Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Article: Why We Need Audiophiles (Read 501287 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Article: Why We Need Audiophiles

Reply #350 –
Quote from: cpchan on
Quote from: Stereoeditor on
The report on the debate at http://www.stereophile.com/news/050905debate/ includes a link to an MP3 of the debate between Mr. Krueger and myself,


Does anyone else find it ironic that for people who are so against lossy codecs that they are offering a mp3 for download on their site? I expected an old fashioned Shorten file. 


If you think that's amusing, you'll find it hilarious that, given that my magazine's name is _Stereo_phile, the recording of the debate is a _mono_ file. :-)

As for Stereophile being against lossy codecs, Colorado-based audiophiles can attend a series of demonstrations I am giving May 5, 6, & 7 comparing hi-rez recordings against Red Book and lossy versions. Details later this weekend on the www.stereophile.com home page.

John Atkinson
Editor, Stereophile

Article: Why We Need Audiophiles

Reply #351 –
Quote from: analogcorner on
Yet no matter what I say, I am stereotyped as someone who "hates digital," or who "doesn't listen to digital," or whatever, by people who are happier to attack me for something I am not.


Unless you are claiming that you actually never had a "Compact Discs Suck" bumper sticker and you are recanting articles like this one: http://www.musicangle.com/feat.php?id=106 then I would argue that you have stereotyped yourself.

Article: Why We Need Audiophiles

Reply #352 –
Quote from: Stereoeditor on
As for Stereophile being against lossy codecs, Colorado-based audiophiles can attend a series of demonstrations I am giving May 5, 6, & 7 comparing hi-rez recordings against Red Book and lossy versions. Details later this weekend on the www.stereophile.com home page.

Will you be judging the lossy files using graphs, or will you be using your ears during controlled tests?

Article: Why We Need Audiophiles

Reply #353 –
Quote from: ShowsOn on
Quote from: Stereoeditor on
As for Stereophile being against lossy codecs, Colorado-based audiophiles can attend a series of demonstrations I am giving May 5, 6, & 7 comparing hi-rez recordings against Red Book and lossy versions. Details later this weekend on the www.stereophile.com home page.

Will you be judging the lossy files using graphs, or will you be using your ears during controlled tests?


Those aren' tthe only options. Read more closely - I used the word "demonstrations." I have a  series of files derived from the original hi-rez data, along the lines of Philip Hobbs' test described at http://www.stereophile.com/asweseeit/1007awsi/ . The downsampled, decimated, lossy compressed versions were all resampled to 24-bit/88.2kHz files, so the only substantive difference will be the coding. Allowing the people who attend my seminars to hear for themselves whether they can hear any differences and if so, what the magnitude of those differences might be, is one of the purposes of my public demonstrations.

I also have files prepared containing just the _differences_ between the lower-rez versions and the original, so listeners can hear what has been discarded by the various codecs. You may feel that use of a lossy codec is justifiable but that doesn't mean you shouldn't be aware of what it is doing, surely?

John Atkinson
Editor, Stereophile

Article: Why We Need Audiophiles

Reply #354 –
There are MP3 codecs I wouldn't even use for my answering machine. What encoder has been used?

Article: Why We Need Audiophiles

Reply #355 –
See footnote 1 at  :  http://www.stereophile.com/features/308mp3cd/index.html

Footnote 1: Something I have rarely seen discussed is the fact is that because all compressed file formats, both lossless and lossy, effectively have zero data redundancy, they are much more vulnerable than uncompressed files to bit errors in transmission.


This is the most funny statement I have read recently from audiofools.
Who is the author of that story ?

John Atkinson !

He reveals great technical knowledge of the basics of maths, PC, audio

Unfortunately he is fooling his readers

What makes me sad, is, he recommends wav uncompressed over compression without losses (so called "Lossless") like FLAC etc.

Not only in this footnote, also above in the text:


So to us at Stereophile, the question of which lossy codec is "the best" is moot. We recommend that, for serious listening, our readers use uncompressed audio file formats, such as WAV or AIF—or, if file size is an issue because of limited hard-drive space, use a lossless format such as FLAC or ALC. These will be audibly transparent to all listeners at all times with all kinds of music through all systems.


This is really sad to read.
it really shows null, zero understanding of technics, underlying simple maths or even usability of wav vs. FLAC.


So, Mr. John Atkinson plus Mr. Michael Fremer,
welcome in your world of fooling other people.

At some points, I am really still undecided, if you 2 believe yourself, what you write.
See the examples above


So, to bring some light into technics for Audiophiles (I count myself into this category, coming from the ancient Greek and Latin meanings):

FLAC offers tags, wav doesn't properly.
FLAC saves space, about 30% to 50% of the wav.
FLAC has no technical disadvantages, only advantages to wav.
FLAC stands as synonym for the other Lossles formats, but FLAC is the only one necessary, other Lossless formats are redundant, as Lossless  = Lossless = WAV = FLAC = APE = WAVPACK = ALAC = etc.

