit carried away in their analysis, claiming things as fact that they possibly don't fully understand.
Cheers,
David.
I think there's some confusion because the UEN (red curve in the graph quoted many times in this thread) is quoted for 1Hz wide bands, and the noise floor in the same graph is quote for 1Hz wide bands. That's fine. The noise will cross the UEN at the same real-world signal level that a one-third octave measure of the same noise would cross the regular (tone referenced) threshold of hearing curve. The 1Hz wide measures are obviously very different from the one-third octave measures, but as long as you compare like with like, you get the same threshold for noise in the range that one-third of an octave is a regional approximation to the width of a critical band (i.e. the filter in the ear). Where it's not, it doesn't work, and an accurate UEN is a better way of doing it than one-third octave measures.
If you don't know what you're doing, it goes "wrong" for tones, or very shaped noise. The fact the curve for tones ends up being 20dB+ higher than the curve for noise shows part of the problem. The fact that a 1Hz bin implies an analysis time of one second is the other problem: any music which changes spectrum within a second isn't being analysed properly.
I'm not suggesting any mistake is being made in the paper (though as ever, I'd like more details). I do think people are getting a bit carried away in their analysis, claiming things as fact that they possibly don't fully understand.
There are several dimensions to things "...they possibly don't understand", and one of them is how our favorite conference participant can be expected to slice and dice quotes to his debating trade advantage. The above is written in such a way that it can reasonably be expected to be the centerpiece of one or a dozen of his flaming posts, as an example of how I don't know what I'm talking about.
You don't have to make a mistake to set the stage for misinterpretation.
We've already seen this chart used to support a number of claims that tax the imagination: http://amirviews.smugmug.com/photos/i-5vBd...O/i-5vBdNfX.png (http://amirviews.smugmug.com/photos/i-5vBdNfX/0/O/i-5vBdNfX.png)
So there is no denying that it is prone to these sorts of things.
One experience that weighs heavy in my mind is that I've done equivalent analysis of the music that the above chart purports to represent. I know it that chart nothing like what you get if you analyze the whole piece or any of the 15-18 second segments of it that the paper mentions. It is probable that the data was cherry-picked to lead to the mistaken conclusion that we are both trying to rebut.
We are not going to succeed in any conventional sense because our correspondent does not admit his mistakes if they are anything but completely trivial, and most of those are obfuscated heavily if pointed out.
Thanks for the MAF/UEN explications.
I'm still wondering if the paper doesn't contradict itself by predicting in Fig. 3 that the dither should be inaudible and at the same time finding in the tests that dither made a significant audible difference.
Another thing that bothers me is that the paper uses the term "quantization" for what seems to be truncation (undithered wordlength reduction). Please correct me if I'm wrong.
It almost seems to me that they presented truncated 24 and 16 bit audio and added dither after the truncation.
Page 10: This could be an effect of quantization distortion; it is interesting that this was audible even in a 24-bit system, and is consistent with the hypotheses of Stuart [29] that 16 bits are not sufficient for inaudible quantization.
It is possible that that these two qualitative effects on sound quality cancelled out each other for the conditions where 16-bit quantization plus RPDF dither was applied, which could explain why these signals were harder to discriminate from their unfiltered counterparts.
[/size]
Are they masking truncation artifacts with dither noise instead of properly (RPDF is better than nothing) dithering before truncation ?
I'd like to see 24-bit quantization noise plotted in Fig.3. It should be far below the thresholds of human hearing, so why would this be audible, even in a 24-bit system ?
Where things are not explained in full, I am giving the authors the benefit of the doubt. The AES does not provide infinite space and something has to give. I'm sure it was dither then truncation.
cheers,
David.
I'm still wondering if the paper doesn't contradict itself by predicting in Fig. 3 that the dither should be inaudible and at the same time finding in the tests that dither made a significant audible difference.
Yeah, assuming this is not a result of chance or bias in their methodology or something like that, the results show that for 22.05 kHz cutoff condition 1 (24 bit TPDF) was easier to distinguish from the original than both condition 2 (16-bit "quantization") and condition 3 (16-bit "quantization and RPDF").
So their claim that 16-bit made a significant audible difference seems to be refuted by their own results.
They really should have called it "16-bit truncation" instead of just "16-bit quantization".
And their phrasing "16-bit quantization and rectangular dither" really does sound like adding dither after truncation, but I doubt that they would make such mistakes. At least I hope it's just bad phrasing.
Why not just use TPDF? Why the weird phrasing? Why do they mix up quantization distortion and dither noise, and audibility of these things?
edit: Actually, the anecdotal subjective impressions would make more sense if they screwed up quantization.
Another thing that bothers me is that the paper uses the term "quantization" for what seems to be truncation (undithered wordlength reduction). Please correct me if I'm wrong.
You are wrong . In the context of peers reading such a paper, saying that going from 24 bits to 16 bit is quantization is correct terminology. A more precise one would be re-quantization but that "re" is understood.
As it happens I had another AES paper open which directly addresses this:
Subtractive Dither for Internet Audio
BEN DENCKLA
Dithered quantization can be useful in preparing audio files for distribution on the Internet, since this often involves quantizing audio down from its original word length (typically 16 bits) to 8 bit in order to achieve a form of lossy compression, for example, offering 2:1 compression of a 16-bit source. This is sometimes referred to as “requantization” to distinguish it from the quantization of a truly continuous source, such as an analog electronic waveform, but we will not bother with that distinction here.So the terminology is fine.
Are they masking truncation artifacts with dither noise instead of properly (RPDF is better than nothing) dithering before truncation ?
No, the context in that quote is the combination of filtering artifacts and requantization combining in some way to generate that listening test result.
I'd like to see 24-bit quantization noise plotted in Fig.3. It should be far below the thresholds of human hearing, so why would this be audible, even in a 24-bit system ?
The quantization in 24 bit domain used TPDF post output of the filter. You are asking for level of that?
I'm still wondering if the paper doesn't contradict itself by predicting in Fig. 3 that the dither should be inaudible and at the same time finding in the tests that dither made a significant audible difference.
Yeah, assuming this is not a result of chance or bias in their methodology or something like that, the results show that for 22.05 kHz cutoff condition 1 (24 bit TPDF) was easier to distinguish from the original than both condition 2 (16-bit "quantization") and condition 3 (16-bit "quantization and RPDF").
So their claim that 16-bit made a significant audible difference seems to be refuted by their own results.
They really should have called it "16-bit truncation" instead of just "16-bit quantization".
And their phrasing "16-bit quantization and rectangular dither" really does sound like adding dither after truncation, but I doubt that they would make such mistakes. At least I hope it's just bad phrasing.
Why not just use TPDF? Why the weird phrasing? Why do they mix up quantization distortion and dither noise, and audibility of these things?
edit: Actually, the anecdotal subjective impressions would make more sense if they screwed up quantization.
The results they achieved were IME close enough to random guessing to not totally eliminate the possibility that they were in fact random guessing. 56% right after working as hard as they did is really pretty disappointing.
Another point is that best practices says that the dither should be TPDF and have perceptually shaped PSD. They had two opportunities to use best practices and blew them both off. Pretty ironic given that Meridian used to promote the benefits of equipment they sold that did both things right.
The results they achieved were IME close enough to random guessing to not totally eliminate the possibility that they were in fact random guessing. 56% right after working as hard as they did is really pretty disappointing.
Good thing that was not their results then.
Another point is that best practices says that the dither should be TPDF and have perceptually shaped PSD. They had two opportunities to use best practices and blew them both off. Pretty ironic given that Meridian used to promote the benefits of equipment they sold that did both things right.
TPDF dither was used in the filtering part of the test. Those generated positive outcomes while the bit stream remained at 24 bits.
TPDF dither was used
Yep, by the BS crew...and here is the result:
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
You are wrong . In the context of peers reading such a paper, saying that going from 24 bits to 16 bit is quantization is correct terminology. A more precise one would be re-quantization but that "re" is understood.
No, he is not. "Truncation" would be a more narrow term that describes better what they apparently did. "Quantization" is a broad term and is actually implied with "truncation", not the other way around though.
As it happens I had another AES paper open which directly addresses this:
Subtractive Dither for Internet Audio
BEN DENCKLA
Dithered quantization can be useful in preparing audio files for distribution on the Internet, since this often involves quantizing audio down from its original word length (typically 16 bits) to 8 bit in order to achieve a form of lossy compression, for example, offering 2:1 compression of a 16-bit source. This is sometimes referred to as “requantization” to distinguish it from the quantization of a truly continuous source, such as an analog electronic waveform, but we will not bother with that distinction here.
So the terminology is fine.
No, you are stepping in exactly the trap that Kees de Visser pointed out, confusing truncation with dithered quantization.
No, the context in that quote is the combination of filtering artifacts and requantization combining in some way to generate that listening test result.
I don't think so.
Or are you seriously suggesting that 16-bit RPDF cancels out the alleged effects of the lowpass filters?
I'd like to see 24-bit quantization noise plotted in Fig.3. It should be far below the thresholds of human hearing, so why would this be audible, even in a 24-bit system ?
If I'm not mistaken then the amplitude spectral density of 16-bit TPDF in a 192 kHz channel is -143 dB, about 2 dB lower for RPDF.
For 24-bit TPDF about -191 dB.
In figure 3 this would be relative to the acoustic gain, so -38 dB for 16-bit TPDF, -86 dB for 24-bit.
The results they achieved were IME close enough to random guessing to not totally eliminate the possibility that they were in fact random guessing. 56% right after working as hard as they did is really pretty disappointing.
Another point is that best practices says that the dither should be TPDF and have perceptually shaped PSD. They had two opportunities to use best practices and blew them both off. Pretty ironic given that Meridian used to promote the benefits of equipment they sold that did both things right.
Small correction: about 60% correct on average.
But it gets worse, since Meridian also uses a slow roll-off filter to resample CD audio that is said to eliminate pre-ringing, removing claimed negative effects, fixing even tracks that were recorded at 44.1 kHz. Everyone can resample 44.1 kHz with a similar filter on the fly in foobar2000 ...
And in the test they use only one filter which has a bandwidth of less than 460 Hz (40 Hz more for the 48 kHz case)?!
Or are you seriously suggesting that 16-bit RPDF cancels out the alleged effects of the lowpass filters?
That is what they are hypothesizing to explain the outcome in the article. That is what the topic was: what the text meant. And I explained it. So no, it is not my suggestion.
TPDF dither was used
Yep, by the BS crew...and here is the result:
TPDF dither was used
Yep, by the BS crew...and here is the result:
This gets so tedius!
From:
The audibility of typical digital audio filters in a high- delity playback system AES paper 9174
Page 5
"
After filltering with either FIR fi lter, the signals
were either unchanged or were quantized to 16-
bit. The quantization either included RPDF (rect-
angular probability density function) dither or did
not.
Page 8
"A within-subjects ANOVA was performed on the
data for all eight listeners with two factors: filter
cutoff frequency (low or high) and quantization
type (none, 16-bit, or 16-bit plus RPDF dither).
Page 9:
"
Post-hoc Fisher tests based on a least-signi cant dif-
ference of means at a 5% level of 0.1394 for quanti-
zation showed that performance was worse for 16-bit
quantization plus RPDF dither (mean=1.773) than
for no quantisation (mean=1.902) or 16-bit quanti-
sation alone (mean=1.909).
Page 9-10
"Every condition where 16-bit quantization was ap-
plied gave performance that was signi cantly better
than chance. Performance was signi cantly worse
for 16-bit quantization plus RDPF dither than for
no quantisation or 16-bit quantization alone. This
suggests that the e ect of adding the RPDF dither
on top of the 16-bit quantization and FIR ltering
was to make it more dicult to identify that process-
ing had been applied to the signal, which is perhaps
counterintuitive. To try to explain this nding, we
turn to subjective descriptions of what listeners de-
scribed hearing in these tests.
"
Page 10
"It is possible that that these two qualitative ef-
fects on sound quality cancelled out each other for
the conditions where 16-bit quantization plus RPDF
dither was applied, which could explain why these
signals were harder to discriminate from their un l-
tered counterparts."
"All forms of processing tested here were audi-
ble, except for one condition where performance was
signi cantly di erent from chance at the 6.7% level,
including emulated downsampling lters at standard
sample rates and 16-bit quantization with or without
RPDF dither. Di erences were demonstrated here
in a double-blind test using non-expert listeners who
received minimal training.
"16-bit quantization with and without RPDF
dither can have a deleterious e ect on the listen-
ing experience in a wideband playback system.
That is what they are hypothesizing to explain the outcome in the article. That is what the topic was: what the text meant. And I explained it. So no, it is not my suggestion.
Once again, the filter test used the above TPDF. Per that same quote, it rules out audibility of quantization distortion in the interim math used for the filter.
They didn't use either TDPF let alone noise shaping with 16 bits, so all their comments on 16 bit being insufficient based on their results are moot. I've also never heard anyone say such types of "errors" don't add up but subtract, such that lower bit depth makes it harder to distinguish the filtered file from the original one. Does that mean that with 15 bits and insufficient dither, they wouldn't even have reached their magic 56.25% statistical significance?
The conclusion: Lower bit depth with inadequate dither may make files sound more hi-res like. Nice.
The score that comes closest to CD audio (44.1/16) is ~60%, which is not much different from their 48 kHz, 24 bit score.
They should have ditched the weird 16-bit conditions, just use TPDF and noise shaping, add a low-anchor (e.g. 12 bit and/or 32 kHz) and a "control", such as 96/24, to the result so that this and the test methodology could be better judged by a single look.
I guess we have to wait for (a lot) more details... I'd be interested especially in the raw data of all the tests they did.
edit:
You seem to have skipped this:
The interesting questions are:
What is the crossover frequency for the tweeter?
How much power does it receive during loud passages?
Where are the measurements for this tweeter?
and this:
4 second short test file (http://www.hydrogenaud.io/forums/index.php?showtopic=107588), what do you hear in your system?
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
Once again, the filter test used the above TPDF
Right. That means there was some other reason for the false positives. Was it the concocted "emulated filters" themselves, or the lack of system transparency due to the speakers, switching software, Time alignment, level alignment, frequency response in-band, etc, etc.
Or was it simply due to the pecuniary interests and vested interest in the Hi-Re$ $cam, of your camp?
Who knows?
cheers,
AJ
This gets so tedius!
You telling me? Here is the relevant part:
2.3. Signal processing and test conditions
Two kinds of linear-phase FIR ( Finite impulse re-
sponse) filter were used, both of which operated at
192 kHz and both of which were implemented us-
ing TPDF (triangular probability density function)
dither at the 24th bit.That processing, filtering down to half bandwidth of 44 Khz, while maintaining 24 bit sample resolution was done with TPDF. The results were positive.
Since any down conversion first requires a filter, then the above demonstrates that the process is not transparent according to this test data.
You telling me? Here is the relevant part:
[...]
That processing, filtering down to half bandwidth of 44 Khz, while maintaining 24 bit sample resolution was done with TPDF. The results were positive.
Since any down conversion first requires a filter, then the above demonstrates that the process is not transparent according to this test data.
You were the one that brought up quantization again in the first place, trying to justify your nonsense about non-linear distortion. You also posted the
nonsense graph here:
(http://oi61.tinypic.com/2ajxc11.jpg)
So don't say "you telling me", to anyone but yourself. Seriously...
You are doing this here again amirm:
(https://i.imgur.com/wJYjGZ4.gif)
As it happens I had another AES paper open which directly addresses this:
Subtractive Dither for Internet Audio
BEN DENCKLA
Dithered quantization can be useful in preparing audio files for distribution on the Internet, since this often involves quantizing audio down from its original word length (typically 16 bits) to 8 bit in order to achieve a form of lossy compression, for example, offering 2:1 compression of a 16-bit source. This is sometimes referred to as “requantization” to distinguish it from the quantization of a truly continuous source, such as an analog electronic waveform, but we will not bother with that distinction here.
So the terminology is fine.
No, you are stepping in exactly the trap that Kees de Visser pointed out, confusing truncation with dithered quantization.
Sorry, no. Dither is an optional processing step in quantization. The term quantization therefore has no dependency whatsoever on dither being used or not.
Truncation to fewer bits is re-quantization. And per above, it is customary to drop the "re" from that.
I'm still wondering if the paper doesn't contradict itself by predicting in Fig. 3 that the dither should be inaudible and at the same time finding in the tests that dither made a significant audible difference.
Kees, I haven't read the paper discussed in this topic, but your quote contains no contradiction: although 16-bit dither is inaudible (@105 dB SPL), the truncation distortion may be audible. In this way, dither does make a difference.
Dither is an optional processing step in quantization.
Here is what BS/Meridian say about dither when not doctoring $cammer results:
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
I'd like to see 24-bit quantization noise plotted in Fig.3. It should be far below the thresholds of human hearing, so why would this be audible, even in a 24-bit system ?
The quantization in 24 bit domain used TPDF post output of the filter. You are asking for level of that?
I'm still trying to grasp the message from the following part:
Listeners described that quantization gave a “roughness” or “edginess” to the tone of the instruments, and that quantization had a significant impact on decay, particularly after homophonic chords, where “decay was sustained louder for longer and then died suddenly”. This could be an effect of quantization distortion; it is interesting that this was audible even in a 24-bit system, and is consistent with the hypotheses of Stuart [29] that 16 bits are not sufficient for inaudible quantization.
[/size]
First I interpreted (apparently wrongly) that 24-bit quantization artifacts were audible in a 24-bit system, since the 24-bit filtered versions (condition nos. 1&4) have more (24-bit) dither than the 24/192 original. I didn't see why that would be consistent with the hypothesis that 16 bits are not sufficient.
OK, so listeners reported roughness or edginess and cut-off reverb tails, even in the 24/48 TPDF dithered version, implying that the filter is responsible for this (since the 24-bit dither shouldn't be audible).
What I would like to see in Fig.3 is the difference signals of the 24/192 original and each of the 6 filtered versions, to verify that those differences are below the threshold curves.
OK, so listeners reported roughness or edginess and cut-off reverb tails...
But there is no indication of causation, as it could be false positives generated by system artifacts, switching, etc.
It's up to Amirs camp pushing the Hi-Re$ $cam to confirm system transparency to test, so no assumptions about cause>filters need to be made as you did above.
No system/setup/test transparency data = no conclusions, only wishful thinking fallacy assumptions by those with pecuniary interests. Like Amir's camp.
cheers,
AJ
(http://amirviews.smugmug.com/photos/i-FKcswgv/0/M/i-FKcswgv-M.png)
Sorry, no. Dither is an optional processing step in quantization. The term quantization therefore has no dependency whatsoever on dither being used or not.
Exactly, so why call it "16 bit quantization" when it could be a more unambiguous "truncated to 16 bits", for example.
Truncation to fewer bits is re-quantization. And per above, it is customary to drop the "re" from that.
Kees de Visser talked about "truncation" but you cite some paragraph about dithered quantization and the term "requantization" which nobody even mentioned. So I have to assume you're confused again.
First I interpreted (apparently wrongly) that 24-bit quantization artifacts were audible in a 24-bit system, since the 24-bit filtered versions (condition nos. 1&4) have more (24-bit) dither than the 24/192 original. I didn't see why that would be consistent with the hypothesis that 16 bits are not sufficient.
OK, so listeners reported roughness or edginess and cut-off reverb tails, even in the 24/48 TPDF dithered version, implying that the filter is responsible for this (since the 24-bit dither shouldn't be audible).
What I would like to see in Fig.3 is the difference signals of the 24/192 original and each of the 6 filtered versions, to verify that those differences are below the threshold curves.
1) See the noise levels in my previous post (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882846). Doing filtering and dithering again changes this density from -191 to -188 dB (-83 dB in the figure given their acoustic gain).
They are purely speculating at this point. There's nothing definitive in their results that would support this speculation. Also, saying that 24 bit is not sufficient is not consistent with saying that 16 bit theoretically isn't.
Also keep in mind that the noise floor of one of the most "revealing" tracks lies a couple dB over even 16-bit dither noise.
2) If they did the filtering right, the difference should be around -110 dB in the critical 4 kHz area (linear spectrum, not density, since the difference is music) given the passband ripple of 0.025 dB which prevents nulling down to the dither.
(Difference here means simply subtracting the original sample by sample from the filtered and quantized version.)
Also keep in mind that the noise floor of one of the most "revealing" tracks lies a couple dB over even 16-bit dither noise.
In a 192 Khz sampling system.
If peak playback level at 16 bit resolution is in fact 120dB SPL, you )(*&(*& well should be able to hear the lack of dither in a piano decay.
You should also be able to hear the difference between TPDF and RPDF dither, too, via noise modulation.
On the other hand who in the name of Monty Python would listen at that level? :horrors:
For that matter, in a decent listening room, someone ought to be able to hear the (*&*(&(*&( dither as well.
Which leads to the question: What in the heck are these people doing? Trying to blow up their equipment, or destroy their hearing, or what?
I don't entirely know what was set up, and I don't want to know at this point.
All I can say is: Are you the type of ruffian that would say "ni" to an old woman?
oh, stop *it*.
In a 192 Khz sampling system.
Yes, and 16-bit TPDF noise density would still be lower in a 44.1 kHz channel, crossing the fig. 3 192 kHz test signal noise floor at ~15 kHz, which is already far below the hearing threshold of humans.
Also, why would it matter? According to their interpretation of the results less bits = more hi-res like sound at 44.1 kHz.
Guess I'll keep my CDs...and my hearing
Damnit, Janet, there is a reason my avatar has a manure fork in hand...
Damnit, Janet, there is a reason my avatar has a manure fork in hand...
I thought that was for the well manicured vegetable garden....
Damnit, Janet, there is a reason my avatar has a manure fork in hand...
We could do with its services here.
(http://amirviews.smugmug.com/photos/i-FKcswgv/0/M/i-FKcswgv-M.png)
I know, it's *something*, but *not much*, it seems.
TPDF dither was used
Yep, by the BS crew...and here is the result:
This gets so tedius!
From:
The audibility of typical digital audio filters in a high- delity playback system AES paper 9174
Page 5
"
After filltering with either FIR fi lter, the signals
were either unchanged or were quantized to 16-
bit. The quantization either included RPDF (rect-
angular probability density function) dither or did
not.
Page 8
"A within-subjects ANOVA was performed on the
data for all eight listeners with two factors: filter
cutoff frequency (low or high) and quantization
type (none, 16-bit, or 16-bit plus RPDF dither).
Page 9:
"
Post-hoc Fisher tests based on a least-signi cant dif-
ference of means at a 5% level of 0.1394 for quanti-
zation showed that performance was worse for 16-bit
quantization plus RPDF dither (mean=1.773) than
for no quantisation (mean=1.902) or 16-bit quanti-
sation alone (mean=1.909).
Page 9-10
"Every condition where 16-bit quantization was ap-
plied gave performance that was signi cantly better
than chance. Performance was signi cantly worse
for 16-bit quantization plus RDPF dither than for
no quantisation or 16-bit quantization alone. This
suggests that the e ect of adding the RPDF dither
on top of the 16-bit quantization and FIR ltering
was to make it more dicult to identify that process-
ing had been applied to the signal, which is perhaps
counterintuitive. To try to explain this nding, we
turn to subjective descriptions of what listeners de-
scribed hearing in these tests.
"
Page 10
"It is possible that that these two qualitative ef-
fects on sound quality cancelled out each other for
the conditions where 16-bit quantization plus RPDF
dither was applied, which could explain why these
signals were harder to discriminate from their un l-
tered counterparts."
"All forms of processing tested here were audi-
ble, except for one condition where performance was
signi cantly di erent from chance at the 6.7% level,
including emulated downsampling lters at standard
sample rates and 16-bit quantization with or without
RPDF dither. Di erences were demonstrated here
in a double-blind test using non-expert listeners who
received minimal training.
"16-bit quantization with and without RPDF
dither can have a deleterious e ect on the listen-
ing experience in a wideband playback system.
On the other hand who in the name of Monty Python would listen at that level? :horrors:
I've expressed my surprise about the high playback levels before.
One of the first recordings I made as a student recording engineer was Bruckner's 9th symphony. 16 bit was the best we had in the 80's. We had made a nice balance in the control room and were quite happy with the playback level. We didn't miss detail. Then the conductor came for a listen. "louder please". So I increased the monitor level by 6 dB, which felt uncomfortable for us, but well, anything for a happy conductor. "more", he said. We ended up with about 12 dB extra gain and still the conductor felt it was too soft, but it would do. We couldn't squeeze more out of our monitors. I guess around 105 dB was the limit, which is definitely not enough when directly compared with a live symphony orchestra.
So was 16 bit enough for the recording? Yes and no. I can't imagine anyone listening at realistic levels in a home situation. It all boils down to the definition of "typical".
Then there's another anecdote about airplane recording
TPDF dither was used
Yep, by the BS crew...and here is the result:
This gets so tedius!
From:
The audibility of typical digital audio filters in a high- delity playback system AES paper 9174
Page 5
"
After filltering with either FIR fi lter, the signals
were either unchanged or were quantized to 16-
bit. The quantization either included RPDF (rect-
angular probability density function) dither or did
not.
Page 8
"A within-subjects ANOVA was performed on the
data for all eight listeners with two factors: filter
cutoff frequency (low or high) and quantization
type (none, 16-bit, or 16-bit plus RPDF dither).
Page 9:
"
Post-hoc Fisher tests based on a least-signi cant dif-
ference of means at a 5% level of 0.1394 for quanti-
zation showed that performance was worse for 16-bit
quantization plus RPDF dither (mean=1.773) than
for no quantisation (mean=1.902) or 16-bit quanti-
sation alone (mean=1.909).
Page 9-10
"Every condition where 16-bit quantization was ap-
plied gave performance that was signi cantly better
than chance. Performance was signi cantly worse
for 16-bit quantization plus RDPF dither than for
no quantisation or 16-bit quantization alone. This
suggests that the e ect of adding the RPDF dither
on top of the 16-bit quantization and FIR ltering
was to make it more dicult to identify that process-
ing had been applied to the signal, which is perhaps
counterintuitive. To try to explain this nding, we
turn to subjective descriptions of what listeners de-
scribed hearing in these tests.
"
Page 10
"It is possible that that these two qualitative ef-
fects on sound quality cancelled out each other for
the conditions where 16-bit quantization plus RPDF
dither was applied, which could explain why these
signals were harder to discriminate from their un l-
tered counterparts."
"All forms of processing tested here were audi-
ble, except for one condition where performance was
signi cantly di erent from chance at the 6.7% level,
including emulated downsampling lters at standard
sample rates and 16-bit quantization with or without
RPDF dither. Di erences were demonstrated here
in a double-blind test using non-expert listeners who
received minimal training.
"16-bit quantization with and without RPDF
dither can have a deleterious e ect on the listen-
ing experience in a wideband playback system.
I guess the question then becomes, exactly how "typical" is RPDF dither in a "typical" HiFi playback system in 2014?
Outside a certain Meridian test lab not using their own 518, etc. players, anyone know?
cheers,
AJ
Damnit, Janet, there is a reason my avatar has a manure fork in hand...
BS level control?
cheers,
AJ
TPDF dither was used
Yep, by the BS crew...and here is the result:
This gets so tedius!
From:
The audibility of typical digital audio filters in a high- delity playback system AES paper 9174
Page 5
"
After filltering with either FIR fi lter, the signals
were either unchanged or were quantized to 16-
bit. The quantization either included RPDF (rect-
angular probability density function) dither or did
not.
Page 8
"A within-subjects ANOVA was performed on the
data for all eight listeners with two factors: filter
cutoff frequency (low or high) and quantization
type (none, 16-bit, or 16-bit plus RPDF dither).
Page 9:
"
Post-hoc Fisher tests based on a least-signi cant dif-
ference of means at a 5% level of 0.1394 for quanti-
zation showed that performance was worse for 16-bit
quantization plus RPDF dither (mean=1.773) than
for no quantisation (mean=1.902) or 16-bit quanti-
sation alone (mean=1.909).
Page 9-10
"Every condition where 16-bit quantization was ap-
plied gave performance that was signi cantly better
than chance. Performance was signi cantly worse
for 16-bit quantization plus RDPF dither than for
no quantisation or 16-bit quantization alone. This
suggests that the e ect of adding the RPDF dither
on top of the 16-bit quantization and FIR ltering
was to make it more dicult to identify that process-
ing had been applied to the signal, which is perhaps
counterintuitive. To try to explain this nding, we
turn to subjective descriptions of what listeners de-
scribed hearing in these tests.
"
Page 10
"It is possible that that these two qualitative ef-
fects on sound quality cancelled out each other for
the conditions where 16-bit quantization plus RPDF
dither was applied, which could explain why these
signals were harder to discriminate from their un l-
tered counterparts."
"All forms of processing tested here were audi-
ble, except for one condition where performance was
signi cantly di erent from chance at the 6.7% level,
including emulated downsampling lters at standard
sample rates and 16-bit quantization with or without
RPDF dither. Di erences were demonstrated here
in a double-blind test using non-expert listeners who
received minimal training.
"16-bit quantization with and without RPDF
dither can have a deleterious e ect on the listen-
ing experience in a wideband playback system.
I guess the question then becomes, exactly how "typical" is RPDF dither in a "typical" HiFi playback system in 2014?
Outside a certain Meridian test lab not using their own 518, etc. players, anyone know?
The dither in modern recordings can come from many places. Its function needs to be part of many common signal processing operations in AVRs such as gain and volume controls.
I did some quick tests to see if one can easily tell the difference between no dither, RPDF dither and TPDF dither being applied to a conversion of a pure -60 dB sine wave from 32 bit floating point to 16 bit fixed point. The effects of no dither is clearly shown in the output 65 k point FFT analysis, but the difference between RPDF and TPDF is not clear at all to me.
If we are talking common signals with noise floors around FS -70 dB, none of the options make any apparent difference due to self-dithering. For example mixing a -60 dB sine wave with -70 dB brown noise (fair simulation of acoustical noise from natural sources) is sufficient to render the 32 bit -> 16 bit conversion free of any of the visible artifacts related to quantization error. Nothing coherent in the 65k point FFT noise floor other than the original 1 KHz signal.
Damnit, Janet, there is a reason my avatar has a manure fork in hand...
We could do with its services here.
