On Sun, 08 Jul 2007 19:46:18 -0700, Jeff Liebermann

Interesting. So, using my original example, if I take two ultrasonic
tones, above human hearing, you suggest that I do *HEAR* a beat, but
that there's no actual component at the beat frequency. The does
present a problem because if this is true, then the mixing has to
occurring somewhere in order for my brain to detect the beat
frequency. Is it mixing in my ear, in the cochlea, in the nerves
going to the brain, or in the brain somewhere? I don't think it's any
of these because when I do this experiment, I don't hear any such beat
note.

---
Agreed. I've done the same experiment, and it seems that if the
non-linear detector is presented with tones it can't recognize then
no cross-products are generated.

Same like a mixer with lowpass filters on its inputs, but I can't
seem to pin down your position.

What point are you arguing?


--
JF