What's the use of a 192 kHz sample rate?
in [DSP]
|
Prev: Is 44.1 KHz sample-rate enough to cover the entire human hearingrange?
Next: Newbie to DSP - Phase shift detection
From: rajesh on 6 May 2008 00:53 On May 5, 10:33 pm, dpierce.cartchunk....(a)gmail.com wrote: > On May 5, 10:09 am, rajesh <getrajes...(a)gmail.com> wrote: > > > On May 5, 7:05 pm, Oli Charlesworth <ca...(a)olifilth.co.uk> wrote: > > > If we can, then of course a higher sampling rate will sound better. > > > But that goes against the premises of the OP, and is nothing to do > > > with the ECC or interpolation that you've been going on about! > > > > -- > > > Oli > > > I said we cant percieve, but i didnt say they arent there.. > > Again, true but irrelevant. > > > i will continue the dicussion on ECC tomorrow. > > Hopefully, you will be much better prepared. > > As a hint: the issue of proper sampling vs bandwidth > is a topic COMPLETELY separate from ECC. You > might want to keep that in mind during your preparations. take for example h.264 video Apart from having many sophisticated techniques it also recommends simple one like repeating packets.
From: rajesh on 6 May 2008 00:54 On May 6, 9:53 am, rajesh <getrajes...(a)gmail.com> wrote: > On May 5, 10:33 pm, dpierce.cartchunk....(a)gmail.com wrote: > > > > > On May 5, 10:09 am, rajesh <getrajes...(a)gmail.com> wrote: > > > > On May 5, 7:05 pm, Oli Charlesworth <ca...(a)olifilth.co.uk> wrote: > > > > If we can, then of course a higher sampling rate will sound better. > > > > But that goes against the premises of the OP, and is nothing to do > > > > with the ECC or interpolation that you've been going on about! > > > > > -- > > > > Oli > > > > I said we cant percieve, but i didnt say they arent there.. > > > Again, true but irrelevant. > > > > i will continue the dicussion on ECC tomorrow. > > > Hopefully, you will be much better prepared. > > > As a hint: the issue of proper sampling vs bandwidth > > is a topic COMPLETELY separate from ECC. You > > might want to keep that in mind during your preparations. > > take for example h.264 video > Apart from having many sophisticated techniques it also > recommends simple one like repeating packets. by sophisticated techniques i mean error ressilince tools
From: rajesh on 6 May 2008 00:57 On May 6, 9:53 am, rajesh <getrajes...(a)gmail.com> wrote: > On May 5, 10:33 pm, dpierce.cartchunk....(a)gmail.com wrote: > > > > > On May 5, 10:09 am, rajesh <getrajes...(a)gmail.com> wrote: > > > > On May 5, 7:05 pm, Oli Charlesworth <ca...(a)olifilth.co.uk> wrote: > > > > If we can, then of course a higher sampling rate will sound better. > > > > But that goes against the premises of the OP, and is nothing to do > > > > with the ECC or interpolation that you've been going on about! > > > > > -- > > > > Oli > > > > I said we cant percieve, but i didnt say they arent there.. > > > Again, true but irrelevant. > > > > i will continue the dicussion on ECC tomorrow. > > > Hopefully, you will be much better prepared. > > > As a hint: the issue of proper sampling vs bandwidth > > is a topic COMPLETELY separate from ECC. You > > might want to keep that in mind during your preparations. > > take for example h.264 video > Apart from having many sophisticated techniques it also > recommends simple one like repeating packets. Correction its not packets , its slices
From: Chronic Philharmonic on 6 May 2008 01:07 > This is not rocket science. The differences may not be audible to > you, but they exist and they are well understood. I'll give you a > hint; do you know what phase distortion is? The 20KHz low-pass FIR filters that they use in all modern CD/DVD players are phase linear. > Also, I would like you to explain what is wasteful about a 192 kHz > sample rate. How much money does it cost to use 192 kHz instead of 96 > kHz or 44.1 kHz. What time is involved? How much extra energy does > the higher sample rate use? Are you aware that there are virtually > infinite amounts of bandwidth not being used every second? What > difference does it make if DVDs underutilize a bit more? The only reason to use anything higher than 44.1KHz for consumer playback is marketing, and evidently, they have bandwidth (storage) to burn, so it