Is anyone here familiar with the "Nyquist Theory"?

  • Thread starter Thread starter Michael Jones
  • Start date Start date
Michael Jones

Michael Jones

New member
A must read for anyone doing digital audio.
http://www.iar-80.com/page25.html

It basically states that a digital frequency can only be replicated by using HALF the sample rate.

If anyone has other links, or other insight, I'd be interested in reading them.
 
Last edited:
Michael....

Pick up a copy of John Watkinson's "The Art of Digital Audio", it's considered the "bible" for digital theory.
 
Mister (G)Q! - Thanks for the link.

Bear - will do. Thanks.
 
Michael,
I know this probably won't sound as convincing as the other resposes, but I just graduated with an engineering degree from Virginia Tech and through all the mechanical design classes, much emphasis was placed on numeric testing and data collection. The nyquist frequency was half the sample frequency and any wavelength greater than the nyquist with essentially a symmetric representation of those frequencies below the Nyquist frequency and didn't truely represent those greater frequencies.

I don't know if this helps, but I guess it's good proof that that's what people are teaching.

Brandon
 
There was a really good thread on RAP about nyquist (its theorem, not theory BTW) and high sample rates. Basically, the conclusions were that 44.1kHz is capable of perfectly reproducing signals up to 20.05kHz and that since most people can't hear above 20kHz (although a few can hear up to 25kHz or so, most people stop at 18kHz) 96kHz sample rates and higher are pretty much pointless and the proliferation of higher sample rates is purely market driven.

I'll post a link if I can find it.
 
Last edited:
plus the digital audio on, for instance, a CD is filltered at 20k to avoid biasing below that from audio above that has been undersampled.
 
You must log in or register to reply here.
Back
Top