As you now see, there is no reason to prefer wav over FLAC.
Audiophiles do the opposite, Audiophiles prefer FLAC over wav for practical reasons.
In a proper system FLAC and wav will play identical sound quality,
in a borked bugged system like with faulty RAM,
both, FLAC and wav, would be bugged.
But that would be similar to vinyl playback at wrong speed, with a ton of dust on the vinyl,
or playback via CD-player with faulty electrics, so that sound gets muted, on and off..



For those listeners with some audio experience it is no question, that vinyl can sound simply great.
But the reason for this was already discussed lentghy here at HA !

Here at HA is the regular publication medium for Dr. David Robinson, inventor of ReplayGain, called 2Bdecided.
Replaygain is a value, which describes the average loudness of a song or an averaged loudness of a complete album.

The reason for modern CDs sounding worse than same new albums released on vinyl is so simple, read here at HA.
You will get to know that already by the replaygain values.
Point to: Loudness race/war.


But this is not the fault of the CD standard at 16 bit stereo, 44,1 kHz sampling.
The technical possibilites as medium to come the live experience of music as close as possible,
these possibilities are owned by the CD.
Depending on the producers....

Nevertheless, vinyl already offers great playback capabilites, but CD offers even more....






my personal footnote 1 to my texts:
Please excuse, if i should have simplified a bit in here,
but I needed to write about technical things that way, that Mr. John Atkinson and maybe even a Michael Fremer, have a little bit of chance, to grasp it.
Also it helps, that people outside of HA, get the points about those Audiofools writing about Lossless or any formats or devices at Stereophile.

2. ABX footnote as additional technical explanation:

A/B/X listening , as test methodoligy , helps to find out, if an individuum can separate between 2 test-setups, be it 2 different formats or 2 different devices like speakers or amps.
ABX can show/prove only, that there is a difference.
ABX might show, that there is no difference in generally, but ABX cannot prove with 100% certainty, that there is no difference at all, eg. for all listeners.
ABX cannot prove the absence of a difference, written more compact.

AB means: sighted comparison.

ABX means:
1. step AB sighted known comparison, to get familiar with (tiny/small) differences between test candidates A + B.
2. step X: hidden/Blinded comparison of X, which is either A or B. X can be tested/listened directly against A or B, to find out, if X is either A or B.
of course, a proper ABX test needs some thinking at beginning, and maybe help by 2nd or 3rd person and/or technical parts like switcher, foobar2000 etc.





footnote 3: the 1.000.000 challenge about audio cables for Michael Fremer

Because Fremer has so many years experience in comparing audio, I suggest, he should take the James Randi challenge to earn the million dollars and to prove finally, that somebody can listen (not see) a difference between different audio cables, as long those are not totally corrupted towards the HiFi setup. (like dust on vinyl).
Unfortunately I recall an ugly conversation between Randi and Fremer, which had shown writing Fremer a lot of in CAPITALS and using some weird words, I did not understand as non-native English speaker.
The capitals did not appear here so many so far, or not so obvious, but I found again some words from Fremer, I wasn't able to grasp.
I cite, give citates:

Fremer:

The psychiatrist line was funny, I have to admit. However I have many in my family and their behavior leaves plenty to desire. As for shoving anything up one's ass, well you should know.

Sorry, I don't understand, shoving ass ?
is it anything audio related ?

THERE I GO AGAIN WITH THE CAR ANALOGIES...

oh, i excuse, there are the capitals. My sensitive ears, they hurt so easily when other shout...



I really couldn't give a shit. I'm outta here. Nice talking with some of you. However, some of you are really among the bitterest, arrogant, condescending assholes I've ever encountered online or anywhere else....

What's the meaning of those 2 words ?
some experts' vocabulary for HiFi or High-End audio ?
I cannot find them in my Oxford Dictionary of 1986.
Probably some new development.



My summary About Stereophile:

Maybe the knowledgeable Mr. Atkinson is the editor, because his ability to remember facts correctly, is better than Mr. Fremer's.

See: http://www.hydrogenaudio.org/forums/index....st&p=629517

QUOTE (analogcorner @ Apr 24 2009, 19:50) *
Krueger was at the AES where I got the 5/5 identifications correct. That part is true....that's the only part.


I don't believe so, Michael. It was David Clark aided by Tom Nousaine, if I remember correctly. Arny Krueger isn't an AES member and doesn't attend the conventions.

John Atkinson
Editor, Stereophile

But even the knowledgeable Mr. Atkinson wasn't good enough to describe some simple basic facts about digital audio (Lossless formats and wav) in an adequate way, see above in my post.
Misinformation instead.

I hope, both improve in their ways,
I recommend as start for reading my introduction tutorials into High Quality Audio at http://www.High-Quality.ch.vu , preferably the Lossless guides at http://www.angelfire.com/magic2/hq-audio/tut-lossless.htm
if then anything is still unclear, read the hydrogenaudio wiki or explanations about special expert terms.
or ask me, or the knowledge group here at HA !