He may well need a shovel in each hand, or a backhoe, before this ones said and done.
cheers,
AJ
I did some quick tests to see if one can easily tell the difference between no dither, RPDF dither and TPDF dither being applied to a conversion of a pure -60 dB sine wave from 32 bit floating point to 16 bit fixed point. The effects of no dither is clearly shown in the output 65 k point FFT analysis, but the difference between RPDF and TPDF is not clear at all to me.
Much of this seems academic to me.
I've lost track of how many "Hi End" audio shows I've attended. I've personally exhibited my own products in at least 10, helped with more. I'm a member of the local audiophile club, with in excess of 100 members, that meets at minimum once a month (including today!) "officially", far more unofficially. Plus we occasionally link up with other clubs within FL.
So I have a rather large exposure to "Hi End" audiophiles who would be the exact target of "Hi-Re$".
I would guesstimate, based on this exposure, that less than 2% of "audiophiles" listen to classical music. The only type of music that might remotely approach the dynamic range/noise limits of 16/44, at absurd levels, in very quiet, highly atypical rooms for said audiophiles.
I've also lost count of how many, upon entering room at shows, have asked, is that "Hi-Re$"?
Only to be told, nope, just good ol' "low res" 16/44...using WMP (thanks JJ...and oh, yes, Amir ). I'll then be asked about 99% of the time, to play something akin to Pink Floyd/Neil Young, NOT classical.
Yes, I know, largely anecdotal....but I seriously wonder just how "typical" these contrived test conditions of the BS paper are, never mind how "typical" the "emulated" filters are, the playback levels for string quartet music, etc, etc....or the lack of any system transparency data.
cheers,
AJ
Sorry, no. Dither is an optional processing step in quantization. The term quantization therefore has no dependency whatsoever on dither being used or not.
Exactly, so why call it "16 bit quantization" when it could be a more unambiguous "truncated to 16 bits", for example.
Truncation to fewer bits is re-quantization. And per above, it is customary to drop the "re" from that.
Kees de Visser talked about "truncation" but you cite some paragraph about dithered quantization and the term "requantization" which nobody even mentioned. So I have to assume you're confused again.
The terms that Amir and the paper's authors used have been used in AES publications for decades. Just because they're not the terms we might use here does not make them wrong. As Amir said, the intended audience would understand them. This quantisation vs truncation argument is misguided IMO.
This isn't targeted at you xnor, but I think
some of the other arguments we've had in this thread show a lack of understanding. When I finally got hold of the paper, I wondered if everyone who claimed to have read it really had.
I'm not trying to sound clever. I'm suggesting that some people on
both sides are so keen to argue that they don't take the time to figure out what the words they've read really mean.
Cheers,
David.
The effects of no dither is clearly shown in the output 65 k point FFT analysis, but the difference between RPDF and TPDF is not clear at all to me.
If we are talking common signals with noise floors around FS -70 dB, none of the options make any apparent difference due to self-dithering. For example mixing a -60 dB sine wave with -70 dB brown noise (fair simulation of acoustical noise from natural sources) is sufficient to render the 32 bit -> 16 bit conversion free of any of the visible artifacts related to quantization error. Nothing coherent in the 65k point FFT noise floor other than the original 1 KHz signal.
Remove any extra noise and compare the RPDF dither noise floor with just a -60 dB sine vs. a -120 dB sine. You should see that the noise changes with signal level with RPDF.
With all the loads of room/acoustic noise in the real-world recording, none of this really seems to make a difference.
... I did some quick tests to see if one can easily tell the difference between no dither, RPDF dither and TPDF dither being applied to a conversion of a pure -60 dB sine wave from 32 bit floating point to 16 bit fixed point. The effects of no dither is clearly shown in the output 65 k point FFT analysis, but the difference between RPDF and TPDF is not clear at all to me.
The problem as I understand it is RPDF dither [unlike TPDF] suffers from noise modulation, key word "modulation", so the problem wouldn't occur from a fixed level recording which is I believe what you attempted.
Remember the rival tape NR system to Dolby, "dbx"? It too was a complimentary, dynamic range "compansion" system [COMPression applied during recording/ re-expANSION in playback] but it worked over a broader range. Although the complaints about it were overblown (in my opinion) it was true that under some odd scenarios it suffered from "pumping and breathing artifacts". The classic one (that comes to mind) was attempting to record the solo heart beat from the opening cut of Pink Floyd's DSotM. Although different in causation, I think a dynamically changing noise floor is RPDF dither's similar issue.
edit to add: If we had access to the tracks used here we'd have some examples to listen to: http://www.coe.montana.edu/ee/rmaher/aes_c...information.pdf (http://www.coe.montana.edu/ee/rmaher/aes_c...information.pdf) My links often fail in this forum, I don't know why, the article is:
Quantization, Dither, and Noise Shaping Demonstrations. by . Stanley P. Lipshitz,
I think I found them! [Exploring them now]: http://www.coe.montana.edu/ee/rmaher/aes_cd/1.2.htm (http://www.coe.montana.edu/ee/rmaher/aes_cd/1.2.htm)Bingo! As I suspected, RPDF dither, for instance used in tracks 18 -22, is similar to "pumping and breathing" artifacts, whereas track 23, made with TPDF dither, is immune to this noise modulation problem. [Track 1 is the clean, unmodified, reference version.]
The terms that Amir and the paper's authors used have been used in AES publications for decades. Just because they're not the terms we might use here does not make them wrong. As Amir said, the intended audience would understand them. This quantisation vs truncation argument is misguided IMO.
This isn't targeted at you xnor, but I think some of the other arguments we've had in this thread show a lack of understanding. When I finally got hold of the paper, I wondered if everyone who claimed to have read it really had.
I'm not trying to sound clever. I'm suggesting that some people on both sides are so keen to argue that they don't take the time to figure out what the words they've read really mean.
Cheers,
David.
Agreed . 'quantization' vs 'requantization' is not the droid we are looking for.
'Quantization' is just the digital representation of amplitude. Going from 24bit to 16 bit, the 're' is hardly necessary, it's still a quantization of amplitudes.
But it would have been good for the Meridian folk to have been more explicit as to the order of downconversion operations (i.e, application of dither). It would hardly have taken up significantly more space, just a minor re-phrasing.
Perhaps when/if this is published formally in JAES, there will be some modifications, or links to supplementary materals.
I think the suggestion that they'd get dither and quantisation the wrong way around, and not mention it, is going too far.
I think the suggestion that they'd get dither and quantisation the wrong way around, and not mention it, is going too far.
Sorry if I misread the phrase (pg.10): "this suggests that the effect of adding the RPDF dither on top of the 16-bit quantization and FIR filtering was to make it more difficult to identify that processing had been applied to the signal, which is perhaps counterintuitive."
Is this unambiguous for a native English reader like you ?
I think the suggestion that they'd get dither and quantisation the wrong way around, and not mention it, is going too far.
Sorry if I misread the phrase (pg.10): "this suggests that the effect of adding the RPDF dither on top of the 16-bit quantization and FIR filtering was to make it more difficult to identify that processing had been applied to the signal, which is perhaps counterintuitive."
Is this unambiguous for a native English reader like you ?
It comes down to how you interpret 'adding....on top of'. In most contexts a native English reader would interpret that as meaning: 'after'. (Well, this native English reader would, at least )
But in the context of an peer-reviewed prize-winning AES presentation, it would be rather remarkable of them to have done it the 'wrong' way around, like that. (Though stranger things have happened in science papers.....) So I guess we should assume that 'adding dither on top of quantization and filtering ' here means, the combined effects of dither, truncation, and filtering [done in the right order] was to make it more difficult etc....
It would be very easy to disambiguate in the next draft.
Without having read the paper, I would take them at their word: dither on top of (ie: after) 16-bit quantization.
But it didn't appear to matter:
(http://amirviews.smugmug.com/photos/i-FKcswgv/0/M/i-FKcswgv-M.png)
Without having read the paper, I would take them at their word: dither on top of (ie: after) 16-bit quantization.
But it didn't appear to matter:
(http://amirviews.smugmug.com/photos/i-FKcswgv/0/M/i-FKcswgv-M.png)
In this interpretation of the phrase on p.10, the Meridian researchers purposely did something 'wrong' that they would *expect to* matter...yet it didn't
That works too.
A little disambiguation would go a long way.
(Amir posts his images to a service called 'smugmug.com'? That's awesome. )
It comes down to how you interpret 'adding....on top of'. In most contexts a native English reader would interpret that as meaning: 'after'. (Well, this native English reader would, at least )
You would unless there was good reason not to, when the phrase would mean "in addition to" without prescribing that the thing "on top of" the others came last.
I would be surprised if the authors would call noise added after quantisation "dither".
Maybe I'm being too generous, but (apart from the conjecture sections) I found the paper cautious, honest and straightforward. It just needs a few more details.
Cheers,
David.
I would be surprised if the authors would call noise added after quantisation "dither".
Indeed. The paper is written in a somewhat academic style and would be primarily aimed at people with a knowledge of the subject area. It is elementary knowledge that for dither to be beneficial it must be added prior to the reduction in bit depth, or in conjunction with it. To have simply added noise to the signal, after an earlier quantisation process by truncation, would have made no sense. It would have been bizarre. Even a non-ideal implementation of dither would not consist merely of adding random noise to an already truncated signal. That would not be adding dither. That would be adding noise.
So, to make sense, the reference to adding dither "on top of" the quantisation to 16 bits must be read as "
as an additional factor present in the processing", not as "
subsequent to earlier processing".
To use a loose analogy, if at a meal a person consumes additional alcohol by way of drinking wine "on top of" consuming a Christmas pudding made with rum, there is no suggestion as to exactly when the person drank the wine in relation to when they consumed the pudding. The important point is that the consumption of the wine was not the only source of alcohol for the meal, whether the wine was drunk early in the meal, or late in the meal. It was "on top of" the effect of the alcohol present in the pudding. It could be that everyone at the meal had the Christmas pudding but only a few chose to have wine "on top of" that.
I see a thesaurus entry at http://www.thesaurus.com/browse/on+top+of (http://www.thesaurus.com/browse/on+top+of) as follows:
on top ofalso adv. in addition to
additionally
again
along
along with
and
as well
as well as
besides
conjointly
further
furthermore
in conjunction with
in like manner
including
likewise
more
more than that
moreover
on top of
over and above
plus
still
to boot
together with Elsewhere in the paper we see further detail. On page 5:
"After filtering with either FIR filter, the signals were either unchanged or were quantized to 16-bit. The quantization either included RPDF (rectangular probability density function) dither or did not. We chose to use undithered quantization as a probe and -- although we would normally recommend TPDF dither for best practice -- we considered rectangular dither to be more representative of the non-ideal dither or error-feedback processing found in some commercial A/D and/A filters."
I note that error-feedback is used in certain dither implementations. What we might question is the claim that rectangular dither would be representative of non-ideal situations. It begs the question how common such non-ideal situations might be.
Maybe I'm being too generous, but (apart from the conjecture sections) I found the paper cautious, honest and straightforward. It just needs a few more details.
I would agree.
I found the conjectural comment on page 10, "this suggests that the effect of adding the RPDF dither on top of the 16-bit quantization and FIR filtering was to make it more difficult to identify that processing had been applied to the signal, which is perhaps counterintuitive", puzzling. If a recording of music has very low noise, then a simple truncation could be expected to give rise to audible distortion at critical points in the recording (e.g. the tail end of reverberation). If even a non-ideal dither is applied at the time of quantization, in this case RPDF dither, that would reduce the distortion. So I am not sure why the authors suggested that RPDF dither apparently aiding transparency, was "perhaps counterintuitive". Perhaps what they had in mind was the possibility of audible noise modulation with RPDF dither.
Make pedantic excuses for what amounts to poor communication all you want, the data suggests the issue is moot. Does it not?
Make pedantic excuses for what amounts to poor communication all you want, the data suggests the issue is moot. Does it not?
I'm not sure I'd agree with some of the posters in this thread that the communication was all that poor, particularly if one is able to read the paper as a whole, but I would agree that the data as presented suggests the question is moot, there being such a low percentage of correct answers (not much above 50%, albeit "statistically significant") when the results from all sections are lumped together.
The real issue to my mind is the suggestion that filtering
by itself was allegedly audible. The question of quantization and whether or not dither was applied becomes secondary. I expressed that view earlier in this thread when I said:
For those who have not read the whole of this thread, I'd note that the question of the use of no dither when quantising to 16-bits (test conditions 2 and 5), or rectangular dither when quantising to 16-bits (test conditions 3 and 6) could be regarded as a subsidiary matter. This is because the paper reports that there was a statistically significant correct identification of an audible difference for test condition 1, i.e. filtering to emulate a resampling to 44.1kHz, and without any quantisation to 16 bits. (As for conditon 4, a filtering to emulate resampling to 48kHz, and without any quantization to 16 bits, "the t-test just failed to reach significance at the 5% level". )
If it is true that a mere filtering at 24-bit depth for the 22.05kHz Nyquisit limit of 44.1kHz sampling was of itself identifiable (in particular for certain "high yield" [easier to spot differences] sections of the music), it becomes a subsidiary matter what effect a subsequent quantisation to 16-bits might have had. The "damage" or "impairment" had already occurred, or so it would appear.
I think it would be helpful if two or three of the "high yield" sections (which are relatively short) could be released in their reference 24/192 form, together with their MATLAB filtered forms (the 48kHz sample rate filter emulation, and the 44.1kHz emulation). Then anyone with a system capable of playing back 24/192 files, with ABX switching, could listen for themselves. The listener could then determine whether they could hear any difference at all; and, if so, whether this made a significant difference to the listening experience or was at the outer limits of their auditory perception. They could also provide subjective comments.
Given that high definition versions of the reference files (but apparently with a reduction of around 2dB in amplitude)
have been available for download free of charge for some years, it is hard to understand why some short sections of the reference files could not be hosted, without objection by the copyright holder.
I think it would be helpful if two or three of the "high yield" sections (which are relatively short) could be released in their reference 24/192 form, together with their MATLAB filtered forms (the 48kHz sample rate filter emulation, and the 44.1kHz emulation). Then anyone with a system capable of playing back 24/192 files, with ABX switching, could listen for themselves.
It will never happen. Making the tests accessable to all isn't in the best interests of the hi-re$ promoters, and besides, even if any of us somehow got a hold of the test segments all failures to obtain any statistically significant differentiations has already been cleverly dealt with
, preemptively, built right into the wording of the 2nd conclusion of the opening abstract itself:
"Two main conclusions are offered: first, there exist audible signals that cannot be encoded transparently by a standard CD;
and second, an audio chain used for such experiments must be capable of high-fidelity reproduction."
Any attempts using a lesser, budget conscious, say, um, $23K speaker instead of his $46K ones used for the test will be dismissed as "not being of high enough fidelity". [From Audiophile 101, RULE#1: Attack your critics' gear as being "pedestrian" and of "low resolution", so they wouldn't hear the difference themselves.]
As I pointed out earlier, there was no evidence to support that second conclusion. A cheap mini-system from a department store might have done just as well for all we know, but putting this line in the paper's opening abstract helps keep the truth seekers at bay.
... I did some quick tests to see if one can easily tell the difference between no dither, RPDF dither and TPDF dither being applied to a conversion of a pure -60 dB sine wave from 32 bit floating point to 16 bit fixed point. The effects of no dither is clearly shown in the output 65 k point FFT analysis, but the difference between RPDF and TPDF is not clear at all to me.
The problem as I understand it is RPDF dither [unlike TPDF] suffers from noise modulation, key word "modulation", so the problem wouldn't occur from a fixed level recording which is I believe what you attempted.
That is correct, and my first set of tests was thus in error. Thank you for the assistance!
My second set of tests prepared 8 32 bit files.
Files 1 & 2 were 32 bits with a -60 dB sine wave
Files 3 & 4 were 32 bits with a -120 dB sine wave
I converted 1 & 3 to 16 bits using RPDF dither
I converted 2 & 4 tp 16 bits with TPDF dither
I converted all files back to 32 bits and notched out the 1 KHz tone in each
The two TPDF files contained noise at the same level -96 dB as expected
The RPDF files with dramatically different signal levels contained noise at dramatically different levels -151 dB and -96 dB
Noise modulation due to RPDF dither was thus confirmed. The noise floor varied dramatically in the RPDF dither files as the signal level changed dramatically. With TPDF dither the noise was greater than with RPDF in one case, but it was consistent. The consistent noise floor would be appreciated by listeners. Audible noise modulation bugs me and most other people that I've seen exposed to it.
Files 5 & 6 were 32 bits with a -60 dB sine wave and -70 dB brown (double pink) noise to simulate architectural noise in recording
Files 7 & 8 were 32 bits with a -120 dB sine wave and -70 dB brown (double pink) noise to simulate architectural noise in recording
I converted 5 & 7 to 16 bits using RPDF dither
I converted 6 & 8 to 16 bits using TPDF dither
I converted all files back to 32 bits and notched out the 1 KHz tone in each.
All 4 files had the same noise level -82 dB.
Thus adding simulated consistent architectural noise at a reasonable level (which constituted relatively high level Gaussian dither) was able to erase any observable difference between TPDF and RPDF dither at the levels they would be used in real world conversions to 16 bits.
The PDF of Gaussian dither is far closer to TPDF dither than RPDF dither. In this example, the RPDF and TPDF dithers had a flat PSD, while the brown noise dither had a PSD with a very pronounced downward slope.
Elsewhere in the paper we see further detail. On page 5:
"After filtering with either FIR filter, the signals were either unchanged or were quantized to 16-bit. The quantization either included RPDF (rectangular probability density function) dither or did not. We chose to use undithered quantization as a probe and -- although we would normally recommend TPDF dither for best practice -- we considered rectangular dither to be more representative of the non-ideal dither or error-feedback processing found in some commercial A/D and/A filters."
I note that error-feedback is used in certain dither implementations. What we might question is the claim that rectangular dither would be representative of non-ideal situations. It begs the question how common such non-ideal situations might be.
Maybe I'm being too generous, but (apart from the conjecture sections) I found the paper cautious, honest and straightforward. It just needs a few more details.
I would agree.
I found the conjectural comment on page 10, "this suggests that the effect of adding the RPDF dither on top of the 16-bit quantization and FIR filtering was to make it more difficult to identify that processing had been applied to the signal, which is perhaps counterintuitive", puzzling. If a recording of music has very low noise, then a simple truncation could be expected to give rise to audible distortion at critical points in the recording (e.g. the tail end of reverberation). If even a non-ideal dither is applied at the time of quantization, in this case RPDF dither, that would reduce the distortion. So I am not sure why the authors suggested that RPDF dither apparently aiding transparency, was "perhaps counterintuitive". Perhaps what they had in mind was the possibility of audible noise modulation with RPDF dither.
In almost every recording that is converted to 16 bits the dither added to the conversion itself is far smaller in amplitude than residual noise that is already in the recording due to other noise sources in the production chain. In the days of analog recording the tape machines themselves were major sources of noise. In modern times the musicians, the room, the microphones, and the other analog components in the production chain are the major sources of noise. There may be some deterministic components to this room tone or background noise, but in a well-controlled environment the deterministic components provide the smaller contributions. Deterministic noises are generally asynchronous with the music and can still help address correlated quantization distortion, which is the problem that dither is there to address.
In any case TPDF dither in appropriate quantities should be added as a safeguard and it generally is inherent in the very common Sigma Delta ADCs, but other natural or at least inherent non-digital sources in the recording process generally render the details of dithering of the digital conversion moot.
Thank you for the academic exercise, speculation about noise and making mention of self-dither.
Let's get back on topic now.
"Two main conclusions are offered: first, there exist audible signals that cannot be encoded transparently by a standard CD;"
The authors cannot conclude that from these tests. There are two possible effects that caused these results:
1. The filters had an effect on the audible portion of the signal.
2. Allowing the ultrasonic frequencies to remain in the signal caused the introduction of audible artifacts due to non-linearities in the processing/amplification/transduction chain.
Until some effort is made to measure and analyze the signal that is actually radiated by the speaker with and without the filter(s) in place it is not possible to eliminate 2 and conclude 1.
"Two main conclusions are offered: first, there exist audible signals that cannot be encoded transparently by a standard CD;"
The authors cannot conclude that from these tests.
Notice the wording Fred. Are "offered". Not "Are".
You're not buying?
cheers,
AJ
I think it would be helpful if two or three of the "high yield" sections (which are relatively short) could be released in their reference 24/192 form, together with their MATLAB filtered forms (the 48kHz sample rate filter emulation, and the 44.1kHz emulation). Then anyone with a system capable of playing back 24/192 files, with ABX switching, could listen for themselves. The listener could then determine whether they could hear any difference at all; and, if so, whether this made a significant difference to the listening experience or was at the outer limits of their auditory perception. They could also provide subjective comments.
But would the Meridian group approve?....the cognitive load might be too high, you see...:
There is a more general problem with listening tests
of this kind [referring to Meyer & Moran 2007], which concerns the testing procedure.
ABX tests are viewed as the "gold standard" for
objective measures of listening. In an ABX test, a
listener is required to listen to two reference sounds,
sound A and sound B, and then to listen to sound
X, and to decide whether sound X is the same as
sound A or sound B. ABX tests have a high sensi-
tivity, that is, the proportion of true-positive results
out of total positive results is high. However, ABX
tests also have low speci city, meaning that the pro-
portion of true-negative results out of total negative
results can be spuriously low. Translating this into
outcomes in psychophysical tests, the proportion of
the time that a listener scores well on an ABX test
by chance is low, but the proportion of the time that
a listener can score poorly on a test in spite of being
able to discriminate the sounds is high. An ABX
test requires that a listener retains all three sounds
in working memory, and that they perform a min-
imum of two pair-wise comparisons (A with X and
B with X), after which the correct response must be
given; this results in the cognitive load for an ABX
test being high.
from Jackson et al, 2014 "The audibility of typical digital audio filters in a high- fidelity playback system"
Their own test was A/B, where A was always the reference (unfiltered) ; the task for listeners was to decide if B was the same as A or not. They were allowed unlimited switching between A annd B , and were also allowed to adjust playback level (it is not clear whether they were allowed to do this once, at the start of the trials, or throughout the test, or even between A and B).
Listeners also had instant feedback on whether each choice was correct or incorrect.
I think it would be helpful if two or three of the "high yield" sections (which are relatively short) could be released in their reference 24/192 form, together with their MATLAB filtered forms (the 48kHz sample rate filter emulation, and the 44.1kHz emulation). Then anyone with a system capable of playing back 24/192 files, with ABX switching, could listen for themselves.
FWIW, the fragments themselves could be recreated using the Appendix chart that gives start and stop times for each one. That just leaves the Matlab-filtered versions...
I thought the claim that their A/B test was better than an ABX test was strange. I mean, you can always participate in an ABX test just playing A and X if you want, and that turns it into their test.
Sometimes I do that, sometimes I don't. It's nice to have the choice. You don't get the choice with their test.
I like the option to know if I get each trial right though. AFAIK as long as you pick the number of trials beforehand and stick to it anyway then knowing the result of each trial doesn't break the test, does it?
Cheers,
David.
I thought the claim that their A/B test was better than an ABX test was strange. I mean, you can always participate in an ABX test just playing A and X if you want, and that turns it into their test.
Sometimes I do that, sometimes I don't. It's nice to have the choice. You don't get the choice with their test.
I like the option to know if I get each trial right though. AFAIK as long as you pick the number of trials beforehand and stick to it anyway then knowing the result of each trial doesn't break the test, does it?
Cheers,
David.
It shouldn't. I'm assuming the number of trials was decided in advance , as per:
The extract presented in each trial was selected ran-
domly from the 17 sections into which the piece had
been divided based on musical phrases. Twelve trials
were presented within a "block", with the results of
the last 10 being counted; the two uncounted trials
were included in order to familiarise listeners with
the task and processing before beginning the test.
For each block, the type of filtering used was the
same. Each listener completed 2 blocks for each con-
dition, giving a total of 12 blocks per subject.
so, number of trials was pre-set at 12 per block (though only the last 10 were counted, the first 2 being 'warm ups') , and 2 blocks per condition, and 6 conditions, and thus , if I understand correctly, the pre-set total was:
12*2*6 = 144 trials per subject in total, of which
10*2*6 = 120 were counted
For each condition, 10*2 = 20 trials per subject were counted towards results. Results for each condition were reported as pooled results from all 8 subjects. Thus for each condition a total of 20*8 = 160 trials were counted. (as per : "160 trials combined across listeners for each condition", p. 8)
It's interesting that results are not broken down by subject anywhere. (NB Meyer and Moran 1997 didn't do that comprehensively either, though they did report some per-subject results: "The “best” listener score, achieved one single time, was 8 for 10, still short of the desired 95% confidence level. There were two 7/10 results. All other trial totals were worse than 70% correct.")
AFAIK as long as you pick the number of trials beforehand and stick to it anyway then knowing the result of each trial doesn't break the test, does it? Cheers, David.
If one is of the mind that, "People will be under tremendous, undo stress using ABX, having to juggle three distinct sounds in their heads to determine which is which" [I'm not one of those people, but Stuart seems to be] then I'd think one could equally argue that giving feedback to the listener may have adverse effects as well.
Here's an example. In taking a test one is on their honor to not only not cheat, but also to honestly give it their focused attention and to do the best they can, rather than randomly selecting answers without giving it a good listen. Everyone with me so far? OK, say while taking a test, three quarters through, you notice your correct number of responses is exactly 50% of the number of trials taken so far, clearly implying no real ability to differentiate A from B. Tell me, can we REALLY expect such test subjects to continue with the remaining trials giving it their "very best" effort? I doubt it.
Although it still exists in the training mode, mid-test feedback has been
removed from the current foobar ABX v.2 testing, now in beta, and when I saw that I thought it was a good idea. Scolding or praising a test subject during a test ["You are doing great!" vs. "You can't hear a thing. You might as well be flipping a coin."] will influence at least the mood and disposition of the test subject, possibly increasing stress, even if it doesn't technically bias the test results in one specific way or another.
Maybe testees could have done better than ~60% had they used Proper protocol.
If one is of the mind that, "People will be under tremendous, undo stress having to juggle three distinct sounds in their heads to determine which is which" [I'm not one of those people, but Stuart seems to be...
What explain this difficulty then? http://www.avsforum.com/forum/91-audio-the...ml#post26141122 (http://www.avsforum.com/forum/91-audio-theory-setup-chat/1532092-debate-thread-scott-s-hi-res-audio-test-87.html#post26141122)
My selections of A and B were done entirely by my hearing alone, of an extremely subtle difference that took me over an hour to pull off, but the important point is I had no outside assistance from dogs, analyzers, etc.. [Using such external tools as these would have been correctly deemed "cheating", since it is then no longer a test of human hearing at all.]
If the difference was always there, why so much difficulty as to take an hour?
In the parallel thread I post the results of a double blind test with this note from the tester: http://www.hydrogenaud.io/forums/index.php...st&p=883067 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107604&view=findpost&p=883067)
(http://amirviews.smugmug.com/photos/i-VP7kdQT/0/O/i-VP7kdQT.png)
Clearly Stuart is right in the stress that triple stimulus creates when differences are small.
Here's an example. In taking a test one is on their honor to not only not cheat, but also to honestly give it their focused attention and to do the best they can, rather than randomly selecting answers without giving it a good listen. Everyone with me so far? OK, say while taking a test, three quarters through, you notice your correct number of responses is exactly 50% of the number of trials taken so far, clearly implying no real ability to differentiate A from B. Tell me, can we REALLY expect such test subjects to continue with the remaining trials giving it their "very best" effort? I doubt it.
Instead of hypotheticals let's look at real results of a real sample:
foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/11 06:18:47
File A: C:\Users\Amir\Music\AIX AVS Test files\Mosaic_A2.wav
File B: C:\Users\Amir\Music\AIX AVS Test files\Mosaic_B2.wav
06:18:47 : Test started.
06:19:38 : 00/01 100.0%
06:20:15 : 00/02 100.0%
06:20:47 : 01/03 87.5%
06:21:01 : 01/04 93.8%
06:21:20 : 02/05 81.3%
06:21:32 : 03/06 65.6%
06:21:48 : 04/07 50.0%
06:22:01 : 04/08 63.7%
06:22:15 : 05/09 50.0%
06:22:24 : 05/10 62.3%
06:23:15 : 06/11 50.0% <---- difference found reliably. Note the 100% correct votes from here on.
06:23:27 : 07/12 38.7%
06:23:36 : 08/13 29.1%
06:23:49 : 09/14 21.2%
06:24:02 : 10/15 15.1%
06:24:10 : 11/16 10.5%
06:24:20 : 12/17 7.2%
06:24:27 : 13/18 4.8%
06:24:35 : 14/19 3.2%
06:24:40 : 15/20 2.1%
06:24:46 : 16/21 1.3%
06:24:56 : 17/22 0.8%
06:25:04 : 18/23 0.5%
06:25:13 : 19/24 0.3%
06:25:25 : 20/25 0.2%
06:25:32 : 21/26 0.1%
06:25:38 : 22/27 0.1%
06:25:45 : 23/28 0.0%
06:25:51 : 24/29 0.0%
06:25:58 : 25/30 0.0%
06:26:24 : Test finished.
----------
Total: 25/30 (0.0%)
Notice what the feedback loop of results allowed me to do. I was able to positively identify a revealing segment and complete the test successfully. Without that feedback I could not determine that and stay with that segment. Doing this in trial mode does not work because once you think you have found the difference, you have to go and run the test again and by then you may forget what you had heard. The newer foobar abx plug-in makes that near impossible anyway because there is no help with re-selection of the precise segment.
Another variation is second guessing yourself which is a serious, serious problem. You identify a difference and you listen and get a bunch of trials right. Without feedback you may wonder, "what if I am getting this wrong?" That is all that is needed to change the perception you had of the difference. Placebo works both ways. It can easily erase differences or make them sound different. Without feedback you would then get a bunch of trials wrong. With feedback you would know that you got off track and get back on and see confirmation of that in correct answer after correct answer.
Our goals in these tests must be to do everything in our power to discover differences. Not see how many ways we could encourage a negative outcome by handcuffing the listeners.
Although it still exists in the training mode, mid-test feedback has been removed from the current foobar ABX v.2 testing, now in beta, and when I saw that I thought it was a good idea. Scolding or praising a test subject during a test ["You are doing great!" vs. "You can't hear a thing. You might as well be flipping a coin." will influence at least the mood and disposition of the test subject, even if it doesn't technically bias the test results in one specific way or another.
