Idiocy alert re: mp3

[Rspaight]

I've heard of these...have you had a chance to see one in action?

Yeah, I saw the Denon 1910 feeding a Samsung 56" DLP. The DVI output was "smoother" than the 480p -- the dot structure was less apparent. Obviously, it can't create detail that isn't there, but it did a good job of making the DVD look good on that huge screen.

I've heard others say that the results are much less noticeable on smaller screens.

(I've got an HD set, but it doesn't have a DVI or HDMI input, so I'm out of the loop on these devices.)

(comparing DVD/component to laserdisc/S-video, which may not be a fair comparison)

IMHO, DVD over S-video stomps LD over S-video, so I'd say a good chunk of that is DVD. Even so, I find color rendition to be noticeably better on component.

My TV has the digital inputs as well, but I'm thinking on holding off on a DVD with digital outs in favor of HD-DVD. But that's probably a long ways off.

Could be as soon as the end of this year. But there's currently two formats out there (HD-DVD and Blu-Ray), so I definitely wouldn't spend any money until that's sorted out (either by one format becoming dominant or by universal players being available).

Ryan

[krabapple]

...people who know nothing of perceptual masking and its crucial role in codec development.

Since I have no idea what that means, I guess it's safe to say I know nothing about it!

Try this:
http://www.soundonsound.com/sos/may00/articles/mp3.htm

Quote:
Moving the sample rate higher during *recording* can have benefits for filter design, and moving to higher bit depths eases digital postprocessing.

In layman's terms, does this mean that SACD can potentially sound better than redbook?

Only if the redbook system has poorly designed antialiasing filters and the SACD system has well-designed ones. There's no *inherent* reason one would be better than another . For elaboration about filters and sampling rates see this post from Dan LAvry's forum on ProSound Web.

No one has ever demonstrated that properly designed CD recording and playback is audibly different from properly designed SACD recording and playback.

[krabapple]

re: upsampling: allow me to quote not one but *two* estimable and highly reliable engineering personages from rec.audio.high-end (jj and Pierce)

the original thread starts here and is worht reading all the way though.

Richard D Pierce May 30 2001, 1:25 pm
In article <9f3a3n02...@enews2.newsguy.com>,
jj, curmudgeon and tiring philalethist <j...@research.att.com> wrote:

>In article <9f1va7$dt...@bourbaki.localdomain>,

>Peter <p_ull...@ix.netcom.com> wrote:

>>Since a number of people within this newsgroup, who I consider technically quite
>>knowledgeable, have written that DACs with "upsampling" theoretically cannot
>>improve the sound of CDs, I am totally perplexed by the above quoted statement.

>Pure "upsampling" should not change the sound of a CD going out a good
>DAC.

To repeat what JJ said in slightly different words with more
details: properly done, distortion free upsampling CANNOT and
DOES NOT add any more information to the stream. It does not
make it any smoother, it does not increase its real resolution:
it cannot becuase to increase the resolution demands that new
information come from somewhere.

Thus, properly done, distortion free upsampling won't sound any
different, assuming the original and upsampled implementations
are both done properly.

>If one does some euphonic stuff in the process, or the DAC performance
>increases due to whatever, then of course one may be able to hear
>a difference.

In other words, IF the upsampling process DOES sound different,
it must be adding something to the sound that simply was not
there to begin with: this constitutes distortionb. That
distortion may be pleasant and it may even be desirable in some
cases, but, nonetheless it is distortion.

[Ess Ay Cee Dee]

In other words, IF the upsampling process DOES sound different,
it must be adding something to the sound that simply was not
there to begin with: this constitutes distortionb. That
distortion may be pleasant and it may even be desirable in some
cases, but, nonetheless it is distortion.

Hmmmm, sounds suspiciously like Grant's precious "ambience retrieval".

[lukpac]

Hmmmm, sounds suspiciously like Grant's precious "ambience retrieval".

