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: "Sizzling sound" of mp3's is preferred (Read 41497 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

"Sizzling sound" of mp3's is preferred

Reply #101 –
And 200 points for Kees de Visser for missing Member No. 22222 by 2! 

"Sizzling sound" of mp3's is preferred

Reply #102 –
...what do points make?


Cheers,
David.

"Sizzling sound" of mp3's is preferred

Reply #103 –
Sparks if the condenser has failed?

Cheers, Slipstreem. 

"Sizzling sound" of mp3's is preferred

Reply #104 –
Prizes!
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

"Sizzling sound" of mp3's is preferred

Reply #105 –
They bloat the thread.
 

"Sizzling sound" of mp3's is preferred

Reply #106 –
Quote from: Nick.C on
Prizes!
It was a little unfair of me - I doubt  Bruce Forsyth (or Humphrey Lyttelton) are that well known outside of the UK!


Back on topic: there's no guarantee that there's duality in this conversion - CD>SACD and SACD>CD will only meet the "lossless" requirements with appropriate noise shaping in both directions. It's the noise shaping that makes or breaks this (both ways).

Cheers,
David.

"Sizzling sound" of mp3's is preferred

Reply #107 –
Quote from: 2Bdecided on
Back on topic: there's no guarantee that there's duality in this conversion - CD>SACD and SACD>CD will only meet the "lossless" requirements with appropriate noise shaping in both directions. It's the noise shaping that makes or breaks this (both ways).


So what does lossless mean?

* Bit perfect?

Kinda tough when bits mean different things!

* No audible differences?

Makes some sense, but I predict objections. ;-)

* Identical bandwidth and dynamic range, or very nearly so?

What does very nearly so mean? ;-)

"Sizzling sound" of mp3's is preferred

Reply #108 –
I don't know what lossless means in this context.

However, if the noise floor of medium B is clearly higher than that of medium A, then converting from medium A to medium B cannot be lossless for all possible signals on medium A.


We could have an interesting discussion on this lossless / not lossless point, but I don't think there's any point. The proposed transform wouldn't satisfy DSD proponents that you were testing DSD any more.

Cheers,
David.

"Sizzling sound" of mp3's is preferred

Reply #109 –
... I had a very witty comment here on the DSD lossless thing, but in light of all the drama in this thread, I think it's probably poor form to openly flout TOS5.

"Sizzling sound" of mp3's is preferred

Reply #110 –

***Please close this thread in violation of TOS42***

"Sizzling sound" of mp3's is preferred

Reply #111 –
42?!

We haven't got anywhere near the meaning of life yet.


If lossless = reversible, we have a problem - since each single conversion isn't even repeatable unless you re-set the noise shapers to the exact same state each time.

Cheers,
David.

"Sizzling sound" of mp3's is preferred

Reply #112 –
Noise shaping is not imperative. If you don't do it, you just waste a bit or two worth of (theoretically achievable) SNR. At 16 bits that's too much, but at 24 bit, for example, no big deal.

"Sizzling sound" of mp3's is preferred

Reply #113 –
True - but DSD is entirely noise shaping.

So DSD > anything > DSD doesn't give you the original signal back unless you work very hard to do so. "anything" really could be almost anything - even a hypothetical analogue system with infinite bandwidth and no noise - and it would still be irreversible unless extreme care was taken with the second DSD stage - and even then, I don't believe you could get the same numbers back.

You could cheat, simply adding extra zero samples or bits to the DSD, and then removing them again - that shows it's possible to convert DSD to something else, and back to DSD again, in a lossless and reversible manner - but it's not very useful.


This is all OTT - it's easy enough to do the ABX properly with real DSD, real PCM, and a hardware switch box.

Cheers,
David.

"Sizzling sound" of mp3's is preferred

Reply #114 –
Quote from: 2Bdecided on
I don't know what lossless means in this context.


Then we shouldn't use words whose meaning we don't know?


Quote
However, if the noise floor of medium B is clearly higher than that of medium A, then converting from medium A to medium B cannot be lossless for all possible signals on medium A.


Cledasrly higher, in dB would be?

Quote
We could have an interesting discussion on this lossless / not lossless point, but I don't think there's any point. The proposed transform wouldn't satisfy DSD proponents that you were testing DSD any more.


