On Sun, Jan 04, 2015 at 09:30:16PM +0000, Aurélien Leblond wrote:
> Would you know how this is handled in AMS (being c++ as well)?
Just look at the source code...
AMS includes <complex.h> before <fftw3.h>, and it compiles
without even a warning. Of course it doesn't use -std=c++11.
So just put the DSP code in a separate file and compile it
without that option.
> > A second source of problems with the vocoder is the mixing
> > of floats and doubles. There is no good reason to use a
> > double FFT here.
>
> I actually did notice that, but again I don't want to "correct" it in
> ams-lv2 because my aim here is to port "as-is".
You don't modify anything fundamental when changing the FFT
to single precision float. And the modifications required can
be automated quite easily in case you ever need to repeat them.
Tschuss,
-- FA A world of exhaustive, reliable metadata would be an utopia. It's also a pipe-dream, founded on self-delusion, nerd hubris and hysterically inflated market opportunities. (Cory Doctorow) _______________________________________________ Linux-audio-dev mailing list Linux-audio-dev@email-addr-hidden http://lists.linuxaudio.org/listinfo/linux-audio-devReceived on Mon Jan 5 04:15:02 2015
This archive was generated by hypermail 2.1.8 : Mon Jan 05 2015 - 04:15:02 EET