[LAU] Studio reverb (impulse)

[Fons Adriaensen]

On Tue, Dec 25, 2007 at 12:31:05AM -0800, Spencer Russell wrote:

> One thing about linear systems though, is that the output doesn't have
> any frequency components that aren't in the input signal, it just
> scales and phase-shifts frequency components according to the
> frequency response. A snare rattle can generate wideband-ish noise in
> response to a single frequency that makes the snare head vibrate, so
> it's a nonlinear effect.

Well, it *is* possible to do this using IRs and convolution, by
having many convolutions in parallel: one working on the signal
x(t), one on x(t)^2, x(t)^3, etc. This allows simulating a time-
invariant nonlinear system. Not so easy in practice. Farina has
a paper on it.

Resonance of the toms: delay + high-Q resonance.

-- 
FA

Laboratorio di Acustica ed Elettroacustica
Parma, Italia

Lascia la spina, cogli la rosa.