Why is SACD not really succeeding?

[Totoro]

Another reason is that bass management can be a real pain using analog connections and usually fall short of the capabilities of a decent AV receiver.

If everyone could use a single digitial connection, I am sure that it would be far more widespread. But, as usual, the music industry has cut off their nose to spite their face.

VERY TRUE

I considered getting into SACD DVD-A since I have a player capable of hi-res playback. But managing the analog because of the stupid copy protection scheme will keep me away from making the investment. Even the sales rep at Cambridge Soundworks said forget it.

 

[Sleestack]

VERY TRUE

I considered getting into SACD DVD-A since I have a player capable of hi-res playback. But managing the analog because of the stupid copy protection scheme will keep me away from making the investment. Even the sales rep at Cambridge Soundworks said forget it.

I am able to manage the bass through AD conversion via the TACT by designing specific target curves for the sub. It is completely transparent and makes an enormous difference, which again, leads me to believe it's more about the mastering.

 

[krabapple]

VERY TRUE

I considered getting into SACD DVD-A since I have a player capable of hi-res playback. But managing the analog because of the stupid copy protection scheme will keep me away from making the investment. Even the sales rep at Cambridge Soundworks said forget it.

There are a few setups taht allow digital passage of SACD and DVD-A --either using ilink (IEEE1394) , or a proprietary link (e.g Denonlink). These rationalize bass management,DSP etc by letting the AVR do it all; the player is just a transport,as far as audio is concerned. I managed to put one such system (AVR + player) together for ~$1600. There's a $150 new Oppo uni-player that reportedly allows both formats to pass over HDMI (whereas previosly HDMI lacked SACD support) but the Audioholics review indicates a problem with the HDMI audio.

 

[Jay_WJ]

Logic also suggests you need to factor in biases associated with sighted comparison into your 'observations'. That itself is enough to explain the consistency. I'ts nothing in the least bit controversial; such psychological effects have been known about for years. It's why you have to do such comparisons 'blind', as one of the controls, to make it fair.

Are you beginnning to see how many issues are involved in what you probably thought was a straightforward observation, that 'HDCDs sounds better than CDs'?

1) Your perception -- and mine, and everyone's -- is inhererntly fraught with sources of error. On its own, without careful comparison setup, it's of limited reliability.

2) No, I don't think every single non-HDCD CD (they are ALL redbook) is poorly remastered, without exception.

3) Your conclusion has leapt ahead of your evidence. If you consistenly preferred the HDCD in a *double-blind or ABX comparison* of the same music, with the *same* mastering except for HDCD encoding, output at the *same* level to within ~ 0.5 dB -- *THEN* you'd have a solid case that you preference is due to the HDCD process and not some other factor. Good luck setting -that- test up.

An AB blind test? Every methodology in science is domain-specific. The method you're supporting is only one among many possible ones. I see that many people in audio hobby community simply accept well-controlled AB blind test as one and only best comparison testing tool. I'm afraid I don't. I don't think AB blind test result tells everything. An AB blind test in a controlled setting involves only a form of perception, I would say, an 'AB blind test' kind of perception. I believe that this kind of test should involve comparison relying on only short-term or registry memory which is heavily supported by bottom-up processing but minimally by top-down processing. Based on my experience in audio hobby, I came to believe that there should exist another kind of perception that effectively utilizes top-down, long-term memory support which has been accumulated by many, many perception trials. I don't think this form of perception is inferior to the 'AB blind test' type of perception but I believe that it could possibly be more sensitive and precise. The fact that it is not easy and simple to operationalize and implement such an experimental tool that can make this kind of perception be involved in a controlled manner, doesn't negate the plausible existence of this type of perception of sound quality.

Please don't try any more to convince me. Please don't try to fool me. You think that I say this because I don't understand your explanations? I UNDERSTAND what you're talking about. But it simply doesn't work for me.

