Skip to main content

This is the question. I have heard rumours, some quite reliable that it has "gone to meet its maker".

Anyone have any info that is "solid"? Also I guess this applies to DVD-A as well.

The big news seems to be HD DVD.

Comments, abuse, invective, seething injustices ?

Topic Tags

Comments

Cucco Mon, 02/14/2005 - 11:30

Nika wrote: My inclination is that we should research what it is that people have found so appealing about their SACD materials and then find a way to incorporate that into more traditional PCM if needed. ...

Nika

Now that's an idea I can get behind! There are a lot of flaws with the theory of DSD as well as (and in my never-humble opinion), more importantly, the poor implementation of it. If there were a way to take the positive sound attributes of DSD and shove that square peg into the round hole that is PCM, it would be phenomenal!

What are your thoughts on DxD? Of course, to deliver a DxD format to the public in stereo would require a DVD-like medium. To do so in multi-channel would require far more - perhaps dual or multi-layer formats.

I'm not one of those that puts that much faith into the extended frequency response of higher SRs (though, I do believe this plays a small role). It's quite obvious that the filters required for low SRs adversely affect the sound - and to some degree the same is true about DSD (though in a different manner).

I would love to put some collaborative heads together, come up with a new High-Def Audio that the general public has access to.

"Beyond HDCD, Beyond SACD, it's HRCD!" (I think XM might sue on that one).

J...

anonymous Mon, 02/14/2005 - 11:56

Cucco wrote: Now that's an idea I can get behind! There are a lot of flaws with the theory of DSD as well as (and in my never-humble opinion), more importantly, the poor implementation of it.

Hmmm. It seems to me that the protocol for DSD forces poor implementation. I'm curious of what your perspective is on where the implementation falls short but is somehow fixable?

What are your thoughts on DxD? Of course, to deliver a DxD format to the public in stereo would require a DVD-like medium. To do so in multi-channel would require far more - perhaps dual or multi-layer formats.

I guess the question is why? Are we certain that these alter-formats will provide something that CANNOT be provided with traditional, low rate PCM systems? Why? How? If it CAN be provided with the traditional systems then wouldn't we be doing the market a bigger favor by merely fixing the implementation used in those systems?

It's quite obvious that the filters required for low SRs adversely affect the sound

Or rather, CAN affect the sound. But does it have to inherently? Why?

I would love to put some collaborative heads together, come up with a new High-Def Audio that the general public has access to.

Why not just fix what we have?

Nika

Cucco Mon, 02/14/2005 - 12:15

"Hmmm. It seems to me that the protocol for DSD forces poor implementation. I'm curious of what your perspective is on where the implementation falls short but is somehow fixable? "

I wouldn't say it "forces it." Rather, the execs at Sony/Philips decided to use the 6 analog outputs as a means of copy protection. What part of DSD "forces" this kind of implementation? The consumer doesn't want to spend ass-loads of money on that many cables. Why not use some kind of optical or fire-wire type transport method?

I guess the question is why? Are we certain that these alter-formats will provide something that CANNOT be provided with traditional, low rate PCM systems? Why? How? If it CAN be provided with the traditional systems then wouldn't we be doing the market a bigger favor by merely fixing the implementation used in those systems?

Well, that's a big "IF..." CD has been around for 22 years now and digital/PCM quite a bit longer than that. Why is it that you think that traditional 16/44 simply needs a little tweaking when the industry as a whole regards higher sampling rates and bit rates as a superior solution? Did we give up hope on improving 16/44 or is it futile? My guess is on the latter. Granted, many advancements have been made to actually make CD sound pretty good, but higher-resolution formats (SACD/DVD-A) are generally accepted as far better than 16/44.

Or rather, CAN affect the sound. But does it have to inherently? Why?

Well Dr. You wrote the book on it, you tell me why you haven't created a filter that isn't degrading to the signal.

Why not just fix what we have?

You know, frankly I hate this mentallity. (Don't take it personally) We as a species are dependent upon our constant improvements. Why should we live with something that is "good enough" when there is better? Do you think that 16/44 PCM is capable of accurately displaying a sonic picture the way it was recorded? If so, then I think your more troubled than my original assessment.