doesn't cost them much. > I am working on a circuit that uses a 192 kHz CODEC to process 1 kHz > signals. Of course I am not using it at 192 kHz, so am I wasting > bandwidth still? Or by using an 8 kHz sample rate, am I conserving > bandwidth and deserve recognition? I like the idea of being "pink" (as > in noise) by conserving precious bandwidth. Actually, I can't wait to > get the thing out on the open road and open it up! I want to put the > pedal to the metal and sample at the full 192 kHz to see just what > sort of analog bandwidth these CODECs really have! I'm not actually > sure they will produce higher than about 40 kHz at the analog output. > I'll have to hack away the 1 pole filter I added to the input and > output. It limits the frequency range to a paltry 30 kHz or so. I > may not hear the difference, but my scope can "see" it! Okay, if you are still working with analog filters, then you need more room for the transition band. Digital FIR filters get you much closer to the theoretical limits much more easily and predictably. And they stay in spec with temperature and age. > BTW, if you aren't looking at your monitor or you aren't sitting close > enough to see every pixel, are you wasting bandwidth on your video > signal? Actually, that is an excellent analogy. If you limit the > video bandwidth to half the pixel rate, what difference will you see > on a CRT display? The difference here is, video bandwidth is chosen for a particular viewing distance, and if you are within (or nearer) the optimal distance, higher bandwidth is easily perceptible. Not so with excessive audio sample rates. > Finally, why do you see the advantage of using 96 kHz sample rate and > not 192 kHz? Isn't 96 kHz wasting bandwidth? Yes, very much so. I daresay you won't find a lot of 96 or 192KHz sample rates where bandwidth is at a premium (broadcast, satellite or streaming media). Quite the reverse; they'll be looking for ways of reducing the bit rate. But if you have bandwidth (storage) to burn, and a lot of unsophisticated consumers who think bigger is better, then 96 or 192KHz it is. Not because good engineering demands it, but because marketing departments know consumers are gullible. The original CD spec was pretty much optimal from an engineering perspective. It delivers just as much as it needs to, and not much beyond. It was beautifully engineered. Many of the original CD players (and recorders) did not actually meet the spec very well, nor did recording engineers and producers understand how to work with the new medium, which gave CDs a bit of a black eye. But the theoretical design was quite optimal.
From: isw on 6 May 2008 23:53
In article <b86829fe-0a50-4bc5-8c2b-4274cfff7035(a)q27g2000prf.googlegroups.com>, rajesh <getrajeshin(a)gmail.com> wrote: > On May 6, 9:53 am, rajesh <getrajes...(a)gmail.com> wrote: > > On May 5, 10:33 pm, dpierce.cartchunk....(a)gmail.com wrote: > > > > > > > > > On May 5, 10:09 am, rajesh <getrajes...(a)gmail.com> wrote: > > > > > > On May 5, 7:05 pm, Oli Charlesworth <ca...(a)olifilth.co.uk> wrote: > > > > > If we can, then of course a higher sampling rate will sound better. > > > > > But that goes against the premises of the OP, and is nothing to do > > > > > with the ECC or interpolation that you've been going on about! > > > > > > > -- > > > > > Oli > > > > > > I said we cant percieve, but i didnt say they arent there.. > > > > > Again, true but irrelevant. > > > > > > i will continue the dicussion on ECC tomorrow. > > > > > Hopefully, you will be much better prepared. > > > > > As a hint: the issue of proper sampling vs bandwidth > > > is a topic COMPLETELY separate from ECC. You > > > might want to keep that in mind during your preparations. > > > > take for example h.264 video > > Apart from having many sophisticated techniques it also > > recommends simple one like repeating packets. > > Correction its not packets , its slices Slice repeat is not for handling errors; it's an efficient encoding technique. Instead of sending all the data a second time whwnever two slices are nearly identical (and that happens fairly often), just say "remember that slice I just sent you? Well, use it again, but make these minor changes to it." Isaac |