Ove Karlsen wrote:
A linear smoother might fix some issues, on filters that are sensitive
to fast changes.
Also for filters that seems to work fine with EQs, highpasses, my own
Minimal-phase IIR Gaussians work fine (Beneficient Open-Source Licence).
Variable order up to 9th, and variable cutoff, with 5 onepoles in
parallel. Which I worked out in my own math, and can be referred to as
"Karlsen Gaussians" if neccesary.
And this connects with the normal theory as... how?
I mean what a non-sense. Anything you do in the digital domain will
require you to resample properly, unless you're sure you're not aliasing
(which more often than not isn't the case). The resample function cannot
be replaced by another form, or theoretically this would be known. What
do Gaussians (integrals ? Stochastic theory ? Nuclear Physics basis
functions ? Or ?) have to do with these filters?
A little theoretical basis training wouldn't hurt, believe me the actual
higher maths aren't available by using these sort of linguistic
monstrosities either, they're well founded on solid mathematics.
T.V.
--
dupswapdrop -- the music-dsp mailing list and website:
subscription info, FAQ, source code archive, list archive, book reviews, dsp
links
http://music.columbia.edu/cmc/music-dsp
http://music.columbia.edu/mailman/listinfo/music-dsp