Nice Review of DSD by Andreas Koch

[MylesBAstor]

It is not. Multi-bit PCM quantization noise can be completely eliminated because we have so much headroom. Indeed, 24-bit audio is "self-dithering" in that there is natural noise in the right order bits so even out of the box, it is distortion free. If one wanted to have lower noise floor still, the same noise shaping used in DSD can be utilized to push the noise to ultrasonics.


The AES paper has good simulations of this. It shows that when driving by one or two pure tones, there are harmonic distortions that are just 32 db below the signal! Of course, the frequency by then is well into ultrasonics. So how audible those distortions are is subject to debate. More on this below. PCM on the other hand, has no harmonic distortion so is completely superior in this regard. The bats get to enjoy PCM far more than they do DSD .


Well, per above, PCM does not have distortion. Due to the fact that we can dither multi-bit PCM, we have no distortion at all. So it is a curious case that a system with distortion would be preferred.


Andreas' article talks about that. But with high sampling rate PCM, we are free of that in the audible band. This is the point I wanted to make earlier. See this graph from the article:

The noise above 20 Khz in DSD escalates at the rate of *140* db per decade! Someone needs to tell me the purpose of this system. If it is there because we supposedly hear ultrasnonics, why would we want to trash everything that is there this way? I mean either frequencies >20K are audible or not. If they are audible, then we should preserve them as we do with PCM. Not throw both noise and distortion in there.


This is the most important point. The AES article touches on it. It says that maybe the reason we like DSD is because the ultrasonic distortion in our speakers and amps, causes them to sound different in normal audio band.

It is also possible that somehow our ear responds differently when excited by ultrasonic content. This would explain why we don't hear it but why it might have an effect. In this scenario, it is entirely possible that what we want to hear is not the original content of music but the distorted version. Perhaps this is why analog sounds good because it probably has a lot of junk going on there too. I am making this stuff up as I go of course . But I can't come up with a different explanation of why I also like *high sampling rate* PCM converted to DSD.


Again, high sampling rate PCM should not have an issue here. And at any rate, DSD is often filtered also at playback. If that is a digital filter, it will also ring.


As I mentioned, the Koch article is marketing material. It makes statements like "it is proven we can hear 100 Khz" impulse or some such thing. What proof? Proof needs to be presented with references and there is none here. And then he moves on to show the above graph. If we care up to 100 Khz, why would we want the system response to look like that? Did the test that prove 100 Khz was audible use such a distorted response?

Bottom line is that DSD system is not linear. And can have tones and noise floor that is modulated by the input signal. Its foundation came from an approach to make DACs/ADC converters cheaper to make. And it came at a time when both Sony and Philips CD patents were expiring and they had to cook up something proprietary.

I was at the AES conference where a panel consisting of the authors in the AES paper (professors Lipshitz and Vanderkooy) and Sony/Philips were present. The content of the paper was shown using proper technical/mathematical analysis. Then everyone turned to the Sony/Philips folks and I was surprised that they provided no defense other than it sounds good anyway! Philips did present a counterpoint in the next conference but I have to say, despite preferring the sound of DSD, it was a let down. It was clear that smart signal processing people were not in the design loop of DSD.

If I surveyed 10 signal processing expert, I suspect every one of them would say DSD is flawed and has no advantage over PCM. I have yet to see anyone other than Sony/Philips folks praise it. There may be some that are out there but I have not seen them.

So to summarize, you prefer the sound of DSD, but the measurements show it's flawed. So maybe we're not measuring the right thing or as Andreas suggests, there's some psychoacoustics at work.

Sorry but I don't see this as marketing material but a basic introduction for the average audiophile on what DSD/SACD is about. BTW, Andreas Is probably one of the foremost experts on subject of DSD (he was also with Meitner---and Ed was a consultant to Sony) and his PD is the best digital playback system I've heard to date. Might I add Andreas has implemented some unique things in his digital gear.