The audio world is full of people who are afraid to move on to bigger and better for a few reasons:
*They can't afford it
*They can't comprehend it
*Some people just don't like change

Face it - CD will not be around forever; there will be a better format. Feel free to cling desperately to your aging and dying 16/44 while the rest of us are recording in a format that people enjoy listening to instead of finding fatiguing after a short while.

J...

DavidSpearritt Mon, 02/14/2005 - 12:37

You know, frankly I hate this mentallity. (Don't take it personally) We as a species are dependent upon our constant improvements. Why should we live with something that is "good enough" when there is better? Do you think that 16/44 PCM is capable of accurately displaying a sonic picture the way it was recorded? If so, then I think your more troubled than my original assessment.

The audio world is full of people who are afraid to move on to bigger and better for a few reasons:
*They can't afford it
*They can't comprehend it
*Some people just don't like change

Face it - CD will not be around forever; there will be a better format. Feel free to cling desperately to your aging and dying 16/44 while the rest of us are recording in a format that people enjoy listening to instead of finding fatiguing after a short while.
J...

Or there is no significant sonic difference. Evidence is there that no-one other than audiophile dudes cares, the public are happy with the old crappy format. You know, why invent a new wheel when the one invented in 2000BC works. I also hate the "if it ain't broke then don't fix it" attitude, but I also think an "improvement" has to be just that, something significant. They need to fix CD first, get good filtering and PSU's into CD players, get jitter immune output stages into them, then its time to move on.

These same audiophiles subscribe to the "intuitive" argument that the little staircases get smaller and smoother with hi-res sampling without understanding Nyquist and the fact that the staircases never get into it, don't seem to realise that the companies who thought this stuff up, a) need to sell new gear to survive and b) they need copy protection to "start" the money flowing again instead of actually creating something new that the public WANTS.

We have been here before. :)

anonymous Mon, 02/14/2005 - 20:46

Cucco wrote: Regarding the conjecture that DSD forces poor implementation:

I wouldn't say it "forces it." Rather, the execs at Sony/Philips decided to use the 6 analog outputs as a means of copy protection. What part of DSD "forces" this kind of implementation? The consumer doesn't want to spend ass-loads of money on that many cables. Why not use some kind of optical or fire-wire type transport method?

Assuredly the issue you speak of is an implementation issue as well, and that one is indeed a choice from the conglomerates that push DSD and codified its format. There are other issues, however, such as the 1-bit problem, the distortion inherent in its data, and the need for an analog low-pass filter (phase shifting) that are an inherent part of this ideology, independent of choices made at Sony of how to codify it.

Nika: If it CAN be provided with the traditional systems then wouldn't we be doing the market a bigger favor by merely fixing the implementation used in those systems?

Well, that's a big "IF..." CD has been around for 22 years now and digital/PCM quite a bit longer than that. Why is it that you think that traditional 16/44 simply needs a little tweaking when the industry as a whole regards higher sampling rates and bit rates as a superior solution?

You keep talking about the "industry as a whole." I don't think that you and I have the same perspective on the whole of this industry. Many of the circles that I am involved in have participants that do not so eagerly endorse that higher bit depths and sample rates are the solution to the audible problems we are trying to address.

... but higher-resolution formats (SACD/DVD-A) are generally accepted as far better than 16/44.

"Generally accepted?" This sounds like conjecture again. By jumping to that conclusion so readily I think you undermine the ability to try to find the true answers. Though those answers may indeed lie in higher sample rates and bit depths (though I have seen no evidence of this to date) we obfuscate the ability to look at the situation objectively by stating as an assertion that it is already widely accepted that this is the case. I can refer you to many, including converter designers, audiologists, otologists, recording engineers, producers, DSP engineers, people at Sony and Philips, experts in psychological testing and more that do not agree with your assertion.

Well Dr. You wrote the book on it, you tell me why you haven't created a filter that isn't degrading to the signal.

Have you read about Bob Katz's filter test with mastering engineers? What about the testing done at Sony's Oxford division in their ABX testing chamber with the converters used on the Oxford console? Despite the sarcastic nature of your remark, it is not in my position to spend my energy designing filters. We all have roles in this industry. While I know a chunk about filter design and while I have Matlab and the filter design software, I am not the best guy for that job.

Nika: Why not just fix what we have?

You know, frankly I hate this mentallity. (Don't take it personally) We as a species are dependent upon our constant improvements.

