Jump to content

CD Quality


mas

Recommended Posts

Oh, we've hit the motherload this month!

For those who are perenially bemoaning the state of CDs, here is a reference to an article that some just might want check out.

But why do I have the feeling that many won't let the test get in the way of their emotional opinions?

In fact, I will go so far as to wonder how many will actually even read the paper before commenting and correcting it!

post-23237-13819349026878_thumb.png

Link to comment
Share on other sites

You should try not to scratch your CD's... (sorry lame pun on pitiful)

I haven't read the article, but the spoiler implies that the A/D/A loop

was done in a little circuit that shared the same clock. When I was

first getting into digital in the studio, I wasn't syncing all of my

clocks between units - which actually became a problem. In the absolute

worst case scenario, one of my digital recorders got off by 3 seconds

after an hour of playback...both recorders were recording the exact

same thing so it was interesting to notice the ever so slight pitch

change between the two. I originally thought it was an issue with DAT

so did some computer based comparisons too - while the syncing didn't

get off by as much, there was still a difference between the two when

played back simultaneously. Getting a master clock in the studio and syncing everything up got rid of the problem.

However, in a consumer playback environment, there's no way to

guarantee that the clock rate is exactly identical. Ignoring possible

jitter issues, you're left with a signal that is ever so slightly

detuned...or at least in theory world. The differences shouldn't be

perceptible, but it seems that even with my crappy enough acoustic

memory that you can tell there is a difference when the original master

clock is being used versus letting the unit play on its own.

I dunno - my personal experiences indicate that 24 Bit 96kHz sounds better in the studio, but perhaps it's because of the rounding errors with DSP and all that. It might also have to do with the potential problems of aliasing introduce by noise in the signal path. A nice tight circuit like that is gonna have a really low noise floor.

I would like to read the whole article though. While I don't disagree with the conclusions of the specific situation, it doesn't seem like it can be applied to every situation. But that's not to say that it is something to get overly worked up about either. I think there are far bigger distortions being introduced than any digital anomalies.

Link to comment
Share on other sites

".....

But why do I have the feeling that many won't let the test get in the way of their emotional opinions?

In fact, I will go so far as to wonder how many will actually even read the paper before commenting and correcting it!

....."

This made the rounds in the AA forum. In answer to your questions, 1) the response was entirely emotional and 2) I suspect less than 10% actually read the document, but that did not get in their way of bashing the research.

Some years ago, a colleague of mine actually did some formal measures (others have also, but frequently the data are not public)) and found that a 14 bit recording/playback was pretty much perceptually equivalent (confusable) with a 16 bit version (in the vast majority of instances).

The caveat was that all 14 bits had to be used. The specifics about the test material, listener experience, set up, & testing protocol were all reasonably done (there were no apparent flaws and this guy was an experienced researcher). However if you mention this finding to "experts outside the field", they will give you a thousand reasons why this can not be true. On top of that they will then tell you all the ways the test was performed incorrectly (the specifics of which they were never told to begin with). I guess being an audiophile requires a very strong knowledge that you are right and everyone else is wrong.

Good Luck,

-Tom

Link to comment
Share on other sites

Hey Tom,

I don't disagree with the findings of your colleague, but I have a hard time believing it...doesn't that mean I disagree? I dunno, surely you can understand a little skepticism. Could you describe it any further or provide the document?

I find the limits of our hearing interesting, but at the same time I don't think we need to be engineering "just enough" solutions. If nothing else, going a little overboard is usually going to reduce mistakes. I guess what I'm saying is that it would probably be rather audible if everyone started going with 14-Bit DAC's...

At this point in the game, going lower isn't making it any cheaper either...

Link to comment
Share on other sites

Geesh!

Obviously NO ONE even tried to search for the article!

(Of course, if you did and found it, you were probably just sitting there quietly enjoying the complaining and lack of initiative!)

Here is some preliminary information.

I will post the article in full in a bit when I gain access to it in a postable form (either via scan or download).

And as providing you with the results wasn't enough, you/you all/yous guys (did I get all of you?) will just have to find someone else to read it for you too!

Sorry, the forum does not support the .mht extensions that allow me to post the document complete with embedded links.

If I get around to it, i will try to assemble the various links to the supporting resources.

post-23237-13819349037548_thumb.png

Link to comment
Share on other sites

Mike,


"I don't disagree with the findings of your colleague, but I have a hard time believing it...doesn't that mean I disagree?"

That isn't really that surprising. 14bit has 12dB lower SNR. Theoretically 84 dB instead of 96dB. In the grand scheme of things how many people have a system (including the room) with 84dB of resolution? Not many that listen at sane volume levels.

Look at vinyl... it has roughly 60dB of SNR. Most don't have an issue with the noise floor of that either.

Shawn


Link to comment
Share on other sites

Interesting that one of the listeners was able to detect the A/D/A chain when listening 10dB above their reference. At the very least I think that shows that we need to use all of the bits available if we're only going to go with 16-Bit. That kind of jives with my experiences in the studio where I'll keep the continuous level about 20dB to 30dB down so that I can capture transients without clipping (depending on the instrument of course). At that point, I think the analog noise of the system is getting down around the digital noise of the system so it should be more audible?

Link to comment
Share on other sites

Mike,

"I don't disagree with the findings of your colleague, but I have a hard time believing it...doesn't that mean I disagree?"

That isn't really that surprising. 14bit has 12dB lower SNR. Theoretically 84 dB instead of 96dB. In the grand scheme of things how many people have a system (including the room) with 84dB of resolution? Not many that listen at sane volume levels.

Look at vinyl... it has roughly 60dB of SNR. Most don't have an issue with the noise floor of that either.

Shawn

That pretty much hit it on the head. In either case these are incredibly low noise floors provided by the quantization (if all the bits are being used). It is quite unlikely that a home audio solid state amp could even keep up with that noise floor, let alone the noise from the neighbor's kids and the refrigerator compressor in the next room etc. The quantization noise is not the limiting factor here.

But Mike the other question you should have asked was about the modulation. Was it PCM or sigma delta?

Take care,

-Tom

Link to comment
Share on other sites

Oh, we've hit the motherload this month!

For those who are perenially bemoaning the state of CDs, here is a reference to an article that some just might want check out.

But why do I have the feeling that many won't let the test get in the way of their emotional opinions?

In fact, I will go so far as to wonder how many will actually even read the paper before commenting and correcting it!

Following up is this article from "Audio Critic"

http://theaudiocritic.com/blog/index.php?op=ViewArticle&articleId=41&blogId=1

The consensus is that muti-channel may be more esthetically pleasing but NOT more accurate in audible sound reproduction.

Link to comment
Share on other sites

"But why do I have the feeling that many won't let the test get in the way of their emotional opinions? "

I tend to find that if I can listen to a piece of music without getting emotionally involved then that is a piece of music I would not normally choose to listen to.

I have long suspected that the digital/analogue conversion is not the problem with CD's - although I could not hazard a guess as to what the problem actually is (for me of course). As others have mentioned this is not the first time I have heard of a analogue/digital/analogue conversion being done in the signal and there being no apparent degradation of quality.

Suffice to say - accepting that listening to music is an emotional experience I get a fuller version of it listening to vinyl than to CD. High Res digital formats have been hugely disappointing to me too which is largely what lead me to my conclusions in the first place.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...