[
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2008, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]
71.188.240.155
In Reply to: RE: I can't tell if you are serious or not posted by BKWells on January 06, 2008 at 13:56:17
Howdy
The practical problems with Nyquist theory isn't so much band-limiting vs. time-limiting (predictability as you describe it) but practical implementations of the anti-aliasing filter and the reconstruction filter.
These days oversampling can really make a difference in a quality reconstruction filter, but using higher sample rates clearly makes it easier to avoid the worst problems with these filters.
Getting back to your questions about predictability, there is more than one approach to such a problem.
Because we are recording we can delay the actual processing of a signal as long as we want and hence in principle don't have to worry about unpredictability: just wait a while and we'll see the context :)
But truth be told the simplest way of looking at it is "What does the system look like with an impulse?" Compared to a steady state signal isn't that the most unpredictable thing you can do? The anti-aliasing filter and reconstruction filters limit the slope of the impulse response and this will smooth it out a little. But some of the worst things you see in, say Redbook, is the pre-ringing that some filters cause: you see the signal starting to wave around before the impulse arrives. Unlike a gong or any other natural signal which has an exponential decay, a filter pre-ring can be an exponential attack: a very unnatural thing. For example see "Fig.1 Typical impulse response of a linear-phase, brick-wall filter" in http://www.stereophile.com/reference/106ringing/
-Ted
Follow Ups:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2008, Audio Asylum, all rights reserved ]