[linux-audio-dev] [ann] unmatched - a LADSPA amp tone

[Steve Harris]

On Mon, Oct 28, 2002 at 07:42:04 +0100, Tim Goetze wrote:
> >I don't know its hard to seperate phase effects from waveshaping, thats
> >why it needs FFTing. Onve we have the aproximate harmonic series we can
> >simulate that with a cheby, and if neccesary put the phase back with an
> >allpass or two.
> 
> ok, i'm slowly beginning to understand this i think. you mean 
> you plan to generate the harmonics with a time domain shaper 
> (like beloved valve_1209) and then 'shape the freq domain' with 
> the chebyshev?

Not really. Both are freq (well, and amplitude) domain only.

The chebyshev is a way of generating specific harmonics, eg. you can
output the fundamental + the 2nd harmonic at -6dB, the 4th at -16, the 6th
at -21, etc. You can't easily control the phase though.

The harmonics track the source as you would expect, regardless of
frequency range (though aliasing is an issue, ofcourse).

> i don't think phase shifting will make a difference (and if
> it does, it's good to know you can fix it ;)

OK, then all I need to do is split out the distorted sinwave and load it
into octave. How Hard Can It Be(TM). ;)

Fix it is probably the wrong term, I can sort of bash it into compliance.

- Steve