In Reply to: Appreciate the Clarification..... posted by Todd Krieger on June 17, 2006 at 15:40:24:
HowdyI am serious that graphs done the ways I've mentioned would be very enlightening for everyone who reads here. I'll prepare them if I can find the time. (Not for a little while, unfortunately.)
Well I don't agree with the more qualified statement for reasons stated earlier. There are different things going on here and it's easier to understand them if we keep them separate.
As mentioned earlier dither works as designed in a pure digital domain. It doesn't need filtering to work in a digital environment and it does encode more information about the predithered signal (at the expense of noise) in the resultant signal. It doesn't require filtering to work as designed there and it doesn't require filtering to work for our ears.
At the levels that dither is reasonably used real signals (as opposed to artificial signals like the sawtooth referred to) already don't look like themselves once they are bandwidth limited and sampled. They look like the bandwidth limited signal :) Dither will add a little visible noise at the 1 or 2 lsb level but you are unlikely to see the difference from an undithered signal. But you will hear it, and the difference isn't trivial.
-Ted
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Thanks - Ted Smith 16:02:50 06/17/06 (0)