Per above, it is only a "good idea" if you want to force more negative outcomes.
A certain amount of stress is beneficial.
"Two main conclusions are offered: first, there exist audible signals that cannot be encoded transparently by a standard CD;"
The authors cannot conclude that from these tests.
Notice the wording Fred. Are "offered". Not "Are".
You're not buying?
cheers,
AJ
Not buying it without more evidence as to what the panel was really hearing. I'd like to see the tests re-run with different hardware (speakers in particular) and measurements of the nonlinearities in the system in the ultrasonic frequency range before the possibility of a conclusion is even raised.
A certain amount of stress is beneficial.
Taking a test by definition is somewhat "stressful" however we generally want to minimize any additional stress as best we can, otherwise any failure to hear distinctions can be dismissed as being due to the "added stress".
This is why when I designed my blind amp testing of my audiophile, expert listener friend [to settle a small bet with 2:1 odds in his favor] I allowed him to decide almost EVERY single aspect of the test: He picked to the who, what ,where, when, why, and how of the entire test. My only provisions were that the number of trials he selected must show statistical significance to my satisfaction [we ended up agreeing on 16 trials total, >12 correct to win] and since he was a part time recording engineer he wasn't allowed to use his own, private recordings and had to limit himself to any commercially released CDs or SACDs of his choosing. [Of course the amps weren't allowed to be driven beyond their safe operational range, at any time, and were level matched.]
Forms of possible added stress I successfully avoided, which this Stuart et al. paper's listeners might have theoretically complained about, include:
- test listeners weren't in control of what music was selected
- test listeners weren't allowed to practice with the music and gear for an indefinite period of time, of their choosing, before the test
- test listeners weren't in control of the test transition points/segments
- test listeners weren't in control of what switching methodology was used
- test listeners, I assume, were told when and where to show up for testing, whereas my guy picked both
- test listeners didn't select the room
- test listeners didn't select the speakers used and other gear.
My guy received no correct answer feedback, mid-test, but never asked for any either. I think I would have allowed it had he asked prior to the start, but I think he would have been shooting himself in the foot by doing so [at least in retrospect], since his results were barely different than random chance in the end, and seeing that he was only guessing correctly about 50% of the time, mid test, might have potentially upset him and put him in a bad mood.
You didn't get the memo:
Slightly better than chance is a huge win for proponents of hi-res.
This is why when I designed my blind amp testing of my audiophile, expert listener friend....
Expert listener? How did you quantify that? Run him through blind tests of amplifier with controls and he passed them?
, I allowed him to decide almost EVERY single aspect of the test:
And with it failed to follow best practices for finding small differences. In drug research we don't ask the patients what the test protocol should be. Why is it that in audio we think it is whatever the person taking the test should be the test material?
There is real science to how one picks listeners and content. You can't throw all of that out of the window and go on to make a point.
My guy received no correct answer feedback, mid-test, but never asked for any either. I would have allowed it had he asked, but I think he would have been shooting himself in the foot by doing so, since his results were barely different than random chance, and seeing that, mid test, might have potentially upset him.
Sounds like your only purpose for running such a test was to make a fool of your friend which seemingly you did. Don't confuse that with investigation of audio science and real audibility differences. So yes, if that is your goal, you want to do everything in your power to reduces the chances of a positive outcome. Non-trained listener. No controls. No training. No screening out the candidate.
Anyone can administer a bad test. The trick is to not do that....
Maybe testees could have done better than ~60% had they used Proper protocol.
Exactly. Added stress is sometimes bad, and sometimes good, hence my initial position it should simply be avoided.
Mid test feedback can go either way, adding stress and frustration that one isn't doing well, or give praise for success, we don't know which, so I'm now thinking, maybe, it might be best for the individual test subjects to decide ahead of time if they want mid-test feedback prior to the test. [I'm open minded to changing how I stand on this and assume in the history of testing this has been ironed out before, but I just can't find it. ]
One disturbing take on how mid test feedback alters the end results are the (in)famous Milgram Experiments of the 60's. Stress, obedience, praise, reward, punishment, etc. all have a strong influence, so my gut tells me we should get rid of them whenever we can. Blind tests should be blind in ALL regards, no?
A certain amount of stress is beneficial.
Don't see that as good practices in any research I have read. Fortunately it is a theory that we can easily test here and now.
Run a double blind test with the old ABX plug-in that gives you feedback and the new one that doesn't.
Here is me doing exactly that:
Back on topic: for those who missed it, a test of filter audibility has already been attempted right here on HA...
http://www.hydrogenaud.io/forums/index.php?showtopic=68524 (http://www.hydrogenaud.io/forums/index.php?showtopic=68524)
Any new takers?
Cheers,
David.
foo_abx 1.3.4 report
foobar2000 v1.3.5
2014/11/13 08:40:05
File A: C:\Users\Amir\Music\HA Forum Tests\limehouse\limehouse_maximum_phase_100.wav
File B: C:\Users\Amir\Music\HA Forum Tests\limehouse\limehouse_reference.wav
08:40:05 : Test started.
08:40:37 : 00/01 100.0%
08:42:19 : 00/02 100.0%
08:43:22 : 00/03 100.0%
08:44:21 : 01/04 93.8% <--- Difference found.
08:45:14 : 02/05 81.3%
08:45:21 : 03/06 65.6%
08:45:34 : 04/07 50.0%
08:45:43 : 05/08 36.3%
08:45:52 : 06/09 25.4%
08:46:00 : 07/10 17.2%
08:46:10 : 08/11 11.3%
08:46:20 : 09/12 7.3%
08:46:29 : 10/13 4.6%
08:46:39 : 11/14 2.9%
08:46:51 : 12/15 1.8%
08:47:00 : 13/16 1.1%
08:47:10 : 14/17 0.6%
08:47:18 : 15/18 0.4%
08:47:26 : 16/19 0.2%
08:47:34 : 17/20 0.1%
08:47:42 : 18/21 0.1%
08:47:49 : 19/22 0.0%
08:47:55 : Test finished.
----------
Total: 19/22 (0.0%)
And with the new plug-in:
And here is the same using the new Foobar ABX plug-in with signatures:
2014-11-13 09:16:06
File A: limehouse_maximum_phase_100.wav
SHA1: 722dc26db8d4ce666dc03875b2c8d4570d22b521
File B: limehouse_reference.wav
SHA1: e8ad96830d23cad4bba5bf822ce875ae452b9e7c
Output:
DS : Primary Sound Driver
09:16:06 : Test started.
09:16:48 : 01/01
09:16:56 : 02/02
09:17:04 : 03/03
09:17:14 : 04/04
09:17:21 : 05/05
09:17:29 : 06/06
09:17:38 : 07/07
09:17:45 : 08/08
09:17:52 : 09/09
09:18:02 : 10/10
09:18:08 : 11/11
09:18:14 : 12/12
09:18:20 : 13/13
09:18:28 : 14/14
09:18:36 : 15/15
09:18:42 : 16/16
09:18:42 : Test finished.
----------
Total: 16/16
Probability that you were guessing: 0.0%
-- signature --
5b42b06c414b6ba77a3998695bf119a2d57663c0
I used the new plug-in after I knew what to listen for. Despite that, it was far more stressful. Not knowing the outcome until the end leaves you on the edge the whole time you are trying to get the job done.
Who wants to repeat one of their tests with old and new plug-in and represent that the new method is just as good?
Instead of hypotheticals let's look at real results of a real sample:
Yeah, a real sample with a couple milliseconds delay in one of the files ... good job showing off false positive ABX logs, amirm.
File A: C:\Users\Amir\Music\AIX AVS Test files\Mosaic_A2.wav
I was able to positively identify a revealing segment and complete the test successfully.
Yeah, set selection until the time delay produces an audible artifacts, that's it. But if you think you heard something else then please tell us what that revealing segment was and what you heard.
When you're at it, tell us what you hear in this 4 second short test file (http://www.hydrogenaud.io/forums/index.php?showtopic=107588).
Without that feedback I could not determine that and stay with that segment. Doing this in trial mode does not work because once you think you have found the difference, you have to go and run the test again and by then you may forget what you had heard. The newer foobar abx plug-in makes that near impossible anyway because there is no help with re-selection of the precise segment.
Yeah, precise selection is needed to make those misaligned files exhibit differences that are really simple to hear. You don't need an ABX for that.
But this automatic re-selection can be suggested as a feature. The new ABX component is still in beta, you know.
Another variation is second guessing yourself which is a serious, serious problem. You identify a difference and you listen and get a bunch of trials right. Without feedback you may wonder, "what if I am getting this wrong?" That is all that is needed to change the perception you had of the difference. Placebo works both ways. It can easily erase differences or make them sound different. Without feedback you would then get a bunch of trials wrong. With feedback you would know that you got off track and get back on and see confirmation of that in correct answer after correct answer.
As you said, it works both ways. You may think you are back on track but could have just made a lucky guess.
If you reliable can detect differences in training mode then you should be able to nail the test without even looking at the computer screen. If you are not confident in your ability to distinguish the files then you should go back to training mode or finally conclude that you can't hear a difference.
Per above, it is only a "good idea" if you want to force more negative outcomes.
Nope.
It can also be argued that it removes stress ("oh no, that last choice was wrong again").
Instead of hypotheticals let's look at real results of a real sample:
Yeah, a real sample with a couple milliseconds delay in one of the files ... good job showing off false positive ABX logs, amirm.
Nope. This is David's test: http://www.hydrogenaud.io/forums/index.php?showtopic=68524 (http://www.hydrogenaud.io/forums/index.php?showtopic=68524)
Maybe testees could have done better than ~60% had they used Proper protocol.
Or worse.
Really, instead of the weird conditions why didn't they just use
a) a filter that should be easily audible (e.g. <16 kHz cutoff)
b) a steep 44.1 kHz filter (like the one they used)
c) a more real-world like filter that is less steep (they could even use the one that they actually use in their products...)
d) a filter that even a super human shouldn't be able to hear (40+ kHz for example)
to demonstrate that their A-X test and methodology, filtering software, switching software ... works properly?
A simple test to check their tweeter without measurements could have been using b) with HF noise added that has a similar spectrum to the original track..
Nope. This is David's test: http://www.hydrogenaud.io/forums/index.php?showtopic=68524 (http://www.hydrogenaud.io/forums/index.php?showtopic=68524)
Nope. Look at your own post: here (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=883073).
I clearly commented on that one, did I not? Or do you now copy-paste posts and re-post with different logs? I wouldn't be surprised..
edit: easy as a pie
foo_abx 2.0 beta 6 report
foobar2000 v1.3.5
2014-12-01 23:05:37
File A: Mosaic_A2.wav
File B: Mosaic_B2.wav
Output:
DS : Primärer Soundtreiber
23:05:37 : Test started.
23:06:22 : 00/01
23:06:33 : 01/02
23:06:39 : 02/03
23:06:50 : 03/04
23:06:57 : 04/05
23:07:04 : 05/06
23:07:09 : 06/07
23:07:14 : 07/08
23:07:19 : 08/09
23:07:29 : 09/10
23:07:40 : 10/11
23:07:45 : 11/12
23:07:50 : 12/13
23:07:55 : 13/14
23:08:11 : 14/15
23:08:17 : 15/16
23:08:17 : Test finished.
----------
Total: 15/16
Probability that you were guessing: 0.0%
-- signature --
03091b9caae2455d289845ccd68536215f94b3fd
due to the delay between the files.
Clearly Stuart is right in the stress that triple stimulus creates when differences are small.
No one is making you listen to B in an ABX test.
(Foobar2k adds "Y" as well, and no one is making you listen to that either.)
Cheers,
David.
A certain amount of stress is beneficial.
Taking a test by definition is somewhat "stressful" however we generally want to minimize any additional stress as best we can, otherwise any failure to hear distinctions can be dismissed as being due to the "added stress".
This is why when I designed my blind amp testing of my audiophile, expert listener friend [to settle a small bet with 2:1 odds in his favor] I allowed him to decide almost EVERY single aspect of the test: He picked to the who, what ,where, when, why, and how of the entire test. My only provisions were that the number of trials he selected must show statistical significance to my satisfaction [we ended up agreeing on 16 trials total, >12 correct to win] and since he was a part time recording engineer he wasn't allowed to use his own, private recordings and had to limit himself to any commercially released CDs or SACDs of his choosing. [Of course the amps weren't allowed to be driven beyond their safe operational range, at any time, and were level matched.]
Forms of possible added stress I successfully avoided, which this Stuart et al. paper's listeners might have theoretically complained about, include:
- test listeners weren't in control of what music was selected
- test listeners weren't allowed to practice with the music and gear for an indefinite period of time, of their choosing, before the test
- test listeners weren't in control of the test transition points/segments
- test listeners weren't in control of what switching methodology was used
- test listeners, I assume, were told when and where to show up for testing, whereas my guy picked both
- test listeners didn't select the room
- test listeners didn't select the speakers used and other gear.
My guy received no correct answer feedback, mid-test, but never asked for any either. I think I would have allowed it had he asked prior to the start, but I think he would have been shooting himself in the foot by doing so [at least in retrospect], since his results were barely different than random chance in the end, and seeing that he was only guessing correctly about 50% of the time, mid test, might have potentially upset him and put him in a bad mood.
This I think comes from a misunderstanding of stress by using it as a single word when physiologically the way it manifests itself in people can range from highly beneficial to highly detrimental. Certainly, if you don't practice or study, it won't matter if you're relaxed or not. You simply won't do well. But there are areas of stress, which can be measured by heart rate and respiration where stress has positively beneficial effects on one's performance.
Consider this excerpt from Malcolm Gladwell's book,
Blink.
http://nemaloknig.info/read-147015/?page=24 (http://nemaloknig.info/read-147015/?page=24)
While it uses law enforcement and sports to make its point, I believe it's equally applicable to other endeavors. Sharpening one's concentration to the task at hand while minimizing distractions seems beneficial.
Nope. This is David's test: http://www.hydrogenaud.io/forums/index.php?showtopic=68524 (http://www.hydrogenaud.io/forums/index.php?showtopic=68524)
Nope. Look at your own post: here (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=883073).
I clearly commented on that one, did I not? Or do you now copy-paste posts and re-post with different logs? I wouldn't be surprised..
Who knows what topic you are on. We are discussing stress in running tests and its impact on results. You jumped in with your cheating accusation?
But congratulation for posting your first ABX test in this thread. It was not so bad, was it? Now you can speak from experience.
Now do David's test with the old and new plug-in so that when you can't follow mzil's cheat, you can produce the same outcome with the same level of stress. I have provided my results already.
BTW, how are you and mzil creating log files that don't have the hash of the input files? Here is mine again:
File A: limehouse_maximum_phase_100.wav
SHA1: 722dc26db8d4ce666dc03875b2c8d4570d22b521
File B: limehouse_reference.wav
SHA1: e8ad96830d23cad4bba5bf822ce875ae452b9e7c
Here is yours:
File A: Mosaic_A2.wav
File B: Mosaic_B2.wav
Here is Wombat's results which like mine shows the signature:
foo_abx 2.0 beta 4 report
foobar2000 v1.3.4
2014-11-14 16:20:39
File A: limehouse_maximum_phase_100.wav
SHA1: 722dc26db8d4ce666dc03875b2c8d4570d22b521
File B: limehouse_reference.wav
SHA1: e8ad96830d23cad4bba5bf822ce875ae452b9e7c
Output:
DS : Primärer Soundtreiber
Are you using an option to suppress that and if so, why? Mine by default shows the hash for both files just like wombat's.
Were the BS tests run supervised at Meridian, using their speakers/software/files, or on Amirs computer screen, unsupervised?
I thought this thread was about the BS paper doctored dither results, not unsupervised Windows pc generated logs??
Who knows what topic you are on.
This one, but it seems like you're confusing yourself with all the places where you post your logs to not even remember what you posted.
But congratulation for posting your first ABX test in this thread. It was not so bad, was it? Now you can speak from experience.
You gotta be seriously deluded if you think this was even close to a first... Btw, as soon as I saw a mention of the files here I replied that there are problems with the files. That was over 4 months ago.
Now do David's test with the old and new plug-in so that when you can't follow mzil's cheat, you can produce the same outcome with the same level of stress. I have provided my results already.
BTW, how are you and mzil creating log files that don't have the hash of the input files?
What stress? There was no stress. As I said easy as pie even with the new beta's long forces fade ins/outs.
The lack of hashes appears to be a bug when the test is started from within training mode. Already reported a few minutes ago.
Consider this excerpt from Malcolm Gladwell's book, Blink.
http://nemaloknig.info/read-147015/?page=24 (http://nemaloknig.info/read-147015/?page=24)
While it uses law enforcement and sports to make its point, I believe it's equally applicable to other endeavors. Sharpening one's concentration to the task at hand while minimizing distractions seems beneficial.
??? You are drawing an analogy from this:
"This is how the human body reacts to extreme stress, and it makes sense. Our mind, faced with a life-threatening situation, drastically limits the range and amount of information that we have to deal with" to what we are doing?
No this is not a life-threatening situation although one read of xnor's replies you think it is.
The feelings that are involved are:
1. Fear. Fear of losing. These tests are not created in a "safe" situation. One outcome can lead to your ridicule by the most obnoxious people on the planet.
2. Suspicion. That the other side has cooked the test to make a fool of you. This one is deadly and I have fallen victim to it many times. I am taking the test, and I know I have heard the difference. But then think, "what if he is tricking me?" Fear kicks in and I vote the opposite of what I heard, only to find out my first instinct was correct.
3. Frustration. This is what we are calling "stress" for the most part. We are having to remember one complex set of notes and freeze it in our mind, and then perform an aural image match. When differences get small this is not easy to do at all. Even when testing big differences such as Harman's double blind speaker tests, I found myself in a tough spot trying to remember what the last speaker sounded like.
We are asking a human to be a machine with precision. This is never easy.
4. Anticipation. This is the point I made in my prior post with new version of ABX. As the trials counted up, you could just hear my heart rate increase. What if I get to the end and I have to repeat this stupid test again?
A little story about my younger years. I was always good with math. Not perfect but one step below. I remember one time getting 4 or 5 math problems to solve as a homework assignment. I sat there and figured them right out. The next day I come to the class and was amazed that I was the only one to have solved all of them! The excitement did not last long. The final exam comes and what is on the test? A couple of the same problem. Momentary excitement turned into sheer panic as I could not solve either one! I tried everything. I wrote pages and pages of stuff and at the end could not remember how to solve the problem. Barely got a passing grade.
I did not become dumb between the home work assignment and final exam. Clearly something about being tested caused me to all of a sudden get frustrated, worried about running out of time, etc.
This is why we need to do everything in our power to make the test simpler for people to take. Expert listeners and revealing content does that.
If peak playback level at 16 bit resolution is in fact 120dB SPL, you )(*&(*& well should be able to hear the lack of dither in a piano decay.
You should also be able to hear the difference between TPDF and RPDF dither, too, via noise modulation.
On the other hand who in the name of Monty Python would listen at that level? :horrors:
For that matter, in a decent listening room, someone ought to be able to hear the (*&*(&(*&( dither as well.
Which leads to the question: What in the heck are these people doing? Trying to blow up their equipment, or destroy their hearing, or what?
The BS paper claims 102db peaks measured at the LP, not 120db.
cheers,
AJ
Clearly Stuart is right in the stress that triple stimulus creates when differences are small.
No one is making you listen to B in an ABX test.
I have passed the tests as written. The issue is what recruited testers do, not me. If you have an ABX test and you explain it as an ABX test, then they will listen to A, B and X. If you write the instructions to just listen to A and X, then why do you have the B in the user interface? You would eliminate that and that is what Stuart's test has done.
How often do you think someone has taken the ABX test as AX?
No this is not a life-threatening situation although one read of xnor's replies you think it is.
No, it's just reason and rational thinking. You really should try it one day.
I'm not nearly stupid enough to participate in your little concocted war game, sorry.
3. Frustration. This is what we are calling "stress" for the most part.
Yes, and the most frustrating thing is getting feedback that tells you that your choice was wrong during a test. I believe that negative feedback has a stronger impact than positive fb.
Clearly Stuart is right in the stress that triple stimulus creates when differences are small.
Unlike during Meridian's A/B test....
...or not.
It was also possible that listeners would tire during testing sessions; listeners were
able to proceed at their own pace, with the only restriction being that breaks could not be taken in the
middle of a block.
The training and test blocks were completed in under four hours by all listeners (not including breaks).
Jackson et al. (2014) "The audibility of typical digital audio fi lters in a high- fidelity playback system" p8
The fact is, Amir, that Stuart and Co. offered their critique of ABX with no citations of evidence from others' work. I found that
quite peculiar, especially in an Introduction section of a research paper.
It's just, like, their
opinion, man.
cognitive *load*, indeed
How often do you think someone has taken the ABX test as AX?
I have certainly taken ABX tests where I have not had to bother switching between A and B after the first trial or two. I'm sure others do it too, when the difference is obvious enough to them. You 'learn' the difference, and from there on, X is either A or not A (and therefore B, you don't even need to actaully hear B) .
Those, btw , are the sort of difference I'd *expect* hi rez to offer, given the relentless, decade-long hype from your crowd. Difference so obvious, an ABX would be practically effortless.
But it's not. It's not in Stuart's test either. It's not in your own tests. The hype is a lie.
Consider this excerpt from Malcolm Gladwell's book, Blink.
http://nemaloknig.info/read-147015/?page=24 (http://nemaloknig.info/read-147015/?page=24)
While it uses law enforcement and sports to make its point, I believe it's equally applicable to other endeavors. Sharpening one's concentration to the task at hand while minimizing distractions seems beneficial.
??? You are drawing an analogy from this: "This is how the human body reacts to extreme stress, and it makes sense. Our mind, faced with a life-threatening situation, drastically limits the range and amount of information that we have to deal with" to what we are doing?
No this is not a life-threatening situation although one read of xnor's replies you think it is.
The feelings that are involved are:
1. Fear. Fear of losing. These tests are not created in a "safe" situation. One outcome can lead to your ridicule by the most obnoxious people on the planet.
2. Suspicion. That the other side has cooked the test to make a fool of you. This one is deadly and I have fallen victim to it many times. I am taking the test, and I know I have heard the difference. But then think, "what if he is tricking me?" Fear kicks in and I vote the opposite of what I heard, only to find out my first instinct was correct.
3. Frustration. This is what we are calling "stress" for the most part. We are having to remember one complex set of notes and freeze it in our mind, and then perform an aural image match. When differences get small this is not easy to do at all. Even when testing big differences such as Harman's double blind speaker tests, I found myself in a tough spot trying to remember what the last speaker sounded like.
We are asking a human to be a machine with precision. This is never easy.
4. Anticipation. This is the point I made in my prior post with new version of ABX. As the trials counted up, you could just hear my heart rate increase. What if I get to the end and I have to repeat this stupid test again?
A little story about my younger years. I was always good with math. Not perfect but one step below. I remember one time getting 4 or 5 math problems to solve as a homework assignment. I sat there and figured them right out. The next day I come to the class and was amazed that I was the only one to have solved all of them! The excitement did not last long. The final exam comes and what is on the test? A couple of the same problem. Momentary excitement turned into sheer panic as I could not solve either one! I tried everything. I wrote pages and pages of stuff and at the end could not remember how to solve the problem. Barely got a passing grade.
I did not become dumb between the home work assignment and final exam. Clearly something about being tested caused me to all of a sudden get frustrated, worried about running out of time, etc.
This is why we need to do everything in our power to make the test simpler for people to take. Expert listeners and revealing content does that.
You didn't practice enough and overestimated your abilities. You choked. Lots of things aren't life and death Amir but that doesn't mean you don't benefit from getting in the zone.
Maybe testees could have done better than ~60% had they been given the option to audition A and/or B when making a selection for X.
Perhaps they'd have done better had Amir been the instructor.
Maybe testees could have done better than ~60% had they been given the option to audition A and/or B when making a selection for X.
Seems like >95% is possible with ABX, when under the enormous stress of taking the test at home via downloadable online files, on a Windows pc, completely unsupervised.
I wonder why the BS paper wasn't done like this instead? I'm sure it would still have garnered those AES awards.
Maybe they just wanted to showcase their products as well?
cheers,
AJ
I'm sure others do it too
<raises hand>
Difference so obvious, an ABX would be practically effortless.
But it's not. It's not in Stuart's test either.
~60% doesn't obviously indicate the endeavor was practically effortless?
I'm sure it would still have garnered those AES awards.
I'm not so sure it wouldn't have.
The fact is, Amir, that Stuart and Co. offered their critique of ABX with no citations of evidence from others' work.
So? The stated a fact: that in ABX tests the listener is instructed to listen to A and X and B and X and vote which association is closer. That creates three stimulus that must be kept in mind:
An ABX test requires that a listener retains all three sounds
in working memory, and that they perform a min-
imum of two pair-wise comparisons (A with X and
B with X), after which the correct response must be
given; this results in the cognitive load for an ABX
test being high.
Explaining what ABX testing is doesn't require prior reference. And the cognitive load is obvious to anyone who has taken the ABX blind tests.
You have some evidence that this is not true?
I found that quite peculiar, especially in an Introduction section of a research paper.
It's just, like, their opinion, man.
cognitive *load*, indeed
You are not in the audio field Steven so what is peculiar in your world is not here.
What is really at issue here is that folks/you don't want to hear any criticism of ABX testing. That is supposed to be sacrilegious. In real world it is not. Facts can be stated about it and expert opinion expressed.
Here is Clark in his JAES paper,
High-Resolution Subjective Testing Using a Double-Blind ComparatorWhen scientific tests have been performed, listeners'
audibility thresholds have appeared to be poorer by
orders of magnitude compared to casual tests.There is no reference provided in this second paragraph in the Introduction of the paper. And it is vague at heck. What on earth is "casual tests?"
If you believe David's theory, then their testing is a degenerate version of ABX with the same reliability. So why the objection?
And the cognitive load is obvious to anyone who has taken the ABX blind tests.
You have some evidence that this is not true?
Only a person devoid of logic would try to prove a negative, so no surprised you asked.
Amir, please cite your valid, scientific evidence regarding cognitive loads of ABX tests.
The onus falls squarely on you.
There was a somewhat recent thread about ABX (http://www.hydrogenaud.io/forums/index.php?showtopic=101354), where someone with expertise, but no strong pecuniary interests in "Hi-Re$" or misrepresenting M&Ms myth buster had this to say:
There is rather some research on discrimination with time lapse, and it is quite true that a break between A/X or B/X (or even A/B) will disrupt the subject, so time proximity WITHOUT glitches is absolutely requisite. This is because partial loudnesses, rather than time domain waveforms, is what must be recalled, and the first level of memory for such is under 200 milliseconds. So proximate IN TIME presentation is the relevant, germane thing to do.
Since people do perform with such testing down to physical limits, I don't think there is a great deal of room left there for problems.
If you want to do an audio test, how else would you do this? By the way, sequential tests with proper windowing are documented as the best way to extract the most reliable answers from subjects.
cheers,
AJ
Only a person devoid of logic would try to prove a negative, so no surprised you asked.
There is no such thing as "you can't prove a negative." It is a mad up debating phrase/talking point and no more. Just because Randi James says it, doesn't mean it has any proper basis in logic.
Steven Hales is professor of philosophy at Bloomsburg
University, Pennsylvania.
http://departments.bloomu.edu/philosophy/p...veanegative.pdf (http://departments.bloomu.edu/philosophy/pages/content/hales/articlepdf/proveanegative.pdf)
But there is one big, fat problem with all this. Among professional
logicians, guess how many think that you can’t prove
a negative? That’s right: zero. Yes, Virginia, you can prove a
negative, and it’s easy, too.
[...]
So we’d better keep induction, warts and all, and
use it to form negative beliefs as well as positive ones. You
can prove a negative — at least as much as you can prove
anything at all.I can show with ease that you know nothing about signal processing Ammar. That is a negative. And it can be proven using inductive reasoning.
Don't confuse reality with what you read online....
And the cognitive load is obvious to anyone who has taken the ABX blind tests.
You have some evidence that this is not true?
Amir, please cite your valid, scientific evidence regarding cognitive loads of ABX tests.
The onus falls squarely on you.
Also, there was a somewhat recent thread about ABX (http://www.hydrogenaud.io/forums/index.php?showtopic=101354), where someone with expertise, but no strong pecuniary interests in "Hi-Re$" or misrepresenting M&Ms myth buster had this to say:
There is rather some research on discrimination with time lapse, and it is quite true that a break between A/X or B/X (or even A/B) will disrupt the subject, so time proximity WITHOUT glitches is absolutely requisite. This is because partial loudnesses, rather than time domain waveforms, is what must be recalled, and the first level of memory for such is under 200 milliseconds. So proximate IN TIME presentation is the relevant, germane thing to do.
Since people do perform with such testing down to physical limits, I don't think there is a great deal of room left there for problems.
If you want to do an audio test, how else would you do this? By the way, sequential tests with proper windowing are documented as the best way to extract the most reliable answers from subjects.
Please do try to comprehend what JJ has written here, thanks.
cheers,
AJ
Only a person devoid of logic would try to prove a negative, so no surprised you asked.
There is no such thing as "you can't prove a negative." It is a mad up debating phrase/talking point and no more. Just because Randi James says it, doesn't mean it has any proper basis in logic..
The only proof of a universal negative is complete testing of everything, everwhere, under every condition.
So, practically speaking, no, you can not prove a universal negative, you can only offer evidence that the result is the same as the negative to some level of confidence. That is the only proof available in the real world for any universal negative.
So I don't think AJ's comment is so terribly wrong.
So? The stated a fact: that in ABX tests the listener is instructed to listen to A and X and B and X and vote which association is closer. That creates three stimulus that must be kept in mind:
An ABX test requires that a listener retains all three sounds
in working memory, and that they perform a min-
imum of two pair-wise comparisons (A with X and
B with X), after which the correct response must be
given; this results in the cognitive load for an ABX
test being high.
Explaining what ABX testing is doesn't require prior reference. And the cognitive load is obvious to anyone who has taken the ABX blind tests.
You have some evidence that this is not true?
And once again, you quote-mine and omit the context, wherein Meridian is claiming ABX is a *problem*...and then you beg the question by declaring the problem 'obvious'.
You are not in the audio field Steven so what is peculiar in your world is not here.
