What is SACD?

[KevinP]

What kind of experience of multichannel sound do you have? Have you ever actually heard SACD multichannel sound on a proper set-up? You are just conservative and you want to stay with stereo sound. Nothing wrong with that except you should know you are being conservative. Just don't judge things you don't understand/know enough about.

Holy cow. I can't believe you wrote that for nothing, accusing me of judging what I won't understand without any basis. Of course I have a multi-channel setup. I love 5.1, just not for ambiance alone. I am not conservative nor am I judging something I don't understand.

Even mono and stereo adds 'competing ambiances', just in flat style. You have the acoustics on the disc. Monophonic recording of organ music has long church reverbation that will add to the acoustics of your room. The placement and directivity of your speakers will determine how much direct sound compared to reverberation of your room you will hear.

Yes, but there isn't a channel or two dedicated to ambiance as there is with some surround mixes. And as I said, my biggest problem is when they add artificial reverb rather than recording the actual ambiance, which is sometimes done when they want to make an an old stereo recording multi-channel.

Multichannel sound is not about "adding echos". It is about giving a stronger feel of the original acoustics. It's no different from the difference between mono and stereo.

Why the quotation marks? Who are you quoting? I hope you don't think it's me.

Yeah, some SACDs are certainly badly done just as some stereo recordings (early ping-pong stereo!) Most SACDs are however well (in many cases superbly) done and there is nothing artificial about them.

Hello, preacher.
Signed,
The choir.

You are right, multichannel is not required but it helps a lot. Black and White movies show colours are not required in films but isn't it nice to have movies with colours?

I like good surrround mixes. I like good surround mixes. I like good surround mixes. I like good surround mixes. I like good surround mixes.

Calling multichannel better doesn't mean stereo is bad. Stereo does amazingly good job! I don't have a problem with people choosing stereo, I have a problem with ignorance. People judge multichannel without having a clue what it is about.

I like good surround mixes. I like good stereo mixes. I like good mono recordings. I have a clue!

[Daverz]

I went googling for info on double blind tests of CD players and didn't find much.  There was an old ABX test of a whopping total of 3 old CD players which did find differences in one case that wasn't very surprising.

I do hear very definite differences between CD players.  A slightly recessed midrange on one or slightly tipped up lower treble on another.  Slightly tighter or warmer bass.  More or less sibilance.  Some seem to be better at spreading the soundstage between the speakers and for others the soundstage doesn't seem to cohere very well.  That sort of thing.  It doesn't necessarily correlate with price or build quality.   Usually I'll listen for a few hours to one player and then put in my old player for contrast.  Still, I'm open to the idea that I could be fooling myself.  I'd love to be able to reach audio nirvana with a $170 Oppo vs. a $2500 player (I may just get the Oppo to try it out; I spend more than that on groceries every week).

Also, in my experience, switching between 30 second snippets isn't adequate for testing.  I think the brain adjusts itself to a different sound, say over the course of an hour or so, so that changes are more obvious when switching after long periods.

[Dancing Divertimentian]

Yes, I agree. So, why do we have colour movies? Why did they bother develop color film?
Why did they develop stereo sound? Extraordinary things can be achieved in monophonic sound.
Now, why did they develop multichannel audio? Are you seeing the pattern?

You've completely missed my point. No matter what the medium - black & white or color - artistic 'perfection' can still be achieved. It all depends on the creative hand at work. Black & white cinema along with black & white photography to this day inspire a great many artists to achieve great things.

This is a lineal progression. No real improvements per se beyond the obvious technical advances but artistically things haven't changed much.

As far as classical music reproduction, sure there have been improvements (stereo is better than mono) but at some point the law of diminishing returns comes into play. I'm not about to spend an obscene amount of money on surround sound for only a tiny incremental improvement in sound. In reality it isn't so much an improvement as it is a way to part me from my money for nothing really beneficial.   

Like I said, for movies surround sound might be preferable but for classical music a good two-channel system is hard to beat. At least for me.

[Senta]

I don't have a surround setup but still enjoy my SACD player a lot. To me there is a noticeable difference in that the sound is more warm and natural in quality, which I suppose has to do with SACD containing more bits of information per second.