 

[shokhead]

There are a few setups taht allow digital passage of SACD and DVD-A --either using ilink (IEEE1394) , or a proprietary link (e.g Denonlink). These rationalize bass management,DSP etc by letting the AVR do it all; the player is just a transport,as far as audio is concerned. I managed to put one such system (AVR + player) together for ~$1600. There's a $150 new Oppo uni-player that reportedly allows both formats to pass over HDMI (whereas previosly HDMI lacked SACD support) but the Audioholics review indicates a problem with the HDMI audio.

So when you use IEEE1394 or a denonlink between a denon receiver and player,the receiver will do the bm for dvd-a and sacd?

 

[Jay_WJ]

You can define sound quality, if what you mean is target-able and quantifiable measure of response characteristics that are preferred by the overwhelming majority of test subjects. Floyd Toole has [1]demonstrated that non-hearing impaired test subjects(large numbers of test subjects) in control listening tests prefer very similar measuring loudspeakers, and rate them the same sound quality, with remarkable consistency.

I don't deny that scientific efforts for defining sound qualities that are favored by people are possible. But how far such efforts can go is a different matter. Probably you are an engineer. It may be possible that engineers develop some useful (yet insufficient) criteria for such a definition by engineering standards. But I doubt that it is currently possible by scientific standards. You probably don't understand how complex the human perception is. As a cognitive scientist, I know that we scientists are still struggling to understand how it really works. Cognitve science in most domains is still at a primitive stage. For example, we don't have a clue to explain how humans can perceive different colors so correctly and consistently in various situations. We are not even able to rule out the possibility that color perception is a totally subjective phenomenon! Perhaps we humans are collectively subjective and happen to be consistent with each other in perceiving colors? We don't know yet.

You see my point? This matter is not as simple as you could imagine. Citation of experimental work? As an experimentalist, I always see that one's experimental results are refuted by other's by changing design, methods, and experimental conditions that are not essential to key assumptions. For phenomena that involve human perception or cognition, there exist only very few scientific findings that have been established by nonarguably consistent empirical evidence in general settings, like ones in physics or biology. So don't be so confident that you know one. We have a long way to go.

Anyway, I really enjoy different views and opinions shared here. I respect and value not only comments by people of intellectual style (like you ) but also opinions by pure audio hobbists based on their long-time experiences. I also enjoy the fact that I can sometimes observe some consistent patterns in this jumble of information. That's why I think a forum like this is useful to do my audio hobby. I do believe and actully use such EXTRACTED information I developed via forum research in my investment in audio.

 

[Jay_WJ]

So when you use IEEE1394 or a denonlink between a denon receiver and player,the receiver will do the bm for dvd-a and sacd?

Yes, it does.

 

[WmAx]

Citation of experimental work? As an experimentalist, I always see that one's experimental results are refuted by other's by changing design, methods, and experimental conditions that are not essential to key assumptions. For phenomena that involve human perception or cognition, there exist only very few scientific findings that have been established by nonarguably consistent empirical evidence in general settings, like ones in physics or biology. So don't be so confident that you know one. We have a long way to go.

It would appear that you are making assumptions of the validity of work that you do not seem to be familiar with. The perceptual work by Toole is peer-reviewed by members of the foremost international audio engineering journal, and his work[on loudspeaker quantitative preferences] has not been successfully refuted with any credible reasoning. The same goes for the original perceptual research performed when CD audio was originally being developed. What is the probable reason to assume the work is flawed at this point? Without sufficient credible evidence to demonstrate that it is probably wrong, how can one [reasonably] assume differently with confidence?

Your reasoning so far, to me, seems to be one of theorized doubt, not one based on probable circumstances that would invalidate the current standing/tenative research.

You want to bring up A/B/X style blind testing, and explain how it is not sufficient. Please offer a better solution, and explain the actual probabilities of current standard being inaccurate, and then relate these issues to actual listening conditions.

A/B/X style listening tests are probably too sensitive, from my observation(s). I can detect differences in A/B/X that were not discernible in normal listening. From this perspective, it appears to be excellent for threshold determination. In addition, I have perceived differences in normal listening that disappeared in A/B/X testing. Seems like purely psychological effect I was perceiving at this point. What do you suggest?

