Re: [Sc-devel] [commit?] StereoConvolution2L

[nescivi <nescivi@xxxxxxxxx>]

Hiho,

On Wednesday 05 December 2007 20:08:32 Scott Wilson wrote:
> I suppose this is technically already possible without the extra FFT
> using PV_Copy? But it certainly makes sense as a package for binaural
> stuff.

No, Convolution is not using the FFT UGen as input.
it calculates its own internally.
I changed them to be identified as normal UGens, instead of PV_ChainUGen, 
since they aren't.
They should still be modified to make use of the FFTW library as well, as that 
would make them faster...
but that's been on the TODO list for quite a while actually..

sincerely,
Marije


>
> S.
>
> On 4 Dec 2007, at 18:36, nescivi wrote:
> > Hiho,
> >
> > I just made a variation of the Convolution2L UGen: a stereo version.
> >
> > This will be useful for for example binaural convolution, where you
> > typically
> > want to convolve one signal with two impulse responses (one for
> > each ear).
> > Having a stereo version saves doing one FFT on the input, compared
> > to using
> > two Convolution UGens.
> >
> > ok to commit?
> >
> > sincerely,
> > Marije
> > _______________________________________________
> > Sc-devel mailing list
> > Sc-devel@xxxxxxxxxxxxxxx
> > http://www.create.ucsb.edu/mailman/listinfo/sc-devel