[Music Fanatic]

Quote:

Originally posted by theaudiohobby Care to give a rigorous theoretical basis for this assertion covering the bit-depth, transient response, impulse response, noise floor and such like to PROVE your confident assertion. I assume this is a theoretical assertion as I am sure you have not heard all the best implementations of both systems. I think it is time we went back to basics and ask for MATHEMATICAL PROOF. I have provided an example to give an idea of what I am thinking of. This paper is the most rigorous paper I am aware written in opposition to DSD and it has a significant flaw that is DSD is not based on a 1-bit quantizer. So there you are, please take the platform and prove your assertion.

Regretably, I am unable to open the paper you cited -- Adobe gives me a "file error" message.

 

[Sovkiller]

Quote:

Originally posted by Music Fanatic Regretably, I am unable to open the paper you cited -- Adobe gives me a "file error" message.

The doc open perfectly, check your settings, or maybe you have an old version of adobe, if you want I can email it to you....

 

[Music Fanatic]

Quote:

Originally posted by Wodgy Yes, strictly speaking, given perfect DACs for each format, the SACD format is inferior to the regular CD format, let alone DVD-A.

I very much disagree with the first assertion that SACD is inferior to regular CDs. Over the range up to 22.05 kHz, the noise floor of SACDs is lower than CDs and resolution is higher. This is far more true over the range where music mostly takes place (say, under 8 kHz).

The argument you cite by Lit****z (that dithering is more difficult with SACDs) is an interesting one, and he does show examples where that is the case, but he most certainly does not prove that is typically the case. For normal cases of music, one would expect that dithering with SACDs would be just fine.

Having said all of this, I must say that there are some sound wave forms that can be represented with 44.1/16 that can not be realized with DSD (and visa versa) so the two formats are not strictly comparable. However, I think it is fair to say that DSD is in most ways superior to CDs.

Your argument about DACs is a bit of a red herring since there is so much phase distortion introduced in the brickwall ADCs required to make CD recordings.

 

[Music Fanatic]

Quote:

Originally posted by Sovkiller The doc open perfectly, check your settings, or maybe you have an old version of adobe, if you want I can email it to you....

I can see the document was just changed, or the web server is at least retaining connections now.

 

[Wodgy]

Quote:

Originally posted by theaudiohobby You do not have to look far,, see my previous post for the link and read the followup reactions.The history and here is a previous thread that the later pages cover a few of the key papers of both sides of the discussion.

Thanks for the links to those threads. I had totally missed them when they came around the first time. They were very interesting reading.

I'm somewhat concerned that most of the data used to support DSD cited in those threads were from white papers from manufacturers, particularly dCS, rather than peer-reviewed scientific work (but read on below...).

Quote:

Edit - Wodgy, sorry I misunderstood your post, however the previous thread does indeed cover some key papers that answer the Lip****z contention, however I have yet to locate a key AES response paper that directly address some of his key contentions, the paper was reported in the press at the time, however I do not think it is in public domain.

Yeah, the Stereophile "rebuttal" is outdated and not really on topic. I was able to find a few follow up papers on both sides of this topic, some from the manufacturers. (I'm only interested in peer-reviewed stuff though.)

It seems that everyone on both sides of the issue agrees that single-bit ADCs are insufficient even if your final goal is DSD (you posted this yourself when you mentioned that good DSD quantizers had to be multibit). Where there is considerable disagreement is whether multibit DSD ADCs can generate an adequate DSD signal. There is a rebuttal paper from Philips published at the AES that gives empirical results showing that multibit DSD ADCs are indeed good enough, but they do not rebut the claim in the final part of the Lipschitz and Vanderkooy paper that DSD can never be perfected, even with multibit techniques (see the Appendix of the original Lip****z paper for the multibit discussion). The math is merely mentioned there, but Lip****z and Vanderkooy have two follow up papers where they rebut the criticisms and work through all the math properly: "Toward a Better Understanding of 1-Bit Sigma-Delta Modulators, Part 2" and "Toward a Better Understanding of 1-Bit Sigma-Delta Modulators, Part 3." Later, some researchers at the University of London came up with a good paper that demonstrates how to construct audio signals that will have audible distortion when encoded as DSD, even using the best multibit techniques:
http://www2.elec.qmul.ac.uk/%7Ejosh/...ons/AES114.pdf

When you step back, there's no contradiction here. The manufacturers are saying that the technology is adequate and perhaps even superior to PCM under certain circumstances and implementations which may be common or more practical, while the scientists are saying that DSD isn't perfect and is indeed techically inferior to PCM assuming perfect implementations. The sticking point which doesn't totally sway the discussion towards PCM entirely, is pointed out in a Philips paper that I found at work but can't find now for some reason, is that DSD is somewhat more information-rich in the time-domain (but weaker in the noise and frequency response areas). So the debate will probably go on forever given the technologies as they stand today. I will concede now (and I didn't realize this earlier) that both SACD and DVD-A have their own advantages and disadvantages, though IMHO DVD-A is still probably the winner overall.