You started this off offering your views based on purely unscientific/uncontrolled listening tests as your evidence. Please review your standards before stating that there is an issue to begin with.

-Chris

 

[Jay_WJ]

It would appear that you are making assumptions of the validity of work that you do not seem to be familiar with. The perceptual work by Toole is peer-reviewed by members of the foremost international audio engineering journal, and his work[on loudspeaker quantitative preferences] has not been successfully refuted with any credible reasoning. The same goes for the original perceptual research performed when CD audio was originally being developed. What is the probable reason to assume the work is flawed at this point? Without sufficient credible evidence to demonstrate that it is probably wrong, how can one [reasonably] assume differently with confidence?

Your reasoning so far, to me, seems to be one of theorized doubt, not one based on probable circumstances that would invalidate the current standing/tenative research.

You want to bring up A/B/X style blind testing, and explain how it is not sufficient. Please offer a better solution, and explain the actual probabilities of current standard being inaccurate, and then relate these issues to actual listening conditions.

A/B/X style listening tests are probably too sensitive, from my observation(s). I can detect differences in A/B/X that were not discernible in normal listening. From this perspective, it appears to be excellent for threshold determination. In addition, I have perceived differences in normal listening that disappeared in A/B/X testing. Seems like purely psychological effect I was perceiving at this point. What do you suggest?

You started this off offering your views based on purely unscientific/uncontrolled listening tests as your evidence. Please review your standards before stating that there is an issue to begin with.

-Chris

You don't see my point. You're still on your own ground. My guess was correct that you're an engineer. Please understand that we have different views. Read my reply again carefully. I didn't doubt that your cited article was a peer-reviewed one. The article you mentioned is an engineering article. It shouldn't address questions regarding human auditory perception that should be involved in the task by standards whereby we cognitive scientists approach problems. It should only answer some engineering-purposed questions, but not scientific questions which should concern human auditory perceptual functioning. Your engineering-oriented mind may not understand the depth of questions I have in mind. Yes, I didn't offer any scientific/controlled listenging test results. But as I said, the reason for that is that I KNOW how complicated the problem IS and can't or don't even dare to do so! I respect your scientific attitude. But don't try to blind people with science. I'm glad that I'm not among those.

A reason why A/B/X testing tool is used is that it can easily provide a controlled condition. I strongly believe that this type of test addresses only one type of perception defined and limited by it. The fact that it is difficult to eliminate confounding factors in testing other types of perception doesn't mean that they don't exist. Please read my another reply to krabapple. If you don't have a sufficient insight for understanding what I stated there, it's alright. I don't have time and duty to convince you. Let's simply think that we are different.

 

[Rock&Roll Ninja]

Why SACD is going nowhere fast....

The Beatles - No albums
The Doors - No Albums
Black Sabbath - No Albums
Led Zeppelin - No Albums
Pink Floyd - Very Limited Catalog
David Bowie - Limited Catalog
U2 - Very Limited Catalog

Sure they have Bob Dylan and The Rolling Stones, but 2 artists do not a format make.

 

[krabapple]

An AB blind test? Every methodology in science is domain-specific. The method you're supporting is only one among many possible ones. I see that many people in audio hobby community simply accept well-controlled AB blind test as one and only best comparison testing tool. I'm afraid I don't.

Well, in the scientific 'domain' of psychoacoustics, dbt is the gold standard.

I don't think AB blind test result tells everything. An AB blind test in a controlled setting involves only a form of perception, I would say, an 'AB blind test' kind of perception. I believe that this kind of test should involve comparison relying on only short-term or registry memory which is heavily supported by bottom-up processing but minimally by top-down processing. Based on my experience in audio hobby, I came to believe that there should exist another kind of perception that effectively utilizes top-down, long-term memory support which has been accumulated by many, many perception trials.