 

[ack]

I rarely engage in debates on forums - a couple of posts per thread will do - and won't again here... But what I want to emphasize, as Myles alludes to, is, let's also spend our energy to try to understand why one technology is ellegedly audibly better than another, even though on paper it seemed "flawed". At the moment, I have no more technical reasons to believe DSD isn't technically flawed, but something's telling me we are perhaps not seeing the elephant in the room. For example, if DSD is also filtered with subsequent pre-ringing, why don't we look at their filters and pre-ringing characteristics of both PCM (even high-sampling versions of it) and DSD??? On the other hand, I suppose if Philips couldn't explain the audible benefits of DSD no one here probably can, or perhaps they just didn't spend the time to research it...

It also seems to me we ought to look at PCM conversion solutions that minimize or eliminate pre-ringing, like Meridian's & Ayre's minimum phase filters, or Spectral's and Berkeley's implementations using those SHARC processors. Could it be that pre-ringing isn't really an issue nowadays when brickwall filters are "done right", and that when someone claims PCM converted to DSD sounds better than the original perhaps what's going on is that they just didn't test with the best PCM conversion implementations out there, while at the same time any DSD pre-ringing is way off into the ultrasound or something??? Therefore, perhaps if we were to compare the best PCM implementations with the best DSD then perhaps we won't hear any differences??? So I would love to get all the details from Bruce on his experiment...

BTW, if I remember correctly, I _think_ there is indeed research that shows we do subconsciously hear up to 80kHz or thereabouts (and I vaguely recall engineers like KOJ quoting it as well), but I need to dig it up...

 

[JackD201]

Digital vs Digital. Ahhhhh, after all these years, one format pronounced dead and buried (DVD-A) who would have thunk it would happen again.

 

[ack]

BTW, if I remember correctly, I _think_ there is indeed research that shows we do subconsciously hear up to 80kHz or thereabouts (and I vaguely recall engineers like KOJ quoting it as well), but I need to dig it up...

Didn't take me too long... here's a quote from the HDCD patent (and I was probably right that KOJ was involved):

Recent research has shown, however, that humans use transient information in sounds with frequencies much higher than [Nyquist] to determine the direction from which the sound has come, and that eliminating those very high frequency components impairs one's ability to locate the source of the sound. The inner ear actually has nerve receptors for frequencies up to about 80 kiloHertz. Therefore, if the "brick wall" low-pass filter, which is a necessary part of all digital recording, removes frequencies above about 20 kiloHertz in transients, it reduces the level of realism in the sonic image.

It would be fairly trivial for me to dig up said medical research on those nerve receptors... Again, the above quote is from my reading of the HDCD patent blog. I am really concerned that we are over-simplifying things and I would have to temporarily agree with Myles that perhaps Koch's article is really a primer into the DSD technology and not the whole story...

 

[Phelonious Ponk]

Amir: Of course, every medium has its own particular/peculiar set of "distortions." It's just which distortions are kinder to the ear

Perhaps, but that's not the point made in the post you quoted from Amir. In that case, one medium turned the distortions into randomized noise and rendered them inaudible. The distortions were elimated. In the other, the distortions were not dealt with, in fact they were arguably maximized by a 1-bit system and rendered incapable of being neutralized.

If the distortions of DSD are kinder to your ear, enjoy them. But it is not a "pick your poison" proposition, it is, rather, that you like the dish better with a bit of the flavor of the poison in it than you do if the poison has been neutralized into something benign and undetectable.

Recent research has shown, however, that humans use transient information in sounds with frequencies much higher than [Nyquist] to determine the direction from which the sound has come, and that eliminating those very high frequency components impairs one's ability to locate the source of the sound. The inner ear actually has nerve receptors for frequencies up to about 80 kiloHertz. Therefore, if the "brick wall" low-pass filter, which is a necessary part of all digital recording, removes frequencies above about 20 kiloHertz in transients, it reduces the level of realism in the sonic image.

a) We're clearly talking about a biased source talking about this "recent research." Did they provide any references?

b) If we "hear" things up to 80k, and that effects our ability to locate sounds, why on earth would DSD be sending all its garbage up there to muck it up?