From my understanding of all this, if there was some mythical 24/392 PCM format (sampling about as twice as fast as DVD-A), there would be little basis for debate at all, and that it would be hard to argue that DSD had any advantages at all there, even given comparable or somewhat higher sampling rates, but alas such a mythical format doesn't exist, and the debate will probably continue. I'm exiting now because I have other stuff to do tonight.

 

[Wodgy]

Quote:

Originally posted by Music Fanatic Your argument about DACs is a bit of a red herring since there is so much phase distortion introduced in the brickwall ADCs required to make CD recordings.

I can't think of any modern ADCs that aren't phase-linear, even at 20kHz, though this was a problem 10 years ago. I'll agree with you though that the digital filters are still problematic, but it seems to be more related to ripple or to the unnatural nature of the rolloff itself (most instruments have harmonics about 20kHz, and throwing them away probably does harm, even if we can't hear them audibly).

And, heck, I won't argue that CDs sound good. They don't. Not today. Maybe one day.

 

[Music Fanatic]

Quote:

Originally posted by Wodgy From my understanding of all this, if there was some mythical 24/392 PCM format (sampling about as twice as fast as DVD-A), there would be little basis for debate at all, and that it would be hard to argue that DSD had any advantages at all there, even given comparable or somewhat higher sampling rates, but alas such a mythical format doesn't exist, and the debate will probably continue. I'm exiting now because I have other stuff to do tonight.

I've seen arguments that preserving audio over 20 kHz is important either because it is audible or to help make a smoother filter. However, this is the first time I've seen an argument that preserving freuencies over 96kHz is useful or important.


And I'm afraid I've seen plenty of recording equipment -- even that used today -- that introduces phase distortion in the anti-aliasing step of A/D conversion. However, it must be said that the level of good equipment today is much less than that introduced by, for example, a typical loudspeaker.

 

[Czilla9000]

http://www.hydrogenaudio.org/index.p...hl=bad+quality

The guys at HA seem rather conviced that the CD iis, for all intents and purposes, perfect.

 

[Wodgy]

The guys at HydrogenAudio are, for the most part, holier-than-thou people without even a basic understanding of signal processing. There are some good guys over there too, but they often get drowned out by the noise.

It's not uncommon over at HA to find people who argue ludicrous points, for instance that it's impossible that there could be any signal degradation at all when you run line-level unbalanced signals over 200 feet of cable, or that if you can't hear the difference between two sources in a six-second sample then there could be no possible difference.

 

[Czilla9000]

Wodgy,



So, in your professional opinion, even with a perfect DAC design, the CD is not perfect.


What about DVDA with a perfect DAC design?

 

[Music Fanatic]

Quote:

Originally posted by Czilla9000 http://www.hydrogenaudio.org/index.p...hl=bad+quality The guys at HA seem rather conviced that the CD iis, for all intents and purposes, perfect.

Well, in fact, I'm pretty sympathetic to the philosophy at HydrogenAudio.

But it would not be possible to logically argue the CD is perfect, if, as I believe is true, experimentally most people could easily hear the difference between SACDs (or DVD-As) and CDs and think the former sound more lifelike. If the CD was mastered just the way the high-resolution format was (and was decimated from that format) this would be a clear indication that there was something better than the CD and thus it could not be perfect.

 

[Glassman]

Quote:

Originally posted by Music Fanatic Well, in fact, I'm pretty sympathetic to the philosophy at HydrogenAudio. But it would not be possible to logically argue the CD is perfect, if, as I believe is true, experimentally most people could easily hear the difference between SACDs (or DVD-As) and CDs and think the former sound more lifelike. If the CD was mastered just the way the high-resolution format was (and was decimated from that format) this would be a clear indication that there was something better than the CD and thus it could not be perfect.

EXACTLY!

I definitely need to rip some 24/192 material off DVD-A and convert it to 16/44 and back to 24/192 and then we can speak about quality of the format

I've got the feeling that the difference will be surprisingly subtle

I read somwhere that the most important think why hi-rez formats sounds better is just due to the possibility to place the LPF far from 20kHz, not that 16/44 is not enough information-wise, but yes, little bit of that..

 

[fewtch]

Quote:

Originally posted by Music Fanatic Well, in fact, I'm pretty sympathetic to the philosophy at HydrogenAudio. But it would not be possible to logically argue the CD is perfect, if, as I believe is true, experimentally most people could easily hear the difference between SACDs (or DVD-As) and CDs and think the former sound more lifelike. If the CD was mastered just the way the high-resolution format was (and was decimated from that format) this would be a clear indication that there was something better than the CD and thus it could not be perfect.

