Phil Leigh wrote: > In terms of AUDIO, 44.1khz sampling is all that is needed to capture the > "information". (refer Shannon et al - Information Theory). The key > assumption being that the Nyquist frequency of 44.1 divided by 2 is > adequate. If we only want frequencies up to a max of 22.050Khz then it > is. >
cliveb wrote: > The only information that 24/192 holds which is not present in 24/96 are > frequencies above 48kHz. > Perhaps that might be of significance to bats - I personally don't care, > since my hearing maxes out around the 16kHz mark. This looks like an old thread, however it turned up on a recent google search, as I was looking for the possibility of 24/192 playback from my Transporter. I thought I'd add some information here and clear up some misconceptions. Nyquist has really no place in any audiophile discussion. As you've stated Phil, the highest frequency that can be produced by a 44.1 khz sample rate, is 22.050 khz. No humans can hear above 22 khz anyhow, so isn't this "good enough"? Short answer is No. Here's why. If you plot out a 22 khz frequency at a 44.1 khz sample rate, result is not a wave - it's a sawtooth. No matter what frequency we're talking about, a sawtooth is a very poor approximation of a wave, and it sounds it. It is generally agreed that to have a decent digital approximation of an analog wave, you need 8 points, not 2. 44.1 khz divided by 8 equals 5.5125 khz. So with a 44.1 khz sample rate, anything above 5.5 khz is noticeably diminished in its accuracy. So Clive, you see now how I'm going to disagree with your statement. You're correct in that we don't care about a 48 khz frequency, as no human can hear it. However the benefits of 192 khz sample rate has nothing to do with the 48 khz theoretical limit. As I described above, 44.1 khz sample rate is highly accurate to only 5.5 khz. However 96 khz sample rate is highly accurate to 12 khz, and 192 khz sample rate is highly accurate to 24 khz. Therefore, frequencies that are in the audible range are considerably more accurate at these higher sampling rates, with 192 khz delivering highly accurate (8+ sample per wave) reproduction throughout the entire range of human hearing. As for the argument that there is no discernible difference between 96 khz and 192 khz, that's clearly false as I've just described, however there is a very real reason for why some would make this assertion. Our high frequency hearing diminishes with age. And lets face it, audiophiles are generally not known for being young sprightly spring chickens! While a teenager can hear an 18 khz tone, it's common for a 50 year old adult hear only up to 12 khz or so - hence the reason that even a discerning ear may not have the ability to distinguish between a 96 khz and a 192 khz recording. Cheers ------------------------------------------------------------------------ lupin..the..3rd's Profile: http://forums.slimdevices.com/member.php?userid=56447 View this thread: http://forums.slimdevices.com/showthread.php?t=69882 _______________________________________________ audiophiles mailing list audiophiles@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/audiophiles