DBT does not require short-term modes. How about listenign to two things long enough until your are quite sure you can tell the difference between them -- and then trying a DBT switching comparison? Are you suggesting that the long-term memory would then suddenly disappear?

I don't think this form of perception is inferior to the 'AB blind test' type of perception but I believe that it could possibly be more sensitive and precise. The fact that it is not easy and simple to operationalize and implement such an experimental tool that can make this kind of perception be involved in a controlled manner, doesn't negate the plausible existence of this type of perception of sound quality.

It isn't the 'long term' aspect that's questionable -- it's the *sighted* aspect, which is *demonstrably* fraught with error.

Please don't try any more to convince me. Please don't try to fool me. You think that I say this because I don't understand your explanations? I UNDERSTAND what you're talking about. But it simply doesn't work for me.

From a scientific POV you're in deep denial. So be it.

 

[krabapple]

It would appear that you are making assumptions of the validity of work that you do not seem to be familiar with. The perceptual work by Toole is peer-reviewed by members of the foremost international audio engineering journal, and his work[on loudspeaker quantitative preferences] has not been successfully refuted with any credible reasoning. The same goes for the original perceptual research performed when CD audio was originally being developed. What is the probable reason to assume the work is flawed at this point? Without sufficient credible evidence to demonstrate that it is probably wrong, how can one [reasonably] assume differently with confidence?

He seems not to be aware of one of the most interesting aspects of Sean Olive's findings about preference for speaker sound: that when tested under blind conditions, the preferences tend to be for those loudspeakers which *measured* best on criteria that Toole et al predicted could be important.

 

[mtrycrafts]

I am a cognitive scientist. Human perception in most domains (especailly visual and auditory) is a very complex function and is still a mystery. For example, despite a lot of research efforts, the mechanism of color perception is still unresolved---visual perception is relatively much more studied than auditory perception.

You CANNOT objectively define sound quality. And to MY ears, a format matters.

We are not talking about mechanisms you seem to be hung up on. We are talking about the ability to differentiate two components in a fair, unbiased test, period. It doesn't matter how you perceive things if you cannot differentiate two components, level matched and bias controlled.
You, on the other hand believe because perhaps, audio is you hobby.
Biased perception is just that, biased and means nothing to others as to what reality is.

As to color, I am sure we can test what people are capable of detecting, differentiate and what the thresholds are for detection. Not knowing the mechanisms for its perception is really irrelevant to threshold detections, color detections, etc.

 

[mtrycrafts]

How do you explain my consistent experience?

Perhaps biased? Perhaps unfair comparisons of unequal quantity? Lots of reasons may be found. Levels mismatched?

And logic plus empirical observations (though not by scientific standard) led to my conclusion.

Oh, but perceptions can be biased and unreliable, no? after all, it is not a consistent yardstick, is it?

Hardware and software manufacturers of SACD, DVD-A, and HDCD may use physical data to say that their technology improves SQ. But for the same reason I stated before, I don't believe their claims until I really perceive the difference.

But, how do you check this, that you in fact really perceive a difference and not just imagined one? After all, humans are known to perceive a difference when the same component is presented twice but they just don't know that tiny bit of critical info. I am sure you know this trait in humans?

Uncontrolled comparisons, sighted listening, unmatched levels are useless in drawing useful conclusions.

 

[mtrycrafts]

Based on my experience in audio hobby, I came to believe that there should exist another kind of perception that effectively utilizes top-down, long-term memory support which has been accumulated by many, many perception trials. I don't think this form of perception is inferior to the 'AB blind test' type of perception but I believe that it could possibly be more sensitive and precise. The fact that it is not easy and simple to operationalize and implement such an experimental tool that can make this kind of perception be involved in a controlled manner, doesn't negate the plausible existence of this type of perception of sound quality.
.

Maybe you have something. One way to find out, experiment, and publish in a peer paper. Until then, it is buy an exercise.

Oh, by the way, why can't you do a long term blind A/B comparison? Dan Shanefield did and wrote about it???
Shanefield, Daniel, " The Great Ego Crunchers: Equalized, Double Blind Testing,: Hi-Fidelity, Mar 80, pg 57-61.