Edit:



I also have files prepared containing just the _differences_ between the lower-rez versions and the original, so listeners can hear what has been discarded by the various codecs. You may feel that use of a lossy codec is justifiable but that doesn't mean you shouldn't be aware of what it is doing, surely?

John Atkinson
Editor, Stereophile

Please,
Mr. Atkinson,
avoid this mistake of listening to difference of wav minus lossy.
At least from a little bit scientific informed people , you will get a laugh.
Listening the difference signals is a total  overamplifying of the lossy "mistakes" on purpose.
Carrying out such way, is against the construction of lossy compression, and does not make sense, simply.
If it would be used as demonstration for difference  original vs. wav, it would be like comparing oranges against lemons. 1 sweet, 1 extremely sour.
Every school boy knows, that oranges taste well, but lemons are quite sour.
No test or demo needed.

Just compare your 88.2 kHz 24 bit preprared stuff by "blinded" listening.
Comparing lossy by graphs or by difference signals, these methods are lame and out since , hm, more than 10 years ?




Just for the personal record,
in living room, I listen either to CD or to Lossless sourced CD copies, ie. also the original music, bits for bits, be them 16  bit, 24 bit or HDCD, or DVD-Audio, or SACD.
No lossy for my HiFi/living room.
I have 2 tasks for lossy mp3 encoded by Lame V5 (130 - 150 k vbr):
a) earphones listening during sports, running 1 hour in the woods or streets, USB-stick/SD-card with analogue output
b) car listening, source a)
here mp3 is fine enough,
both tasks have too much background noises, as that Lossless / original CD would offer more sound quality compared to good mp3.

Article: Why We Need Audiophiles

Reply #356 –
Quote from: Stereoeditor on
The downsampled, decimated, lossy compressed versions were all resampled to 24-bit/88.2kHz files, so the only substantive difference will be the coding.

Sorry... what?  88khz 24bit -> 44Khz 16bit -> mp3 -> 88khz 24bits? And you say "the only difference will be the coding"?

Quote from: Stereoeditor on
I also have files prepared containing just the _differences_ between the lower-rez versions and the original, so listeners can hear what has been discarded by the various codecs. You may feel that use of a lossy codec is justifiable but that doesn't mean you shouldn't be aware of what it is doing, surely?


Yes. this is the exact reason why it can't be considered serious. It is the equivalent of looking at graphs.
I will not explain you why, because I have already contributed in this thread once, just to be told wrong on a thing i was right.


@ Moderators, I think it is really time to close this thread. There have been too many wrongdoings, for both usual members and newcomers, but in the end, it has not provided any goodness at all, just a lengthly thread of such a quality that doesn't belong to hydrogenaudio.

Not only that, but I fully think that this thread is attacking at the bases of hydrogenaudio.

Article: Why We Need Audiophiles

Reply #357 –
Quote from: Stereoeditor on
Quote from: ShowsOn on
Quote from: Stereoeditor on
As for Stereophile being against lossy codecs, Colorado-based audiophiles can attend a series of demonstrations I am giving May 5, 6, & 7 comparing hi-rez recordings against Red Book and lossy versions. Details later this weekend on the www.stereophile.com home page.

Will you be judging the lossy files using graphs, or will you be using your ears during controlled tests?

Those aren' tthe only options. Read more closely - I used the word "demonstrations."

Oh OK, you'll be doing uncontrolled tests. Whatever floats your boat.
Quote from: Stereoeditor on
The downsampled, decimated, lossy compressed versions were all resampled to 24-bit/88.2kHz files, so the only substantive difference will be the coding. Allowing the people who attend my seminars to hear for themselves whether they can hear any differences and if so, what the magnitude of those differences might be, is one of the purposes of my public demonstrations.

And the only way you can do this conclusively must be a double blind test - the founder of the magazine you edit says so.
Quote from: Stereoeditor on
I also have files prepared containing just the _differences_ between the lower-rez versions and the original, so listeners can hear what has been discarded by the various codecs.

This statement is just bizarre, and demonstrates you don't understand how lossy encoders work.  They are designed to take into account what sounds will be masked (and thus not audible) when the entire audio stream is played back. Performing tests that involve cancelling out parts of the audio to 'reveal' sonic artefacts isn't a real world test. It would be the same as a magazine publishing spectral graphs of lossy files, which seems to assume that people listen to the files with their eyes.

A lossy encoder BY DEFINITION removes parts of the audio that humans can't hear WHEN ALL OF THE AUDIO is being played back. That is part of how they achieve transparency, they take advantage of the fact human hearing has certain flaws and limitations that can be exploited. Sort of like how optical illusions exploit limitations in the ways our brains and eyes interact to produce visual perception.

