I'll deal with all of your comments here...
As for references, I did a quick search on the web, but didn't find anything. The test I read about was in an issue of Sympatico Netlife magazine, can't remember which one.
But if I'm dealing with an instrument that is sensitive enough to detect something smaller than 5 feet, then yes, I do need that extra quantization accuracy, regardless of how impressive the comparison to the distance to the sun is.
Don't forget that it's not just your ears that matter here. The CD/DVD player's analog electronics, the amplifier, and your speakers/headphones must all be up to the task as well.
The point is not whether you can come up with an analogy that has a big visceral impact, it's whether or not the human ear is sensitive enough to hear the difference. And it doesn't matter if some or even most people fail to notice the difference some or most of the time. The accuracy of the quantization should be so high that nobody can hear the difference under any circumstances. Why settle for less?
Heck, then, lets go for 64-bit, 512KHz digital audio :)
One thing I've never understood is the utter contempt some people seem to have for those who don't like digital. I'm not saying you're one of those people, but I do sense a bit uppitiness.
I'm certainly not against digital. What I am against is throwing a lot of bits (about 3x as many) to solve what, even at worst, is a very minor problem.
Technically literate people who try to remain unbiased will usually admit that complaints about CDs have a basis in scientific fact, and are not just some crackpot fantasy.
Interestingly enough, the problem mentioned by your link would tend to have the biggest effect on classical music. But it isn't usually the classical music people that I hear complaining about CD audio quality. :)
There is also the "headroom" issue that I brough up earlier. If the system has just barely enough bits to cover the limits of human perception, then we become too dependent on people mastering the recording to maximize the potential of the medium. Making sure that there is actually room for error is a good thing. 24bit/96Khz may be a bit of overkill (then again, it may not; I'm not convinced that the details and limits of human auditory perception are fully understood), but if it will help reduce the number of recordings coming out that fail to maximize the potential of today's audio systems, what's the harm in that?
It certainly makes sense to make the master recording at 96k/24, and, AFAIK, that is how it is done today. But the extra headroom is only really useful in preproduction. Once the recording has been produced, there's no real advantage in it -- indeed, there's often too much as is, which sometimes necessitates fiddling with the volume control to keep the sound at a comfortable level.
Perhaps DVD-A will catch on; it's too early to say for sure. If it does, I doubt it will be because of the increased audio quality -- most people will not have either the expensive equipment or the 'golden ears' necessary for them to percieve any difference at all, and even then, the difference will be slight.
Personally, though, I wouldn't object to being able to store 3.5 hours of CD-quality music on a DVD, though.
[ Parent ]