You're funny. So, Meridian gets a pass? Hmm. Let's come back to that.
What is really at issue here is that folks/you don't want to hear any criticism of ABX testing. That is supposed to be sacrilegious. In real world it is not. Facts can be stated about it and expert opinion expressed.
ah, the Amir Pirouette. One of your favorite steps.
Here is Clark in his JAES paper, High-Resolution Subjective Testing Using a Double-Blind Comparator
When scientific tests have been performed, listeners'
audibility thresholds have appeared to be poorer by
orders of magnitude compared to casual tests.
There is no reference provided in this second paragraph in the Introduction of the paper. And it is vague at heck. What on earth is "casual tests?"
I would make a
wild guess that 'casual tests' means 'sighted tests'...you know,the ones audiophiles 'conduct' all the time when they say things like, high resolution 'lifted the veil' that lies over Redbook. And when blind ('scientific') testing doesn't give the same results, they claim that the method (and equipment) inhibits the *truth*. They did that then; they still do it now. Which is what Clark reports.
Let's observe that Clark's 1982 paper had rather less sensory testing literature to draw on than Meridian's 2014 paper. Still, Clark
should have backed up his claims, too. Clarks' intro is very brief and has no cites; Meridian's 2014 Intro is much longer, and actually *is* cite-filled...*except* for that paragraph where he claims ABX is problematic. Clark's 1982 convention paper was an *engineering report* whereas Meridian's 2014 convention paper is a *prize winning peer reviewed research paper*, as you told us several times, like a child parading his new birthday toy. So that omission of cited for that quite bold claim of theirs, seems quite strange. But hey, I only deal with peer-reviewed scientific papers -- reading, writing, editing, reviewing -- on a regular basis between doing science, what do I know? You run a high end audio shop, and invest your time, emotions, and boundless energy in googling and copy-pasting tidbits from the Internet, championing 'high resolution' digital formats that offer miniscule audible difference, and ankle-biting Arny Kruger. Which makes you an expert in
something, I guess.
But whatever it is, no one's buying it here.
Amir, you might also want to address what 'stress' that Meridian's A/B trials entailed, given that they reportedly took over three hours per subject, plus breaks, to accomplish.
Is 'cognitive load' the only stressor? And might there be stressors associated with A/B that are lessened by A/B/X? Meridian at least should have hedged their bets and stuck in some 'more research is needed' boilerplate on that score. Instead they simply recommend 'consideration' of the use of tests that 'minimize cognitive load' -- as if the supposed 'problem ' had been demonstrated and the 'solution' found.
Maybe they'll fix it in the next draft?
The only proof of a universal negative is complete testing of everything, everwhere, under every condition.
We don't live in idealistic world. My toaster is not unsafe because it has been through a set of UL tests that give high confidence of it not catching on fire. I certainly don't worry about it combusting when I put toast in it every morning.
All of these audio tests with negative outcomes are used exactly the same way. We use them to inform us that there is no difference. In other words, we rely on negative proof all the time in these discussions. Yet for some reason are quick to state, "you can't prove the negative." Why do we say that when we are considering negative outcomes sufficient to have created negative proof?
So, practically speaking, no, you can not prove a universal negative, you can only offer evidence that the result is the same as the negative to some level of confidence. That is the only proof available in the real world for any universal negative.
That's right. Ditto for positive outcome. There is no universal positive proof either. Newton thought he had it and then came Einstein. We thought we had it with Einstein only to have "spooky action at a distance" come true. Yet I sit there relying on Newton having been right enough. Lack of its universality is not an issue. The only time Einstein comes into play is when I use my navigation system.
When I look at audio listening tests, there are so many factors that could render their outcomes unreliable whether positive or negative. Both outcomes are interesting equally to examine and learn from.
So I don't think AJ's comment is so terribly wrong.
Wonder what he thinks of being damned by faint praise this way. I hope you know he has no technical intent in any of these discussions....
Edit:typos.
And the cognitive load is obvious to anyone who has taken the ABX blind tests.
You have some evidence that this is not true?
Amir, please cite your valid, scientific evidence regarding cognitive loads of ABX tests.
The onus falls squarely on you.
(http://amirviews.smugmug.com/photos/i-D6dPFbG/0/O/i-D6dPFbG.png)
The only proof of a universal negative is complete testing of everything, everwhere, under every condition.
We don't live in idealistic world. My toaster is not unsafe because it has been through a set of UL tests that give high confidence of it not catching on fire. I certainly don't worry about it combusting when I put toast in it every morning.
All of these audio tests with negative outcomes are used exactly the same way. We use them to inform us that there is no difference. In other words, we rely on negative proof all the time in these discussions. Yet for some reason are quick to state, "you can't prove the negative." Why do we say that when we are considering negative outcomes sufficient to have created negative proof?
How inaccurate you become when it suits you!
Just looking at the supposed 'negative proof' papers at hand -- Meyer & Moran 2007 in JAES on 'hi rez' vs Redbook, Meyer 1991 in Stereo Review on amplifier differences -- neither one makes simple, adamantly conclusive negative 'proof' claims of the sort you're propping up to knock down. They inform us (in one case) that no significant difference was found in their trials (except when it
was, when playback levels were high), and that the onus is on high rez proponents to find one (Meridian's paper is of course a direct response to that); and in the other case, we are informed quite politely that that difference *was* found, under certain circumstances (amplifiers reacting differently to a difficult load), but that most 'good' solid state amps 'probably' sound identical 'or at least very much alike' , 'within their power limits'.
You, once again, are painting 'objectivist' views as being simplistic, sweeping and over-general, rather than being qualified and modified with adjectives like 'likely and 'probably' and pointing out the conditions under which audible differences
do manifest. It's the old subjectivist canard that 'objectivists' believe 'all X sound the same', period, full stop.
Now, why would you propagate such a canard? Is there a 'technical intent' there?
I can't help thinking that a lot of listening test stress could be reduced by contributing anonymously and not turning the test into a whosbest competition.
In most cases it is the device that's under test, not the listener. Participants who don't share their result because they know (feedback) it's negative and think they have failed can have an effect on the accuracy of the test.
IMO the most interesting part of the paper is the audibility of time domain effects of filters. AFAIK the authors only give "potential" explanations. They try to explain why the 44.1 kHz filter was less audible than the 48 kHz filter with the difference in pre- and post-ringing. If that's true, doesn't that mean that a ringing filter in a very high resolution ADC (e.g. DXD) will have a deleterious effect that can't be undone ?
If so, it would be interesting to know how much ringing is tolerable, if this is linked at all to sampling rate, and how cascading filters effects audibility. In a typical recording chain there are three filters: 1) ADC (hi-res), 2) mastering filter (AKA SRC to redbook), 3) DAC (anti-imaging filter).
DSD seems to be immune to this "problem", but I suppose that can only be verified with non-pcm-speakers.
No it is not.
Yes, it is. You defend that position and with your typical amir spiel tried to, once again, shift the burden of proof demanding from others to disprove your position.
The problem here is not proving a negative, but assuming you are right because there is no strong evidence that your position is wrong. That's a pretty absurd and fallacious position to take, but hey, we're not surprised.
If so, it would be interesting to know how much ringing is tolerable, if this is linked at all to sampling rate, and how cascading filters effects audibility. In a typical recording chain there are three filters: 1) ADC (hi-res), 2) mastering filter (AKA SRC to redbook), 3) DAC (anti-imaging filter).
The last filter with the lowest cutoff is what matters. The "mastering filter" can be extremely steep since the anti-imaging (or resampling in your player) filter dictates how an impulse will finally look like.
Just use a slower roll off, min phase resampling filter that allows imaging (aliasing) and all the alleged pre-ringing problems in your CDs will be gone. That's what Meridian says and actually uses in their products...
Maybe testees could have done better than ~60% had they used Proper protocol.
Or worse.
Really, instead of the weird conditions why didn't they just use
a) a filter that should be easily audible (e.g. <16 kHz cutoff)
b) a steep 44.1 kHz filter (like the one they used)
c) a more real-world like filter that is less steep (they could even use the one that they actually use in their products...)
d) a filter that even a super human shouldn't be able to hear (40+ kHz for example)
to demonstrate that their A-X test and methodology, filtering software, switching software ... works properly?
A simple test to check their tweeter without measurements could have been using b) with HF noise added that has a similar spectrum to the original track..
I suspect they didn't have a "low anchor" like (a) because it wouldn't have been a low anchor for the older listeners, but where it worked as a low anchor for the younger listeners it would have shown what easily audible differences look like in this test context. It would have highlighted how weak the audibility of this steep 22kHz filtering is.
There's a possible interesting follow up to all of this, and given there seems little point repeating all the dither tests, it would be a shorter test. First is your option c). Second is their really steep filter, but put essentially all the original HF content back afterwards. So you keep the horrible ringing, but keep almost all of the original HF too.
As a separate test, just play back what's missing from each version and check if that's audible on its own.
Some ultra high quality test recordings of what comes out of the speakers under each condition would be helpful too.
That would help to disentangle high frequency content vs ringing vs intermodulation as the cause of these barely audible differences.
FWIW I would guess that they've listened to option c) plenty, and know the answer already.
Cheers,
David.
IMO the most interesting part of the paper is the audibility of time domain effects of filters. AFAIK the authors only give "potential" explanations. They try to explain why the 44.1 kHz filter was less audible than the 48 kHz filter with the difference in pre- and post-ringing. If that's true, doesn't that mean that a ringing filter in a very high resolution ADC (e.g. DXD) will have a deleterious effect that can't be undone ?
If so, it would be interesting to know how much ringing is tolerable, if this is linked at all to sampling rate, and how cascading filters effects audibility. In a typical recording chain there are three filters: 1) ADC (hi-res), 2) mastering filter (AKA SRC to redbook), 3) DAC (anti-imaging filter).
DSD seems to be immune to this "problem", but I suppose that can only be verified with non-pcm-speakers.
Yes I thought the most interesting aspect the [apparent] reduction in transparency with a filter to emulate a reduction in sample rate. Not so much for the 48kHz emulation (condition 4, where "the t-test just failed to reach significance at the 5% level"), but particularly for the 44.1kHz emulation (condition 1) which -- especially for the "high yield" sections -- was apparently quite an obstacle to transparency.
This BS study lacked investigatory drive in my view. It was very passive. It seems that the study was conceived and implemented, full stop. I note that its two main conclusions are expressed vaguely, or even cryptically.
I'd have thought that preliminary informal tests should have been done with subjects with outstanding hearing and listening skills. This presumably would have shown up high yield (revealing) sections. Then a variation could have been done by way of substituting a different loudspeaker system, or even headphones to see whether a difference was still in evidence.
And if differences could
still be heard, then experimentation with different MATLAB settings for example with a wider transition band for the Nyquist filtering for 44.1kHz, to see whether that aided transparency. Such courses of action would have been constructive.
The way the two "main" conclusions "offered" are expressed is also passive, and not particularly helpful. It is obvious that audible sounds exist that cannot be transparently encoded at 16/44. Some children can hear a loud sine wave at 23kHz and a standard CD cannot render 23kHz at all. Also if a CD is played back with ear-splitting gain, then dither noise may be audible even to middle-aged adults. It is accordingly quite cryptic and unhelpful to offer as a conclusion: "first, there exist audible signals that cannot be encoded transparently by a standard CD". Perhaps what was meant was "audible signals contained in conventional music recorded with high definition and listened to at a normal listening level".
As for the second conclusion "offered", "an audio chain used for such experiments must be capable of high-fidelity reproduction", as someone else has already pointed out in this thread there is no sign the study investigated use of different loudspeaker systems, including hi-fi speakers of modest performance. In any case, the phrase "high-fidelity reproduction" is too vague to be helpful.
In summary,
- the BS experimental study appears to have been executed in a passive way, not taking advantage of preliminary informal testing to pinpoint what was causing an apparent lack of transparency in the filter emulations (particularly the emulation for a sample rate reduction to 44.1kHz), and
- the two main conclusions offered are vaguely expressed and unhelpful, and in the case of the second conclusion (about the need for a hi-fidelity reproduction chain) not actually supported by experiment.
Further commentsFor someone familiar with MATLAB audio filters, it is a walk in the park to alter the parameters. The BS study could have been so much more. In my opinion, it was a narrowly focussed endeavour, with vaguely expressed conclusions.
For high definition proponents the subjective comments included in two consecutive paragraphs on page 10 of the report might lead to wild rejoicing. However it is not clear in the report wording that the comments were screened so as to be limited to those test subjects who scored over 50% correct, with statistical significance, for the particular test condition commented on. The two paragraphs begin, respectively:
It was reported that filtering gave "softer edges" to the instruments, and "softer leading edges" to musical features with abrupt onsets or changes.
Listeners described that quantization gave a "roughness" or "edginess" to the tone of the instruments, ...
It appears that the second quote is in relation to the so-called "probe" test conditions involving quantization to 16 bits
without dither. (Straw man conditions, these ones, as dithering is routine for preparing a CD master from a higher bit depth version.) The last sentence of the paragraph begins :
This could be an effect of quantization distortion; it is interesting that this was audible even in a 24-bit system,...".
Why wouldn't it be? This is a very strange remark. A 32-bit system would still potentially reveal distortion if feed
truncated 16-bits from a low noise undithered source.
The sentence continues:
... and is consistent with the hypothesis of Stuart [29] that 16 bits are not sufficient for inaudible quantization.
Whilst it is true (well established in fact) that truncation to 16-bits has been found to be inadequate, this comment in my opinion is passive, and misdirected. Real life CDs are
not produced from undithered low-noise truncations to 16 bits. A lay reader could easily fail to realise that this comment is in relation to a mere "probe" condition, not a realistic condition. I suggest it would have been helpful in the interests of academic neutrality and clarity to have continued with qualifying sentences along these lines:
[Hypothetical sentences:] We note that in practice standard CDs are not produced with quantization alone (that is to say discarding all bits beyond the 16th by way of truncation) but with dither. Dither allows bits beyond the 16th to be reflected in the audible signal, despite the signal being distributed to the consumer in a 16-bit format. We would stress that the test conditions that involved quantization without accompanying dither were merely a "probe" to serve as a reference point: there is no suggestion that standard CDs are produced in this way.
Alas, no such qualifying sentences appear in this part of the report. I think it is this tenth page of the report that will be fodder for high definition proponents not averse to letting technical accuracy get in the way of condemning the standard CD format for its "insufficient" 16 bits; or simply not understanding the benefits of dither.
If you look at the final results, fig. 4, it just doesn't make much sense.
If you focus on the arithmetic mean results with the 16-bit conditions, the 44.1 kHz filters were steeper, with longer and louder ringing and cutoff closer to the audible range, but they were harder to detect than the 48 kHz ones. So the filter seems to make little difference. 24-bit scored better for 44.1 kHz but worse for 48 kHz, so quantization doesn't seem to have a consistent effect either.
But on the other hand all the means are well within 1 standard error of each other. Condition 4 is as much an outlier as is condition 1 in that respect.
Conclusion: their test resulted in an average 60% score pretty much regardless of filter or quantization used. They supposedly heard something, the question is what.
Did I mention that more information and data is needed?
All of these audio tests with negative outcomes are used exactly the same way.
Wrong. It simply means you don't comprehend their meaning, that the particular test failed to yield the claimed differences, real or imaginary. That the burden of proof still lies with your sides constant claims of audibility, real or imagined.
But that isn't an excuse to blatantly fabricate positive results, be they ABX logs, amp distortions or 16/44 non-transparency, just to sell $50k bling and "Hi-Re$" hardware/media, etc.
Further, you have presented zero evidence that ABX creates higher cognitive loads than A/B, or your claimed gold standard, ABC/HR (perhaps +/-10% shell game more appropriately).
In this regard the BS paper contains a load of it, but it isn't cognitive.
Once more:
There is rather some research on discrimination with time lapse, and it is quite true that a break between A/X or B/X (or even A/B) will disrupt the subject, so time proximity WITHOUT glitches is absolutely requisite. This is because partial loudnesses, rather than time domain waveforms, is what must be recalled, and the first level of memory for such is under 200 milliseconds. So proximate IN TIME presentation is the relevant, germane thing to do.
Since people do perform with such testing down to physical limits, I don't think there is a great deal of room left there for problems.
If you want to do an audio test, how else would you do this? By the way, sequential tests with proper windowing are documented as the best way to extract the most reliable answers from subjects.
The conclusion is that only those with vested pecuniary interests would draw any conclusion from the BS paper, other than it was highly contrived and inconclusive.
cheers,
AJ
If you look at the final results, fig. 4, it just doesn't make much sense.
Agreed.
If you focus on the arithmetic mean results with the 16-bit conditions, the 44.1 kHz filters were steeper, with longer and louder ringing and cutoff closer to the audible range, but they were harder to detect than the 48 kHz ones. So the filter seems to make little difference. 24-bit scored better for 44.1 kHz but worse for 48 kHz, so quantization doesn't seem to have a consistent effect either.
Yes, in figure 4 the scores for the rectangle symbol (for 16-bit truncation) and square symbol (for dithered to 16 bits) are slightly lower on the graph for the 22050Hz cutoff than for the 24000Hz cutoff; suggesting
it was slighter easier to hear the effect of the emulation of a 48kHz sample rate; under those conditions of quantization. Yet
with no quantization, it was the emulation of a 44.1kHz sample rate that was slightly easier to detect (scores marked with a circle)!
But with overall results hovering around only 60% correct and variations about that of only a few percent in the overall scores, it is hard to draw conclusions.
Did I mention that more information and data is needed?
I wouldn't have minded seeing a similar graphical representation to Figure 4 of the results for the "high yield" sections in isolation, though there is already some narrative about statistical outcomes for that [which I found a bit difficult to follow] in part 3.2 of the paper. It could be suggested that the low yield sections are a distraction given the scores for those sections were so very poor (and commensurate with pure chance). They arguably are generating statistical noise, though academically it may be a conservative statistical approach not to exclude them.
And what about that specific paragraph about their post-hoc analysis that showed almost identical mean for no quantization and 16-bit quantization?
The standard deviations all overlap. Doesn't that mean we shouldn't be drawing any conclusions?
(stats isn't my thing)
Cheers,
David.
No it is not.
Yes, it is.
No it is not. Or else you wouldn't feel the urge to keep posting and posting and get unprofessional to boot.
You defend that position and with your typical amir spiel tried to, once again, shift the burden of proof demanding from others to disprove your position.
No I didn't shift the burden of proof. I stated the reality rather than some made up talking points we use to get out of providing a technical explanation.
Both sides are in equal position to prove their point in a forum discussion. There is no overriding rule that says you can sit back and let me post and post as evidenced by the fact that you post more than I do on this topic. You are acting as I say, but saying otherwise. That is not logical and I won't accept it as being such.
You don't know what my position is anyway.
The problem here is not proving a negative, but assuming you are right because there is no strong evidence that your position is wrong. That's a pretty absurd and fallacious position to take, but hey, we're not surprised.
Once again you don't know my position so your foundation is wrong. The position I take is very simple and cannot be denied:
1. High resolution audio is coming to market and no one cares how much anyone jumps up and down here.
2. This is not 2007 where we were discussing whether two new physical formats (DVD-A and SACD) had enough merit to force us all to buy all new players and deal with their strict copyright to allow us to make convenient copies. High-resolution files today are 100% compatible with all of our music servers and many appliances.
3. Playing back high-resolution audio has zero, zero cost associated with it. It does not enrich any major company with patents (Sony/Philips or DVD Forum contributors). The notion then that anything good about high resolution audio is motivated by money is totally absurd.
4. The 16/44 is 99.99999% of time a down converted version of a 24-bit higher sampling rate file. Per #3, I don't need anyone to screw around with my bits. If you or anyone else want 16/44, you can generate them from high resolutions stereo master using whatever dither, resampler, etc. you want. I have no reason to follow your preference there, nor does anyone who wants the best quality. I know the best quality will always be a generation up from 16/44. Want the original bits thank you very much.
5. By a miracle, the music labels have decided to license their studio masters to us. You don't see that in movies, do you? This is a gift to be embraced.
6. CD as a format will decline and go away as an option for your music purchases in a few years. There is no longer any reason to stamp plastic and spin it with a motor to read and play digital bits. You will as sure as sun comes out the east, face the situation of wanting your favorite music yet it will only be available in MP3 or AAC.
Last night I went searching high and low to find the CD version of one of my favorite soundtrack composers, Max Richter's album: Perfect Sense: Original Film Soundtrack. Yet Amazon only lists the stupid MP3 version. This is the first time I have seen a movie soundtrack skip CD release.
Wake up and smell the coffee please. You need to be 1000% in favor of more consumer choice. You don't have to buy high-res but you better join me in supporting it as otherwise we will all be screwed with the first and biggest jump back in fidelity.
These are the facts. None of it is impacted by any logical fallacy or debating terms. None of it is impacted of what you think of Stuart's paper or what you think you know about listening tests or signal processing.
Don't confuse me engaging you on the technical side as it being my "position" that I have to win such arguments. I just find it fun to turn the tables on people who keep demanding double blind tests exactly that and see them squirm. You squirm because you held an unreasonable position that such tests could not be passed. Well, they have been passed and if that is what you want to base your belief on it instead of the reality of the marketplace, then you suffer and must defend your position. The burden is on your shoulder and big time.
Don't go responding to this post because per above, you don't matter. AJ doesn't matter. Krab doesn't matter. None of you do. Get a sense of reality, conduct yourself more professionally and join the audiophiles elsewhere in increases the choice of formats we have. Heaven knows I am getting bored interacting with you all.
No it is not. Or else you wouldn't feel the urge to keep posting and posting and get unprofessional to boot.
Wow, childish behavior and irony. I guess that is how "professional experts" act normally?!
There is no overriding rule that says you can sit back and let me post and post as evidenced by the fact that you post more than I do on this topic. You are acting as I say, but saying otherwise. That is not logical and I won't accept it as being such.
You don't know what my position is anyway.
Oh amirm, have you already forgotten what you posted yesterday?
Here's what you posted:
"And the cognitive load is obvious to anyone who has taken the ABX blind tests.
You have some evidence that this is not true?"and
"What is really at issue here is that folks/you don't want to hear any criticism of ABX testing."Of course you'd turn the attention away from this ridiculous statement and turn it on me. We've seen this throughout this entire thread many, many times.
You've also stated your position before, and even told us several stories that were completely off-topic and a waste of Internet traffic. I guess you forgot that as well.
Once again you don't know my position so your foundation is wrong. The position I take is very simple and cannot be denied:
Yes, and my position on rocks falling to the ground cannot be denied either. See, I can chance topics too.
You need to be 1000% in favor of more consumer choice.
I am, and we can only hope that hi-res will not replace CDs because then we will not only be stuck with bigger files, but much more importantly with files that sound as horrible as CDs do today.
You don't have to buy high-res but you better join me in supporting it as otherwise we will all be screwed with the first and biggest jump back in fidelity.
I'd never support a
demonstrably intellectually dishonest person. Also, there is no big jump in fidelity going from CD to hi-res. The big jump is not in the format, but in the recording, mixing, mastering.
We've had this discussion already, you also forgot that?
These are the facts. [...] I just find it fun to turn the tables on people who keep demanding double blind tests exactly that and see them squirm. [...] The burden is on your shoulder and big time.
Don't go responding to this post because per above, you don't matter. AJ doesn't matter. Krab doesn't matter. None of you do. Get a sense of reality, [...]
I think we call that being despicable and delusional.
1. High resolution audio is coming to market and no one cares how much anyone jumps up and down here.
Is anybody denying that here? We're denying the
necessity of hi-res audio format delivery to the end-user. People here noticed that the hi-res craze will happen years ago, and pointed out the stupidity of the endeavour back then already. Nothing has changed since then.
2. This is not 2007 where we were discussing whether two new physical formats (DVD-A and SACD) had enough merit to force us all to buy all new players and deal with their strict copyright to allow us to make convenient copies. High-resolution files today are 100% compatible with all of our music servers and many appliances.
Are they? It's not unthinkable or impossible for a company to use or create an audio container with DRM and special hardware or software requirements.
3. Playing back high-resolution audio has zero, zero cost associated with it. It does not enrich any major company with patents (Sony/Philips or DVD Forum contributors). The notion then that anything good about high resolution audio is motivated by money is totally absurd.
Monty has pointed out that in practice hi-res playback
might have a lower audio fidelity due to distortions coming from high-frequency content (http://xiph.org/~xiphmont/demo/neil-young.html).
The 16/44 is 99.99999% of time a down converted version of a 24-bit higher sampling rate file. Per #3, I don't need anyone to screw around with my bits. If you or anyone else want 16/44, you can generate them from high resolutions stereo master using whatever dither, resampler, etc. you want. I have no reason to follow your preference there, nor does anyone who wants the best quality. I know the best quality will always be a generation up from 16/44. Want the original bits thank you very much.
People like you always know that the "best quality" is up from whatever the standard is currently. That's the oldschool audiophile delusion.
5. By a miracle, the music labels have decided to license their studio masters to us. You don't see that in movies, do you? This is a gift to be embraced.
I still wonder why you assume hi-res masters to be inherently superior to CD masters. What prohibits engineers from fucking up those, too? They managed to make 16/44 sound like shit, it's just as easily possible with 24/96. Audio technology is
not the problem, it's
human error.
6. CD as a format will decline and go away as an option for your music purchases in a few years. There is no longer any reason to stamp plastic and spin it with a motor to read and play digital bits. You will as sure as sun comes out the east, face the situation of wanting your favorite music yet it will only be available in MP3 or AAC.
Lossy delivery can be acceptable with proper mastering. The CD releases you get today are (depending on genre) sometimes hardly more than 12bit/32khz anyway.
Wake up and smell the coffee please. You need to be 1000% in favor of more consumer choice. You don't have to buy high-res but you better join me in supporting it as otherwise we will all be screwed with the first and biggest jump back in fidelity.
See above. The conjecture that
audio engineers will magically improve their craft just due to the existence of a hi-res audio format is absurd. Again, it is already possible to produce proper releases
on CD today (which has been demonstrated in the classical and soundtrack genres), it's just not done.
Don't go responding to this post because per above, you don't matter. AJ doesn't matter. Krab doesn't matter. None of you do. Get a sense of reality, conduct yourself more professionally and join the audiophiles elsewhere in increases the choice of formats we have. Heaven knows I am getting bored interacting with you all.
Get lost already. Nobody gains any insight from reading your FUD nonsense and propaganda.
Don't go responding to this post because per above, you don't matter. AJ doesn't matter. Krab doesn't matter. None of you do.
In the sense that none of us are the intended target of the $hysters and $cammers peddling $50k bling and "Hi-Re$", of course not Amir. We are certainly not going to be fleeced by shell games and contrived or outright fabricated "tests" posited as evidence of audibility.
The target there is the $cam addicted audiomorons, who crave that stuff, not logical, rational folks. The irony of course, is that those same $cam addicts reject any form of honesty controls/blind testing as relevant (see latest Stereophile). So exactly how this BS paper will be sold to them is fascinating.
However, this is HA Amir, so as members, we in fact, do matter.
Get a sense of reality, conduct yourself more professionally and join the audiophiles elsewhere in increases the choice of formats we have.
You advising on getting a sense of reality is comical. No joining your camp for rational people and much $avings for the exact same sound quality music. $42 per album? Pleeease.
Heaven knows I am getting bored interacting with you all.
With nearly 40,000 posts on just 2 forums, somehow I doubt that statement. The preponderance of the evidence suggests otherwise.
cheers,
AJ
Thinking about Eva Green listening to the recomposed 4 seasons is heaven!
6. CD as a format will decline and go away as an option for your music purchases in a few years. There is no longer any reason to stamp plastic and spin it with a motor to read and play digital bits. You will as sure as sun comes out the east, face the situation of wanting your favorite music yet it will only be available in MP3 or AAC.
Last night I went searching high and low to find the CD version of one of my favorite soundtrack composers, Max Richter's album: Perfect Sense: Original Film Soundtrack. Yet Amazon only lists the stupid MP3 version. This is the first time I have seen a movie soundtrack skip CD release.
Wake up and smell the coffee please. You need to be 1000% in favor of more consumer choice. You don't have to buy high-res but you better join me in supporting it as otherwise we will all be screwed with the first and biggest jump back in fidelity.
Wanna bet?
Amirm,
in my opinion the part I highlighted in your post in red is mere conjecture. Or worse, scaremongering.
Louspeakers have extremely different sounds; professional microphones have slightly different sounds (even before the positioning of the mike is considered). Where you sit in an auditorium (even a small auditorium to suit a chamber orchestra) has a pronounced effect on what you hear.
The music I like is classical, including opera. Why that would be released in the inefficient mp3 format is very unclear. If it were released as AAC (or some future more efficient codec) at a bitrate transparent for my ears I might feel a degree of dissatisfaction, but practically it would suffice. If indeed transparent for my ears.
I would be more interested in improvements in loudspeaker technology, possible improvements in use of multi-channel in recording and reproduction, and possible improvements in microphones used for recording, and mike placement and mixing.
These days many audiophiles are embracing room correction processing (Audyssey, etc). The effects of that would far, far outweigh any use of 16/44 in lieu of 24/96, or 32/384. There's also the question of acoustic treatments for rooms and partial soundproofing.
It is easy to focus on the question of high definition in the distribution format, as this is under the control of the party releasing or distributing the recording, and yet it is perhaps the least significant aspect of the whole process in terms of the audible result. No, I will go further: for stereo sound it
is the least significant aspect impinging on sound quality, despite whatever doubts may have been raised in the BS paper under discussion in this thread of possible effects (apparenty at the outer limits of audibility).
To resort to the claim you have made at 6 in your post is to present a strategic, marketing related, argument that could conceivably have some weight or may have no validity at all! To compare it with the sun rising in the east is in my view hyperbole on a grand scale.
I would be very surprised if recordings of classical music and opera ceased to be available in either PCM format or lossless equivalents, and if they were instead released in a lossy form of compression only, I cannot see that being in a low bitrate mp3 version. It it were lossy, it would be with an efficient codec, at a decent bitrate.
Actually I think the future of recorded classical music (including opera) is video with lossless audio, such as occurs already to some extent on Blu-ray. YouTube is so readily available that many people not attending live concerts are becoming accustomed to seeing music performed as well as hearing it. I am sorry if this post is a bit off-topic but I felt I couldn't let your claim at point 6 pass without challenge.