Do we know that this "ambience retrieval" is simply added "euphonic distortion"? All I've seen here so far is speculation along those lines.

[krabapple]

one might try this: record a signal of known character -- say, test tones -- apply 'ambience retrievial' methods to it, and see if it changes the signal.

[lukpac]

Well, that's the thing - is it trying to recreate something that was already "lost". Or is it trying to preserve something so it doesn't get "lost"?

[Crummy Old Label Avatar]

Have any of you ever heard Pioneer's "Legato Link" circuit? I don't know if they still use this in their players today, but it was a much-touted feature on their Elite series players in the early 90s. I have one of these (I use it now as a transport only). Anyway, to my ears, the upsamplers I've heard produce a sound suspiciously similar to Legato Link. I guess the best way to describe Legato Link is "on the surface, a bit warmer, a bit fuller". Also a bit fuzzier, "softer", less defined. That was the advertising hook with Legato Link at the time: "Analog warmth!"

In other words, it's simply a filter similar to a loudness booster, along with a bit of spatial expansion. I think that's what these "upsamplers" are also doing. This sonic effect is coming not from the actual upsampling process but rather the filter they've added to the unit. Now whether you find the sound pleasant or not is an individual choice, but accurate it most definitely is not. I believe Krabapple is correct. If they didn't add the filter to the unit, if it only offered "pure upsampling", you'd hear no difference whatsoever. People like their sound coloration, don't they?

I'm waiting for someone to start selling a component that claims to "restore" the missing info from mp3's for "audiophile sound quality". Hey! I ought to keep that idea to myself! Imagine how rich I could get if I constructed a spatial expansion filter into a nice looking chassis, stuck a couple of tubes in it, and, voila! .......... "At last! Compressed Audio Now Uncompressed for the Audiophile! At $15,000 it's a steal!"

[krabapple]

Well, that's the thing - is it trying to recreate something that was already "lost". Or is it trying to preserve something so it doesn't get "lost"?

One way to test that is generate the test signal entirely digitally -- there's no 'recording' involved.
Pure upsamping shouldn't have any effect. High-end 'upsampling'....?

[lukpac]

One way to test that is generate the test signal entirely digitally -- there's no 'recording' involved.
Pure upsamping shouldn't have any effect. High-end 'upsampling'....?

Unless I'm missing something, how do we know this "ambience retrieval" is upsampling?

[Crummy Old Label Avatar]

Just what the hell is this "ambience retrieval" supposed to be? What claims are the adherents making about it? From what little I've seen, it appears that they're talking about it strictly as a benefit of "hi-rez". More bits and all that. Cue mumbo jumbo 'bout frequencies and harmonics no one can hear that add "presence". Jump to "Nyquist is wrong, see." Blahblahblah. When I pressed the issue last year, Hoffman huffily snapped something about "well, you need to have this $8000 DAC to properly hear all the extra 'air' in hi-rez formats." Pshaw!

Is "ambience retrieval" the new fancy-schmancy term for "air"?

[krabapple]

I'm gonna guess it's either euphonic distortion, or bullshit (e.g. bias effect).

[Dob]

The idea of 'inaudible compression' is only unacceptable to people who know nothing of perceptual masking and its crucial role in codec development. This isn't brand new stuff, nor is it controversial in science.

I read the excellent Sound On Sound article that you linked to, and most of it did actually sound familiar.

My problem with perceptual masking is that it depends on perception. The scientific data backing it up is only as good as the ears that were providing it. I'm sure that extensive double blind testing was done, but that doesn't necessarily make the case air tight for me (unless I was one of the ones participating...and maybe not even then).

Let me give an example. A few years back I was reading an interview with a building inspector (here in Detroit) who casually mentioned that he could (if he really tried) smell carbon monoxide. The interviewer pressed him on it, but he insisted that he could, even describing what it smelled like (I forget his description), and the interview moved on.