But I accept that this would be a difficult thing to appreciate for people who don't think lossy encoders can ever be transparent BECAUSE they throw away parts of the signal (i.e. when an encoder is working properly, the parts humans can't hear). People of this ilk make unsubstantiated blanket statements such as "So to us at Stereophile, the question of which lossy codec is "the best" is moot."
http://www.stereophile.com/features/308mp3cd/index.html

I think the issue is that most audiophiles spend so much time building systems that will supposedly provide the most complete reproduction of a signal, that they can't fathom that human hearing can be tricked after removing a heap of signal that is never heard.
Quote from: Stereoeditor on
You may feel that use of a lossy codec is justifiable but that doesns't mean you shouldn't be aware of what it is doing, surely?

Where is Sterophile's complex understanding of how lossy encoders work? I mean something more sophisticated than the biased statement "The MP3 codec, and others that achieve similar reductions in file size, are "lossy"; ie, of necessity they eliminate some of the musical information. The degree of this degradation depends on the data rate. Less bits always equals less music. " (From the same link as above)

Have you published interviews with people who work on developing lossy codecs, or engineers and mathematicians that developed the principles that underpin how they work? Have you provided a mathematical explanation to your readers of what inverse discrete consine transformation does, or Huffman coding (shock horror, a LOSSLESS process!), or the use of sophisticated polyphase filtering, or what features AAC has that make it a technological progression beyond MP3, and how AAC fixes flaws that are common in (improperly encoded) MP3s?

Beyond these omissions, I think the worst thing is you write articles as if all versions of encoders are essentially the same. You think the FORMAT, e.g. MP3 or AAC or just the category "lossy" is more important than actual encoder. This is a tell-tale sign of people who are inherently biased against the capabilities of modern lossy encoders, because it assumes that all encoders of a format will produce the same quality.

So why don't you publish a report on the quality of as many different MP3 encoders as you can find from early implementations to the latest version of LAME (3.98.2)?  Or do you think that there haven't been any improvements to MP3 and other lossy encoders over the last 18+ years?

Article: Why We Need Audiophiles

Reply #358 –
Quote from: [JAZ] on

Quote from: Stereoeditor on
The downsampled, decimated, lossy compressed versions were all resampled to 24-bit/88.2kHz files, so the only substantive difference will be the coding.

Sorry... what?  88khz 24bit -> 44Khz 16bit -> mp3 -> 88khz 24bits? And you say "the only difference will be the coding"?

What does it matter? He has already concluded that lossy files are all inherently flawed because "Less bits always equals less music." (My emphasis) so why would he even bother applying the same resampling to the lossless source file? I mean that would be kind of J. Gordon Holt principled and scientific like, Stereophile gave up on that in the 1970s.

Article: Why We Need Audiophiles

Reply #359 –
Quote from: [JAZ] on

Not only that, but I fully think that this thread is attacking at the bases of hydrogenaudio.


Why? Fremer has shown in this thread, for everybody to read, that he cannot answer the simple question, why he requires his eyes for audio testing, with anything else than a childish car analogy. We should not underestimate our readers. They can surely differentiate without the mods' help how telling that is.

Whole cars can't be ABXed for obvious reasons, audio can, if you are just willing to. Fremer isn't and his reasons might be that he needs optical self deception to write the kind prose that he is bought for. His customers also enjoy their equipment with eyes and ears. But letting himself get involved into the latter makes him much more of a style guide than a trustable source about sound quality, even if he would not want to hear that.

Article: Why We Need Audiophiles

Reply #360 –
I wonder how many people can hear differences between original hi-rez version and its downsampled (to 44.1/16) version.

Article: Why We Need Audiophiles

Reply #361 –
Quote from: ShowsOn on
This statement is just bizarre, and demonstrates you don't understand how lossy encoders work.  They are designed to take into account what sounds will be masked (and thus not audible) when the entire audio stream is played back.


We should not bash everything they say, that makes HA look one sided. I have listened to difference files myself. It is just interesting to hear what's inside them, that should not be prohibited.

The sample rate conversion to 88.2 khz was also an attempt to increase objectivity and we shouldn't bash that just for the cause of it. An upsampling DAC doesn't do much else. 88.2 instead of 96 khz is also a sensible choice to keep SRC conversion artifacts as low as possible. From this side the test setup looks good and we should not bark at those points just because we might not trust the organizer.

There may be flaws (choice of dither, MP3 codec used, etc..) but we don't know that. When we do, we can bark. When they aren't disclosed we can bark, but just don't let us bark at everything.

Article: Why We Need Audiophiles

Reply #362 –
Quote from: rpp3po on
But letting into the latter makes him much more of a style guide than a trustable source about sound quality, even if he would not want to hear that.

What you're saying is he should be entitled to his own opinion (like everyone else) but not his own facts.

I think that is the issue here, the constant assertion that opinions (e.g. all MP3s sound bad, and can never be transparent) as if it is a fact is what is killing the audiophile industry, because if all opinions are facts, then why should we care what anyone writes about audio?