You are implying that more bits and more samples will somehow be an improvement. This is a conjecture that you have failed to validate as of yet. I have as much interest as anyone else in providing the best system - through whatever means are necessary. Should that include higher sample rates or higher bit depths then I get behind that. As of yet your assumption is not validated and in order to take a truly objective approach we MUST first start with what is wrong with the current system. Once we understand that we may indeed find that merely fixing it is best. We may not. We can't know that until we approach this using the scientific method, however - an approach that jumping to conclusions does not support.

Why should we live with something that is "good enough" when there is better? Do you think that 16/44 PCM is capable of accurately displaying a sonic picture the way it was recorded? If so, then I think your more troubled than my original assessment.

From the information I have now, yes. And I'll take your ad hominem attack at face value, thank you.

Tell me, why do you think it is not?

Feel free to cling desperately to your aging and dying 16/44 while the rest of us are recording in a format that people enjoy listening to ....

Ahh, I just have to chuckle as I think of the first available, cynical reply:

"What is that? Straight to iPod?"

Cheers!
Nika

Cucco Tue, 02/15/2005 - 07:37

Nika,

For every single "scientist" you point out, or every single technical engineer at Sony or Philips, I could point out 100 musicians who differ in opinion. Frankly, I could give 2 sh*ts about the information on the paper if in fact, the people who are listening hear a substantial difference and they find it virtually universally better.

If you put any stock into the industry rags from both fronts recording and consumer (Home Theater, Stereophile, etc.) you will see that, as representatives of the industry we serve, these customers highly enjoy SACD and DVD-A. So what, your papers say there should be distortion - I could care less. What I hear is clearer, warmer music, much more analog in style.

As for my unqualified opinion - I have been involved in many market surveys - some of which I created/conducted myself (which, if you dig a little on this board, you will read about) and others that I was involved in that I am not at liberty to disclose until released to do so. Overwhelmingly, in every one of the projects which I've worked on, the average Joe hears a significant improvement and likes it.

So, as for science - sure, it's important, but frankly, you can tell me all day long that science has proved the sky isn't blue, but if it looks blue to me, it's blue. The problem is, you are making an awful lot of assumptions based on the scientific findings of others. Frankly, people can't hear the miniscule phase shiftings inherent in an analog filter which affects the signal so significantly outside the range of critical listening. The distortion to which you refer is again, well below the perception of human hearing and again, outside the critical range.

There is a fine line between the consumers' choice between convenience and quality. Do you think that, if someone created a format that sounded as good or better than PCM and was able to implement it easily with an IPod that people wouldn't flock to the new device? Now what if it cost $25 more, but the difference in quality were evident? People would still buy it.

Sony has an excellent product on their hands. Flawed? Yes. Better sounding than CD? Yes. Less convenient and more expensive? Yes.

Here's my point - don't sit here and tell me that SACD/DSD doesn't sound as good as PCM just because some numbers on paper "prove" that it isn't perfect. Damnit, I hear a better sound and a more lifelike sound as do pretty much anyone else who listens.

The original question - why is SACD failing (paraphrased) - poor implementation and even worse marketing. Sony thought that by including a free SACD of music no one wants to listen to with their products and advertising in magazines focusing on audiophiles that their product would take off. Instead, they needed to create a player that was easy to hook up, affordable and advertise during the Super Bowl.

This is my last comment on this, so if you would like to waste your time typing a response to show how the numbers mean that music doesn't sound like music, go ahead. But frankly, I'm growing tired of listening to people like you telling us that the world is broken but not providing any clear manner in which to fix it. So, my suggestion, stop criticizing unless you have something positive as a solution.

Exsultavit Tue, 02/15/2005 - 12:05

Dear Nika and Cucco-

"Just for the sake of the casual observer, Jeremy and I had a comprehensive conversation on the phone today and, as I'm sure we both assumed, our perspectives/positions are much closer than might appear in public."

Well, I am certainly happy to hear that!

I enjoy a good discussion between mutually respectful posters. The exchange between the two of you, however, was less than polite and NOT representative of the kind of exchange I am used to reading on this forum. In short-- I myself don't want it.

I am not interested in Who Was Rude First, or making one of you out as the baddie. Each of you has your own style of posting and I find each of you as both loveable and frustrating in your own special ways :?

