7 Jul
2004
7 Jul
'04
8:11 p.m.
Kjetil Svalastog Matheussen wrote:
My impression is that the more maths an audio
professional knows, the
more
sure the audio professional is that
higher sampling rates is a
bad thing. (unless you are recording sounds that is later going to be downsampled a lot of course)
Perhaps its impossible for us
non-skilled-mathematicians to
understand properly why 96 kHz is a bad thing...
96kHz is not bad, 192kHZ IS bad. 96 pushes side effects of filters in
the A/D/A chain beyond out hearing. However, it's suspected that 192 is
not so good as there isn't sufficient time between samples to allow
components (e.g. caps) to function correctly. I think I've said this
before but read Dan Lavry's paper on the subject at
http://lavryengineering.com/documents/Sampling_Theory.pdf
Yes, Dan Lavry does makeand sell ADCs but this subject has been
discussed to death on a number of audio lists and forums frequented by
many eminent engineers. It is the concensus that 96kHZ is a Good Thing
(tm).
Greg
___________________________________________________________ALL-NEW Yahoo! Messenger -
sooooo many all-new ways to express yourself http://uk.messenger.yahoo.com
7 Jul
7 Jul
8:50 p.m.
New subject: [linux-audio-user] New Asus Ath. XP 2500 Delta 1010LT performance
On Wed, 2004-07-07 at 15:01, Greg Murphy wrote:
Kjetil Svalastog Matheussen wrote:
My only, somewhat less than engineering level thoughts were that some of
the fidelity on the reverb and delay tails might be enhanced with a
little additional fidelity. I use decent quality live speakers at
reasonable non ear splitting levels and generally keep them close enough
to maintain a stereo image. Again, I don't want to spur the argument
that "Only my Dog can hear that damn noise!" :)
My only real concern here was with my new box. I figured it should run
this no problemo on the 2.4 kernel with Xruns. I read the other
responses from Jack and co (thanks) and need to take a closer look at
tweaking my box for optimum performance.
R~
My impression is that the more maths an audio
professional knows, the
more
sure the audio professional is that
higher sampling rates is a
bad thing. (unless you are recording sounds that is later going to be downsampled a lot of course)
Perhaps its impossible for us
non-skilled-mathematicians to
understand properly why 96 kHz is a bad thing...
96kHz is not bad, 192kHZ IS bad. 96 pushes side effects of filters in
the A/D/A chain beyond out hearing. However, it's suspected that 192 is
not so good as there isn't sufficient time between samples to allow
components (e.g. caps) to function correctly. I think I've said this
before but read Dan Lavry's paper on the subject at
http://lavryengineering.com/documents/Sampling_Theory.pdf
Yes, Dan Lavry does makeand sell ADCs but this subject has been
discussed to death on a number of audio lists and forums frequented by
many eminent engineers. It is the concensus that 96kHZ is a Good Thing
(tm).
Greg
___________________________________________________________ALL-NEW Yahoo! Messenger -
sooooo many all-new ways to express yourself http://uk.messenger.yahoo.com
8 Jul
8 Jul
5:30 a.m.
New subject: [linux-audio-user] New Asus Ath. XP 2500 Delta 1010LT performance
Oa , Jul 07, 2004 at 03:40:39PM -0700, Russell Hanaghan wrote:
On Wed, 2004-07-07 at 15:01, Greg Murphy wrote:
That's a plausible argument for more bits, but not for faster sampling.
Somebody mentioned anti-aliasing filters: That's where the benefit
lies. When you go from from 48kHz to 96kHz sampling rate, the gap
between passband and stopband for your antialiasing filters goes up
from 8kHz to 56kHz (assuming a 20kHz audio band) so you can make much
better filters: flatter response to 20kHz, less phase error in that
range, less overshoot on transients, much smoother clipping if you ever
go over range on the A-D, and less interference from spurious HF signals
appearing at the input.
--
Anahata
anahata(a)treewind.co.uk Tel: 01638 720444
http://www.treewind.co.uk Mob: 07976 263827
It is the concensus that 96kHZ is a Good Thing
(tm).
My only, somewhat less than engineering level thoughts were that some of
the fidelity on the reverb and delay tails might be enhanced with a
little additional fidelity.
7490
days inactive
7491
days old
linux-audio-user@lists.linuxaudio.org
2 comments
3 participants
participants (3)
-
Anahata -
Greg Murphy -
Russell Hanaghan