There is also the issue that new technologies have made better quality audio available at cheaper prices (e.g. the CD, and the transistor), but instead sections of audiophilia pretend that this inventions have only reduced sound quality. Or to have this point put another way:
Quote
John Atkinson (Stereophile Editor): As you were so committed to surround, do you feel that the commercial failures of DVD-Audio and SACD could have been avoided?

J. Gordon Holt (Stereophile Founder):  I doubt it. No audio product has ever succeeded because it was better, only because it was cheaper, smaller, or easier to use. Your generation of music lovers will probably be the last that even think about fidelity.

http://www.stereophile.com/asweseeit/1107awsi/

The blame for the decline of sound quality should be directed at some idiotic record producers, bands, and sell out mastering 'engineers' who have spent the last decade completely under utilising cutting edge analog, and more importantly, digital technologies. It has nothing to do with technologies themselves, which are as good as, if not far better than what was affordable for the average person 20 or 30 years ago.

Article: Why We Need Audiophiles

Reply #363 –
Quote from: rpp3po on
We should not bash everything they say, that makes HA look one sided. I have listened to difference files myself. It is just interesting to hear what's inside them, that should not be prohibited.

It may be interesting, but what can you definitively conclude from it? I propose nothing.
Quote from: rpp3po on
The sample rate conversion to 88.2 khz was also an attempt to increase objectivity and we shouldn't bash that just for the cause of it. An upsampling DAC doesn't do much else. 88.2 instead of 96 khz is also a sensible choice to keep SRC conversion artifacts as low as possible. From this side the test setup looks good and we should not bark at those points just because we might not trust the organizer.

Sure, but the point was if the MP3 is resampled, then the source file should be put through the same resampling routines too, else the experiment would be uncontrolled. (Are any differences attributable to the lossy encoding, or to the resampling, or to both?)

Article: Why We Need Audiophiles

Reply #364 –
Quote from: analogcorner on
Quote from: cpchan on
Quote from: Arnold B. Krueger on
Unlike most people who post here, I've had the *privilege* of meeting Michael Fremer in the flesh. Well sort of. This was in 2005.  After a few seconds in my presence, he started loudly screaming profanities about an ABX demo that some of my friends did at an AES meeting back in the very early 1990s.  His friends had to forcably restrain him and drag him out of the room.


  I am totally speechless. I don't know whether to laugh or cry at such ignorance.


Cry: you're believing a total liar. Krueger lied. No such encounter ever occurred.


Typical of the usual posturing we see from some people.

If I was a "total liar" then every statement I ever made would be false, and known to be false by me when I posted it.

I'll stand on just my posting record on HA - how many true, verifiable statements have I made here?

At least one, right? ;-)

If I've ever made any true statements at all, then the claim that I am a "total liar" has been falsified.

I suspect that even the person who claimed that I am a "total liar" knows that *everything* I say is not false.

Therefore he made a knowingly made a false statement. Technically he was lying.

Of course he didn't mean to lie - it was all just his usual hyperbole.  ;-)

The real problem with people like him is knowing when its hyperbole, and when it is something that you can "take to the bank".  :-(

If I say it without obvioius intent at humor, I would hope that you can take it to the bank.

Article: Why We Need Audiophiles

Reply #365 –
Quote from: ShowsOn on
Sure, but the point was if the MP3 is resampled, then the source file should be put through the same resampling routines too, else the experiment would be uncontrolled. (Are any differences attributable to the lossy encoding, or to the resampling, or to both?)


They also upsample the Redbook PCM to 88.2 khz just like the MP3 as I read it, so that's fine. You cannot demand that high rez audio is resampled to Redbook and back first (it wouldn't be high rez anymore) if you want to compare MP3 to high rez.

I don't believe that a difference will be perceivable anyway for their choice of music, if the setup was 100% proper (choice of encoder, no truncation) and blind. Of course, having an admired Stereophile 'star' in your city who asks the audience "did you hear the difference, did you hear it? much more lively, wasn't it", a lot of people would hear a difference - no matter how proper the comparison was setup otherwise.

Article: Why We Need Audiophiles

Reply #366 –
Quote from: analogcorner on
I am not a psychiatrist. But people who spend their time counting other people's money can probably use one.


You heard it here. According to this post, *all* people who "who spend their time counting other people's money can probably use (a psychiatrist)"

So who counts other people's money?

Bank tellers
Other cashiers
Accountants
Criminal investigators
People who help blind people manage their personal affairs.
etc. etc.

So does the poster actually believe what he said?

Obviously not.

Obviously, he writes in the *Hyperbole* dialect of the English language! ;-)

Article: Why We Need Audiophiles

Reply #367 –
Quote from: rpp3po on
They also upsample the Redbook PCM to 88.2 khz just like the MP3 as I read it, so that's fine. You cannot demand that high rez audio is resampled to Redbook and back first (it wouldn't be high rez anymore) if you want to compare MP3 to high rez.

So any test won't be able to explain if any differences are a result of the resolution or the lossy compression.