My player only cost me $150, and it's a Sony, it also plays DVDs too and has HDMI capability which I haven't used yet. It sounds great and I love it, in fact it was recommended to me by a friend who is an audiophile. I don't think you have to play tons to have good sound. 

[Al Moritz]

I don't have a surround setup but still enjoy my SACD player a lot. To me there is a noticeable difference in that the sound is more warm and natural in quality, which I suppose has to do with SACD containing more bits of information per second.

My player only cost me $150, and it's a Sony, it also plays DVDs too and has HDMI capability which I haven't used yet. It sounds great and I love it, in fact it was recommended to me by a friend who is an audiophile. I don't think you have to play tons to have good sound. 

In the $ 150 price range, a not so surprising result for SACD vs. CD, in view of what was discussed here.

[Al Moritz]

         Why don't you give up the idea that you're protesting methods and just come out and say you don't like the result? I'll admit my bias. I'm glad that bits is bits and volts are volts.

Well, you can discuss the merits of listening tests whatever you want, but it is not simply that "bits is bits and volts are volts". As I pointed out before:

²⁾ there are those that claim that any CD player is technically good enough but this is not true. Deficiencies in CD players can be easily tracked by, for example, measuring their pulse behaviour and linearity (no "non-scientific" listening tests required, even though these reveal differences between CD players as well). Crude standard measurements, like frequency characteristics and dynamic range, do not tell the whole story. They may be "science", but definitely science that is not good enough.

In that sense, listening just confirms what measurements show. Thus, to claim that a $ 100 CD player should sound just as good as a $ 1000 player is laughable and foolish.

[Iconito]

[...] to claim that a $ 100 CD player should sound just as good as a $ 1000 player is laughable and foolish.

Yup. And no doubt the one who laughs louder than anyone else is the one who sells the $1,000 CD player...

[...] To me there is a noticeable difference in that the sound is more warm and natural in quality, which I suppose has to do with SACD containing more bits of information per second. [...]

OK, everybody: Once more, Da Capo!   

[Daverz]

Yup. And no doubt the one who laughs louder than anyone else is the one who sells the $1,000 CD player...

Building a player for a $100 price point does involve making a lot of compromises in the quality of parts used.

[DavidW]

Al can you make a case for those deficiencies that you speak of being (a) audible, and (b) occur often enough to be important?

This sounds like making a mountain out of a molehill.  I'm sure an audible thinks it's very important, but what actual impact does pulse behavior and linearity have on everyday listening?  Is this one of those silly, dumb jitter things where you claim to hear differences in interpolation between the 1/44.1 kHz bins?  Just to put that in actual numbers that's like saying that you can hear the difference in sounds that have a duration of only 10^(-5) s.  That really does not make one bit of sense to me.

[Al Moritz]

David,

Al can you make a case for those deficiencies that you speak of being (a) audible, and (b) occur often enough to be important?

This sounds like making a mountain out of a molehill.  I'm sure an audible thinks it's very important, but what actual impact does pulse behavior and linearity have on everyday listening?  Is this one of those silly, dumb jitter things where you claim to hear differences in interpolation between the 1/44.1 kHz bins?  Just to put that in actual numbers that's like saying that you can hear the difference in sounds that have a duration of only 10^(-5) s.  That really does not make one bit of sense to me.

10E-5 seconds (ten microseconds) is a significant time in music. A short musical signal lasts no more than fractions of a tenth of second, so we are getting down to the 10E-2 second range. Don't tell me that differences in thousands of such a signal cannot be heard.

For pulse response, see below links:

http://www.digitalaudio.dk/technical_papers/axion/dxd%20Resolution%20v3.5.pdf
(Figure 1)

http://www.tnt-audio.com/clinica/convertus2_e.html
(Figure 9; from the text: ""What's worse, the pulse response of a digital filter, and hence of a normal CD player, has a deeply non-natural behaviour, that is it has a ripple that begins before the real pulse, which is really hard to find in any real word musical instrument behaviour."

http://www.headwize.com/projects/showfile.php?file=meier4_prj.htm

Obviously, this is a serious issue. Music is not a continuous signal, and pulse behavior shows that the timing of the music signal is affected (again, I do think in those time frames we are dealing with something substantial). Crude standard measurements like frequency response and dynamic range work with continuous signals. While these are important measurements too, I can safely assume that you will agree that trying to deduce behavior from a continuous signal to something as complex and everchanging as music is bad science, or at least, incomplete science. I find it ludicrous when some people try to sell simplistic models as "scientific reality" (in the field of audio this happens all the time, and there are enough gullible "objectivists" who buy into this). Certainly, science needs to work with simple models at times, but good scientists never mistake their extrapolations from these simple models for reality itself -- they are constantly aware of the limitations of their extrapolations, even when they know that they can obtain partial information about reality this way.