The psychology of this debate is more facinating than the digital science. As in the old analog/tube arguments, I wonder why it is so hard for people to admit that they might like a coloration (distortion), to actually accept that coloration might be the difference they hear? They will speculate about the undiscovered, unmeasured and unlikely to avoid that possibility. Why? I like a slice of lemon in my ice water. The traces of lemon juice are a "distortion" of that water, but I'm not compelled to imagine scenarios in which as of yet undiscovered science will someday declare that it wasn't the lemon that I tasted, it was pure water, and my preference was the way all water should have allways tasted.

I'm content to just enjoy how refreshing it is. YMMV.

Tim

 

[JackD201]

What we have is two certified experts on separate ends of the same field. Andreas and Amir certainly have the credentials. THAT is interesting. What would be even more interesting is if Andreas would join the discussion himself. Although it would inevitably end up about implementation, getting to that point would be priceless.

Would make for some great stuff on the invited debate section of the forum, yes?

 

[microstrip]

Although I would prefer that PCM would be the winner in both fields (specification and sound quality), as it was what I have access currently (the ARC DAC8 returned for a few weeks), I can not see how one can conclude anything of practical interest to users from the presented graph of noise threshold versus frequency.

Both techniques present results that are bellow audibility and no other parameters relevant to sound quality are discussed. As Jack said, I would also conclude that the process of selection the winner would be temporally implementation dominated.

 

[MylesBAstor]

What we have is two certified experts on separate ends of the same field. Andreas and Amir certainly have the credentials. THAT is interesting. What would be even more interesting is if Andreas would join the discussion himself. Although it would inevitably end up about implementation, getting to that point would be priceless.

Would make for some great stuff on the invited debate section of the forum, yes?

Well Andreas lives near Steve Perhaps Steve could entice Andreas!

 

[Steve williams]

He lives a mile away from me but I am sure he is at CES now. I leave tomorrow for CES and will try to catch up with him. I am betting however that Jonathan is reading this and will alert Andreas

 

[MylesBAstor]

He lives a mile away from me but I am sure he is at CES now. I leave tomorrow for CES and will try to catch up with him. I am betting however that Jonathan is reading this and will alert Andreas

Me too

Oh yes. Sitting in O'scare waiting for another two hrs for connecting flight to Vegas

 

[Steve williams]

Myles

We should try to connect at CES.

Do you still have my cell number. Where are you staying. PM me

 

[Bruce B]

It also seems to me we ought to look at PCM conversion solutions that minimize or eliminate pre-ringing, like Meridian's & Ayre's minimum phase filters, or Spectral's and Berkeley's implementations using those SHARC processors. Could it be that pre-ringing isn't really an issue nowadays when brickwall filters are "done right", and that when someone claims PCM converted to DSD sounds better than the original perhaps what's going on is that they just didn't test with the best PCM conversion implementations out there, while at the same time any DSD pre-ringing is way off into the ultrasound or something??? Therefore, perhaps if we were to compare the best PCM implementations with the best DSD then perhaps we won't hear any differences??? So I would love to get all the details from Bruce on his experiment.....

The experiments I have done over the past 4-5 years have been capturing analog sources, be it either live or from an analog tape machine, into multiple formats. I have always tried to stay on the cutting edge of technology to give my audiophile colleagues the best the music has to offer. Music has an emotion that the artist tries to convey to their audience. Up until the past 10 years, I don't feel digital could do that.
When I'm working with an audiophile label, I'm always caught between what is resolution/clarity and what is musical/euphoric. Over the years I've used the PM2, EMM Labs, dCS, Prism, Digital Audio Denmark, Grimm, Tascam, Korg, Forssell, Lynx, Avid/Digidesign, StageTech, MSB, Apogee, Weiss, Lavry, Universal Audio and finally the Alesis. That's quite the list. These are just A-D converters I've used going into a plethora of workstations. Certainly, you can see that I've exhausted about every known high-end converter available. With all the testing I've done for FIM, HDtracks, WMG and Concord, I think I have a pretty good handle on what all formats of digital sound like and what each brings to the table. As with all my projects, redundancy is the key.

