[linux-audio-dev] Re: [Jackit-devel] Re: [linux-audio-user][ANNOUNCE] jack_convolve-0.0.1

Florian Schmidt mista.tapas at gmx.net
Sun Jan 30 22:56:50 EST 2005


On Mon, 31 Jan 2005 14:43:55 +1100
Erik de Castro Lopo <erikd-lad at mega-nerd.com> wrote:

> On Sun, 30 Jan 2005 21:22:02 -0600
> "Reuben Martin" <MartinR at jbu.edu> wrote:
> 
> > Forgive my ignorance, but is there any significant difference in the
> > sound produced by this complex convolve and BruteFIR's half-complex
> > convolve? (I'm speaking in a theoretical sense rather than comparing the
> > actual implementations.)
> 
> There shouldn't be. There are however significant savings (approx 50%) in 
> computation with the half complex version.

Yes, i adopted this scheme. I use the r2c and c2r versions though. Only
difference is how they store the values (they transfor N real input
values into N/2+1 complex outputs. This is 1 more than the r2hc
versions, but i can access memory in forward direction in this case w/o
reordering). 

Flo

-- 
Palimm Palimm!
http://affenbande.org/~tapas/



More information about the Linux-audio-user mailing list