While it is obvious that some audiophiles believe in wizardry and audio voodoo and things that don't really exist, simplistic "measurement freaks" miss reality just the same. We also should not forget that when it comes to music, in principle our ears are much better instruments to measure than standard audio measuring instruments and methods -- even when in practice they sometimes betray us and subjectivity runs amok. As so often, reality lies in the middle between extremes, in this case the extreme subjectivist and the extreme objectivist stance.


Linearity:
http://zone.ni.com/devzone/cda/tut/p/id/4472

Inaccuracies tend to occur especially at weak signals; non-linear digital-analog conversion will, at the minimum, lead to loss of musical detail in the reproduction.

[Al Moritz]

David,

as for your question if the deficiencies occur often enough to be important: Yes, obviously they occur all the time, given the complexity that music as a signal represents.

[DavidW]

David,

10E-5 seconds (ten microseconds) is a significant time in music. A short musical signal lasts no more than fractions of a tenth of second, so we are getting down to the 10E-2 second range. Don't tell me that differences in thousands of such a signal cannot be heard.

I'm sorry but I placed the burden on your shoulders, putting it back onto mine is not really an adequate response.  By a "short musical signal" do you mean a note?  What do you mean exactly?  And why do you think it's reasonable to resolve 1 part in a 1000 of it?

http://www.tnt-audio.com/clinica/convertus2_e.html
(Figure 9; from the text: ""What's worse, the pulse response of a digital filter, and hence of a normal CD player, has a deeply non-natural behaviour, that is it has a ripple that begins before the real pulse, which is really hard to find in any real word musical instrument behaviour."

The plot is deceitful, there is no horizontal axis and hence no scale.  If you look elsewhere in the article the author says that the differences are on the scale of picoseconds!  Picoseconds!  And not once but several times in several ways he suggested that those are the durations for which there are discrepancies.

I looked at the other articles, but it would be nice if there was something that concisely argues for the relevance of these issues, and those are articles are not primarily about that it seems.  I'll look at them again later though.

[71 dB]

Yes, but there isn't a channel or two dedicated to ambiance as there is with some surround mixes. And as I said, my biggest problem is when they add artificial reverb rather than recording the actual ambiance, which is sometimes done when they want to make an an old stereo recording multi-channel.

Making a multichannel SACD out of and old stereo recording makes no sense. How could it ever work? Why buy old stereo recordings in SACD format? The CD "only" version is cheaper and sound as good as it can. When I say that multichannel sound is the real asset of SACD I mean those recordings that have been made multichannel in mind (=real multichannel recordings).

Why do people talk about crappy recordings when the discussion is on audio formats? No format do well when used badly.

[Al Moritz]

I'm sorry but I placed the burden on your shoulders, putting it back onto mine is not really an adequate response.  By a "short musical signal" do you mean a note?  What do you mean exactly?  And why do you think it's reasonable to resolve 1 part in a 1000 of it?

For example the attack phase of a note on the piano. Why don't you think it's reasonable to resolve 1 part in a 1000 of it? You can measure it, the burden is on you to prove that you cannot hear it. With something as complex as music (brutally complex compared to a simple sine wave signal), I would be very careful to argue what is audible and what is not.

The plot is deceitful, there is no horizontal axis and hence no scale.  If you look elsewhere in the article the author says that the differences are on the scale of picoseconds!  Picoseconds!  And not once but several times in several ways he suggested that those are the durations for which there are discrepancies.

I looked at the other articles, but it would be nice if there was something that concisely argues for the relevance of these issues, and those are articles are not primarily about that it seems.  I'll look at them again later though.

The other articles show scales, and they are in 10-100 microseconds.