Agreed that discussion with most DSD proponents I've had the displeasure of conversing with turns futile when rational discussion is attempted.

"Sizzling sound" of mp3's is preferred

Reply #115 –
Quote from: Arnold B. Krueger on
Quote from: 2Bdecided on
I don't know what lossless means in this context.
Then we shouldn't use words whose meaning we don't know?
It was useful to use it. It made an unknown unknown into a known unknown  I lean towards the definition that lossless = reversible. Not every one is so strict. I guess in that sense it's a useless word - except for situations where it's reasonable to believe that something is reversible, but it's very difficult to do so in practice. In such a case, practically, it's not reversible, but maybe it is still lossless?

Quote
Quote
However, if the noise floor of medium B is clearly higher than that of medium A, then converting from medium A to medium B cannot be lossless for all possible signals on medium A.
Cledasrly higher, in dB would be?
I don't know - let's discuss it in English when you've got a new keyboard

Seriously, are you asking for maths or concepts? Because there are lots of transforms, even on PCM data, that appear to conserve all the information yet are near impossible to reverse (unless specifically designed to be reversible). DSD is far worse. So I'm uncomfortable with saying that something with X or Y parameter is a lossless representation of something else. However, I am comfortable with saying that something isn't lossless.

So, clearly higher, in dB, is probably any positive number.

In many scenarios, even an equal noise floor is not lossless, since the transform adds more noise at the same level, and the noise powers add making the situation lossy. Whereas 16-bit > 16-bit is a (null) transform where the noise floor is equal yet there is no loss.

I'm waiting for SebG to wade in and dazzle us all with the maths behind this vague discussion.

Cheers,
David.

"Sizzling sound" of mp3's is preferred

Reply #116 –
Quote from: 2Bdecided on
Quote from: Nick.C on
Prizes!
It was a little unfair of me - I doubt  Bruce Forsyth (or Humphrey Lyttelton) are that well known outside of the UK!


Back on topic: there's no guarantee that there's duality in this conversion - CD>SACD and SACD>CD will only meet the "lossless" requirements with appropriate noise shaping in both directions. It's the noise shaping that makes or breaks this (both ways).

Cheers,
David.



Did Meyer &  Moran's  DSD-->Redbook setup use noise shaping?  Their conversion was only audible at elevated listening levels (during quiet parts).

"Sizzling sound" of mp3's is preferred

Reply #117 –
At one point in time I listened to a borrowed casette. It was badly played out but I liked the sound alot. Then I later borrowed a CD from which it was made and came up really disappointed. The sound was much rougher (although it was way cleaner), especially the treble was ear-splitting compared to the casette (where the sound was "mild" and all the roughness was washed out). It, in some sense, reversed the effects of aggressive-sounding recording/mastering (it was metal genre).
I'm not defending lossy sources or lossy playback chains, just pointing out what might have been the reason for the better perceived sound.

About that 15y old girl. Until my 17, I listened mostly to electronic stuff like breakbeat, rave, techno etc. Then I was literally broken by Sepultura's Roots album and two years later I ended up listening to black and death metal... The conclusion is that you can't really foretell anything about the children's music preference the same way you can't predict much about their grown-up personality.
I'm approaching 30 and perhaps I will switch to brass music over time... (I have to be kidding actually!  )

"Sizzling sound" of mp3's is preferred

Reply #118 –
Jumping on late I bumped into this story of the "Sizzling sound" on a German public radio station (source).

I thought immediately that this is a bogus story. Especially since they were referencing Green Day as an example. Without running into TOS8 I think this would be a less well chosen style for for the proof of audio compression artifacts. There is a full mp3 interview at CBC Radio with Mr. Berger available here (from which I believe the excerpts for the German radio story where taken from). He said that the last tests where conducted about a year ago, so he could possibly not know of the latest listening test results at HA

I agree that this article in my point of view shows eventually how badly journalism can be conducted at times. Of course Mr. Berger can state anything he wants but it should be the job of the journalists to get at least a competent second opinion.
At the end of the interview he said that he is mostly listening to aac, because he listens to it on an ipod and most of it is actually lossless. That'll be the "Sizzling sound" of the next generation then

