Carlo Capocasa <capocasa@xxxxxxx> writes:
> how useful is 192kHz
If you want to recreate a clear signal, the higher the better of
course. The Nyquist sampling theorem is flawed in that sense; it's a
very minimum requirement to be able to sample frequencies of SR/2. The
way I understand it, is that Nyquist says that to be able to get all
the frequency content by means of Fourier transform, the sampling
frequency must be atleast twice the highest frequency. Think f.ex,
sampling at 40kHz on a 20kHz signal, you would get 2 samples per
period. How close do you think you get to a sine function with 2
values?.
The worst condition would be of course that the sample point would be
the zero crossing and you'd hear nothing;).
Even if it's SR/4 (80kHz), best would be looking like a pyramid at
20kHz.
And, there is no perfect low pass filter. With a higher sampling rate
the filter can do better job of not tainting the audible frequencies.
It's time for Fons to step in;).
--
Esben Stien is b0ef@e s a
http://www. s t n m
irc://irc. b - i . e/%23contact
[sip|iax]: e e
jid:b0ef@ n n