Just to set the record straight -- HA is really not about something being 'better' or 'worse' than something else. The HA philosophy is about first proving differences -- if a difference can be established with statistical validity, then nobody there really gives much of a damn whether someone might find some sonic difference personally/subjectively "better" than another. Or alternately, "better" means lower distortion and a truer representation of the original signal, the "classic" hi-fi approach.

It's a scientifically-oriented objectivist forum, and the interest is in provable facts (insofar as such things can reasonably be proven) and not subjective opinion. That's how it fundamentally differs from Head-Fi... FWIW I find merit in both the subjective and objective approaches to audio, and enjoy each board for different reasons.

There is in fact a fundamental weakness to the subjective approach: Some fairly dramatic claims are made sometimes with nothing backing them up but the ears and tastes of the person(s) posting, or maybe a review in Stereophile or manufacturer/dealer marketing claims... as a result, a lot of people may end up wasting money (errr... I mean upgrading *cough*

), only to get rid of the item a month or two later in search of something they like better.

IMO, the DIY community here has probably done the best job of combining the two approaches (subjective good sound + objective/measurements) and the results of the synthesis (Meta42, PIMETA, PPA) are here and available for all to benefit from.

 

[theaudiohobby]

Quote:

Originally posted by Wodgy It seems that everyone on both sides of the issue agrees that single-bit ADCs are insufficient even if your final goal is DSD (you posted this yourself when you mentioned that good DSD quantizers had to be multibit). Where there is considerable disagreement is whether multibit DSD ADCs can generate an adequate DSD signal. There is a rebuttal paper from Philips published at the AES that gives empirical results showing that multibit DSD ADCs are indeed good enough, but they do not rebut the claim in the final part of the Lipschitz and Vanderkooy paper that DSD can never be perfected, even with multibit techniques (see the Appendix of the original Lip****z paper for the multibit discussion). The math is merely mentioned there, but Lip****z and Vanderkooy have two follow up papers where they rebut the criticisms and work through all the math properly: "Toward a Better Understanding of 1-Bit Sigma-Delta Modulators, Part 2" and "Toward a Better Understanding of 1-Bit Sigma-Delta Modulators, Part 3." Later, some researchers at the University of London came up with a good paper that demonstrates how to construct audio signals that will have audible distortion when encoded as DSD, even using the best multibit techniques: http://www2.elec.qmul.ac.uk/%7Ejosh/...ons/AES114.pdf When you step back, there's no contradiction here. The manufacturers are saying that the technology is adequate and perhaps even superior to PCM under certain circumstances and implementations which may be common or more practical, while the scientists are saying that DSD isn't perfect and is indeed techically inferior to PCM assuming perfect implementations. The sticking point which doesn't totally sway the discussion towards PCM entirely, is pointed out in a Philips paper that I found at work but can't find now for some reason, is that DSD is somewhat more information-rich in the time-domain (but weaker in the noise and frequency response areas). So the debate will probably go on forever given the technologies as they stand today. I will concede now (and I didn't realize this earlier) that both SACD and DVD-A have their own advantages and disadvantages, though IMHO DVD-A is still probably the winner overall.

The second time out on this discussion and it is certainly a lot more productive, thanks for digging out the AES papers, I could not locate them the last time round, though I notice that one the papers is quite recent and maybe only recently got posted. I agree mostly with your post, however a few things, more accurate in the time domain is probably a more correct description. Secondly, as sampling frequency increases PCM begins to suffer from a number of practical limitations one of them being noise another being onerous storage requirements, IOW the sampling frequency cannot rise infinitely in a practical application because of the practical unweildiness of handling 24bits at very high sampling frequencies. In fact, I think that the industry itself is mostly limiting itself to 96kHz for these reasons as the DVDA specification only specifies 192/24 for 2CH and requires MLP for MCH 96/24. My take, from an implementation point of view, DSD runs rings round DVDA and is almost certainly an industry approach and DVDA an academic approach since former's advantages are really only fully appreciated in real-life commercial applications.

 

[Music Fanatic]

Quote:

Originally posted by theaudiohobby My take, from an implementation point of view, DSD runs rings round DVDA and is almost certainly an industry approach and DVDA an academic approach since former's advantages are really only fully appreciated in real-life commercial applications.

Well, if you are going to say that we may as well point out that as a simple matter of marketing, SACDs tend to be more numerous and have a much more diverse set of content than DVD-A. I have little doubt that as of today, SACD is winning the marketing battle (although, perhaps losing the war, since I'm not sure either format has enough traction to last.)