On 2003-07-08, Major Variola (ret) uttered to cypherpunks@lne.com:

I haven't, but it does ring true. You'd get 2 Khz as well as other intermodulation products.

Provided there's a nonlinearity, effective in the ultrasonic range, somewhere. Mere interference (which is what we usually refer to as "beats") doesn't give rise to intermodulation. The beat, it isn't an audible frequency per se, but double the frequency you'd need to amplitude modulate a sinusoid halfway between the original sinusoids to get an equivalent result.

You've read about the company trying to sell highly localized speakers? They modulate two intense ultrasound beams, and the air does the nonlinear mixing where they meet.

You can do it with a single beam, too. MIT's Sonic Spotlight is one example, but there are better developed applications on the market. However, you need huge amplitudes to get the air to distort. (I've heard numbers in the 130-150dB range.)

In the audiophile, lower-intensity case, the ears' nonlinearity would do it.

I don't think it would. Before the nonlinearity gets to do its job, the sound needs to be conducted to the inner ear. But it probably won't be -- our ossicles and the tympanic membrane are too massive to operate in that frequency range. So I agree if the amplitudes are extreme, but otherwise I doubt it. -- Sampo Syreeni, aka decoy - mailto:decoy@iki.fi, tel:+358-50-5756111 student/math+cs/helsinki university, http://www.iki.fi/~decoy/front openpgp: 050985C2/025E D175 ABE5 027C 9494 EEB0 E090 8BA9 0509 85C2