I still remember those glichy artifacts that my first Philips mp3-cd player used to make, I never got used to it
-Chris

"Sizzling sound" of mp3's is preferred

Reply #119 –
Quote from: j8ee on
From the gizmodo post, this appears to be a quote from Berger:
Quote
Students were asked to judge the quality of a variety of compression methods randomly mixed with uncompressed 44.1 KHz audio. The music examples included both orchestral, jazz and rock music. When I first did this I was expecting to hear preferences for uncompressed audio and expecting to see MP3 (at 128, 160 and 192 bit rates) well below other methods (including a proprietary wavelet-based approach and AAC). To my surprise, in the rock examples the MP3 at 128 was preferred. I repeated the experiment over 6 years and found the preference for MP3 - particularly in music with high energy (cymbal crashes, brass hits, etc) rising over time.



Does anyone know more details about this study: the music samples, test methodology, MP3 encoder used? I'd like to do a similar test using some high school students.

Cheers
Sean
Audio Musings

"Sizzling sound" of mp3's is preferred

Reply #120 –
Quote from: solive on
Does anyone know more details about this study: the music samples, test methodology, MP3 encoder used? I'd like to do a similar test using some high school students.


I believe that that is one of the issues with the 'study' itself -- no details, no specifics.

"Sizzling sound" of mp3's is preferred

Reply #121 –
You may have to email him or something. I don't know if you read the Gizmodo article in which your quote originated, and is introduced as follows:
Quote
Each year, Stanford Professor of Music Jonathan Berger does an informal test of his students by playing a bunch of different music in a bunch of different formats. Over email, here's how he told me performs the informal study:

"Sizzling sound" of mp3's is preferred

Reply #122 –
Quote from: dv1989 on
You may have to email him or something. I don't know if you read the Gizmodo article in which your quote originated, and is introduced as follows:
Quote
Each year, Stanford Professor of Music Jonathan Berger does an informal test of his students by playing a bunch of different music in a bunch of different formats. Over email, here's how he told me performs the informal study:



I did email him after the Gizmodo article appeared months ago, as did Floyd Toole and many other people I know. The result:  no response.  I also found no publications describing the study. 

Cheers
Sean

"Sizzling sound" of mp3's is preferred

Reply #123 –
Quote from: j8ee on
Some of you may have read it already on Slashdot or elsewhere, but I thought it would be good to have a post here about an interesting article that suggests that more and more people seem to prefer music with encoding artifacts over the original audio or better encodes. Mp3's with noises from bad encodes are apparently so common and widespread that it's much of a norm for how music should sound.

"The Sizzling Sound of Music" by Dale Dougherty - http://radar.oreilly.com/2009/03/the-sizzl...d-of-music.html The article builds upon work and tests by a professor of music at Stanford, Jonathan Berger.

I don't know if this has been posted already, I presume it should have been because it's blatantly obvious, this seems like precisely the reason why some people seem to prefer the sound of LPs over that of digitised Red Book CD music - they're familiar with the artefacts of music on LPs and better sound seems strange to them.

"Sizzling sound" of mp3's is preferred

Reply #124 –
Quote from: doccolinni on
Quote from: j8ee on
Some of you may have read it already on Slashdot or elsewhere, but I thought it would be good to have a post here about an interesting article that suggests that more and more people seem to prefer music with encoding artifacts over the original audio or better encodes. Mp3's with noises from bad encodes are apparently so common and widespread that it's much of a norm for how music should sound.

"The Sizzling Sound of Music" by Dale Dougherty - http://radar.oreilly.com/2009/03/the-sizzl...d-of-music.html The article builds upon work and tests by a professor of music at Stanford, Jonathan Berger.

I don't know if this has been posted already, I presume it should have been because it's blatantly obvious, this seems like precisely the reason why some people seem to prefer the sound of LPs over that of digitised Red Book CD music - they're familiar with the artefacts of music on LPs and better sound seems strange to them.


This analysis is exactly consisten with scientific knowlege about how music activates the brain's pleasure centers. A key to the connection witht he pleasure centers is memories about how the music sounded in the past. The pleasure comes not from accurate reproduction or even just sounding good, but rather from something that sounds the same as it did in the past.