My initial thoughts were "he's full of it." My whole life I have been assured by every textbook, every teacher, everything I've ever heard or read, that carbon monoxide is totally odorless.

But then I started to think about it. What if carbon monoxide was just mostly odorless? What if someone could gradually learn to detect it? The building inspector said that's what happened to him, because of his job. Thinking more, I realized that (probably) no one has ever thoroughly tested the odor of CO. Who would volunteer for a study like that, testing the smell over and over? Even if someone did, a short term study wouldn't account for exposure to trace amounts over many years. That possibility was probably dismissed out of hand...but did they ever think to ask any building inspectors?

Perception is such a difficult thing to nail down. Remember those "magic eye" books that were so popular about 10 years ago, with the 3-D pictures? When I first saw them I thought they were total BS and continued to think so...until I learned how to see the picture. Then it suddenly became the easiest thing in the world! I went from 0% double blind identification to 100% within a span of a few seconds.

[krabapple]

My problem with perceptual masking is that it depends on perception. The scientific data backing it up is only as good as the ears that were providing it. I'm sure that extensive double blind testing was done, but that doesn't necessarily make the case air tight for me (unless I was one of the ones participating...and maybe not even then).

If you won't even accept the existence of perceptual masking if you youself were to experience it in a controlled test...then there' s really *no* convincing you, is there? Still, I hope you';ll try the discs I sent anyway.

Perception is indeed a difficult thing to nail down, but that doesn't mean we know *nothing* about it. Advances of compression codecs has, unlike high-end audio, been driven almost entirely by controlled testing involving careful listening. That has to count for something, unless you consider the scientiric method itself irredeemably flawed because it relies on careful observation.

Your what ifs are just...*what ifs*, unless and until that CO guy, or someone else, actually demonstrates his ability in a controlled test. Seems to me that would be a rather *valuable* thing to have, if it could be demonstrated. Do you know for a fact how thoroughly the odorlessness of CO *has* been tested?

[Dob]

If you won't even accept the existence of perceptual masking if you youself were to experience it in a controlled test...then there' s really *no* convincing you, is there?

Well, I'd certainly admit that I can't hear the masking. But that doesn't mean that no one else can...or that I wouldn't be able to once I was taught what to listen for. As for the science behind perceptual masking...again, I'd admit that it works for me, but I would be reluctant to dismiss any audibility claims on purely scientific principles. Unless I'm missing something, these principles are completely dependent on testing for validation. Can you prove "on paper" that something is imperceptible? All it takes is one gifted person to prove you wrong.

But I wouldn't deny the usefulness of masking. If the data shows that out of hundreds of people none could hear it, that's good enough, I would think!

Advances of compression codecs has, unlike high-end audio, been driven almost entirely by controlled testing involving careful listening. That has to count for something, unless you consider the scientific method itself irredeemably flawed because it relies on careful observation.

Sure it counts for something. There are many, many people who can't hear masking and can therefore take advantage of the benefits offered by the MP3 format. It's tough to deny the utility of that.

Do I consider the "scientific method" flawed? No, but I don't think enough emphasis is placed on this:

"It is often said in science that theories can never be proved, only disproved. There is always the possibility that a new observation or a new experiment will conflict with a long-standing theory."

Your what ifs are just...*what ifs*, unless and until that CO guy, or someone else, actually demonstrates his ability in a controlled test.

Yes...but until we can test this guy, what do we do? Dismissing his claim, or allowing his one claim to cast doubt on CO odor, both seem to be overreactions. It's an awkward situation.

Do you know for a fact how thoroughly the odorlessness of CO *has* been tested?

I have no idea...but it seems reasonable to assume that the odor properties of CO are so firmly accepted that no scientist would even question it. And it would be problematic, I would think, to organize a study based on what some "guy in Detroit" claims...unless that guy made himself available for testing.

Don't they already have badges or something that detect the presence of CO?

BTW, I am looking forward to hearing those MP3s...with an open mind.