This fresh, young forum has had a great S/N ratio until now. I am concerned that this quality might shift to that of the more typical boards where the flames and the BS must be slogged through to read the good stuff. Hoping that this thread will not set a precedent where we lowered the bar.

respectfully,

Exsultavit

Cucco Tue, 02/15/2005 - 12:47

The truth is Exsultavit, I got defensive. I saw Nika's comments as more than they were and took it as an attempt to "stir the pot" for one's own personal gain. After speaking with Nika, I found this to be untrue and unfounded. As a matter of fact, he and I agree on several of the finer objective points of digital.

I've lurked in the audio forums for as long as they've been around and I found this forum to be the most friendly. Not just person-to-person interaction, but easy to approach for the layman as well as the seasoned audio professional. I do take it personally when I feel that it's being threatened.

Again, after a lengthy and quite positive conversation with Nika, I think we are more than past any apparent issues on this board and I would like to extend a professional and personal apology to any of you who may have found our bantering unprofessional or annoying. I would also like to extend that apology directly to Nika and encourage him to continue his residency on this board.

I do want to make it clear though, that always, my intentions are to keep this forum relatively easy to approach by all both personally and professionally.

Thanks!

Jeremy

JoeH Tue, 02/15/2005 - 18:22

I think we're in good shape here, regardles, but I appreciate X's concerns. (Don't worry, Exult; we'll all make sure it stays civil and informative here. 8-) )

I was reading the latest back and forth initially with some concern and a little horror, but then I decided to imagine Nika and Jeremy as two college buddies sitting around in a dorm/lounge, having a coffee (or something stronger) and read it that way: An impassioned, argumentative back and forth, but in spite of the harsh attitudes, no real damage being done. In that light, it was almost humorous, or at least a lot lighter than it seemed. (just me?)

I have to add that Jeremy is a stand up guy (as is, I'm sure, Nika). He's accessable, too, should you want to deal with him and any point of clarification - drop him an email and you'll see what I mean.

The beauty of this place is that a strong opinion doesn't HAVE to lead to flames or combative behavior; everyone in this small, new community sure seems to know that, and we're richer for it.

anonymous Tue, 02/15/2005 - 18:50

Cucco wrote: To me and thousands other, it's very clear that the sound of DSD is superior to that of PCM. You can sit here and write about all of its deficiencies all you want, it's not gonna change that fact.

Well, call me lead-ears, for I am NOT one of the thousands who clearly hear and agree with this "fact." Sounds very good, but so does PCM when done well.

I think it would make a very interesting poll to see who has made the considerable investment neccesary to actually record in DSD. I am not really thinking of the new TASCAM box, because once you've recorded to your external drive or DVD that's the end-- no editing or DSP possible, and that is like going back to the days of the wire recorder.

So, anyone here have their Sonoma on order? Or use PMX or SADiE in DSD? Or own a Genex? Authoring SACDs from pure DSD?

One very famous Brit engineer put it this way:"Using DSD as an interim format before resampling to 44k1 is close to being a complete waste of time."

I completely agree.

Rich

JoeH Tue, 02/15/2005 - 20:28

So, anyone here have their Sonoma on order? Or use PMX or SADiE in DSD? Or own a Genex? Authoring SACDs from pure DSD?

Nope, nope, nope and nope.

I'll get into DSD as soon as my clients start asking (and are willing to pay) for it. Until then, it's just a theoretical discussion for me.

Cucco Wed, 02/16/2005 - 07:04

I had Pyramix on order, but I cancelled the order. It's important to understand that no editing can take place on a Pyramix machine within the DSD world. It is sampled to PCM and then resampled back to DSD.

Sadie and Sonoma are the only two systems I know of where you can do anything to a native DSD signal.

While I whole heartedly agree that standard PCM can sound fantastic - I have been involved in some research projects in conjunction with a few universities (Once the project is released, then I will gladly share all the details) where a signal was recorded in DSD and in PCM from the same source through the same mixer and no additional editing was done. Hearing the difference was not difficult. To say that though, I should mention that not all of the panelists heard the difference.

To suggest that there is a massive difference, I would be in the wrong. But I do hear a difference, and though subtle, I found it and still do find it to be a significant one.

With all that being said, I felt confident enough in PCM to invest in Sequoia - a system incapable of DSD and a right expensive system to boot.

