"Browntimdc" wrote in message

"Erik Squires" wrote in
ervers.com:

So, here's my question. If I digitize a 15 kHz signal, using a 44.1
kHz sampling rate, I'm going to get about 3 samples / cycle.

Are the normal digital filters good enough to reproduce a 15 kHz
signal with varying amplitude? How accurate is that signal, is there
no lag in the reconstructed signal? I mean, if the amplitude of the
original changes, is the reconstructed signal as true at 15 kHz as at
4 kHz?

This to me is a far more important concern than whether I can hear
20+kHz signals.

Thanks for your intelligent and well thought out replies. The rest
of you can suck my electric outlet.

Erik



First, with 44.1kHz sampling & 20-22kHz bandwidth a 15kHz signal can
only be a sine wave.

Good point #1

Second, the 15kHz signal cannot be one isolated
sine wave, that would require more bandwidth. It will be a burst of
waves with a minimum buildup and decay. So it will be more than 3
samples of one cycle.

However, there need not ever be more than two samples per cycle, over
whatever period the sine wave actually extends.

The closer the sine wave is to 22.05kHz the
longer this build up and decay must be to meet the bandwidth
constraint.

A very observable effect. By the time one gets to 22.000 KHz, tone bursts
get pretty distorted. However, this is not a situation that is unique to the
digital domain. Same thing happens in the analog domain when the test
frequency is close to a sharp cut-off.