This BS study lacked investigatory drive in my view. It was very passive. It seems that the study was conceived and implemented, full stop. I note that its two main conclusions are expressed vaguely, or even cryptically.
I'd have thought that preliminary informal tests should have been done with subjects with outstanding hearing and listening skills. This presumably would have shown up high yield (revealing) sections. Then a variation could have been done by way of substituting a different loudspeaker system, or even headphones to see whether a difference was still in evidence.
Yes, it was interesting that they didn't break out individual results, or do this sort of pre-testing to identify the best performers. Then again they only tested 8 subjects! They acknowledge the preliminary nature of the work in conclusions , referring to it as a 'small pilot study'. Which is one reason that trumpeting it as a game-changer is a bit premature.
The way the two "main" conclusions "offered" are expressed is also passive, and not particularly helpful. It is obvious that audible sounds exist that cannot be transparently encoded at 16/44. Some children can hear a loud sine wave at 23kHz and a standard CD cannot render 23kHz at all. Also if a CD is played back with ear-splitting gain, then dither noise may be audible even to middle-aged adults. It is accordingly quite cryptic and unhelpful to offer as a conclusion: "first, there exist audible signals that cannot be encoded transparently by a standard CD". Perhaps what was meant was "audible signals contained in conventional music recorded with high definition and listened to at a normal listening level".
Indeed that first conclusion--and btw, the two conclusions in the Abstract are *not* the same as the ones in the 'Conclusion' section -- is really rather curious. It was certainly already known that there are audible signals that CD cannot encode transparently. So why trumpet that as a 'conclusion' here? This is what some of us here wondered about as soon as we read the abstract, even before the paper was made public.
In summary,
- the BS experimental study appears to have been executed in a passive way, not taking advantage of preliminary informal testing to pinpoint what was causing an apparent lack of transparency in the filter emulations (particularly the emulation for a sample rate reduction to 44.1kHz), and
- the two main conclusions offered are vaguely expressed and unhelpful, and in the case of the second conclusion (about the need for a hi-fidelity reproduction chain) not actually supported by experiment.
But note again, those 'conclusions' are not offered in the actual conclusion section of the paper -- there, there are 5 conclusions that are more specific. If I reviewed this I would have told the authors to bring their Abstract more in line with the Conclusions, or vice-versa.
CONCLUSIONS
1. FIR filters that emulate downsampling for sam-
ple rates of 44.1 kHz and 48 kHz can have a dele-
terious eff ect on the listening experience in a
wideband playback system.
2. 16-bit quantization with and without RPDF
dither can have a deleterious eff ect on the listen-
ing experience in a wideband playback system.
3. Our fi ndings are consistent with the idea that
filters with long impulse responses blur fine tem-
poral details of signals.
4. Not all pieces of music contain musical features
that demonstrate these losses of transparency.
Possible important features include echoes that
give a sense of the physical space around the
performers.
5. Consideration should be given to the use of psy-
chophysical tests that minimise cognitive load
in studies of this kind.
Further comments
For someone familiar with MATLAB audio filters, it is a walk in the park to alter the parameters. The BS study could have been so much more. In my opinion, it was a narrowly focussed endeavour, with vaguely expressed conclusions.
It is meant as a riposte to Meyer and Moran 2007 -- the dragon the authors spend much text trying to slay in the Introduction. There is a 'history' there.
Whilst it is true (well established in fact) that truncation to 16-bits has been found to be inadequate, this comment in my opinion is passive, and misdirected. Real life CDs are not produced from undithered low-noise truncations to 16 bits.
[Hypothetical sentences:] We note that in practice standard CDs are not produced with quantization alone (that is to say discarding all bits beyond the 16th by way of truncation) but with dither. Dither allows bits beyond the 16th to be reflected in the audible signal, despite the signal being distributed to the consumer in a 16-bit format. We would stress that the test conditions that involved quantization without accompanying dither were merely a "probe" to serve as a reference point: there is no suggestion that standard CDs are produced in this way.
Alas, no such qualifying sentences appear in this part of the report. I think it is this tenth page of the report that will be fodder for high definition proponents not averse to letting technical accuracy get in the way of condemning the standard CD format for its "insufficient" 16 bits; or simply not understanding the benefits of dither.
More odd is that IIRC Stuart himself acknowledged much the same point in his older Coding for High Resolution Audio' review paper. And of course the promotional material AJinFla has quoted multiple times show that Meridian has previously touted the 'transparency' of TBDF dither.
Is anybody denying that here? On the contrary, everyone here noticed that years ago, and pointed out the stupidity of the endeavour back then already. Nothing has changed since then.
They must be denying it if they think it is a stupid thing. More consumer choice is never a stupid thing. And getting the original instead of the converted one by who knows what, is not a stupid thing.
Are they? It's not unthinkable or impossible for a company to use or create an audio container with DRM and special hardware or software requirements.
Complain when that happens. Right now, the miracle has happened. We are getting high-resolution masters with absolutely no copy protection.
Monty has pointed out that in practice hi-res playback might have a lower audio fidelity due to distortions coming from high-frequency content (http://xiph.org/~xiphmont/demo/neil-young.html).
I have known Monty more than many of you. He is a super smart guy but has little expertise in this area. I have debated his write-up with him on WBF Forum and have shown the many flaws in his blog on AVS Forum. He has improved his write-up some but at the end of the day, it is yet another person writing about a field that is not his specialty. He doesn't understand the business side of this field, nor does he have electrical engineering experience to understand the hardware side. You want to the right version of that, read Stuart's JAES paper, coding for high-resolution audio. Anything beyond that which Monty says is yet another online blogger making arguments with no personal data or professional experience.
As to his theory of IM distortion, it is just that: some theory. Arny posts test files on AVS to bring out such distortion and I and a number of other people reported no problems at all. And at any rate, as I mentioned, if you think there is a problem there, you can always filter the files yourself. You can't do the reverse.
People like you always know that the "best quality" is up from whatever is the current standard. That's the oldschool audiophile delusion.
What do you mean people like me? I am an objectivist, engineer and professional who has worked in this field for many years. I am not some subjectivist you would throw that line at and sit back.
You are wrong and that can be trivially shown. There are many high-resolution stereo masters that are distributed prior to loudness compression of CD/MP3. There is nothing delusional about their improved fidelity. You speak from point of view of forum facts, not real life facts.
Even if they are identical mastering, there is no way you can tell me someone is doing me a favor by converting to 16/44. I can do that myself and don't need someone whose expertise is to create music to know the TPDF dither from a hole in the wall.
5. By a miracle, the music labels have decided to license their studio masters to us. You don't see that in movies, do you? This is a gift to be embraced.
I still wonder why you assume hi-res masters to be inherently superior to CD masters. What prohibits engineers from fucking up those, too? They managed to make 16/44 sound like shit, it's just as easily possible with 24/96. Audio technology is not the problem, it's human error.
I assume it because I personally know the people who create them. And the founder of the biggest distributor of such (HD tracks). You don't have that kind of real life data and relationship, right? So what you hypothesize is just FUD and random arguments thrown out there. What if I got a steak house and they feed me hamburger they bought from Burger King. Yes, it could happen but come back when you know it has.
The market for high resolution audio has been established. Content distributors know that people will immediately run spectrum analyzer and compare CD to high-res versions. QC has improved substantially and I expect over time for high resolution audio to almost always being the same or better fidelity than the CD. The notion that it can be worse is absurd.
Lossy delivery can be acceptable with proper mastering. The CD releases you get today are (depending on genre) hardly more than 12bit/32khz anyway.
Lossy if it becomes the only copy available, is anything but acceptable. I am not going to endorse going backward in fidelity in the name of technological advancement. If you want to go there, do but don't tell me it should be acceptable to me. And the other statement is just absurd and is never backed by any references or real data. But maybe you surprise us and give us such and not come across as copious producer of FUD.
See above. The conjecture that audio engineers will magically improve their craft just due to the existence of a hi-res audio format is absurd. Again, it is already possible to produce proper releases on CD today (which has been demonstrated in the classical and soundtrack genres), it's just not done.
The don't need to improve their craft because you have some made up fantasy argument on forums that effective resolution of music today is 12 bits/32 Khz. Do you even know how loud the channel noise at 12 bits is? That is signal to noise ratio of cassette tape.
So , no magical improvement is required. All that needs to happen, which fortunately already has, is to get the upstream bits prior to CD/MP3 mastering. I gave evidence of this earlier:
=====
See this from many such examples, for one of the award winning mastering engineers, Doug Sax: http://mixonline.com/news/profiles/masteri...h.cHY5BQ9r.dpuf (http://mixonline.com/news/profiles/mastering-cutting-edge-quality-age-low-res-low-budgets-and-multiple-formats/366464#sthash.cHY5BQ9r.dpuf)
When he was invited to sit on Ludwig’s Platinum Mastering Panel this year, Doug Sax suggested they discuss mastering for multiple formats. “Ten years ago, you made a CD master and you were done,” Sax says. “Now you do a CD master, a master for iTunes, a high-resolution master for HDtracks, and a cutting master for vinyl.” =====
See the separate mastering for HD Tracks? What we want is happening.
[audiophile gibberish]
Get lost already. Nobody gains any insight from reading your FUD nonsense and propaganda.
As if your reply to me had an ounce of insight or data? All you spread is FUD and turn around and say that to me?
But sure. I will be leaving shortly as I have never seen so many unprofessional posters in any audio forum.
Don't go responding to this post because per above, you don't matter. AJ doesn't matter. Krab doesn't matter. None of you do. Get a sense of reality, conduct yourself more professionally and join the audiophiles elsewhere in increases the choice of formats we have. Heaven knows I am getting bored interacting with you all.
And when you are king.... oh wait, in your mind, you already *are*.
You are wrong and that can be trivially shown. There are many high-resolution stereo masters that are distributed prior to loudness compression of CD/MP3. There is nothing delusional about their improved fidelity. You speak from point of view of forum facts, not real life facts.
You're comparing different masters.
That is signal to noise ratio of cassette tape.
Exactly. Again, it's
not the technology but the
actual delivered recording. Numbers don't mean anything if they are not used. That is the point which we try to hammer in your head for days now. Engineers are not using CDs to their full theoretical capacity, so why on earth do you assume giving them even more bits will help with anything?
See the separate mastering for HD Tracks? What we want is happening.
There is no need for
separate mastering. All you need to do to create a CD master from a hi-res master is to decimate to 16bits using proper noise shaped dithering and resample to 44.1kHz. Anybody who's doing it differently
doesn't have a clue what he's doing.
As if your reply to me had an ounce of insight or data?
I have two CDs here. One is the Gladiator soundtrack. The other is Californication. It's trivially obvious to people who know the latter CD that it suffers from distortion and extreme dynamic range compression, so that the end-result is hardly bearable. The former CD is intended as a high-quality soundtrack, and the production shows. Now comes the clue for you, which you have to try to sink into your mind once and for all. Both releases are delivered via redbook audio CD. That means they use the same digital audio technology. Hence any difference in dynamics and distortions are a product of the production. There is simply no other possibility left. That means any distortions and DRC are due to the
production quality and decisions and not inherent to the delivery format. q.e.d.
Monty has pointed out that in practice hi-res playback might have a lower audio fidelity due to distortions coming from high-frequency content (http://xiph.org/~xiphmont/demo/neil-young.html).
I have known Monty more than many of you. He is a super smart guy but has little expertise in this area. I have debated his write-up with him on WBF Forum and have shown the many flaws in his blog on AVS Forum. He has improved his write-up some but at the end of the day, it is yet another person writing about a field that is not his specialty. He doesn't understand the business side of this field, nor does he have electrical engineering experience to understand the hardware side. You want to the right version of that, read Stuart's JAES paper, coding for high-resolution audio. Anything beyond that which Monty says is yet another online blogger making arguments with no personal data or professional experience.
LOL! You're always *what's best* Amir. I'm sure your parents told you that all the time. So we'll tell you too, if it keep s you calm.
But Monty is hardly the only one to point out that hi-rez could produce distortions in some setups. Indeed, SACD players have built-in lowpass output filters after the DAC (at 50 or 100kHz) to try to ameliorate that.
I still wonder why you assume hi-res masters to be inherently superior to CD masters. What prohibits engineers from fucking up those, too? They managed to make 16/44 sound like shit, it's just as easily possible with 24/96. Audio technology is not the problem, it's human error.
I assume it because I personally know the people who create them. And the founder of the biggest distributor of such (HD tracks). You don't have that kind of real life data and relationship, right? So what you hypothesize is just FUD and random arguments thrown out there. What if I got a steak house and they feed me hamburger they bought from Burger King. Yes, it could happen but come back when you know it has.
Actually HD tracks has been dunned for offering tracks that aren't really 'HD'. Heck, you've seen Bruce from Puget Sounds apologize for this on your own forum! And of courss DVDA and SACD releases *also* have been revealed as not necessarily being 'hi rez'. But 'it's just business' right?
Amir, the real 'good fight' isn't for 'high res' or 'consumer choice' of formats. You know that. The good fight would be the one for *high quality mastering*. And then *promotion of better acoustics at home*.
Stop playing the clown for the 'high res' faction in industry. Fight the good fight.
I will be leaving shortly as I have never seen so many unprofessional posters in any audio forum.
Bye. Give Dunning-Kruger my regards.
See the separate mastering for HD Tracks? What we want is happening.
There is no need for separate mastering. All you need to do to cerate a CD master from a hi-res master is to decimate to 16bits using proper noise shaped dithering and resample to 44.1kHz. Anybody who's doing it differently doesn't have a clue what he's doing.
amirm's argument is that you should increase bits and sampling rate so high, such that it doesn't matter anymore what the clueless person mastering the music is doing.
What he doesn't seem to get in his head is that it's the same person that is responsible for the bad sound quality.
Also, as soon as hi-res slowly becomes the "norm", the bad sound quality will shift from CD audio to hi-res.
Don't go responding to this post because per above, you don't matter. AJ doesn't matter. Krab doesn't matter. None of you do. Get a sense of reality, conduct yourself more professionally and join the audiophiles elsewhere in increases the choice of formats we have. Heaven knows I am getting bored interacting with you all.
And when you are king.... oh wait, in your mind, you already *are*.
In the land of the blind...
I have known Monty more than many of you. He is a super smart guy but has little expertise in this area. I have debated his write-up with him on WBF Forum and have shown the many flaws in his blog on AVS Forum. He has improved his write-up some but at the end of the day, it is yet another person writing about a field that is not his specialty. He doesn't understand the business side of this field, nor does he have electrical engineering experience to understand the hardware side. You want to the right version of that, read Stuart's JAES paper, coding for high-resolution audio. Anything beyond that which Monty says is yet another online blogger making arguments with no personal data or professional experience.
Did you really just say the creator of Ogg and Vorbis (https://en.wikipedia.org/wiki/Chris_Montgomery) has "little expertise" in (digital) audio, and "no professional experience"? That is beyond ridiculous. And, fun fact for you, the "business side" is completely unimportant in technical and scientific arguments. But I guess that was just another weak attempt at trolling, like your whole paragraph.
I have known Monty more than many of you. He is a super smart guy but has little expertise in this area. I have debated his write-up with him on WBF Forum and have shown the many flaws in his blog on AVS Forum. He has improved his write-up some but at the end of the day, it is yet another person writing about a field that is not his specialty.
Ok, we are now officially in the Twilight Zone. That coming from Amir, who on a public forum admitted this: (http://www.avsforum.com/forum/86-ultra-hi-end-ht-gear-20-000/1136745-establishing-differences-10-volume-method.html#post16216826):
Originally Posted by amirm
If what I present doesn't meet some high bar, so be it....
The comparison I performed was using a Mark Levinson No360S against the on-board DACs in five to six DVD-A and SACD players, all playing the same time sync'ed CD. In other words, I would listen to the analog output of the player while its digital output would feed the ML DAC. All front panel lights were turned off in addition to video circuits (yes, all of that made a difference in fidelity).
The two sources were fed to the dual inputs of a Stax "earspeaker" electrostatic headphone amp. If you are not familiar with Stax, you can read learn more about them here: http://www.stax.co.jp/Export/ExportProducts.html (http://www.stax.co.jp/Export/ExportProducts.html). I have three of their units and results are consistent across the board although the highest end unit does make the job a bit easier. Using headphones allowed me to completely eliminate the room and take advantage of the amazing transparency of these headphones to listen for the slightest differences. To latter point, I would often listen to material at levels well above what I would use for listening to music, allowing me to hear detail that would otherwise be lost.
I then picked material that made it easier to detect differences between DACs. I am not going to disclose what constitutes such content. Without such material, the job can range from difficult to impossible. One has to know what could be damaged by a DAC and then use music that has such content. To give you an example, when you compress music, it is the transients that suffer. So something like guitar music is much more revealing than say, violin as the latter is much more harmonic than the sharp impulses of a guitar. Voices play the same role. None of these are useful for testing DACs though so don’t use that as a hint to the question posed . You can’t test the cornering of a car if you just drive it straight….
The comparison was then conducted without knowing which input is which, sitting in front of the headphone amp and toggling back and forth. When necessary, I would go back and re-listen. Once I found which one sounded worse, I would then repeat the exercise by randomizing the inputs and seeing if I could still identify which one was worse. My success rate was 100% in the second test (i.e. could always verify that the first result was not by chance). This testing was repeated a number of times comparing the different sources against each other and the ML.
I did not level match anything. However, once I found one source was worse than the other, I would then turn up the volume to counter any effect there. Indeed, doing so would close the gap some but it never changed the outcome. Note that the elevated level clearly made that source sound louder than the other. So the advantage was put on the losing side.
The results above were later objectively shown to be backed by some science in Stereophile magazine.
Rank amateur, that Monty.
He doesn't understand the business side of this field
That might be the truest statement you have ever uttered.
But How Does it Sound - By Amir Majidimehr
In comparison testing I have done, switching amplifiers using the classic class D configuration always sport incredible low frequency control and power. They beat out linear class AB amplifiers almost regardless of price. What they give up though is high frequency fidelity which I find somewhat harsh. The distortion is highly non-linear and challenging to spot but it is there. The Mark Levinson No 53 is the first switching amplifier I have heard which does not have this compromise. Its bass is amazingly authoritative: tight and powerful. Yet the rest of the response is absolutely neutral and pleasant.
If you have not heard these unique $50k amplifiers, I highly encourage you to come into our showroom for a listen. We have a pair on hand driving our Revel speakers. I am confident that they will improve the sound of your current speakers given the ease with which they can drive any load regardless of how difficult they might be (and many high-end speakers are difficult to drive). We are happy to let you evaluate them with your own system to see the benefits of this technology. Hearing this amplifier was an eye-opener for me. I think it will be for you too.
I find it quite refreshing that you have admitted why you are so enamored with this "Hi-Re$" BS paper Amir.
It's understanding the business side of this field....ya understand.
cheers,
AJ
Indeed that first conclusion--and btw, the two conclusions in the Abstract are *not* the same as the ones in the 'Conclusion' section -- is really rather curious. It was certainly already known that there are audible signals that CD cannot encode transparently. So why trumpet that as a 'conclusion' here? This is what some of us here wondered about as soon as we read the abstract, even before the paper was made public.
...
But note again, those 'conclusions' are not offered in the actual conclusion section of the paper -- there, there are 5 conclusions that are more specific. If I reviewed this I would have told the authors to bring their Abstract more in line with the Conclusions, or vice-versa.
Indeed yes. There is quite a disconnect. Perhaps the peer reviewers didn't read to the end of the paper. (!)
I'll quote the text in the form you provided it, and add in some comments though I think others may already have made some similar comments in this thread:
CONCLUSIONS
1. FIR filters that emulate downsampling for sam-
ple rates of 44.1 kHz and 48 kHz can have a dele-
terious effect on the listening experience in a
wideband playback system. [This is a bit of a leap. It is one thing to have established that a difference was correctly identified some 60% of the time. It is another to accept subjective remarks made by one or more of the 8 test subjects that the effect was of a degree and nature to be fairly described as having a "deleterious effect on the listening experience". The report provides no detail on how many of the eight test subjects made negative comments, and whether they were test subjects with high scores.]
2. 16-bit quantization with and without RPDF
dither can have a deleterious effect on the listen-
ing experience in a wideband playback system. [Again, "deleterious effect on the listening experience" may be too strong a phrase. "From time to time perceptible effect, that some listeners may describe as deleterious to their listening experience" might have been more appropriate.]
3. Our findings are consistent with the idea that
filters with long impulse responses blur fine tem-
poral details of signals. [The idea is diminished in persuasiveness given that the filtering was of ultrasonic frequencies using linear phase.]
4. Not all pieces of music contain musical features
that demonstrate these losses of transparency.
Possible important features include echoes that
give a sense of the physical space around the
performers. [It might have assisted to try a less aggressive MATLAB filter, and if results were still audible, look for other possible explanations, e.g. some instability in the tweeters triggered by high frequency content. On the other hand, if this was intended very much as a narrow, pilot study, the limited rather passive approach taken is more understandable; and the paper should not spoken of as a "sea change", but rather as "raising questions for further study".]
5. Consideration should be given to the use of psy-
chophysical tests that minimise cognitive load
in studies of this kind. [Certainly it could help if subjects could control which parts of a musical passage (or “section”) they listened to, just as is done routinely by some members of this forum when using foobar and its ABX plug-in! This begs the question why distinguishable passages have not been identified in the past in this forum when applying a filter for a 44.1kHz, or even 48kHz sample rate, for members with a "wideband playback system" (whatever that is intended to mean exactly, in the context of the BS paper).]
"wideband playback system" (whatever that is intended to mean exactly, in the context of the BS paper).
[a href="https://www.meridian-audio.com/collection/loudspeakers/" rel="nofollow"]https://www.meridian-audio.com/collection/loudspeakers/[/a]
Indeed that first conclusion--and btw, the two conclusions in the Abstract are *not* the same as the ones in the 'Conclusion' section -- is really rather curious. It was certainly already known that there are audible signals that CD cannot encode transparently. So why trumpet that as a 'conclusion' here?
That's easy to explain:
the peer reviewers never even saw the abstracts' two "offered conclusions" because it was written (or modified) after the fact! They thought they were reviewing a paper about how filters can sometimes alter the sound, hence the paper's title, rather than an attack against 44.1 transparency in some situations followed by an immediate, shameless plug for the authors' "truly high fidelity speakers, which are of course necessary to discern such differences", not that any
actual evidence to support this second conclusion was presented by way of testing if lesser speakers/gear would
fail to show audible distinctions.
the peer reviewers never even saw the abstracts' two "offered conclusions" because it was written (or modified) after the fact!
That's a pretty bold claim to be making without offering any proof.
... I still wonder why you assume hi-res masters to be inherently superior to CD masters. What prohibits engineers from fucking up those, too? They managed to make 16/44 sound like shit, it's just as easily possible with 24/96. Audio technology is not the problem, it's human error. ...
No, it's not human error. They do it deliberately and with full knowledge of what they are doing.
No, it's not human error. They do it deliberately and with full knowledge of what they are doing. sad.gif
There is no rule stipulating that all human errors must be unintentional.
The sound quality deficiencies of contemporary CD recordings are due to human error (intention or ignorance) and have nothing to do with the technological means available.
Furthermore, what constitutes an error often qualifies as a subjective matter, this situation being very much the case.
the peer reviewers never even saw the abstracts' two "offered conclusions" because it was written (or modified) after the fact!
That's a pretty bold claim to be making without offering any proof.
Especially as the usual procedure in scientific conferences I've attended, is to submit the abstract *first*. That's what gets you in the door, and if it's interesting/appropriate enough, you get to present a poster. If the abstract is considered especially interesting, you get asked to give a spoken presentation, which could become or be accompanied by a 'conference paper'.
So I would consider mzil's theory to be highly speculative.
the peer reviewers never even saw the abstracts' two "offered conclusions" because it was written (or modified) after the fact!
That's a pretty bold claim to be making without offering any proof.
Allowing an original author to slightly modify the wording of a paper's descriptive blurb, without any change to the actual contents of the paper itself (and without having to resubmit it to a
new peer review
just because of such a slight modification to the blurb), doesn't seem hard for me to believe at all.
It is speculation, admittedly, sorry if my wording didn't make that clear. I should have written: "
That's easy to explain There's an easy way to explain that:
the peer reviewers never even saw the abstracts' two "offered conclusions" because it was written (or modified) after the fact!"
"Might" or "maybe" or "perhaps" or "could" should be your first choice of words when launching into conspiracy theories.
I've made a short audio file (http://www.galaxyclassics.com/public/Sine1kHz@-80dB-160dB_4versions_24bit,16bitTPDF,RPDF,truncated,+60dBgain.wav) with 4 different dither formats to illustrate the differences.
The source is a 1kHz sine fading from -80dBFS to -160dBFS during 8 seconds. The 4 dither versions are:
1) 24-bit TPDF dither
2) 16-bit TPDF dither
3) 16-bit RPDF dither
4) 16-bit truncated
60dB gain was applied to make the result easier to hear at normal playback levels.
Monitor gain has to be reduced by 60 dB to get the original level back.
I've made a short audio file (http://www.galaxyclassics.com/public/Sine1kHz@-80dB-160dB_4versions_24bit,16bitTPDF,RPDF,truncated,+60dBgain.wav) with 4 different dither formats to illustrate the differences.
The source is a 1kHz sine fading from -80dBFS to -160dBFS during 8 seconds. The 4 dither versions are:
1) 24-bit TPDF dither
2) 16-bit TPDF dither
3) 16-bit RPDF dither
4) 16-bit truncated
60dB gain was applied to make the result easier to hear at normal playback levels.
Monitor gain has to be reduced by 60 dB to get the original level back.
How did you apply gain? This alone may add own dither or noise. I suggest shifting it by 10 bit.
How did you apply gain? This alone may add own dither or noise. I suggest shifting it by 10 bit.
No dither was added, as can be seen from the last seconds. It is just an illustration of the visible and audible effect of various dithers. In the RPDF version the noise modulation is audible and the waveform shows the non-linear behavior at the end.
I've made a short audio file (http://www.galaxyclassics.com/public/Sine1kHz@-80dB-160dB_4versions_24bit,16bitTPDF,RPDF,truncated,+60dBgain.wav) with 4 different dither formats to illustrate the differences.
The source is a 1kHz sine fading from -80dBFS to -160dBFS during 8 seconds. The 4 dither versions are:
1) 24-bit TPDF dither
2) 16-bit TPDF dither
3) 16-bit RPDF dither
4) 16-bit truncated
60dB gain was applied to make the result easier to hear at normal playback levels.
Monitor gain has to be reduced by 60 dB to get the original level back.
Thanks for this. Nice and obvious effects.*
__________
*
Thanks for this. Nice and obvious effects.*
Please note that "obvious" and "audibility" are related to playback levels. I wanted to illustrate that audibility of RPDF dither depends on the input signal, which makes predictions based on static measurements (as done in the paper) unreliable IMHO.
The submission process for the 137th AES convention is still publicly available on line...
http://www.aes.org/events/137/authors/137thCallForPapers.pdf (http://www.aes.org/events/137/authors/137thCallForPapers.pdf)
Note the different categories, timescales, and the strong indication that the paper should not exceed 10 pages.
The text at the very top of the paper under discussion matches category 2.
The text at the very top of the other Bob Stuart paper in the same session matches category 1.
The paper under discussion is 12 pages long (though before the header was added, it would have been 10 pages + references and appendix. I don't know if that would have helped or not.)
The abstract had to be delivered two months before the full paper.
Cheers,
David.
As to his theory of IM distortion, it is just that: some theory.
That's a surprising statement in this context!
IM distortion from ultrasonic signals, measurable in most hi-fi speakers on the planet, and laughably easy to hear with various signals, is just
some theory? Yet the audibility of ultrasonic ringing, suggested but not proven as the reason a filtered signal could be picked out 10% better than random choice in one study, is the important thing?
I would suggest that you can
either pick evidence to suit your claims while ignoring evidence that does not,
or complain that people aren't being objective. But you can't get away with doing both. In the same thread.
It's really disappointing, because it looked like you wanted to explore all avenues and possibilities when you started posting, but it seems that you weigh evidence by how much it supports your pre-decided conclusions, rather than by any objective criteria.
Cheers,
David.
Does anyone know the crossover frequency for the tweeter in the speaker that was used in this paper?
As to his theory of IM distortion, it is just that: some theory.
That's a surprising statement in this context!
IM distortion from ultrasonic signals, measurable in most hi-fi speakers on the planet, and laughably easy to hear with various signals, is just some theory? Yet the audibility of ultrasonic ringing, suggested but not proven as the reason a filtered signal could be picked out 10% better than random choice in one study, is the important thing?
I would suggest that you can either pick evidence to suit your claims while ignoring evidence that does not, or complain that people aren't being objective. But you can't get away with doing both. In the same thread.
It's really disappointing, because it looked like you wanted to explore all avenues and possibilities when you started posting, but it seems that you weigh evidence by how much it supports your pre-decided conclusions, rather than by any objective criteria.
Cheers,
David.
Wow David, can't you see how clear the objective answer to your IM audibility/false positive concerns, is.....the BS paper won an award!!! From the AES!!
What does the possibility of false positives matter, when the BS paper won an AES award from 2 reviewers who didn't review the entire manuscript?
cheers,
AJ
It really can't be said often enough to ask about more data especialy about the metal tweeter.
I am not part of amirs professionel parallel universe but i often had surprises with building speakers with metal domes.
There was some self-sound mostly unpleasant that was not to measure with simple frequency responses.
It wouldn't come as a surprise to me if the tested filters only triggered a material resonance differently.
Does anyone know the crossover frequency for the tweeter in the speaker that was used in this paper?
Probably not too much different from other speakers in their product line:
http://www.stereophile.com/content/meridia...-specifications (http://www.stereophile.com/content/meridian-dsp8000-digital-active-loudspeaker-specifications)
"Crossover frequencies: 200Hz, 2.6kHz
Looking at the lab tests, those numbers might be approximations:
http://www.stereophile.com/content/meridia...er-measurements (http://www.stereophile.com/content/meridian-dsp8000-digital-active-loudspeaker-measurements)
(http://www.stereophile.com/images/archivesart/merdsp80001.jpg)
Suggests an actual crossover to the midrange @185 Hz
and
(http://www.stereophile.com/images/archivesart/merdsp80003.jpg)
suggests a crossover to the tweeter @ 3 Khz (a surprisingly amateurish off-axis dip is the tell.