Do not take all of my banterings as me being the proponent-on-high for DSD, but I do think it is necessary for us as musicians to seek out the best possible sound and technology. Agreed, DSD/SACD may not be it, but I'm not entirely convinced (nor will I ever be) that PCM (specifically 16/44) is the best solution either.

I doubt any of us are old enough to recall this, but, back in the day, when Edison invented his nifty phonograph, the music industry was quaking in their boots. They felt for certain that the phonograph sounded so realistic and lifelike that they thought people would stop going to concerts and enjoy their music in the comfort of their own home.

Aren't we all glad the iPod hadn't been invented yet???

J...

anonymous Wed, 02/16/2005 - 07:23

Cucco wrote: I had Pyramix on order, but I cancelled the order. It's important to understand that no editing can take place on a Pyramix machine within the DSD world. It is sampled to PCM and then resampled back to DSD.

Sadie and Sonoma are the only two systems I know of where you can do anything to a native DSD signal.

There's a bit of a misnomer I see passed every now and then that I think is worth bringing up.

DSD is PCM.

DSD is 1 bit PCM, just like if you took your signal in your DAW software, bounced it to disk at -138dB so that you only had 6dB of dynamic range (1 bit) and used some sort of noise-shaping on it. That's DSD, only it's done several octaves higher.

Because of this, when we talk of editing DSD as PCM, we have to acknowledge that it happens by default - because it IS PCM! The difference here is that some software first downsamples it to edit it - down to 768kS/s or 384kS/s or whatever and other software keeps it at the higher rate and edits it.

If the issue is that they are turning the data into greater than 1 bit, that happens by default, as a part of the mathematical process. Think, for a moment, about how you design a 1 bit summing bus, never allowing the data to grow to 2 bits!? :) The same occurs when we filter, compress, or any other process - the data grows, just like it does in any OTHER PCM system. Because of this (and to avoid distortion related to this) Sony has put out an editing/processing spec called DSD-Wide in which the DSD data is padded with 7 zeros and then simply processed as 8 bit data throughout the system until the final product is finished, at which point it is bit-reduced to 1 bit just like you reduce 24 bits to 16 bits in your own DAW.

DSD is PCM, so any talk of them "turning DSD into PCM for processing" must be about them downsampling for ease on the processor, I'm assuming. After all, what is the benefit of keeping it at the higher rate for processing rather than downsampling and upsampling back when the processing is done?

Nika

Cucco Wed, 02/16/2005 - 07:44

True points. In general, the misnomer that you speak of is commonly seperated from PCM (despite its obvious relation) just because we live in a TLA world (Three Letter Acronym...)

When I say that it is changed from DSD to PCM, what I am suggesting is that the SR is changed as is the bit depth, both to more commonly seen 24bit 384kHz sampling. (Which, ironically, is in many ways far superior to DSD -- on paper :wink: )

Simply put, Pyramix cannot edit anything in the single bit/2.xxx gHz sampling range, but technically as Nika suggests, neither can any other system. However, they at least perform these edits in a system which does not create such a drastic change as that of switching from "DSD" to traditional "PCM."

In any case, if Sony or Meridian want their stuff to succeed, which was, as I recall the original topic of discussion - they will need to prove to the consumer that it is FAR superior to their iPods or current CDs and then make it easily accessible. Notice, I didn't say that they had to prove anything to the AE community. We live in a customer driven environment. If High-Res audio is to come, it's cuz the people demand it, not because the studios "can do it."

Soooo...once Bose Lifestyle systems can play back SACD or DVD-A, then the general public will assume it's the best and they will have to have it! :wink: :?

J...

anonymous Wed, 02/16/2005 - 07:50

Cucco wrote: ...both to more commonly seen 24bit 384kHz sampling. (Which, ironically, is in many ways far superior to DSD -- on paper :wink: )

Only if it doesn't come from DSD material. If it comes from the current standard of multi-bit or multi-level A/D signal generation at higher modulator rates (basically a multi-bit version of the delta sigma modulator used for DSD generation) then it is superior to the DSD signal. If it hails from the same single bit modulator as DSD then it will not be able to overcome the errors that DSD provides. This is why the "traditional PCM" world shifted to multi-bit or multi-level modulators about 5 years ago (many years after Sony had already gotten the ball rolling with DSD).