Article: Why We Need Audiophiles

Reply #368 –
Yes, else your supposed "high rez" would just be Redbook padded with zeroes.

PS I would have expected some heavier artillery than sophistry from Arnold. He must be tired today.
 

Article: Why We Need Audiophiles

Reply #369 –
Quote from: analogcorner on
There was an AES meeting back in the early '90s and an ABX test of amplifiers. I was involved because I claimed that the idea that all amplifiers sound alike, or pretty much alike (that measure the same) is kind of foolish, since we really don't measure every aspect of performance and I told David Clark to produce an ABX test and I'd be happy to take it.


So far so good.

Let's review for a second exactly what a relevant ABX test is or would be in the context presented above.

(1) An ABX test is an open-ended event where people can compare two different things (A & B) by means of a third unknown thing (X) which is one of the two,  as many times as they want to. They are forced to summarize their judgements as one decision as to the identity of X  which is called a "trial". The comparison process is repeated as required  to obtain a sufficient number of trials.

(2) An ABX test is composed of as many trials as is established before the first trial. The number of trials has to be sufficient that the usual statistical tests do not run afoul of the problems related to a small number of samples. The traditional number of trials is on the order of 16. 5 or 6 trials is always totally inadedequate for a serious test. Nothing of importance or reliability can logically be determined from such a small number of trials.

(3) Obviously, it would take far more than one comparision of just  two amplifiers to provide compelling evidence related to such a large issue as "the idea that all amplifiers sound alike, or pretty much alike (that measure the same) is kind of foolish, since we really don't measure every aspect of performance".

(4) Any such comparison needs to be done under conditions that are far more favorable to the performance of the listener.  5-6 trials done with  stone cold listeners who had no opportuinty for listener familiarization and training, in a conference room at an AES convention is just a demonstration, and not a proper test.


Quote
I took the ABX test as devised and produced by the group and I got five of five identifications correct. My editor, John Atkinson got 4 of 5 correct.


Based on what I know about ABX testing (and I'm merely the inventor of ABX ;-) )  There never was any such ABX test.  Whatever happened, based on the best evidence available to me, was a demonstration of what an ABX test might be like.

End of story.

This is BTW, the information that I tried to convey to Mr. Fremer at HE2005, and it is my best recollection of what I said or partially said that provoked his regrettable outburst. I do not know if he heard me say any or all of the above, due to his outburst.

I presume that his denials of my account of his actions are in their way not lies, because he was obviously way too distressed at the time to have an accurate recollection of what really happened.

My recollection is that I was pretty much alone in the room when this happened, other than Fremer and anybody who might have walked in with him. I think it was well before the debate. I'm pretty sure that the person who posts here as krabapple was not present.

Ordinarily, an important  issue such as  "The idea that all amplifiers sound alike, or pretty much alike (that measure the same) is kind of foolish, since we really don't measure every aspect of performance" would involve more than two amplifiers, more than two listeners, and more than 5 or 6 trials.

Considerable evidence related to the above question are presented quite clearly in the Stereo Review article that has been linked to out of this thread. If others wish to assert that they are actually just as sincere and energetic in their intentions to help resolve this quesiton, one would think that sometime in the past 15 or more years since the SR article, they would have done an equal amount of similar work of their own and published the same. In fact they have spent their time posturing against any kind of reliable subjective testing efforts. They have even questioned the applicability of science as we know it.

Note that there are more ways to scientifically study this question than just ABX. So, I'm not promoting ABX at all.

I would be happy to work as I have the resources to serve, to help anybody who was seriously interested in the above issue, gather such evidence as they would wish to gather in a scientific way. I believe that my behind the scenes support for many persons who have made publications and presenations related to this question, the 8 years of operation of the now-departed www.pcabx.com web site, and my posts on HA are consistent with and demonstrations of the reality of my desires along those lines.

Article: Why We Need Audiophiles

Reply #370 –
Quote from: user on
See footnote 1 at  :  http://www.stereophile.com/features/308mp3cd/index.html

Footnote 1: Something I have rarely seen discussed is the fact is that because all compressed file formats, both lossless and lossy, effectively have zero data redundancy, they are much more vulnerable than uncompressed files to bit errors in transmission.


This is the most funny statement I have read recently from audiofools.
Who is the author of that story ?

John Atkinson !

He reveals great technical knowledge of the basics of maths, PC, audio

Unfortunately he is fooling his readers

What makes me sad, is, he recommends wav uncompressed over compression without losses (so called "Lossless") like FLAC etc.

Not only in this footnote, also above in the text:


So to us at Stereophile, the question of which lossy codec is "the best" is moot. We recommend that, for serious listening, our readers use uncompressed audio file formats, such as WAV or AIF—or, if file size is an issue because of limited hard-drive space, use a lossless format such as FLAC or ALC. These will be audibly transparent to all listeners at all times with all kinds of music through all systems.