 

[shokhead]

How do you explain my consistent experience?

Perhaps biased? Perhaps unfair comparisons of unequal quantity? Lots of reasons may be found. Levels mismatched?

And logic plus empirical observations (though not by scientific standard) led to my conclusion.

Oh, but perceptions can be biased and unreliable, no? after all, it is not a consistent yardstick, is it?

Hardware and software manufacturers of SACD, DVD-A, and HDCD may use physical data to say that their technology improves SQ. But for the same reason I stated before, I don't believe their claims until I really perceive the difference.

But, how do you check this, that you in fact really perceive a difference and not just imagined one? After all, humans are known to perceive a difference when the same component is presented twice but they just don't know that tiny bit of critical info. I am sure you know this trait in humans?

Uncontrolled comparisons, sighted listening, unmatched levels are useless in drawing useful conclusions.

DVD-A and SACD are in a section by themselfs and HDCD shouldnt even be in the same sentence,imo of course.

 

[Jay_WJ]

Okay, I realized that you guys are right and I am wrong. Then, let's talk about somewhat different matters.

Say that there is no conclusive evidence yet that a format (CD, HDCD, SACD, DVD-A) matters. In this situation, we can have three different personal positions:

a) I don't know because there is no evidence yet (analogous to agnostics).
b) I believe that a format does not matter (athiest).
c) I believe that a format matters (believer).

Apparently, you guys' choice is a) and mine is c). Right?

Then suppose that there are two players both of which are of very good quality in every way but one of which can play every format but the other can play only CDs. The former is a little more expensive than the latter, of course. We have to make a buying decision on either of these.

In reality, either that a format matters or that a format does not matter are possible, right?---let's make it simple (no in-between situation). With no conclusive evidence available, this is now a probablistic matter. In this situation, which player will you choose personally?

I, for one, will choose the player that plays all. How about you? And why? Because there is no evidence yet, you will not prabably use your personal experience with these formats for this decision making. Right?

 

[krabapple]

Sorry, this choice isn't the 'gotcha' you seem to believe it is.

I choose the 'universal player' because the *mastering* is likely to be audibly different on different releases (or even on CD vs DSD layer of an SACD), and thus there's a chance it's *better* on one than the other (e.g., SACD mastering spec does not allow digital clipping, and thus the large amounts of dynamic range limitation that often is applied to modern CD remasters BY CHOICE, not necessity, of the producer/engineer, is less likely to occur on a DSD remaster - unless the album is tinkered with in the PCM domain first!)

I need the 'uni player' to be able to compare *remastering choices*. And of course to play surround mixes, on an SACD.

 

[Jay_WJ]

Sorry, this choice isn't the 'gotcha' you seem to believe it is.

Come on. I just asked for your personal opinion and reasoning. I had no other intention.

 

[Totoro]

There are a few setups taht allow digital passage of SACD and DVD-A --either using ilink (IEEE1394) , or a proprietary link (e.g Denonlink). These rationalize bass management,DSP etc by letting the AVR do it all; the player is just a transport,as far as audio is concerned. I managed to put one such system (AVR + player) together for ~$1600. There's a $150 new Oppo uni-player that reportedly allows both formats to pass over HDMI (whereas previosly HDMI lacked SACD support) but the Audioholics review indicates a problem with the HDMI audio.

That assumes specialized technology that is not standard nor available on a wide variety of players like TOSlink or standard digital out. Since this thread is why hi-res formats are not succeeding, this only reinforces the point. Copy protection schemes invariably reduce the marketability of products. You would think companies would have learned these lessons based on past experience with proprietary video and computing formats. They are worried about people making illegal copies of their music, but by limiting the usefulness of their products, they are also preventing people from BUYING their music as well. Another case where lawyers and IP specialists have killed innovation.

Also, there is no reason why hi-res formats should cost any more than regular CDs. Printing costs are almost exactly the same. It only discourages people from buying their products.