[LAU] Distorded Jamin?

[Fons Adriaensen]

On Tue, Jul 10, 2007 at 09:51:34AM +0300, Sampo Savolainen wrote:
> On Sat, 2007-07-07 at 12:20 +0200, joel silvestre wrote:
> > Hi,
> > 
> > I've just seen something strange, Jamin seems to generate some
> > distorsion even with eq, limiter and compressors bypassed. This
> > distortion disappear when Jamin is completly bypassed.

There could be something wrong with the way 'op' is used in
process_signal().

Suppose that run_eq() or run_eq_iir() is called with in_ptr == 0.
That means that the window is applied without any offset w.r.t.
the signal buffer. After the FFT, samples 0..127 in mid_buf have
their final value, accumulated over 16 windows, and are ready for
further processing. They are read via the index 'op' which
therefore should be equal to in_ptr, and not have the 'latency'
offset applied. 

-- 
FA

Follie! Follie! Delirio vano è questo !