This is really sad to read.
it really shows null, zero understanding of technics, underlying simple maths or even usability of wav vs. FLAC.


I would be prone to call the rather grotesque misapprehension embodied above, a simple ignorant mistake.

Of course you are absolutely right that  compressed (lossy or losless) are no more susceptible to transmission errors than uncompressed files.

Let me count the (most obvioius) ways that Atkinson is oh, so wrong.

(1) The compressed file is shorter. Error processes being mostly random, will therefore add fewer errors to less data.

(2) The compressed file is more likely to be noticably degraded by any errors that take place.

(3) The process of expanding the file may do additional validity checking and again be more revealing of any errors that take place.

Of course it is too bad that there is insufficient technical fact checking at Stereophile to keep embarassing stuff like this from getting published.

As many of us know, if there was adequate technical fact checking at Stereophile, he magazine's content would have to undergo a rather significant and potentially financially disasterous change.

Article: Why We Need Audiophiles

Reply #371 –
Quote from: rpp3po on
PS I would have expected some heavier artillery than sophistry from Arnold. He must be tired today.


What would constittue "heavy artillery"?

I may have some, I'm just warming up! ;-)

Atrkinson and Fremer are zillions of words ahead of me on this topic, so far...

Article: Why We Need Audiophiles

Reply #372 –
Quote from: Frumious B on
Quote from: analogcorner on
Yet no matter what I say, I am stereotyped as someone who "hates digital," or who "doesn't listen to digital," or whatever, by people who are happier to attack me for something I am not.


Unless you are claiming that you actually never had a "Compact Discs Suck" bumper sticker and you are recanting articles like this one: http://www.musicangle.com/feat.php?id=106 then I would argue that you have stereotyped yourself.



That bumper sticker was on my car in the early 1980s when compact discs did suck. They sounded awful. The transfers from analog were uniformly poorly done from questionable sources with overuse of CEDAR and Sonic Solutions. So called DDD discs actually went through multiple D/A A/D conversions since there were no digital mixing boards. Many factors contributed to what was awful sound. Those who called that sound "pristine" and a "big advancement" were wrong. History has proven me correct as virtually the entire catalog of what had been issued back then has been reissued using better sources, better associated gear, especially better converters and fewer attempts at lopping off the top end along with supposedly offensive tape hiss.

Article: Why We Need Audiophiles

Reply #373 –
Quote from: Arnold B. Krueger on
What would constittue "heavy artillery"?


I don't know. I just thought that you might be really pissed after all what they have said about you last night. But it really may be the right strategy to keep the ball low. Just politely tacking them down to central inconsistencies can unveil much more than endless battles about details and personal attacks. It's really an enrichment for HA that they come here to defend themselves instead of us having to argue with zealots as B0RK. I really thought Fremer is a total nut until he showed up here personally!

They don't sell a perspective that many around here would buy, myself included. It's not scientific and it's not suitable to make objectively verifiable statements about the sound (and only sound) of audio gear. But as I understand it now, their world has much more inner consistency than I had thought before. These are sane people. Their subjective approach is so extreme, that I really don't think that they would call anything they do objective even in the slightest sense (they might not admit that in public). Their world is a huge circus to love and celebrate music. It doesn't matter a shit if unit A is really objectively better than unit B, it's a preposterous question in their world. They fill audio gear, even from our point of view idiotic things as cables, with life and story and that alone can make a huge actually perceivable difference for their followers.

Article: Why We Need Audiophiles

Reply #374 –
Quote from: Arnold B. Krueger on
Quote from: analogcorner on
There was an AES meeting back in the early '90s and an ABX test of amplifiers. I was involved because I claimed that the idea that all amplifiers sound alike, or pretty much alike (that measure the same) is kind of foolish, since we really don't measure every aspect of performance and I told David Clark to produce an ABX test and I'd be happy to take it.


So far so good.

Let's review for a second exactly what a relevant ABX test is or would be in the context presented above.

(1) An ABX test is an open-ended event where people can compare two different things (A & B) by means of a third unknown thing (X) which is one of the two,  as many times as they want to. They are forced to summarize their judgements as one decision as to the identity of X  which is called a "trial". The comparison process is repeated as required  to obtain a sufficient number of trials.

(2) An ABX test is composed of as many trials as is established before the first trial. The number of trials has to be sufficient that the usual statistical tests do not run afoul of the problems related to a small number of samples. The traditional number of trials is on the order of 16. 5 or 6 trials is always totally inadedequate for a serious test. Nothing of importance or reliability can logically be determined from such a small number of trials.

(3) Obviously, it would take far more than one comparision of just  two amplifiers to provide compelling evidence related to such a large issue as "the idea that all amplifiers sound alike, or pretty much alike (that measure the same) is kind of foolish, since we really don't measure every aspect of performance".

(4) Any such comparison needs to be done under conditions that are far more favorable to the performance of the listener.  5-6 trials done with  stone cold listeners who had no opportuinty for listener familiarization and training, in a conference room at an AES convention is just a demonstration, and not a proper test.