I don't feel PCM captures the nuances of the emotion. This pertains to low level music like classical, acoustical jazz and piano. On the other hand, I don't feel DSD sounds good when something is recorded loud, like a rock band. DSD excells when you record at lower levels. Anyone can try this if they have a Korg unit. Capture one of your vinyl records or tapes at 2 levels. Record one pass where the levels are almost clipping and capture again where you have at least -6dB of headroom. Now compare the 2, bringing up the level of the quieter one.
With PCM, if you record something at a low level and you bring up the volume, I feel you have a higher noise floor and the nuances/emotion get lost in the muck.

 

[Bruce B]

He lives a mile away from me but I am sure he is at CES now. I leave tomorrow for CES and will try to catch up with him. I am betting however that Jonathan is reading this and will alert Andreas

And Jonathan will tell him to not get involved. You remember what happened to Dan Lavry/Apogee on the old rec.audio forums.

 

[MylesBAstor]

And Jonathan will tell him to not get involved. You remember what happened to Dan Lavry/Apogee on the old rec.audio forums.

Yeah but that was wreck.audio.com

 

[Ron Party]

Bruce, do you have any opinion one way or the other about the audibility of ultrasonic freqs?

 

[ack]

With PCM, if you record something at a low level and you bring up the volume, I feel you have a higher noise floor and the nuances/emotion get lost in the muck.

Thank you; this last statement says a lot to me - when we try to compare PCM and DSD, doing so solely based on some technical data isn't enough. The two technologies have apparently different distortion and noise characteristics. Are these fair statements to make?

 

[docvale]

Bruce, do you have any opinion one way or the other about the audibility of ultrasonic freqs?

Having followed a number of audio forums (all Italians, before this one), it has happened to me to read about inner ear capabilities a lot of times.
I have no doubt about Ack's citation, while I have doubts about the scientific references in that technical paper.

My knowledge in neuro-anatomy and neuro-physiology is not updated (I attended the classes in year 2000). Notwithstanding, I remember that the limit in inner ear detection is not a trivial thing. Neurons transduce at a frequency of action potential that corresponds to the acoustic frequency. Being no neurons that can fire at >800Hz, we can detect till ~17kHz (even 20kHz is over-optimistic) because neurons fire in clusters and they "share" the burden. Having clusters that support till 80kHz is not properly realistic, IMHO.

Anyway, I can believe that sounds that are >20kHz can affect audio perception. While I have read, on an Italian forum, that someone believes that audiophile listeners somehow develop super-human audio skills (which is totally ridiculous), I think that those ultrasounds interact with listenable sounds before those one stimulate the inner ear, thus interfering with them and modifying their detection.

 

[Ron Party]

Thanks, Valerio. Nice post.

As I previously posted, I am not aware of any repeatable scientific study confirming humans do have such ability. This topic has been beaten to death on numerous fora so we're not breaking any new ground here. Of course the fact there hasn't been any such study is not dispositive. Given that Bruce plays with toys that us mere mortals do not have access to I was hoping he might offer his observations.

 

[docvale]

For DAC's you'll also need to include the Mytek, Digital Audio Denmark, Prism and the EMM Labs which were overlooked.

Maybe I'm wrong, but I think that also the about-to-be-released DAC from Mark Levinson could be able to accept DSD files. What seems sure is that it'll get DSD through HDMI inputs.

 

[Bruce B]

Thank you; this last statement says a lot to me - when we try to compare PCM and DSD, doing so solely based on some technical data isn't enough. The two technologies have apparently different distortion and noise characteristics. Are these fair statements to make?

That is correct. They both have their strengths/weaknesses and you need to use them accordingly.