It really can't be said often enough to ask about more data especialy about the metal tweeter.
I am not part of amirs professionel parallel universe but i often had surprises with building speakers with metal domes.
There was some self-sound mostly unpleasant that was not to measure with simple frequency responses.
It wouldn't come as a surprise to me if the tested filters only triggered a material resonance differently.
The worst thing I see in many metal domed speaker drivers is that massive high Q resonance in the 20-25 KHz range. Engineering 101 - high Q resonances ruin transient response. Even if a listener could hear pre-ringing at 22 KHz due to a linear phase reconstruction filter, how would one do so in the face of massive oscillation of the tweeter dome?
Dome tweeters can be pretty counter-intuitive. There are numerous examples of soft dome tweeters with on-axis response that is only a few dB down @ 40 KHz.
https://www.madisoundspeakerstore.com/ring-...diator-tweeter/ (https://www.madisoundspeakerstore.com/ring-radiator-tweeters/vifa-xt25tg30-04-1-ring-radiator-tweeter/)
https://www.madisoundspeakerstore.com/ring-...diator-tweeter/ (https://www.madisoundspeakerstore.com/ring-radiator-tweeters/vifa-xt25tg30-04-1-ring-radiator-tweeter/)
One of the best tweeters ever build imho.
As to his theory of IM distortion, it is just that: some theory.
That's a surprising statement in this context!
IM distortion from ultrasonic signals, measurable in most hi-fi speakers on the planet, and laughably easy to hear with various signals, is just some theory?
Yes because if you read Monty's blog, it has no measurements or test cases whatsoever. If this is a real problem, real examples need to be given.
The theory also ignores the amplitude of ultrasonics in music. We can see that in the sample music in Stuart's paper:
https://www.madisoundspeakerstore.com/ring-...diator-tweeter/ (https://www.madisoundspeakerstore.com/ring-radiator-tweeters/vifa-xt25tg30-04-1-ring-radiator-tweeter/)
One of the best tweeters ever build imho.
Interestingly enough there are a number of versions of it some very conventional-looking:
https://www.madisoundspeakerstore.com/soft-...e-dome-tweeter/ (https://www.madisoundspeakerstore.com/soft-dome-tweeters-vifa/peerless-ox20sc00-04-19mm-textile-dome-tweeter/)
(https://www.madisoundspeakerstore.com/images/OX20SC00-04-curve.jpg)
which have similar measured frequency response. The above unit is only 3 dB down from its 3 KHz level on axis @ 30 KHz.
I have a pair of Primus PC351s with this tweeter, forced on me by the failure of the OEM drivers. The OEM drivers had exotic hard diaphragms and tests show the not-unexpected HF resonance. These are soft domes and are pretty flat. I make no special claims.
There are some funny stories about the origin of soft domes. For one, it is claimed that the first soft domes were made as show exhibit pieces to avoid the fact that nobody can avoid poking at domes and denting them. They guy who built them figured that no way could they have any treble, but just for fun he measured them. They were about as smooth and extended as anything.
Yes because if you read Monty's blog, it has no measurements or test cases whatsoever. If this is a real problem, real examples need to be given.
Some recent tests using Monty's IM files: https://www.gearslutz.com/board/10576574-post9.html (https://www.gearslutz.com/board/10576574-post9.html)
CONCLUSIONS
3. Our findings are consistent with the idea that
filters with long impulse responses blur fine tem-
poral details of signals. [The idea is diminished in persuasiveness given that the filtering was of ultrasonic frequencies using linear phase.]
Agreed.
This finding may be further diminished by the fact that the test involved filters with unrealistically narrow transition bands that lead to unrealistically large amounts of ringing, and that the metal dome tweeters used may have masked the behavior of the filter with their own ringing.
the fact that the test involved filters with unrealistically narrow transition bands that lead to unrealistically large amounts of ringing,
A fact that you doubtless verified, e.g. by visiting src.infinitewave.ca (http://src.infinitewave.ca), assessing the transition band width of those SRCs that are 1) commonly used in music production and 2) not obviously broken by ^*%&$^ design.
the fact that the test involved filters with unrealistically narrow transition bands that lead to unrealistically large amounts of ringing,
A fact that you doubtless verified, e.g. by visiting src.infinitewave.ca (http://src.infinitewave.ca), assessing the transition band width of those SRCs that are 1) commonly used in music production and 2) not obviously broken by ^*%&$^ design.
I did it the old fashioned way - read the Meridian AES conference paper (which said that the transition bands they used were ca. 500 Hz) and studied a number of spec sheets for DAC chips that are likely to be used in modern medium and high quality gear.
I also measured the transition bands in the SRC I used the most - CoolEdit Pro 2.1.
CoolEdit has a quality adjustment for its SRC. The lowest quality filtering @ 44.1 KHz gives a transition band of about 1.8 KHz - a lot like a real world DAC, while the highest quality setting gave a transition band of about 5 Hz.
Quote from the Meridian paper:
"The frequencies of the transition bands were 23500-
24000 Hz and 21591-22050 Hz, corresponding to the
standard sample rates of 48 kHz and 44.1 kHz respectively.4
"
A fact that you doubtless verified, e.g. by visiting src.infinitewave.ca (http://src.infinitewave.ca), assessing the transition band width of those SRCs that are 1) commonly used in music production and 2) not obviously broken by ^*%&$^ design.
Hopefully the masters use ultra steep linear phase settings, because then your DAC's or player's (resampling) filter will completely dominate the output.
This also explains why they had to use 192 kHz all the way... otherwise their 44.1 kHz apodizing filter would have eliminated 21+ kHz pre-ringing during playback.
Yes because if you read Monty's blog, it has no measurements or test cases whatsoever. If this is a real problem, real examples need to be given.
Some recent tests using Monty's IM files: https://www.gearslutz.com/board/10576574-post9.html (https://www.gearslutz.com/board/10576574-post9.html)
And we should care about near clipping distortion at 30/33Khz why?
(http://people.xiph.org/~xiphmont/demo/intermod.png)
The point is that a great many mastering tools have a transition band narrower than the paper's 500Hz. Often much narrower.
The point is that a great many mastering tools have a transition band narrower than the paper's 500Hz. Often much narrower.
So?
The paper is about simulating real-world anti-imaging filters, not resampling filters.
The point is that a great many mastering tools have a transition band narrower than the paper's 500Hz. Often much narrower.
Examples?
Not that I doubt it, as some of the tools I have are exactly thtat way.
The bad news for the Meridian boys is if the mastering people put ringing into the masters, what takes it out? As I read their materials, their Apodizing filters span of correction is the ringing in the filters in their own equipment.
Quite frankly, unless you're an outright con man/shyster type selling $50k+ audio jewelry and trying to further profit from the Hi-Re$ $cam, the BS paper has more holes than swiss cheese.
Completely contrived dither, zero data on system transparency, a high likelihood of system generated artifacts (especially metal dome tweeter driven to 108db), zero switching transparency data, time alignment, etc, etc.
As I noted earlier, it will be interesting how High End Shysters Inc. sells this BS test to the $cam addicts, when they reject any honesty controls/level matched blind tests as valid. Instead opting for purely subjectivist delusion filled sighted "comparisons" of amps, DACs, etc. and some rather hilarious stuff like +/-10% level fiddling "tests" to determine audibility of said delusions.
cheers,
AJ
As to his theory of IM distortion, it is just that: some theory.
That's a surprising statement in this context!
IM distortion from ultrasonic signals, measurable in most hi-fi speakers on the planet, and laughably easy to hear with various signals, is just some theory?
Yes because if you read Monty's blog, it has no measurements or test cases whatsoever. If this is a real problem, real examples need to be given.
The theory also ignores the amplitude of ultrasonics in music. We can see that in the sample music in Stuart's paper:
(http://amirviews.smugmug.com/photos/i-2qR2C5k/0/X2/i-2qR2C5k-X2.png)
The spectrum above 22/24 Khz is what, 90 db lower than peak in the in-band? IM distortion would be a percentage of this so it will be at even lower levels.
Here is the IM distortion test tracks Arny had us run which we passed with no audible IM detected:
(http://amirviews.smugmug.com/photos/i-RNhgjCT/0/X2/i-RNhgjCT-X2.png)
His ultrasonic test tone pairs on the right are nearly clipping level. Yet they still did not garner detection in the sampling of people who ran it.
I'm beginning to think that there were two mechansims that dominated these experiences:
(1) Audio interfaces and PC's that did internal downsampling and thus killed the test tones before they tested anything. I've found a lot of these in my explorations.
(2) People whose hearing or perceptions that were such that they didn't perceive troublesome artifacts that many experienced listeners would find to be highly audible.
The claim that artifacts due to the HF test tones were totally undetected is questionable. I surely heard artifacts due to the test tones in my tests, and I seem to recall that there were some people who switched their test setups after I put out the first set of HF IM tests. I have at least one private report that sounds all the world to me like a monitoring system that failed the test, along with strong assertions that what they heard was OK.
What takes the ringing out? Well, their apodizing filter would if the data were fed to their DSP at 44.1 kHz.
The bad news for the Meridian boys is if the mastering people put ringing into the masters, what takes it out? As I read their materials, their Apodizing filters span of correction is the ringing in the filters in their own equipment.
No, if the ringing is at the Nyquist frequency of the delivered digital audio, the Apodizing filter will remove it. As would any low pass filter that cuts below the original Nyquist frequency when upsampling. However, the point of the Apodizing filter is that it doesn't pre-ring itself. Most upsampling filters would.
I don't claim it's audible. I'm just explaining how it works.
Cheers,
David.
If we are objectivists, we better be objectivists all the time.
Good. Then can we agree on these things? Or not?
1. The BS paper reports a listening test where the filtering was detected 10% better than chance (which, in the number of trials performed, is statistically significant)
2. The different scenarios tested all gave broadly similar results overall.
3. Some differences between scenarios became larger when the data was post-sorted and re-analysed, but the rigorous way of investigating this is to re-test these apparently better conditions, not post-process the data looking for the better conditions after the fact.
4. There was no re-test, so at this stage we shouldn't claim differences between filter types, dither, etc.
5. The theories as to why/how the filtered samples were differentiated from the original samples are just that: theories. It would be possible to design listening tests to probe these theories, but this has not been done. At this stage, we shouldn't claim the filtering was audible because of X, Y or Z.
6. No one has done a listening test to probe the audibility of ultrasonic ringing itself. (AFAIK - do you know different?)
7. Audible IMD from ultrasonic signals is readily demonstrable under certain circumstances.
8. The worst audible IMD from ultrasonic signals is far more easily audible than the worst effects of ultrasonic ringing from the "worst" "CD" sinc filter.
9. Measurable IMD from ultrasonic signals reduces with ultrasonic signal level, often almost proportional to some positive integer power of the signal level, rather than linearly. (i.e. it falls away surprisingly quickly).
My objective conclusion from all the evidence is that neither mechanism should be readily audible, but in a test where people were just about able to hear some kind of difference, that difference could be down to one, the other, both, or neither - pending further investigation.
As someone else has already mentioned, Sony put (switchable!) filters on the output of their SACD players. This is an admission that it will sometimes be better to remove ultrasonic signal components, long before Monty mentioned it.
The fact ultrasonic signal components can have unintended consequences in the audible range due to equipment non-linearities is basic maths and well understood. People have known this for decades. Monty didn't make it up!
The possibility that pre-ringing in ultrasonic filters could be audible is possibly related. It's also something some people have been saying for a decade or so. There's no maths or (currently) psychoacoustics to back it up. This is the first test to properly look at it, and we can't reach any conclusions yet.
Please note which one of these is established fact, and which one is as-yet unproven theory.
Cheers,
David.
If we are objectivists, we better be objectivists all the time.
Good. Then can we agree on these things? Or not?
1. The BS paper reports a listening test where the filtering was detected 10% better than chance (which, in the number of trials performed, is statistically significant)
2. The different scenarios tested all gave broadly similar results overall.
Agree.
3. Some differences between scenarios became larger when the data was post-sorted and re-analysed, but the rigorous way of investigating this is to re-test these apparently better conditions, not post-process the data looking for the better conditions after the fact.
It is routine to perform such post-process analysis. Excluding Simpson's paradox is a good thing, not bad.
4. There was no re-test, so at this stage we shouldn't claim differences between filter types, dither, etc.
But we can say that the claim that "no double blind test has shown differences of this type" or that "the distortions are below JND" have been falsified.
5. The theories as to why/how the filtered samples were differentiated from the original samples are just that: theories. It would be possible to design listening tests to probe these theories, but this has not been done. At this stage, we shouldn't claim the filtering was audible because of X, Y or Z.
Correct.
6. No one has done a listening test to probe the audibility of ultrasonic ringing itself. (AFAIK - do you know different?)
This may be indicative of that.
7. Audible IMD from ultrasonic signals is readily demonstrable under certain circumstances.
In double blind tests with music as opposed to near clipping tones?
8. The worst audible IMD from ultrasonic signals is far more easily audible than the worst effects of ultrasonic ringing from the "worst" "CD" sinc filter.
There is no data on record to represent this. The testing in Stuart's paper used music. The only test provided for IM distortion has near clipping ultrasonic tones which has no music profile that matches it.
9. Measurable IMD from ultrasonic signals reduces with ultrasonic signal level, often almost proportional to some positive integer power of the signal level, rather than linearly. (i.e. it falls away surprisingly quickly).
Correct. And given the fact that the level of ultrasonics also drop sharply on their own, makes the IM distortion an unlikely cause.
My objective conclusion from all the evidence is that neither mechanism should be readily audible, but in a test where people were just about able to hear some kind of difference, that difference could be down to one, the other, both, or neither - pending further investigation.
Agree.
As someone else has already mentioned, Sony put (switchable!) filters on the output of their SACD players. This is an admission that it will sometimes be better to remove ultrasonic signal components, long before Monty mentioned it.
Sorry no. Removing quantization noise shoved in ultrasonic range is a good thing. It can cause amplifier oscillation and potential damage to the tweeter.
That is not the same as chopping off spectrum of music at the low levels that exists in ultrasonic range.
.
Actually Monty (and Jean-Marc Valin of xiph.org) rebutted you (and patiently rebutted all manner of tendentious nonsense from your acolytes) over and over on your forum; you simply refused to accept it (an admitted at one point that you tend not to acknowledge error because 'it's two males') and performed your usual dance steps.
Monty nailed the fallacy in your overall thesis re: the 'need' for hi rez, *back in 2012*:
http://www.whatsbestforum.com/showthread.p...ll=1#post108311 (http://www.whatsbestforum.com/showthread.php?6484-Why-24-192-is-a-bad-idea&p=108311&viewfull=1#post108311)
What you are advocating is being unable to sell better masters to an audiophile without the added expense of senseless recording overkill, because said audiophiles have been 'educated' that they need a gold plated Hummer with artillery mount to drive to the corner store.
Is this a benefit to the industry? to anybody?
I found your dance steps about wanting to hear the *ultrasonics-induced intermodulation distortion that was heard and approved in the studio* (!) particularly amusing.
Actually Monty (and Jean-Marc Valin of xiph.org) rebutted you over and over on your forum; you simply refused to accept it (an admitted at one point that you tend not to acknowledge error because 'it's two males') and performed your usual dance steps.
Monty nailed the fallacy in your overall thesis re: the 'need' for hi rez, *back in 2012*:
What, are you documenting yet another example of Amir being proven wrong and then he just wrapped himself into a tight little ball of denial?
Just goes to show that he's tried that trick with far better persons than I! ;-)
I also loved this anecdote (http://www.whatsbestforum.com/showthread.php?6484-Why-24-192-is-a-bad-idea&p=108490&viewfull=1#post108490)of Monty's, about Neil Young's Pono initiative
When Neil's group contacted me, the first thing they wanted to talk about was 24/192. I replied I probably wasn't the person they wanted to talk to, because I'm a skeptic on the subject. They replied that they didn't believe in high-res either, but they couldn't sell to audiophiles unless their product was high res. So then they wanted to know what other hooks could serve the same marketing purpose. We went back and forth on that a bit, also discussed FLAC, and the conversation eventually petered out.
It's also around this time that Monty appears to start losing patience with Amir's shucking and jiving and shilling and threats-made-with-smileys and 'endless baiting via logical fallacy (http://www.whatsbestforum.com/showthread.php?6484-Why-24-192-is-a-bad-idea&p=108493&viewfull=1#post108493)". I notice he hasn't posted to WhatsBest since 2013...
Hell of a story. Maybe now we know why Young ducked the question posed at SXSW as to what his cut would be.
The bad news for the Meridian boys is if the mastering people put ringing into the masters, what takes it out? As I read their materials, their Apodizing filters span of correction is the ringing in the filters in their own equipment.
No, if the ringing is at the Nyquist frequency of the delivered digital audio, the Apodizing filter will remove it. As would any low pass filter that cuts below the original Nyquist frequency when upsampling. However, the point of the Apodizing filter is that it doesn't pre-ring itself. Most upsampling filters would.
I don't claim it's audible. I'm just explaining how it works.
I did a little research and was surprised to find how sharp and severe their Apodizing filter actually is - approx 100 dB attenuation at Nyquist. Now you and I would probably call 100 dB attenuation removing it, but I know from past experience that Amir believes that artifacts that are 100 dB down can be serious audible problems.
As it stands, there doesn't seem to be much magic - the Apodizing filter seems to be very much like the minimum phase reconstruction filters that we have done listening tests with and found to contain no magic.
It's also around this time that Monty appears to start losing patience with Amir's shucking and jiving and shilling and threats-made-with-smileys and 'endless baiting via logical fallacy (http://www.whatsbestforum.com/showthread.php?6484-Why-24-192-is-a-bad-idea&p=108493&viewfull=1#post108493)". I notice he hasn't posted to WhatsBest since 2013...
The list of well known people who have abandoned Amir's conference site for similar reasons has to be very long and impressive by now.
Actually Monty (and Jean-Marc Valin of xiph.org) rebutted you (and patiently rebutted all manner of tendentious nonsense from your acolytes) over and over on your forum; you simply refused to accept it (an admitted at one point that you tend not to acknowledge error because 'it's two males') and performed your usual dance steps.
Monty nailed the fallacy in your overall thesis re: the 'need' for hi rez, *back in 2012*:
http://www.whatsbestforum.com/showthread.p...ll=1#post108311 (http://www.whatsbestforum.com/showthread.php?6484-Why-24-192-is-a-bad-idea&p=108311&viewfull=1#post108311)
You mean like this?
M&M [Meyer and Moran] were using fantastically expensive equipment. There was no audible IMD. That's not a big surprise.
M&M system was fantastically expensive? Stuart's system was more expensive yet. We take that as his system not producing IM either?
You proved so clearly what I said before: that this is not his domain of expertise. That you think he actually was winning some argument means you must be confused just the same.
My message to him and here have been the same: http://www.whatsbestforum.com/showthread.p...ll=1#post108430 (http://www.whatsbestforum.com/showthread.php?6484-Why-24-192-is-a-bad-idea&p=108430&viewfull=1#post108430)
What you are advocating is being unable to sell better masters to an audiophile without the added expense of senseless recording overkill, because said audiophiles have been 'educated' that they need a gold plated Hummer with artillery mount to drive to the corner store.
Is this a benefit to the industry? to anybody?
I found your dance steps about wanting to hear the *ultrasonics-induced intermodulation distortion that was heard and approved in the studio* (!) particularly amusing.
This is my exact answer to that quote per above link:
I am not advocating anything. The recording world has already decided to use high sample rate and resolution. Your beef therefore needs to be with them. Go and see if the Mix magazine will take your article and "educate" that crew. Folks here have nothing to do with that. What they want is what the talent and engineer heard when the final mix was created. Nothing less, nothing more. You pushing for 16/44.1 means getting the CD masters which is post crazy loudness wars as the Meyer and Moran paper clearly articulated with listening tests.
Nothing about loudness wars is about gold plated hammers. It is about a hammer that has a head that is metal vs one that is plastic. Folks are so wrapped around the axle on the bits and bytes that they forget the reality of how music is produced. Or worse yet, don't know how the music is produced.
Please don't put such spins on my answers when they can be refuted so easily.
I did a little research and was surprised to find how sharp and severe their Apodizing filter actually is - approx 100 dB attenuation at Nyquist. Now you and I would probably call 100 dB attenuation removing it, but I know from past experience that Amir believes that artifacts that are 100 dB down can be serious audible problems.
100 db? I wish. No, here is sample distortions from my measurements which I am sure you recall:
.
It's also around this time that Monty appears to start losing patience with Amir's shucking and jiving and shilling and threats-made-with-smileys and 'endless baiting via logical fallacy (http://www.whatsbestforum.com/showthread.php?6484-Why-24-192-is-a-bad-idea&p=108493&viewfull=1#post108493)". I notice he hasn't posted to WhatsBest since 2013...
Ditto for Dr Earl Geddes: http://www.whatsbestforum.com/showthread.p...s-forum-closing (http://www.whatsbestforum.com/showthread.php?2652-Geddes-forum-closing) in 2011.
Sean Olive (http://seanolive.blogspot.com/2010/06/science-in-service-of-art.html) has bailed since 2013 (http://www.whatsbestforum.com/forumdisplay.php?233-Science-In-The-Service-Of-Art-Dr-Sean-Olive) also. Real scientists wouldn't want to be associated with Shyster Con Central, so no surprise.
Here is what passes as "Expert" advice on "listening" to DACs (http://www.whatsbestforum.com/showthread.php?3211-Review-Benchmark-DAC1-%28Modified%29&p=48449&viewfull=1#post48449) under the watchful eyes of the WTF?? moderators.
This may sound silly..... but try putting the DAC1 upside down and have a listen.
Needless to say, would love to see how this BS paper is spun to those who outright reject blind, level matched testing as valid for discerning audible differences.
I guess even false positives are ok when they help peddle "Hi-Re$".
cheers,
AJ
Amir,
I don't understand how you can see someone demonstrating IMD and say (truthfully) these are exceptional circumstances so I will reject this but see someone else demonstrating bad filters and wrong dither and say this result is really important.
it's true that the former is more exceptional than the latter, but the audible result is dramatically clearer too.
these are almost like two sides of the same coin.
Cheers,
David.
You proved so clearly what I said before: that this is not his domain of expertise. That you think he actually was winning some argument means you must be confused just the same.
that's what so likable about you Amir: your humbleness.
Really, anyone can read the thread and decide for themselves what *substantive* points were made by whom. And who has 'domain expertise' in digital audio/signal processing
My message to him and here have been the same: http://www.whatsbestforum.com/showthread.p...ll=1#post108430 (http://www.whatsbestforum.com/showthread.php?6484-Why-24-192-is-a-bad-idea&p=108430&viewfull=1#post108430)
What you are advocating is being unable to sell better masters to an audiophile without the added expense of senseless recording overkill, because said audiophiles have been 'educated' that they need a gold plated Hummer with artillery mount to drive to the corner store.
Is this a benefit to the industry? to anybody?
I found your dance steps about wanting to hear the *ultrasonics-induced intermodulation distortion that was heard and approved in the studio* (!) particularly amusing.
This is my exact answer to that quote per above link:
I am not advocating anything. The recording world has already decided to use high sample rate and resolution. Your beef therefore needs to be with them. Go and see if the Mix magazine will take your article and "educate" that crew. Folks here have nothing to do with that. What they want is what the talent and engineer heard when the final mix was created. Nothing less, nothing more. You pushing for 16/44.1 means getting the CD masters which is post crazy loudness wars as the Meyer and Moran paper clearly articulated with listening tests.
Nothing about loudness wars is about gold plated hammers. It is about a hammer that has a head that is metal vs one that is plastic. Folks are so wrapped around the axle on the bits and bytes that they forget the reality of how music is produced. Or worse yet, don't know how the music is produced.
Please don't put such spins on my answers when they can be refuted so easily.
Shill, baby, shill.
Monty said, as I have, that he doesn't particularly care if you or anyone offers 'hi rez' releases. I've bought 'em, he probably has too. He doesn't think they make a difference in themselves, he thinks they're wasteful container formats, but he's not calling for a ban on them. Same here. What he *is* against is the following , AS I AM:
1) absurd hype about hi rez sound-- that is, the sonic differences *due necessarily to hi rez* -- all out of proportion to its actual benefits. The industry , and you, *are* engaging in this, -- you with your incessant claims that it's all about 'consumer choice', the industry with the sort of shenanigans that Monty noted in his anecdote about Neil Young's project. About which you maintained a notable silence.
2) diverting the conversation from the *
sources of truly significant audible defects* , in either Redbook OR hi rez releases:
poor mastering, and poor playback setups. Consumers should be alerted to how their 'choice' is being restricted THERE. We should not *have to * buy hi rez releases in order to get good mastering. And we aren't even guaranteed good mastering with hi rez releases!
Hi rez is a sideshow. For which you are a shill.
Kurosawa said: "Once you listen to high-res audio, you won't go back to CDs." – Rappler.com
Monty said, as I have, that he doesn't particularly care if you or anyone offers 'hi rez' releases. I've bought 'em, he probably has too. He doesn't think they make a difference in themselves, he thinks they're wasteful container formats, but he's not calling for a ban on them. Same here. What he *is* against is the following , AS I AM:
As I mentioned Steven, your opinion doesn't matter. Neither does Monty's. But since you want to continue to appeal to authority with Monty (yeh baby! ), here are two out of three introductory paragraphs in his blog:
"Unfortunately, there is no point to distributing music in 24-bit/192kHz format. Its playback fidelity is slightly inferior to 16/44.1 or 16/48, and it takes up 6 times the space.
There are a few real problems with the audio quality and 'experience' of digitally distributed music today. 24/192 solves none of them. While everyone fixates on 24/192 as a magic bullet, we're not going to see any actual improvement.If you have been saying these things, then you are both dabbling in voodoo audio science/industry. Declaring 24/192 Khz to have slightly inferior playback fidelity? I suppose both of you have double blind tests that demonstrate that? Or do we run with that with a fictitious IM distortion caused by near clipping ultrasonic tones?
Also incorrect is the implication that all high resolution releases are at 24/192 Khz. I believe Arny would call this "excluded middle." Or that anyone would care whatsoever how much more space it takes. The cost to store the bits in either case is a fraction of the cost of acquiring the content itself.
As to his next paragraph, he is mistaken because he doesn't have any experience or knowledge of the business dynamics in music industry. He doesn't know that by having three paths to release of digital audio: MP3/AAC, CD and High-res, we get to have different treatments applied to them as far as mastering. No longer do recording engineers have to convince the labels/talent to not apply loudness compression to CD/MP3s. They can do that for them as much as they like. And opt to give the people who care about fidelity better quality music in the form of "high resolution" downloads.
The people he is criticizing, the "Young Group," are proving exactly that. By having a distribution channel for 24/192 Khz for Pono, they can do whatever they want as far as mastering. How he can say he met with them and not appreciate this fact is unfathomable.
1) absurd hype about hi rez sound-- that is, the sonic differences *due necessarily to hi rez* -- all out of proportion to its actual benefits. The industry , and you, *are* engaging in this, -- you with your incessant claims that it's all about 'consumer choice', the industry with the sort of shenanigans that Monty noted in his anecdote about Neil Young's project. About which you maintained a notable silence.
2) diverting the conversation from the *sources of truly significant audible defects* , in either Redbook OR hi rez releases: poor mastering, and poor playback setups. Consumers should be alerted to how their 'choice' is being restricted THERE. We should not *have to * buy hi rez releases in order to get good mastering. And we aren't even guaranteed good mastering with hi rez releases!
Hi rez is a sideshow. For which you are a shill.
I hear you Steven. Here is the problem and there is no getting around it. You wouldn't accept my opinion of a hot topic in your professional field. Likewise as a person with no educational or professional experience in this topic, I can't accept your position because it is not an informed one. Monty knows a lot more than you but likewise is operating outside of his expertise. He has one meeting with the "Young Group" and thinks all of a sudden he is qualified to give us state of union for audio market? I don't think so. That is why he is making such obvious mistakes per above.
Your arguments as his, are wrong. There is no reason to take any action against development of high resolution audio distribution. There is goodness to every aspect of it. No harm comes to any of you happy with MP3 or the CD. The CD may decline but that decline will come from MP3, not from high-resolution. This is why the market is moving forward over your and Monty's strenuous objections. The dynamics are not as you envision as a person standing outside of it
Edit: typos again.
Kurosawa said: "Once you listen to high-res audio, you won't go back to CDs." – Rappler.com
That's
Taku Kurosawa, who works at a hi-re$ music download site. What would you expect him to say? "I don't think there any audible consequences, but I feel it is important to maintain my site so people have access to the tracks for experimentation"?!
The people he is criticizing, the "Young Group," are proving exactly that. By having a distribution channel for 24/192 Khz for Pono, they can do whatever they want as far as mastering. How he can say he met with them and not appreciate this fact is unfathomable. ]
The 'Young Group' proved that they are willing to shuck and jive the consumer to sell a new product, as the industry is wont to do.
1) absurd hype about hi rez sound-- that is, the sonic differences *due necessarily to hi rez* -- all out of proportion to its actual benefits. The industry , and you, *are* engaging in this, -- you with your incessant claims that it's all about 'consumer choice', the industry with the sort of shenanigans that Monty noted in his anecdote about Neil Young's project. About which you maintained a notable silence.
2) diverting the conversation from the *sources of truly significant audible defects* , in either Redbook OR hi rez releases: poor mastering, and poor playback setups. Consumers should be alerted to how their 'choice' is being restricted THERE. We should not *have to * buy hi rez releases in order to get good mastering. And we aren't even guaranteed good mastering with hi rez releases!
Hi rez is a sideshow. For which you are a shill.
I hear you Steven.
Oh,
do you now? Ok, let's dance. Do you
agree with what you hear from me? Namely :
-that oft-claimed audible improvements due to hi rez are vastly inflated;
-that a hi rez delivery format
has not,
does not , and without a change in industry practices,
will not assure high-quality recordings for the consumer;
-and that the actual bottlenecks for good sound for the modern audio consumer are, by far, the recording/production/mastering of the product, and their own home listening setups?
There is goodness to every aspect of it. No harm comes to any of you
You mean like this?