Quote
I took the ABX test as devised and produced by the group and I got five of five identifications correct. My editor, John Atkinson got 4 of 5 correct.


Based on what I know about ABX testing (and I'm merely the inventor of ABX ;-) )  There never was any such ABX test.  Whatever happened, based on the best evidence available to me, was a demonstration of what an ABX test might be like.

End of story.

This is BTW, the information that I tried to convey to Mr. Fremer at HE2005, and it is my best recollection of what I said or partially said that provoked his regrettable outburst. I do not know if he heard me say any or all of the above, due to his outburst.

I presume that his denials of my account of his actions are in their way not lies, because he was obviously way too distressed at the time to have an accurate recollection of what really happened.

My recollection is that I was pretty much alone in the room when this happened, other than Fremer and anybody who might have walked in with him. I think it was well before the debate. I'm pretty sure that the person who posts here as krabapple was not present.

Ordinarily, an important  issue such as  "The idea that all amplifiers sound alike, or pretty much alike (that measure the same) is kind of foolish, since we really don't measure every aspect of performance" would involve more than two amplifiers, more than two listeners, and more than 5 or 6 trials.

Considerable evidence related to the above question are presented quite clearly in the Stereo Review article that has been linked to out of this thread. If others wish to assert that they are actually just as sincere and energetic in their intentions to help resolve this quesiton, one would think that sometime in the past 15 or more years since the SR article, they would have done an equal amount of similar work of their own and published the same. In fact they have spent their time posturing against any kind of reliable subjective testing efforts. They have even questioned the applicability of science as we know it.

Note that there are more ways to scientifically study this question than just ABX. So, I'm not promoting ABX at all.

I would be happy to work as I have the resources to serve, to help anybody who was seriously interested in the above issue, gather such evidence as they would wish to gather in a scientific way. I believe that my behind the scenes support for many persons who have made publications and presenations related to this question, the 8 years of operation of the now-departed www.pcabx.com web site, and my posts on HA are consistent with and demonstrations of the reality of my desires along those lines.


Oy vay. First of all, this useless blather above is absurd. It's not on subject. It's a smokescreen. As I recall it, I ran into David Clark at a CES. I told him I disagreed with his contention regarding amplifiers sounding alike. He challenged me to a blind test. Whatever the hell you want to call the test I didn't set it up. Others did. As agreed to the challenge, I took the test, along with dozens of other AES attendees. The amplifiers used sounded very different from one another, which is why I got all five identifications correct. I was challenged. I took the test set up by others. I passed it 100%. That's what I know. For my efforts I was declared a "lucky coin" and dismissed. Now that pissed me off. I had been "set up" because had I gotten it all wrong, it would have been used as "proof" against me. I failed. But having succeeded the only thing to do was to A)Dismiss me as a "lucky coin," and B) Declare the test not really valid..never mind that I didn't set it up, though it was, of course a 'set-up.' Of course I was pissed. I'd been conned and set up. Had I failed, we all know what the result would have been. But having passed the test, the results were dismissed! What a con job!

One thing that can be said about that test, whatever you want to call it, is that 5 very different sounding amplifiers could not be reliably identified by a statistically significant percentage of the participants. That tells me, that that particular test, and I'd say this kind of test generally (though that's just my supposition) produced statistically insignificant results, not because "it all sounds the same," but because the test is not necessarily a valid way to to evaluate the sonic performance of audio gear.

At the very least, there was something wrong with that test, since the test takers--mostly recording engineers-- couldn't identify what clearly were 5 very different sounding amplifiers. My conclusion is that these kinds of tests produce confusion and performance anxiety that masks obvious differences for inexperienced test takers.

BTW: one more point: aside from getting that test 100% correct, and doing very well in a series of Floyd Toole designed speaker identification blind tests at Harman's factory, there was an interesting cable blind test run by The Wall Street Journal's Portals columnist Lee Gomes that you can read here: http://online.wsj.com/article/SB1200446920...main_promo_left

Gomes took a room at "The Show" (a parasitic audio show that runs concurrent to CES) where he haphazardly set up two identical, modest systems with only the speakers visible. He put a sign on the window asking for volunteers to take an admittedly non-scientific blind test. I took it along with others. Gomes didn't say what the test was about but he intimated that it was a comparison of either an iPod versus a CD player, or redbook versus MP3.. (in fact he run a number of different tests).in any case, it was obvious to me that the two systems sounded very different and one sounded better than the other, though both sounded pretty poor compared to what I was used to.  When it was over I told him I was puzzled because it didn't sound like an iPod versus CD player test or a redbook versus MP3 test......he then took me around back to show me that in fact, it was a cable test---a comparison of, hardware store cable (14 gauge lamp cord I think) versus Monster cable. I preferred the Monster Cable, which sounded much better....

So now, let the angry, bitter, petty, infantile name calling and sniping begin! The more it's done here, the more entertained I am.