If you want 384kS/s to be better than DSD you have to start with a signal that is better than DSD, such as a multi-bit version of DSD.

Nika

anonymous Mon, 02/21/2005 - 09:25

Just thought I'd share an anecdote. I went to a lecture last night at a local AES chapter meeting. The presenter was Todd Whitelock of Sony Studios in New York. He's a multi-grammy award winning engineer for primarily jazz and classical work. He uses primarily DPA mics, Millennia preamps, et al, and he played us several samples of his work.

At one point he played us two samples of the same work (mind you, Genelec 1038 monitors in a relatively nice room) - one of the DSD track of a classical piece and the other merely the 44.1kS/s PCM version of the same thing. He asked the room to compare and the reaction was worthy of some sort of clinical paper - very interesting, but we'll leave that aside.

What was interesting was Todd's claim that DSD is his absolute favorite recording medium these days. He likened it to a 2" machine, and claimed that, just like with old analog tape, whatever you fed into the system always sounded better coming out the other side. Everything sounded bigger and warmer, etc. For this reason it was his favorite medium to record onto.

He was very quick to state (though I'm not sure he recognized the impact) that it is definitely not more accurate - just preferred. He then went on to discuss the shortcomings of using DSD (lack of tools, etc).

This seemed to fit into the discussion somehow.

Nika

anonymous Mon, 02/21/2005 - 10:14

Nika wrote: At one point he played us two samples of the same work (mind you, Genelec 1038 monitors in a relatively nice room) - one of the DSD track of a classical piece and the other merely the 44.1kS/s PCM version of the same thing. He asked the room to compare and the reaction was worthy of some sort of clinical paper - very interesting, but we'll leave that aside.

i think it would be pertinent to tell us the room reaction, especially with that audience.

Rich

anonymous Mon, 02/21/2005 - 10:18

Oh, they "heard" a difference. But it would be unfair to state this without painting the background - "Isn't that amazing? (lots of talking up DSD) - how many of YOU hear that?" Enthusiastically they seemed to hear these amazing differences.

I should also point out that he played a 7' piece. Then stopped and talked for a couple minutes. Then played it again in the other format. The notion that this test is at all representative is laughable. This is why I said it would be worthy of a clinical paper on listener bias, etc. I was amused.

Nika

anonymous Mon, 02/21/2005 - 10:29

Sacrilege! How dare you, Rich, make such an accusation!?

Actually, it was not intentionally "structured to yield maximum positive reaction," as this was a minor part of the presentation. Indeed, however, he was from Sony studios, no less. But the structure of this part of the presentation indeed (especially amongst students - did I mention that this was a student chapter meeting?) happened to be the most likely to yield that reaction. In other words, if I WANTED that reaction, this is how I would have done it, but I believe it was coincidence that it happened that way last night - it was not intentionally "structured" that way.

Nika

Cucco Mon, 02/21/2005 - 18:20

I suspect you're right about their apparent "salesmanship." Harkening back to my days in sales, if I wanted to sell something to someone and really make them feel like there was a difference, I would simply point out the difference where I felt it would be most obvious. ("Do you hear how warm the strings sound on these Infinity Reference Series Speakers???") And agreed, the test is laughable.

That being said, there is a sonic difference between the two formats, that fact can't be disputed. Whether those differences are viewed as subjectively "better" or "worse" is up to the listener. Of course, I'm sure the Sony rep being present and "selling" the DSD didn't hurt the SACD phenom.

As for the thought that it's financially out of reach, I would take a little issue with this. I was in the process of purchasing a DSD system recently until I essentially changed my mind for various reasons.

The converters, available from people like Genex or Prism, aren't really that extremely priced. Yeah, 10K for the Prisms ain't cheap, but that's 8 channels of A/D and D/A and some damn fine conversion at that (even - or especially for that matter - at 24/44.1). Then of course, you can look at the Genex system - about $7k for the 8 channel DSD version. This is no higher priced than a Sequoia system. Then, to edit, you would have to look at something like Sonoma - a reasonable 3K. Of course, you could go Sonoma the whole way, but then you're stuck with Meitner converters (I know, what a damn shame...) The whole thing with 8 channels of conversion, input and editing would set you back roughly $19,000.

Let's call it the BMW dealer visiting the college campus rather than the Rolls dealer.