Subject: Re: [linux-audio-dev] spiced 12AX7, plots, clipper.so
From: Steve Harris (S.W.Harris_AT_ecs.soton.ac.uk)
Date: Fri Nov 08 2002 - 12:50:27 EET
On Fri, Nov 08, 2002 at 01:24:30 +0100, Tim Goetze wrote:
> it may not be a true amp model, but to me ... it rocks.
Yeah it sounds really good.
> i'm still pondering whether antialiasing is needed, and if so,
> how to do it (fit in a windowed sinc where the signal hits
> the clipping threshold? thoughts welcome).
Instead of going into hard clipping with a threshold, you could do a two
stage thing:
if (x < low)
y = x
if (low < x < high)
y = f(x)
if (high < x)
y = high
where f(x) is some function such that f(low) = low, f(1.0) = high,
f'(low) = 1 and f'(1.0) = 0 [I'm too lazy to solve it, but its only a
quadratic]
high is your old clip point, moving low towards high will increase the
"hardness", moving it away will reduce aliasing.
> for building a decent 'virtual amp', i imagine some combination
> of Steve's excellent valve and valve_rect for a good, smooth
> and warm lowish-gain tone, a bit of tunable eq and this kind of
> hard-clipping for hi-gain, harsh distortion topped off with a
> cabinet response will just about do it for me.
Yup, thats sounds like its going to rule. Its not really fair to call the
valve mine though, I just copied it from someones thesis!
- Steve
This archive was generated by hypermail 2b28 : Fri Nov 08 2002 - 12:54:49 EET