(http://i1057.photobucket.com/albums/t386/k2head/eden_zpse80ec435.jpg)
There is no reason to take any action against development of high resolution audio $10k ML DACs and $50k ML amplifier distribution. There is goodne$$ to every aspect of it. No harm comes to any of you happy with MP3 an audibly indistinguishable $150 ODAC or the CD $3k NAD M22 Hypex NCore amp.
FIFY
Yes Amir, it's clear that Krab doesn't understand the busine$$ side of this field, does he?
Of course, only your overheated imagination strawman "takes action" against $cams. Rational folks simply don't buy into the $cam and of course object to the B$ claims.
No rational, objective person "takes action" against the guy buying the Rolex over the Casio watch. But if the audiomoron Rolex buyer or $cam peddler claims the Rolex "tells better time" than the Casio, well, here we are.
cheers,
AJ
I hear you Steven.
Oh, do you now? Ok, let's dance. Do you agree with what you hear from me? Namely :
-that oft-claimed audible improvements due to hi rez are vastly inflated;
Yes. The difference in specification alone has very small incremental fidelity difference.
-that a hi rez delivery format has not, does not , and without a change in industry practices, will not assure high-quality recordings for the consumer;
No. "Excluded middle" again. We don't need assurance. It is like saying the government should have never mandated seatbelts because it could not assure zero fatalities in cars. Availability of a new distribution branch whose sole reason for existence is improved fidelity above CD, has and will continue to drive production of better sounding masters for that channel. It need not achieve 100% success for it to be hugely valuable for audiophiles.
-and that the actual bottlenecks for good sound for the modern audio consumer are, by far, the recording/production/mastering of the product, and their own home listening setups?
Moot per above.
No rational, objective person "takes action" against the guy buying the Rolex over the Casio watch. But if the audiomoron Rolex buyer or $cam peddler claims the Rolex "tells better time" than the Casio, well, here we are.
The only moron is the one thinking your analogy is correct. We are not dealing with Casio. We are dealing with a Chinese knock-off that instead of a quartz oscillator, has an RC resonator because they think the customer wanted clocks that look like clock but don't keep accurate time. Perhaps you have heard of loudness compression of CD/MP3s?
The Rolex as it turns out in this situation, is indeed the more accurate clock. And it hardly costs the same premium over the Casio. It is the master that was used to make the cheap knock off above.
We are not dealing with Casio.
Yep, 16/44 playback is the Casio of music distribution. Tells time just as audibly-to-human-ears-accurate (but maybe not fabricated online logs) as the Rolex, for a lot less money. It's just that $50k $cam amp buyers would rather have a Rolex version of the music for bragging rights and less "worry" about losing street cred about their 12KHz "listening" skills.
Perhaps you have heard of loudness compression of CD/MP3s?
Perhaps you'll never know what a Red Herring is? Wait, never mind...
The Rolex as it turns out in this situation, is indeed the more accurate clock.
Nope. Just the more expensive version of that same 2ch 10% soundfield, or worse, Neil Young and GAGA studio electronic music, that audiomoron suckers and $cam peddlers think, needs "more resolution".
But as you said Amir, few understand this $cam busine$$ field.
cheers,
AJ
Casio:
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
Street hustler fake Rolex rip-off:
http://www.aes.org/e-lib/browse.cfm?elib=17497 (http://www.aes.org/e-lib/browse.cfm?elib=17497)
cheers,
AJ
No rational, objective person "takes action" against the guy buying the Rolex over the Casio watch. But if the audiomoron Rolex buyer or $cam peddler claims the Rolex "tells better time" than the Casio, well, here we are.
The only moron is the one thinking your analogy is correct.
I am totally amazed by the utter respect that you show for people who disagree with you Amir. ;-)
We are not dealing with Casio.
Another example of the well known lack of reading comprehension. We're not talking about action with respect to the Casio side of the argument, we're talking about the Rolex side.
We are dealing with a Chinese knock-off that instead of a quartz oscillator, has an RC resonator because they think the customer wanted clocks that look like clock but don't keep accurate time.
So says someone who wants us to believe that switchmode power supplies always have poorer regulation and more noise than legacy iron transformer based power supplies.
Reality is that MEMS (non-quartz) resonators outperform quartz for time keeping:
http://electronicdesign.com/analog/look-in...ems-clock-chips (http://electronicdesign.com/analog/look-inside-programmable-mems-clock-chips)
Perhaps you have heard of loudness compression of CD/MP3s?
And that has what to do with the Audibility of Typical Digital Filters in a Hi-Fi Playback Systems? Loudness compression is a common production procedure that can be applied to any recording regardless of its digital format. How ignorant does one have to be to claim that all CDs and MP3s are loudness compressed? Seems like an idea that a rank amateur that had no idea of modern audio technology might use in an argument and give every knowlegeable reader a big laugh at their expense.
The Rolex as it turns out in this situation, is indeed the more accurate clock. And it hardly costs the same premium over the Casio. It is the master that was used to make the cheap knock off above.
Please demonstrate this claim with real world examples.
Please demonstrate this claim with real world examples.
Recording done at 24/96 Khz. Converted to 16/44.1. There is no way on earth the 24/96 is less accurate than the 16/44.1 as the analogy of Rolex vs Casio would tell you.
But as you said Amir, few understand this $cam busine$$ field.
Ammar, you went to a high-end audio show and wired up your speaker this way:
(http://audiograb.files.wordpress.com/2014/05/ct6a6429.jpg)
Multi-thousand dollar flat ribbon cable used in parallel with yet another cable? You have discovered something about audio science there Ammar? Or was it a scam of sorts?
Please demonstrate this claim with real world examples.
Recording done at 24/96 Khz. Converted to 16/44.1. There is no way on earth the 24/96 is less accurate than the 16/44.1 as the analogy of Rolex vs Casio would tell you.
Obviously the scrobblings of a poorly educated subjectivist amateur audiophile who believes high end audio hype that bitrate is everything. Professionals with relevant industry experience and training know that producing recordings is a multi step process and resolution and accuracy can be lost at any step along the way.
Thus it is easy for a 24/96 recording to be less accurate than a 16/44 recording, and is even probably true in many cases.
But as you said Amir, few understand this $cam busine$$ field.
Ammar, you went to a high-end audio show and wired up your speaker this way:
(http://audiograb.files.wordpress.com/2014/05/ct6a6429.jpg)
Multi-thousand dollar flat ribbon cable used in parallel with yet another cable? You have discovered something about audio science there Ammar? Or was it a scam of sorts?
Heavens no, that's called synergy!
The paper is about simulating real-world anti-imaging filters, not resampling filters.
This beggars belief. Please go back and re-read the paper's abstract. Do you actually understand anything of this?
The psychoacoustics per slides I post from JJ (lack of pre-temporal masking) is there.
That is only part of the story. There is more JJ to read on this topic (*).
It is an accepted fact that linear-phase filtering-type pre-ringing is readily audible, when the pertinent filter's cut-off falls within the audio band. Do the experiment. Learn what pre-ringing sounds like. Then slowly move up the transition point. You'll notice that the artefact disappears once you get high enough. This suggests that ultrasonic ringing is not audible.
But even when the ringing exists in-band there is a suggestion (or established practice? Ask JJ) to make it harmless. Make the filter's transition band somewhat wider than the ear's critical band width at the cut-off frequency. For CD mastering, and assuming the existence of bat-like people who detect - and rejoice in - 22kHz signals, this amounts to 2-3 kHz critical band, leading to a filter transition band of 4 kHz, give or take. Go out. Ensnare a bunch of teenagers. Do the test.
(* And not because JJ is infallible, but because he did us the enormous favour of compiling all of this widespread knowledge and making it available in one single place.)
Please go back and re-read the paper's abstract.
You are right regarding the paper, they specifically set out to test "typical" A/D filters.
But that still makes the whole paper kinda irrelevant to how such files will sound with a real-world DAC, DACs with apodizing filters, players with resamplers with arbitrarily configurable filters ...
As I said before, the A/D or mastering filter can be extremely steep. Normally it will run through at least another filter which usually dominates how an impulse will finally look like.
You have discovered something about audio science there Ammar?
No, we've discovered you claim x-ray vision along with Amir Geller hearing.
Or was it a scam of sorts?
Amir, you're peddling the Hi-Re$ scam to the wrong folks here.
cheers,
AJ
The paper is about simulating real-world anti-imaging filters, not resampling filters.
This beggars belief. Please go back and re-read the paper's abstract. Do you actually understand anything of this?
The paper may not be well written:
Its title is: "The audibility of typical digital audio fi lters in a high- fidelity
playback system."
"This paper describes listening tests investigating the audibility of various lters applied in a high-resolution wideband digital
playback system."
As most of us know, a digital playback system must contain a DAC, and is very unlikely to contain an ADC. ADCs are well known to be components of recording systems.
However, later the paper says:
"Filter responses tested were representative of anti-alias filters used in A/D (analogue-to-digital) converters or mastering processes."
Regrettably, this is not the only confusion in the paper.
The psychoacoustics per slides I post from JJ (lack of pre-temporal masking) is there.
That is only part of the story. There is more JJ to read on this topic (*).
It is an accepted fact that linear-phase filtering-type pre-ringing is readily audible, when the pertinent filter's cut-off falls within the audio band. Do the experiment. Learn what pre-ringing sounds like. Then slowly move up the transition point. You'll notice that the artefact disappears once you get high enough. This suggests that ultrasonic ringing is not audible.
But even when the ringing exists in-band there is a suggestion (or established practice? Ask JJ) to make it harmless. Make the filter's transition band somewhat wider than the ear's critical band width at the cut-off frequency. For CD mastering, and assuming the existence of bat-like people who detect - and rejoice in - 22kHz signals, this amounts to 2-3 kHz critical band, leading to a filter transition band of 4 kHz, give or take. Go out. Ensnare a bunch of teenagers. Do the test.
(* And not because JJ is infallible, but because he did us the enormous favour of compiling all of this widespread knowledge and making it available in one single place.)
Ah, it is refreshing to run into another knowledgeable member here . In my defense , I was simply answering the point that there was no psychoacoustics involved in this area. So I gave the example of temporal masking. On the larger point, I have only stated that as thinking out loud as an explanation of the results in this listening test/paper. I am not taking the position that there is causation here.
My read of JJ's position is that he is leaving the door open for potential audibility here;
(http://amirviews.smugmug.com/photos/i-kbsQ4Q4/0/X2/i-kbsQ4Q4-X2.png)
Followed by this slide:
(http://amirviews.smugmug.com/photos/i-r8cBMRt/0/X2/i-r8cBMRt-X2.png)
Given the latest data from Stuart, more testing here may prove useful. But again, I am not taking a position that this is audible.
I hear you Steven.
Oh, do you now? Ok, let's dance. Do you agree with what you hear from me? Namely :
-that oft-claimed audible improvements due to hi rez are vastly inflated;
Yes. The difference in specification alone has very small incremental fidelity difference.
Let's all highlight that with our markers.
Now the next step in the dance:
Do you specifically agree that when the industry promotes hi rez to consumers, it touts *quite notable* or even *obvious* 'improved fidelity' --
not a 'very small incremental' improvement? This is 'inflation' from the industry side.
If your answer is 'yes', then 'hi rez' specification really
is a sideshow to the main event, and your answer here is akin to carnival barking.....:
-that a hi rez delivery format has not, does not , and without a change in industry practices, will not assure high-quality recordings for the consumer;
No. "Excluded middle" again. We don't need assurance. It is like saying the government should have never mandated seatbelts because it could not assure zero fatalities in cars. Availability of a new distribution branch whose sole reason for existence is improved fidelity above CD, has and will continue to drive production of better sounding masters for that channel. It need not achieve 100% success for it to be hugely valuable for audiophiles.
....because if we have no 'assurances' that the 'improved fidelity' *won't* just be just the 'very small, incremental' bump that the specification alone gives us, then why are we being promised otherwise? It has broken that promise before, with DVD-A and SACDs and HD tracks that were not notably better sourced or mastered than the CD version -- and sometimes were even 'worse'.
Why not 'drive' the industry
explicitly to live up to its promises? Because that would involve focusing on delivering *better sounding masters*, not *high rez*? Because it would mean acknowledging that 'better sounding masters' offering 99.9% of the audible quality that well-mastered high rez releases do, could be delivered to consumers on CD, on mp3 too?
And thus your answer here needs revisiting; the issue is hardly 'moot' when 'hi rez' won't deliver what it promises:
-and that the actual bottlenecks for good sound for the modern audio consumer are, by far, the recording/production/mastering of the product, and their own home listening setups?
Moot per above.
BTW, I think we've finally pinned you down on the *real* core issue here -- which is not the small, incremental audible differences between digital filters. If we continue from this point and reach resolution, we can close this thread down.
-and that the actual bottlenecks for good sound for the modern audio consumer are, by far, the recording/production/mastering of the product, and their own home listening setups?
However, the nature of the problems with the recording/production/mastering of the product is not in the electronics. Rather the problem is in the very acoustics of picking up sound with microphones. In fact very serious irreversible damage to sound quality takes place by the time the signal gets to the electrical terminals of the microphone.
Thererfore further improvements in the signal processing quality of the electronics will not address the problem.
Obsessing over and fiddling with the electronics, the filters, and the like is just obfuscating the real problem. This is where the narrow minds of Amir, the other golden ears/subjectivists, the mice at Meridian, etc. get derailed.
This article describes where audio went off the rails, and where we need to go back to (ca. 1664), and start correcting it:
http://en.wikipedia.org/wiki/Tin_can_telephone (http://en.wikipedia.org/wiki/Tin_can_telephone)
Can we at least agree that the vast majority of the listening public are justifiably quite happy with CD quality, or even lossy, and that only a small segment needs, or thinks that it needs, hi-rez?
Now the next step in the dance:
Can't wait...
Do you specifically agree that when the industry promotes hi rez to consumers, it touts *quite notable* or even *obvious* 'improved fidelity' -- not a 'very small incremental' improvement? This is 'inflation' from the industry side.
What "industry?" Audio/Video? If so, marketing is the name of the game. This is a money losing, high-competitive business. Every TV manufacturer sells "LED TVs" yet there is no such animal. It is an LCD TV that its backlight has been replaced with LED. They pretend the rest of is also changed as to entice you to buy a new TV. They talk about 1,000,000:1 contrast ratio where the reality is 1,000:1 due to the obvious way they cheat there. Do we go and pollute every technical video discussion by ranting about this? I sure hope not. I assume this is a technical form. People come here to learn about how the technology works. They don't come here for the millionth time to read about your angst with high-end marketing in general, and high-resolution marketing in the specific.
This is a very technical topic. One that is outside of your area of expertise. But you still want to be known as a mover and shaker so you have decided to make it cause célèbre. Ah yes, let's cry over how high-res is marketed. Every high-end hotel I go to has a bottle of Fiji water in the room now that it says is there for my "convenience." Fine print says they will charge me $5 for that small bottle of water. It is what they do to make money. It is like expensive popcorn in theaters. That is the way they make money. Who asked you to pick up the cause to complain about equiv. in audio in every
technical thread?
And why in the heck do you think by doing so with me you are scoring some point? You think I have not heard this argument? You think I don't know the argument? I Know it. And despite that, I hold the strong position that this development, availability of high-resolution audio download is good for us. Your myopic, self-serving attitude of propping your position in audio forums is not material to me.
If your answer is 'yes', then 'hi rez' specification really is a sideshow to the main event, and your answer here is akin to carnival barking.....:
How fast you prove my point above. You go from a technical statement to a PR spin. Said spin is wrong anyway Steven. You have to listen for a moment what is being explained. T
hat by establishing a new distribution channel, and one whose sole reason for existence is satisfying high-fidelity customers, the requirement for loudness compression goes away. No longer do we have to settle for what the mass consumer wants. Toyota created the Lexus brand so that it can put luxury features in its cars. And not care about absolute price elasticity. You want to highlight something with a magic marker, the above is it. Everything else you say is in the service of trying to keep yourself and your arguments relevant.
....because if we have no 'assurances' that the 'improved fidelity' *won't* just be just the 'very small, incremental' bump that the specification alone gives us, then why are we being promised otherwise? It has broken that promise before, with DVD-A and SACDs and HD tracks that were not notably better sourced or mastered than the CD version -- and sometimes were even 'worse'.
So because you don't have an assurance you want the new option to be foreclosed for all distributors and all listeners? What illogical thinking.
There are no assurances anywhere. You have to a smart shopper. Nissan makes an SUV that is very unreliable compared to the rest of their fleet. Do we get to write-ff Nissan and Japanese car reliability because the "assurance" was broken?
Of course not. You can get what you want. Or I should say what I want. THat is, instead of spending all of our time wasting it with your lay, self-serving arguments, we use it to discuss with each other the fidelity of new releases. We can machine analyze them and compare them subjectively. Then we can choose to spend our money or not.
And it is not like anyone gives a you know what, what is being discussed on this topic anyway. The industry marketing is not exactly monitoring idiot rants on forums and go and change what they do. So you are wasting everyone's time with tired lay arguments of no value.
Can we at least agree that the vast majority of the listening public are justifiably quite happy with CD quality, or even lossy, and that only a small segment needs, or thinks that it needs, hi-rez?
Absolutely.
Can we at least agree that the vast majority of the listening public are justifiably quite happy with CD quality, or even lossy, and that only a small segment needs, or thinks that it needs, hi-rez?
We can't even get agreement on what exactly is being simulated, let alone whether it represents what would be found in a typical listening environment.
Interesting that you think technical facts "pollute" a technical discussion.
This is a very technical topic. One that is outside of your area of expertise. But you still want to be known as a mover and shaker so you have decided to make it cause célèbre. Ah yes, let's cry over how high-res is marketed. Every high-end hotel I go to has a bottle of Fiji water in the room now that it says is there for my "convenience." Fine print says they will charge me $5 for that small bottle of water. It is what they do to make money. It is like expensive popcorn in theaters. That is the way they make money. Who asked you to pick up the cause to complain about equiv. in audio in every technical thread?
Who asked you to be the official Fiji water audio salesman? For Fiji water audio distribution, aka "Hi Re$"? Or Fiji water amplification, DACs, etc?
Especially here, where folks aren't dumb enough to fall for same ol' tap water in a $5 bottle, unlike "Audiophiles", who will.
Great analogy btw, thanks.
Maybe we
do understand the busine$$ aspect of the Hi-Re$ $cam better than you think and have decided that 16/44 tap water is just fine for us, without the "worry free" $5 packaging.
cheers,
AJ
Interesting that you think technical facts "pollute" a technical discussion.
Even more interesting is you thinking this statement is technical:
Do you specifically agree that when the industry promotes hi rez to consumers, it touts *quite notable* or even *obvious* 'improved fidelity' -- not a 'very small incremental' improvement? This is 'inflation' from the industry side.
If this is technical, we need to give honorary engineering degrees to every corporate PR person...
Or this one.
This is a very technical topic. One that is outside of your area of expertise. But you still want to be known as a mover and shaker so you have decided to make it cause célèbre. Ah yes, let's cry over how high-res is marketed. Every high-end hotel I go to has a bottle of Fiji water in the room now that it says is there for my "convenience." Fine print says they will charge me $5 for that small bottle of water. It is what they do to make money. It is like expensive popcorn in theaters. That is the way they make money. Who asked you to pick up the cause to complain about equiv. in audio in every technical thread?
Who asked you to be the official Fiji water audio salesman? For Fiji water audio distribution, aka "Hi Re$"? Or Fiji water amplification, DACs, etc?
Especially here, where folks aren't dumb enough to fall for same ol' tap water in a $5 bottle, unlike "Audiophiles", who will.
Great analogy btw, thanks.
Maybe we do understand the busine$$ aspect of the Hi-Re$ $cam better than you think and have decided that 16/44 tap water is just fine for us, without the "worry free" $5 packaging.
cheers,
AJ
Is there no minimum standard for technical discussions in this forum?
Now the next step in the dance:
Do you specifically agree that when the industry promotes hi rez to consumers, it touts *quite notable* or even *obvious* 'improved fidelity' -- not a 'very small incremental' improvement? This is 'inflation' from the industry side.
What "industry?" Audio/Video? If so, marketing is the name of the game. This is a money losing, high-competitive business. Every TV manufacturer sells "LED TVs" yet there is no such animal. It is an LCD TV that its backlight has been replaced with LED. They pretend the rest of is also changed as to entice you to buy a new TV. They talk about 1,000,000:1 contrast ratio where the reality is 1,000:1 due to the obvious way they cheat there. Do we go and pollute every technical video discussion by ranting about this? I sure hope not. I assume this is a technical form. People come here to learn about how the technology works. They don't come here for the millionth time to read about your angst with high-end marketing in general, and high-resolution marketing in the specific.
This is a very technical topic. One that is outside of your area of expertise. But you still want to be known as a mover and shaker so you have decided to make it cause célèbre. Ah yes, let's cry over how high-res is marketed. Every high-end hotel I go to has a bottle of Fiji water in the room now that it says is there for my "convenience." Fine print says they will charge me $5 for that small bottle of water. It is what they do to make money. It is like expensive popcorn in theaters. That is the way they make money. Who asked you to pick up the cause to complain about equiv. in audio in every technical thread?
And why in the heck do you think by doing so with me you are scoring some point? You think I have not heard this argument? You think I don't know the argument? I Know it. And despite that, I hold the strong position that this development, availability of high-resolution audio download is good for us. Your myopic, self-serving attitude of propping your position in audio forums is not material to me.
I'll take that as a 'yes', then.
If your answer is 'yes', then 'hi rez' specification really is a sideshow to the main event, and your answer here is akin to carnival barking.....:
How fast you prove my point above. You go from a technical statement to a PR spin. Said spin is wrong anyway Steven. You have to listen for a moment what is being explained. That by establishing a new distribution channel, and one whose sole reason for existence is satisfying high-fidelity customers, the requirement for loudness compression goes away. No longer do we have to settle for what the mass consumer wants. Toyota created the Lexus brand so that it can put luxury features in its cars. And not care about absolute price elasticity.
You want to highlight something with a magic marker, the above is it. Everything else you say is in the service of trying to keep yourself and your arguments relevant.
Ah, so, 'satisfying high fidelity customers' apparently means, pandering to whatever silly, *technically unsound* beliefs they hold dear about the audible improvement a high rez 'specification' brings. Not just pandering to them...fostering them.
So you don't mind getting behind a *fib* like that?
(You have been less than diligent in correcting the nutty things some people on your own forum believe to be true abount digital audio, so it's not surprising.)
And btw, in my experience, when *car* analogies are introduced to audio arguments, someone is getting desperate. Not to mention that you're asking us to admire that the audio industry wants to emulate *car salesmen*.
....because if we have no 'assurances' that the 'improved fidelity' *won't* just be just the 'very small, incremental' bump that the specification alone gives us, then why are we being promised otherwise? It has broken that promise before, with DVD-A and SACDs and HD tracks that were not notably better sourced or mastered than the CD version -- and sometimes were even 'worse'.
So because you don't have an assurance you want the new option to be foreclosed for all distributors and all listeners? What illogical thinking.
What new option? 'High rez' audio is already here , as physical media and downloads. It's been around for more than a decade now. Same bullshit marketing now as then.
There are no assurances anywhere. You have to a smart shopper. Nissan makes an SUV that is very unreliable compared to the rest of their fleet. Do we get to write-ff Nissan and Japanese car reliability because the "assurance" was broken?
But anyway: an SUV that gets a reputation for being 'unreliable' is going to make Nissans' rep suffer. And if Nissan is smart they will fix it, or phase out that model. Toyota took a big hit to their rep a few years ago, and you can *bet* they've worked hard to address that.
Will the industry you're shilling for work hard to keep the mastering quality of 'high rez' releases' high? If not, will we at least see some fine print absolving them of any such promise?
Of course not. You can get what you want.
If I want good mastering, how do I know that I'm going to get it? Wait for an objective review in Stereophile?
Or I should say what I want. THat is, instead of spending all of our time wasting it with your lay, self-serving arguments, we use it to discuss with each other the fidelity of new releases. We can machine analyze them and compare them subjectively. Then we can choose to spend our money or not.
And it is not like anyone gives a you know what, what is being discussed on this topic anyway. The industry marketing is not exactly monitoring idiot rants on forums and go and change what they do. So you are wasting everyone's time with tired lay arguments of no value.
Idiot rants? Is that your *professionalism* showing again, Amir? Or is this you demonstrating 'technical discussion'
Or do we run with that with a fictitious IM distortion caused by near clipping ultrasonic tones?
Unfortunately the paper made no attempt to determine what the cause of the audible difference actually was, so no plausible source can be dismissed as "fictitious." We simply do not know.
And without access to the actual files used there are avenues that can never be explored.
Or do we run with that with a fictitious IM distortion caused by near clipping ultrasonic tones?
Unfortunately the paper made no attempt to determine what the cause of the audible difference actually was, so no plausible source can be dismissed as "fictitious." We simply do not know.
That is true but we are not totally in the dark due to measurement provided:
(http://amirviews.smugmug.com/photos/i-2qR2C5k/0/X2/i-2qR2C5k-X2.png)
Who here wants to represent that such distortion product driven by that low level of energy in orange would be audible?
And without access to the actual files used there are avenues that can never be explored.
The files are readily available.
LOL! The other penny drops!
Meridian's new proprietary hi rez delivery format
http://musicischanging.com/ (http://musicischanging.com/)
HA thread about it here:
http://www.hydrogenaud.io/forums/index.php?showtopic=107666 (http://www.hydrogenaud.io/forums/index.php?showtopic=107666)
Or do we run with that with a fictitious IM distortion caused by near clipping ultrasonic tones?
Unfortunately the paper made no attempt to determine what the cause of the audible difference actually was, so no plausible source can be dismissed as "fictitious." We simply do not know.
That is true but we are not totally in the dark due to measurement provided:
(http://amirviews.smugmug.com/photos/i-2qR2C5k/0/X2/i-2qR2C5k-X2.png)
Who here wants to represent that such distortion product driven by that low level of energy in orange would be audible?
You realize that the ringing that this paper in the end is all about occurs exactly at that low level where these filtered frequencies begin?
And without access to the actual files used there are avenues that can never be explored.
The files are readily available.
Link to the high yield portions as they were processed by Meridian, please?
And without access to the actual files used there are avenues that can never be explored.
The files are readily available.
Link to the high yield portions as they were processed by Meridian, please?
There are no processed files. The transforms were inside Matlab.
LOL! The other penny drops!
Meridian's new proprietary hi rez delivery format
http://musicischanging.com/ (http://musicischanging.com/)
HA thread about it here:
http://www.hydrogenaud.io/forums/index.php?showtopic=107666 (http://www.hydrogenaud.io/forums/index.php?showtopic=107666)
Is it a bird? Is it a plane? Is it PSR? No, it's super Hi-Re$ man to the rescue, with.....compressed Hi-Re$?????
WTH? CD sized file streaming of 10% 2ch studio constructs. Wow. I wonder what this will cost vs say, Tidal?
Where have I seen this before? Oh yes
(http://tribkiah.files.wordpress.com/2013/06/fiji.jpg)
Looks like there is some BS appreciation of the business side of this industry. I can hardly wait to hear Neil Young in "studio quality" MQA.....
cheers,
AJ
There are no processed files. The transforms were inside Matlab.
Ok, where does one acquire these exact MATLAB "transforms" used, to repeat the experiment?
You realize that the ringing that this paper in the end is all about occurs exactly at that low level where these filtered frequencies begin?
Perhaps easier seen than heard?
They did crank those 1" DR dome beryllium tweeter speakers up to 108db, near breaking point. Plus there is no info on the switching software transparency. Makes one wonder exactly what it is they thought they heard?
"Look, I know the supernatural is something that isn't supposed to happen, but it
does happen"
cheers,
AJ
Where have I seen this before?
Perhaps you are experiencing deja vu from when Oohashi released his second paper on the "benefits" of the hypersonic effect just in the nick of time to promote the release of the only movie soundtrack ever to be released on Blu-ray which had an actual hypersonic musical score, recorded by none other than [...drum roll please] Oohashi himself!
http://www.hydrogenaud.io/forums/index.php...st&p=873474 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=106744&view=findpost&p=873474)
What a startling coincidence *this* paper, the topic of our thread, also just happens to support a questionable hi-re$ audio technology which similarly financially benefits the very authors of the paper, just in the nick of time, in the promotional release of THEIR new audio format! *CHA-CHING*
...but the abstract is AE$-award winning!
Where have I seen this before?
Perhaps you are experiencing deja vu from when Oohashi released his second paper on the "benefits" of the hypersonic effect just in the nick of time to promote the release of the only movie soundtrack ever to be released on Blu-ray which had an actual hypersonic musical score, recorded by none other than [...drum roll please] Oohashi himself!
http://www.hydrogenaud.io/forums/index.php...st&p=873474 (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=106744&view=findpost&p=873474)
What a startling coincidence *this* paper, the topic of our thread, also just happens to support a questionable hi-re$ audio technology which similarly financially benefits the very authors of the paper, just in the nick of time, in the promotional release of THEIR new audio format! *CHA-CHING*
Right... I am sure in their PR packet forCES they will include a copy of this paper. "Reporter: so how do we know high-resolution audio sounds good?" "Meridian: here. Just read the paper and see."
You guys are this out of touch with reality or just say these things for effect?
And without access to the actual files used there are avenues that can never be explored.
The files are readily available.
The actual snippets they used? Links if you would.
And without access to the actual files used there are avenues that can never be explored.
The files are readily available.
The actual snippets they used? Links if you would.
Appendix A in the paper.
So, this thread falls under ToS #14 now?
LOL! The other penny drops!
Meridian's new proprietary hi rez delivery format
http://musicischanging.com/ (http://musicischanging.com/)
HA thread about it here:
http://www.hydrogenaud.io/forums/index.php?showtopic=107666 (http://www.hydrogenaud.io/forums/index.php?showtopic=107666)
Well, if you guys had taken my hint and gone and read the other paper ...
http://www.hydrogenaud.io/forums/index.php...mp;#entry882550 (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&st=700&p=882550&#entry882550)
...you would have seen this coming.
Cheers,
David.
Interesting that you think technical facts "pollute" a technical discussion.
Even more interesting is you thinking this statement is technical:
I wasn't implying anything about what anyone else said.
I was commenting on what you said...
