[Fluxus] fft audio fixes

[Arxontis Politis]

Hi Dave,

>From the little time I have spent going through the fluxus manual and
examples, I haven't understood exactly how (gh) is working yet? Could you
explain how the values are computed? Is it like for an incoming buffer size
N samples from jack:

- perform an FFT of N points
- get the absolute of N complex numbers
- from the N bins get a set of logarithmically (or linearly) spaced bins,
depending on the parameter (set-num-frequency-bins) you mention
- normalise?

Is that it or I am way off?

cheers,
aKis~

On 9 March 2010 17:45, Dave Griffiths <dave at pawfal.org> wrote:

> Hi all,
>
> I've just taken out a lot of nonsense from the fft audio in an attempt
> to make it better. There are no magic numbers in there any more, but I'm
> getting a bit more response from the treble end than the bass end at the
> moment, so it's not perfect yet. If you play in white noise you'll see
> what I mean.
>
> I've also made the number of bins (gh) uses configurable, up to 128,
> with (set-num-frequency-bins).
>
> cheers,
>
> dave
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.pawfal.org/pipermail/fluxus-pawfal.org/attachments/20100310/6bace5e5/attachment-0001.html>