[Herman]

Making a multichannel SACD out of and old stereo recording makes no sense. How could it ever work? Why buy old stereo recordings in SACD format? The CD "only" version is cheaper and sound as good as it can. When I say that multichannel sound is the real asset of SACD I mean those recordings that have been made multichannel in mind (=real multichannel recordings).

Why do people talk about crappy recordings when the discussion is on audio formats? No format do well when used badly.

Hate to tell you, but a lot of 1960's recordings sound a lot better than digital recordings, and I'm not even mentioning multichannel crap.

I know this doesn't accord with your "every technical revolution is better" theory, but if you'd just listen to a Artur Rubinstein RCA recording  -  it's very hard to beat.

A lot of people prefer analogue to digital  -  how's that for progress?

[drogulus]

Also, in my experience, switching between 30 second snippets isn't adequate for testing.  I think the brain adjusts itself to a different sound, say over the course of an hour or so, so that changes are more obvious when switching after long periods.

      You're absolutely right. The brain does adjust and that's why the test should be short. Remember, "adequate for testing" doesn't mean adequate to achieve a particular result, like hearing a difference. And the brains ability to adjust will tend to obscure rather than clarify. That adjustment is what the test must eliminate, since the brains ability to adjust can create or eliminate the very differences we're trying to find.

   

[drogulus]

Hate to tell you, but a lot of 1960's recordings sound a lot better than digital recordings, and I'm not even mentioning multichannel crap.

I know this doesn't accord with your "every technical revolution is better" theory, but if you'd just listen to a Artur Rubinstein RCA recording  -  it's very hard to beat.

A lot of people prefer analogue to digital  -  how's that for progress?

      Who cares what people prefer? If the subject is the audible difference, if any, between 2 formats (rather than how they are most often used) then preferences mean nothing. We're trying to get beyond preference to what people can actually hear. And every technical revolution does not have to be an audible improvement, though they frequently are. In the case of digital (CD) vs. analog (LP) it is an audible improvement, whatever preferences may say to the contrary. In the case of CD vs. stereo SACD the tests indicate no detectable difference under test conditions, so no improvement.

     Technical improvements you can't hear don't count. This is the governing principle behind the mp3 revolution as well, that undetectable differences don't matter, and I note with a certain amount of glee that the intuitionists have exactly the same type of objections. Compressed audio is bad in principle. And of course that was the original objection to CD sound, that chopping audio into bits was unnatural. In each case the a priori objection trumps real world results for the audiophile.

     I have preferences, too, and want to treat them as self-validating facts. I try not to have them, but I'm human so I tend to think this is better than that, even if there's no good reason to think so. Sometimes, though, large amounts of money are at stake, and then I get serious. I recommend that everyone do likewise before they buy that $5,000 player, unless they just like buying pretty, expensive things.

[DavidW]

You can measure it, the burden is on you to prove that you cannot hear it.

You made a positive assertion, the burden is on you.  I don't have to prove that I have a good reason to doubt you.  You know very well that you made a boo-boo.  You know it's the same thing as the invisible pink unicorn existence assertion fallacy.  And you knew that you weren't going to fool me.  Why did you waste both of our time even going there?

With something as complex as music (brutally complex compared to a simple sine wave signal), I would be very careful to argue what is audible and what is not.

So what?  If you can't hear the difference between two sine waves, what chance would you possibly have of picking up differences in a complex signal when you can't get it in the easiest case?  It seems like what you're saying doesn't support your case at all.

[drogulus]

     

invisible pink unicorn existence assertion fallacy

      No, no, no.....you must prove they don't exist, otherwise I'll all have to live with this terrible doubt hanging over me. That could mess up my whole (after)life. 

[Al Moritz]

You know very well that you made a boo-boo. 

No I don't.

You know it's the same thing as the invisible pink unicorn existence assertion fallacy.

That is new to me. I don't see why it would have anything to do with that. I don't believe in the pink unicorn anyway.

And you knew that you weren't going to fool me.

I didn't intend to, so yes, I didn't know.

Why did you waste both of our time even going there?

Well, I now see that it is a waste of time indeed. Especially in light of the following which shows that you don't even try to understand my argument:

If you can't hear the difference between two sine waves, what chance would you possibly have of picking up differences in a complex signal when you can't get it in the easiest case? 

Huh? Conversation over. Have a nice night, David.

Al