Every TV manufacturer sells "LED TVs" yet there is no such animal. It is an LCD TV that its backlight has been replaced with LED. They pretend the rest of is also changed as to entice you to buy a new TV. They talk about 1,000,000:1 contrast ratio where the reality is 1,000:1 due to the obvious way they cheat there. Do we go and pollute every technical video discussion by ranting about this?
Maybe stating facts multiple times is what you count as ranting. I don't know. I think most of us, certainly including you , have made the same points over and over in this thread. I shall try to avoid repeating myself in future.
Cheers,
David.
And without access to the actual files used there are avenues that can never be explored.
The files are readily available.
False claim.
What is alleged to the original source files appear to be available from its commercial source.
The processed files that they were compared to in the test are exactly where?
I think most of us, certainly including you , have made the same points over and over in this thread. I shall try to avoid repeating myself in future.
The only reason why I have repeated anything in this thread is because a certain person makes numerous ludicrous false claims, irrefutable counter evidence is provided, and he ignores the irrefutable evidence and makes the same ludicrous claims over and over again.
Who here wants to represent that such distortion product driven by that low level of energy in orange would be audible?
If the stronger signals at lower frequencies are pushing the tweeter into non-linear operation it is possible that some of the ultrasonic signal is being mixed down to the audible range. You cannot consider only the ultrasonic signal in isolation. (I'm a microwave engineer...is the term "mixed/mixer" used by audio engineers.?)
(I'm a microwave engineer...is the term "mixed/mixer" used by audio engineers.?)
Not in the same way, no.
Mixing is weighted summing in audio land, not multiplication. IM is the term they'll likely use for sum and difference products.
(I'm a microwave engineer...is the term "mixed/mixer" used by audio engineers.?)
Not in the same way, no.
Mixing is weighted summing in audio land, not multiplication. IM is the term they'll likely use for sum and difference products.
Thanks. In hindsight I should have known that.
And without access to the actual files used there are avenues that can never be explored.
The files are readily available.
The actual snippets they used? Links if you would.
Appendix A in the paper.
Since I have the paper:
7. APPENDIX A
segment start stop
1 0.5 13.5
2 13.5 25.5
3 25.5 35.4
4 35.4 44.8
5 44.8 58.3
6 58.3 67.7
7 67.7 78.1
8 78.1 92.3
9 92.3 106.6
10 106.6 119.2
11 119.2 129.9
12 129.9 144.93
13 144.7 154.2
14 154.2 165.02
15 165.1 176.68
16 176.68 188.45
17 188.6 202.0
Table 3: Details of how we split the recording into 17 sections.
Lets say that they wanted to actually document the processing that they did. They would give us the Matlab code?
Who here wants to represent that such distortion product driven by that low level of energy in orange would be audible?
If the stronger signals at lower frequencies are pushing the tweeter into non-linear operation it is possible that some of the ultrasonic signal is being mixed down to the audible range. You cannot consider only the ultrasonic signal in isolation. (I'm a microwave engineer...is the term "mixed/mixer" used by audio engineers.?)
The two main ways this happens with loudspeakers:
(1) The ordinary way - cross-modulation due to amplitude modulation. All real world speakers are somewhat nonlinear (on the order of 0.1% to 10% or more), due to inconsistencies in the electrodynamic motor (voice coil/field magnet) and mechanical nonlinearities in the diaphragm's suspension. The former is often dominant.
(2) Doppler distortion due to the motion of the diaphragm.
Lets say that they wanted to actually document the processing that they did. They would give us the Matlab code?
There is no code.
Lets say that they wanted to actually document the processing that they did. They would give us the Matlab code?
There is no code.
There's no spoon either.
Guess repeatability wasn't one of the goals of the BS paper.
Just a farce to prime the pump for MQA eh?
cheers,
AJ
Lets say that they wanted to actually document the processing that they did. They would give us the Matlab code?
There is no code.
There's no spoon either.
Guess repeatability wasn't one of the goals of the BS paper.
And your guess would be dead wrong. Once again, the paper is not written for lay people who don't know a FIR filter from a Fir tree.
Once again, the paper is not written for lay people
We know. It was written for shyster peddlers of the Hi-Re$ $cam, who have knowledge in the business field of selling Fiji water.
Repackage this:
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
In square $5 bottle per song, pitch via shysters to the audiomorons who crave $cams.
Got it.
Once again, the paper is not written for lay people
We know. It was written for shyster peddlers of the Hi-Re$ $cam, who have knowledge in the business field of selling Fiji water.
Repackage this:
(http://i216.photobucket.com/albums/cc73/AJinFLA/TriangularDither.jpg)
In square $5 bottle per song, pitch via shysters to the audiomorons who crave $cams.
Got it.
David, do you approve of these posts?
Same question for our kind moderators: is this post which has been repeatedly posted compliant with forum terms of service and spirit of the same?
Lets say that they wanted to actually document the processing that they did. They would give us the Matlab code?
There is no code.
There would be. Even if you drive it through the GUI, it will generate an equivalent MATLAB command line which calls MATLAB functions which in turn generate filter coefficients.
The command line would be sufficient for anyone with MATLAB and the appropriate toolbox. Everyone else would need the coefficients. There are 1629 of them for one filter, and 1496 for the other. They will be symmetric, so you only strictly need half. You wouldn't write them in a paper but you might put them on a website. If you wanted to make sure something was proerly documented and repeatable you would do both (e.g. see penultimate section) (http://replaygain.hydrogenaud.io/proposal/equal_loudness.html).
Cheers,
David.
Lets say that they wanted to actually document the processing that they did. They would give us the Matlab code?
There is no code.
There would be. Even if you drive it through the GUI, it will generate an equivalent MATLAB command line which calls MATLAB functions which in turn generate filter coefficients.
The command line would be sufficient for anyone with MATLAB and the appropriate toolbox. Everyone else would need the coefficients. There are 1629 of them for one filter, and 1496 for the other. They will be symmetric, so you only strictly need half. You wouldn't write them in a paper but you might put them on a website. If you wanted to make sure something was proerly documented and repeatable you would do both (e.g. see penultimate section) (http://replaygain.hydrogenaud.io/proposal/equal_loudness.html).
Cheers,
David.
Ah but we wouldn't understand. We don't know an IIR filter from an IRS audit. ;>
David, do you approve of these posts?
HA is a reactively moderated forum. I am neither moderator nor deity. I don't need to approve anything.
However, I fear your question and my response are both off topic.
Cheers,
David.
Lets say that they wanted to actually document the processing that they did. They would give us the Matlab code?
There is no code.
There would be. Even if you drive it through the GUI, it will generate an equivalent MATLAB command line which calls MATLAB functions which in turn generate filter coefficients.
That's not code. One line invocation of the filter subroutine in Matlab is not code. The filter response and all the parameters are given. And the tool used to generate the filter, Matlab filter GUI, specified. In addition the response of the filter is graphed. Anyone who knows how to use matlab can recreate the same. People who can't, need not apply.
David, do you approve of these posts?
HA is a reactively moderated forum. I am neither moderator nor deity. I don't need to approve anything.
However, I fear your question and my response are both off topic.
Cheers,
David.
I assumed it was not the last go around when you comment thusly:
Interesting that you think technical facts "pollute" a technical discussion.
Does his post constitute polluting a technical discussion? You objected to my remark and I am trying to get calibrated on what you mean.
Lets say that they wanted to actually document the processing that they did. They would give us the Matlab code?
There is no code.
There's no spoon either. Guess repeatability wasn't one of the goals of the BS paper. Just a farce to prime the pump for MQA eh? cheers, AJ
But isn't reproducibility one of the main principles of the scientific method? Oh, it is:
http://en.wikipedia.org/wiki/Reproducibility (http://en.wikipedia.org/wiki/Reproducibility)
The AES gives
awards to such papers?
Lets say that they wanted to actually document the processing that they did. They would give us the Matlab code?
There is no code.
There's no spoon either. Guess repeatability wasn't one of the goals of the BS paper. Just a farce to prime the pump for MQA eh? cheers, AJ
But isn't reproducibility one of the main principles of the scientific method? Oh, it is:
http://en.wikipedia.org/wiki/Reproducibility (http://en.wikipedia.org/wiki/Reproducibility)
The AES gives awards to such papers?
There is no requirement for any schmuck to be able to reproduce a scientific experiment. The requirement is for others schooled in the art. Not lay people pretending to understand the topic.
So, this thread falls under ToS #14 now?
The AES frowns on papers used exclusively for describing promotional or commercial products
. They have to show other "merit":
"Commercialism: A manuscript which is based on a commercial product should be reviewed extremely carefully to determine the real scientific content. If a manuscript has no other merit than as a description of the product, it is not acceptable. This requirement is especially important in articles that provide technical results without an adequate description of a device's operation."
http://www.aes.org/journal/authors/guidelines/ (http://www.aes.org/journal/authors/guidelines/)
So perhaps this is why their MQA technology had to be announced
after these related BS papers got published and not before.
all the parameters are given.
They're not. The documentation for that toolbox is freely available on line. Take a look. You will find that several user adjustable parameters are not documented in the paper.
Cheers,
David.
There is no requirement for any schmuck to be able to reproduce a scientific experiment. The requirement is for others schooled in the art.
So shysters only. Those who understand the business field of Hi-Re$. Hmmm, doesn't sound very scientific or repeatable.
But we wouldn't want repetition prior to the launch of MQA anyway, the $cam works better that way.
cheers,
AJ
There is no requirement for any schmuck to be able to reproduce a scientific experiment. The requirement is for others schooled in the art. Not lay people pretending to understand the topic.
Which you've demonstrated that you do not (need I remind you of LTI system basics?), so why talk down on more knowledgeable people? Oh right, it's you.
But since you have matlab, please post the 1-2 lines of code here that produce the exact same 44.1 kHz filter.
You will find that several user adjustable parameters are not documented in the paper.
David, does "filter builder" apply dither or is a separate dither toolbox required ?
Lets say that they wanted to actually document the processing that they did. They would give us the Matlab code?
There is no code.
There's no spoon either. Guess repeatability wasn't one of the goals of the BS paper. Just a farce to prime the pump for MQA eh? cheers, AJ
But isn't reproducibility one of the main principles of the scientific method? Oh, it is:
http://en.wikipedia.org/wiki/Reproducibility (http://en.wikipedia.org/wiki/Reproducibility)
The AES gives awards to such papers?
There is no requirement for any schmuck to be able to reproduce a scientific experiment.
Of course not. Your answers to reasonable questions about the reproducability of the experiment in this paper shows that you don't seem to know what it would take even a very knowledgeable person to accurately reproduce this experiment.
Here is the command syntax that would likely be used:
http://www.mathworks.com/help/signal/ref/f...gn.lowpass.html (http://www.mathworks.com/help/signal/ref/fdesign.lowpass.html)
fdesign.lowpass
Syntax
D = fdesign.lowpass
D = fdesign.lowpass(SPEC)
D = fdesign.lowpass(SPEC,specvalue1,specvalue2,...)
D = fdesign.lowpass(specvalue1,specvalue2,specvalue3,specvalue4)
D = fdesign.lowpass(...,Fs)
D = fdesign.lowpass(...,MAGUNITS)
Description
D = fdesign.lowpass constructs a lowpass filter specification object D, applying default values for the default specification string 'Fp,Fst,Ap,Ast'.
D = fdesign.lowpass(SPEC) constructs object D and sets the Specification property to the string in SPEC. Entries in the SPEC string represent various filter response features, such as the filter order, that govern the filter design. Valid entries for SPEC are shown below. The strings are not case sensitive.
'Fp,Fst,Ap,Ast' (default spec)
'N,F3db'
'N,F3db,Ap' *
'N,F3db,Ap,Ast' *
'N,F3db,Ast' *
'N,F3db,Fst' *
'N,Fc'
'N,Fc,Ap,Ast'
'N,Fp,Ap'
'N,Fp,Ap,Ast'
'N,Fp,Fst,Ap' *
'N,Fp,F3db' *
'N,Fp,Fst'
'N,Fp,Fst,Ast' *
'N,Fst,Ap,Ast' *
'N,Fst,Ast'
'Nb,Na,Fp,Fst' *
etc.
Here's a chance to redeem yourself, Amir. Tell us what parameters we need and what their values are, based on the article.
Amir, no answer or an incorrect answer is a strong indication that you have described yourself with the following somewhat indelicate wording: "There is no requirement for any schmuck to be able to reproduce a scientific experiment."
You will find that several user adjustable parameters are not documented in the paper.
David, does "filter builder" apply dither or is a separate dither toolbox required ?
Sorry Kees, I don't know. Dither is absent from the equivalent tool box in my 2009 release, but may have been added since. You would normally use double precision floating point in matlab, so don't dither until you need integers for playback or integer wave files.
Cheers,
David.
This whole BS test farce is now moot IMO, given that we now have the real McCoy from the BS company: MQA.
Now we need to see them demo their new MQA $cam at shows with the same Meridian DSP7200 rig, vs the TPDF 16/44 version of the master, so that the vast improvement in SQ can be "heard".
Of course, part of the con involving this BS paper, is that there is no TPDF 16/44 "CD" version of the 2L master. What a coincidence.
cheers,
AJ
Now, thanks to this paper, they can point to it should anyone contest that the difference offered by MQA isn't actually audible. They'll contend: "But didn't you get the memo: the 2nd offered conclusion in the AES paper's abstract? It noted that only truly high fidelity audio systems can reveal the difference, so if your pedestrian $23K speakers don't cut it, that's because they are half the quality of the ones we used."
There is no requirement for any schmuck to be able to reproduce a scientific experiment. The requirement is for others schooled in the art. Not lay people pretending to understand the topic.
Hmm, well, JJ is a member of this forum as well as AES and is, safe to say, at least *passingly* familiar with MATLAB. I would imagine some other members of the AES are too. Are they 'schmucks'? They are, after all , the primary audience for that peer-reviewed prize winning convention paper.
There's other people at HA who know MATLAB too. Hell, I have labmates who use it, and we're just a lowly *biologists*.
So, who here at HA *do* you think is qualified to reproduce this experiment and results from the information given? Anyone besides yourself?
Now, thanks to this paper, they can point to it should anyone contest that the difference offered by MQA isn't actually audible. They'll contend: "But didn't you get the memo: the 2nd offered conclusion in the AES paper's abstract? It noted that only truly high fidelity audio systems can reveal the difference, so if your pedestrian $23K speakers don't cut it, that's because they are half the quality of the ones we used."
The paper looks to me like a PR coup for Meridian in particular, and the golden eared segment of the audio industry in general.
Unfortunately past bastions of scientific approaches to audio have such as Dolby have passed to a new generation of leadership who buy into this sort of thing.
I'm "Looking Forward" to several of Stereophile's columnists taking a few victory laps, based on it.
There is no requirement for any schmuck to be able to reproduce a scientific experiment. The requirement is for others schooled in the art. Not lay people pretending to understand the topic.
Hmm, well, JJ is a member of this forum as well as AES and is, safe to say, at least *passingly* familiar with MATLAB. I would imagine some other members of the AES are too. Are they 'schmucks'? They are, after all , the primary audience for that peer-reviewed prize winning convention paper.
There's other people at HA who know MATLAB too. Hell, I have labmates who use it, and we're just a lowly *biologists*.
So, who here at HA *do* you think is qualified to reproduce this experiment and results from the information given? Anyone besides yourself?
I think that we may already have a real world answer to that, based on Amir's response to http://www.hydrogenaud.io/forums/index.php...st&p=883605 (http://www.hydrogenaud.io/forums/index.php?showtopic=107124&view=findpost&p=883605)
It appears that Matlab licenses are just under $2K, so there is an element of elitism to its use.
Whether the same thing can be done with Octave or some other alternative becomes an interesting question:
http://octave.sourceforge.net/signal/overview.html (http://octave.sourceforge.net/signal/overview.html)
Of course the golden ears will argue that no open source software can really duplicate the output of the high priced spread.
OTOH, part of affirming results by means of an independent experimenter can/should include using other, comparable tools. If only a certain tool provides the same results, it is possible that the observed results with it are the result of some glitch within the tool.
Given the fact that Meridian's use of Matlab inexplicably diverged from the stated purpose of the paper...
Now, thanks to this paper, they can point to it should anyone contest that the difference offered by MQA isn't actually audible. They'll contend: "But didn't you get the memo: the 2nd offered conclusion in the AES paper's abstract? It noted that only truly high fidelity audio systems can reveal the difference, so if your pedestrian $23K speakers don't cut it, that's because they are half the quality of the ones we used."
This whole BS test farce is now moot IMO, given that we now have the real McCoy from the BS company: MQA.
You too look like you are going to blow a blood vessel over this .
So let me share with you some good news, bad news. Which one do you want to hear first? Ah, I figured you would pick the bad news first.
Bad News:This is a perceptually lossless scheme for high-resolution audio. Stated (example) efficiency is in the order of 10:1 with respect 24/192Khz, bringing its data rate from nearly 10 mbit/sec to less than that of the CD, i.e. 1 mbit/sec. So there goes the argument that it takes 6 times more to store high resolution audio. The encoding (I think) is layered allowing a lossy 16-bit base layer without a special decoder. The efficiency comes from going from the "rectangle" of bit depth and sampling rate to a much more customized shape where only the data that is a) not noise and b) is important perceptually is saved. There is no reliance on masking and such but rather preserving the "timing" of the signal (xnor: you get to go nuclear now ) in addition to frequency domain.
They have been working with both Sony Music and WMG (Warner) on testing and development of the technology so I would expect both of those labels to be present at their CES announcement.
Good news:The world needs a new audio format like a hole in the head. The audience for high-resolution audio is enthusiasts who a) don't care about bandwidth or storage costs and b) won't accept any argument of perceptual equality. They will insist on getting the original bits.
Getting a new decoder into all manner of devices and players when it comes from a tiny competitive company called Meridian is a non-starter.
While the format is also targeted at archiving, I expect zero uptake there as the cost of storage there compared to production is negligible.
There is this inside story of tech company executives who dream of waking up one day and being in "content business" and rubbing shoulders with the stars in music/movies. The calling is quite strong and folks jump at any idea, no matter how stupid from economical point of view, and jump with both feet. Who doesn't want to stand on the stage with Sony and Warner executives on stage. No longer do you show up with some boxes on a tech show where there is no attention from mainstream press.
Music labels only care about MGs (minimum guarantees) these days. They will sleep with AJ and Mzil if they showed up with a $500,000 check. They will hand you their catalog, wish you good luck and hope the next tech exec falling victim to above.
So Meridian will put up some tracks. Build the decoders and such for this. A couple of years from now hopefully they wake up and realize what a distraction and financial drain this has been and mothball the efforts. At least I hope it only takes that long.
So don't have a heart attack over this. There is no "there there." It is a solution looking for a problem that the end customer does not have. Yet it creates friction in delivering the same.
I told you there is no money in audio business, didn't I?
all the parameters are given.
They're not. The documentation for that toolbox is freely available on line. Take a look. You will find that several user adjustable parameters are not documented in the paper.
Cheers,
David.
Take a look? I already post filters I created in Matlab. I don't need to look at help files online. The design parameters of the filter are documented. That doesn't mean they are going to hold your hands and give you everything you need to type in Matlab.
What parameters are you worried about that impact audibility of the filters?
Stated (example) efficiency is in the order of 10:1 with respect 24/192Khz, bringing its data rate from nearly 10 mbit/sec to less than that of the CD, i.e. 1 mbit/sec.
Isn't it ironic how you doubted the 12bit/32kHz figure I gave for the
actual information content of contemporary music earlier, and now you yourself say it's a good thing that you can compress 24bit/192kHz by a factor of ten, which fits pretty well to something like 12-13 bits and 32-38 kHz. Keep in mind that according to my information the 13(?) MSBs of the new format store the lossless information, so the lossy part will in most cases only contain noise anyway. That you can compress the files that much without damaging the perceptual result shows that the
actual information content of (some) contemporary music is in the range I stated earlier, and that delivery of 24bit/192kHz is wasteful and pointless. The compression efficiency shows that in the most obvious manner.
Good news:
The world needs a new audio format like a hole in the head. The audience for high-resolution audio is enthusiasts who a) don't care about bandwidth or storage costs and b) won't accept any argument of perceptual equality. They will insist on getting the original bits.
And why should we accept a worse implementation of lossyWAV which delivers intentionally crippled data to non-Meridian customers?
all the parameters are given.
They're not. The documentation for that toolbox is freely available on line. Take a look. You will find that several user adjustable parameters are not documented in the paper.
Cheers,
David.
Take a look? I already post filters I created in Matlab. I don't need to look at help files online. The design parameters of the filter are documented. That doesn't mean they are going to hold your hands and give you everything you need to type in Matlab.
What parameters are you worried about that impact audibility of the filters?
I don't have the list on my phone, but you ask the wrong question. We're talkng about re-creating the first experiment that showed this class of filter is just about audible in some specific circumsances, and you want me to guess which filter parameters I can change without impacting audibility? Two months ago there was no evidence that any parameter related soley to ultrasonic performance was audible. Even now, there's no evidence that the exact same filters, equipment and listeners can recreate the experiment. It would be ridiculous to second guess what changes might be audible when designing a repeat experiment, unless you want to intentionally test that difference. We don't. We want to recreate the original experiment as closely as possible.
It would be possible to get close based on the published data, but in this context that's not good enough IMO.
Cheers,
David.
I already post filters I created in Matlab. I don't need to look at help files online. The design parameters of the filter are documented. That doesn't mean they are going to hold your hands and give you everything you need to type in Matlab.
Please post the URL of the post with your Matlab documentation, Amir.
Amir, why aren't you just honest for one single time and just say that you don't know what the precise parameters are to reproduce the exact same filters?
Anyway, as WernerO pointed out, they basically simulated a mix of A/D and mastering filters. Both don't tell you how it will sound with a less steep filter in the D/A conversion process.
You can use a ultra steep lowpass filter when exporting a CD mixdown .. so what? Whenever someone is listening to it, it will have to have run through a filter that has a lot less ringing, and potentially no pre-ringing.
The only way to directly "get" (let's not say hear) such filters is by applying them manually on material with higher sampling rate, e.g. 192 kHz. That has nothing to do with real-world playback.
I already post filters I created in Matlab. I don't need to look at help files online. The design parameters of the filter are documented. That doesn't mean they are going to hold your hands and give you everything you need to type in Matlab.
Please post the URL of the post with your Matlab documentation, Amir.
What?
Amir, why aren't you just honest for one single time and just say that you don't know what the precise parameters are to reproduce the exact same filters?
Lot of good it did me the last time I shared personal data related to this conversation. You guys are not interested in such data. Chu who brought up this topic says can't run any test because he doesn't even have a computer. Arny has only run one double blind test in months of these discussions. That one test was his own bloody test which he would not run after hundreds of posts and requests. Steven (Krab) has not run any tests. Mzil has cheated on the one test he did run. AJ has not run any test. You xnor, have refused to run all tests on moral grounds. I came here and even ran David's test which he himself has not run. Our two kind moderators have not run or reported on any of these tests either.
So no, when you all start to show some interest in actual data as opposed to searching high and low for your future talking points to dismiss these results, I will join you. For now you can pursue the witch-hunt yourself.
"Don't know the exact filter parameters..." Are you kidding me?
Anyway, as WernerO pointed out, they basically simulated a mix of A/D and mastering filters.
No, what he pointed out was this:
The paper is about simulating real-world anti-imaging filters, not resampling filters.
This beggars belief. Please go back and re-read the paper's abstract. Do you actually understand anything of this?
Hey I can do the 'Amir' too!
Next time someone tells me 'Amir thinks hi rez is a good thing' I'll write:
No, what he pointed out was this
The difference in specification alone has very small incremental fidelity difference.
So no, when you all start to show some interest in actual data as opposed to searching high and low for your future talking points to dismiss these results, I will join you. For now you can pursue the witch-hunt yourself.
There is no witch hunt, just growing impatience with irrelevant interjections and obfuscation.
Lot of good it did me the last time I shared personal data related to this conversation. You guys are not interested in such data.
Nobody is asking you for stories about how your wife told you how to program the lights at your house, your business flights, your preferred car type or how you fill out questionnaires.
You said all that is needed is a matlab line and the parameters given.
You also wrote:
There is no requirement for any schmuck to be able to reproduce a scientific experiment. The requirement is for others schooled in the art. Not lay people pretending to understand the topic.
Instead of admitting that you're that "schmuck" and that lay person, you dance around, turn things around by requesting others to give you data ...
Typical case of zero amir credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
You xnor, have refused to run all tests on moral grounds.
This isn't about me, or any of the other names you dropped.
I actually have run many tests, even posted a log in here which was
completely off-topic. Typical amir
lies guys, see amir credibility (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=882029).
So no, when you all start to show some interest in actual data as opposed to searching high and low for your future talking points to dismiss these results, I will join you. For now you can pursue the witch-hunt yourself.
You gotta take responsibility for the asinine nonsense you post here regularly, instead of foisting it off on others.
"Don't know the exact filter parameters..." Are you kidding me?
No, just post the matlab line(s) that produce the same filter.
It could be as easy and simple as that, but you seem to turn everything into these painful back and forths with lots of dancing around, excuses and blaming others. You seem to have been doing this for too many years to be able to participate like a normal human being in a discussion.
No, what he pointed out was this:
His harsh response to my mistake doesn't change the problem at all:
"But that still makes the whole paper kinda irrelevant to how such files will sound with a real-world DAC, DACs with apodizing filters, players with resamplers with arbitrarily configurable filters ...
As I said before, the A/D or mastering filter can be extremely steep. Normally it will run through at least another filter which usually dominates how an impulse will finally look like."Why don't you for once just address what was said, and not who said it how and what the person posted before (sometimes even on other forums) ... ?
This is clearly a hopeless case.
More specifically, it is a matter of "He doesn't fit" with regard to the matter of agreeing to TOS8.
Let me close the chapter "amirm" (all based on evidence found in this thread (http://www.hydrogenaud.io/forums/index.php?s=&showtopic=107124&view=findpost&p=881982)):
- stereotyping people, which he even admitted to
The most dramatic example of this being the matter of his attempts to stereotype himself as being a science-abiding objectivist, which totally fails any sniff test. It might work except for his own efforts at defeating himself.
- keeps on making straw men and red herrings among countless other fallacies
Good example being our recent attempts to square his subjectivist, bad science, golden-eared articles on his for-profit commercial web site with his claims to be a science-loving objectivist. You can't have it both ways, Amir!
- keeps shifting the burden of proof but doesn't even notice it
I disagree. His response to my sequence of repetitions of the question: "Amir, please confirm that you actually understand the irrelevancy of the above comments to our discussion of your proud flaunting of bogus sighted evaluations of power amplifiers." is "I won't confirm anything" which asks even begs for the interpretation that he notices his dishonest arguments, makes them consciously, and is even proud of them.
- sets up flawed arguments where he either wins or wins (that is all he seems to care about anyway)
I think that's the whole point. Winning is his prime directive but he can't win any fair contest so he tries to set up these bogus contests that he can't lose no matter how poorly he plays.
- dares to speak about logic and science when he has demonstrated (dare I say it) willful ignorance
I think the phrase "Willful ignorance" may be an oxymoron. You can't will away things you can easily remember.
I agree with what I perceive to be the meaning of your statement, even though I dispute the logic of the phrase. The facts clearly support the common meaning of that phrase. It is not clear to me that Amir has the technical knowledge, as well as language and logic skills required to play the audio debate game well. Therefore, the ignorant false claims aren't willful. He can't help himself because he lacks the language skills required to read and comprehend correctly. He can't help himself because he lacks the technical background to critically examine many technical claims, whether his own or others. He can't help himself because in ordinary circumstances he can and will avoid taking responsibility for his own statements and their logical consequence.
- thinks this is a war
More specifically total, no-holds-barred war. Any false claim no matter how ridiculous can be proffered. Any irrelevant fact or theory can be trotted out. Compromising his personal integrity to the point of extinction seems to be OK with him.
- admitted to not hearing differences but still magically produces ABX logs
- rather posts multiple pages of nonsense than answering simple questions, one of which he finally explicitly evaded with a lame excuse
- poor reading comprehension (fun fact: I am from Australia according to him)
- even defends fallacies (!)
... and loads of other nonsense.
By now, and this is probably the worst, I think that most if not everyone here can see that "he has no interest in intellectual honesty or scientific rigor".
I take the paper for what it is. amir on the other hand blindly accepts whatever fits his agenda. Amir, do you know that there are peer-reviewed papers for homeopathy, and actually quite a bit more than just 1 paper, but the scientific consensus is still: nonsense, quackery, a sham?
If you had any intellectual honesty, you would have to admit to believe in all kinds of nonsense based on some peer-reviewed papers.
More the point we have his own writings proudly posted on his corporate web site which are far more flawed than the most if not all of the writings of our favorite Golden Eared Clowns like Harley and his old boss.
This all boils down to: no credibility, no honesty, no interest in truth. I will simply link back to this post whenever it fits, and trust me, this will be every other post if he continues like that.
I am sorry that it came to that.
I might quibble with some of the small details but the basic thrust of your comments are just fine with me. It is how things are and it is nothing that we are going to change or even nudge.
His harsh response to my mistake doesn't change the problem at all:
It doesn't change the problem. That is true. But he is saying you don't know which end is up:
The paper is about simulating real-world anti-imaging filters, not resampling filters.
This beggars belief. Please go back and re-read the paper's abstract. Do you actually understand anything of this?
Instead of constantly posting 12-part responses, getting emotional in each one of them, and making up yet another challenge for me to pass, reflect a bit on the statement above. Let the more knowledgeable people carry the torch.
Let the more knowledgeable people carry the torch.
And the cash to the bank selling $50k $cam-amps, $cam DACs and the Hi-Re$ scam. IOW, people who know the busine$$ side of these issues.
Hey Amir, good luck with your $5 square bottle Fiji water MQA distribution, which is what this contrived dither doctored BS paper is all about. I'm sure Madrona will carry this hardware eventually, like the Berkley etc, to compliment the $50k $cam-amps.
cheers,
AJ
I think it was well established in this thread that the authors of this paper (intentionally?) used inappropriate practices to push Meridian's agenda to sell expensive gear and their new format MQA.
This is it, please find a new playground for trolling and feeding. This thread will close as per ToS #5 and ToS #2. If I see any of you engaging in further silly verbal fisticuffs appropriate administrative action will be taken. This affects all of the involved, and we will try to make sure that neither of you will not be able to derail any more topics.