So I've been reading all day about this and this is what my intuition tells me:
the filters from the objetcs I mentioned ([lop~], [hip~], [bp~], [lores~] & [reson~]) and the ones described at "Cookbook formulae for audio EQ biquad filter coefficients" are not butterworth/chebyshev/eliptic/bessel/etc. They're just what they are. Seems these filter topologies are some special cases that don't apply to these more general filters. Does anyone back this up? Moreover... I've figured out how to work with butterworth filters. I had the wrong impression you could start from a second order filter and cascade it a few times to achieve higher order filters, but that's not the way it is. thanks 2014-12-15 15:00 GMT-02:00 Alexandre Torres Porres <por...@gmail.com>: > > "I'm not sure how much difference it makes for low-order filters (biquad > has 2 poles and 2 zeros). The differences really matter more for higher > order filters, say 8 poles or more..." > > That's really relevant information, thanks :) > > > I'd just use the ones already made in iemlib... > > Sure, but it's more of theoretical doubt and concern. I just wanna learn > more about it, even if I just use what's there in iemlib. > > > There's also a 3rd order BW in the docs > > Sure, I know about them and I'm studying it. And I'm also now checking > http://msp.ucsd.edu/techniques/latest/book-html/node145.html > > I assume it's a matter of how to calculate the coefficients for you to get > filters that are butterworth/chebyshev/etc, and the text from miller > point to how to get butterworth coeficients. But it doesn't point to how to > get Chebyshev 1 pr 2, Eliptic, Bessel and whatever. Any hints? > > And I still wonder what's the kind of filter from objects like [lop~], > [hip~], [bp~], [lores~] and [reson~] for instance (cause I know how to get > those with biquad coefficients). And also about the filters described at > "Cookbook formulae for audio EQ biquad filter coefficients" > http://www.musicdsp.org/files/Audio-EQ-Cookbook.txt > > Do any of you know what kind of filter it is? > > Thanks > > > 2014-12-15 13:59 GMT-02:00 Frank Barknecht <f...@footils.org>: > >> Hi, >> >> On Mon, Dec 15, 2014 at 03:41:45PM +0000, Claude Heiland-Allen wrote: >> > On 15/12/14 14:49, Alexandre Torres Porres wrote: >> > >Do any of you know if they're Butterworth or what? >> > >> > I'm not sure how much difference it makes for low-order filters >> > (biquad has 2 poles and 2 zeros). The differences really matter >> > more for higher order filters, say 8 poles or more... >> > >> > You could waste time calculating coefficients for a biquad cascade >> > (the process is quite involved - see the mkfilter package online), >> > but I'd just use the ones already made in iemlib... >> >> There's also a 3rd order BW in the docs, if you're interested in the inner >> workings: 3.audio.examples/H13.butterworth.pd >> >> Ciao >> -- >> Frank Barknecht _ ______footils.org__ >> >> _______________________________________________ >> Pd-list@lists.iem.at mailing list >> UNSUBSCRIBE and account-management -> >> http://lists.puredata.info/listinfo/pd-list >> >
_______________________________________________ Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
