Kurt Hamster wrote:

On Fri, 12 Nov 2004 06:42:56 +0000 (UTC), Stewart Pinkerton used
to say...



Of course it can, you ****. You amplify and store the incoming sound,
then compare it to the database of known signatures. As ever, you are
indulging in ignorant ranting on a subject about which you know
nothing.


Must have a hell of a good pair of speakers to reproduce exactly eh?

Was it a set of Magneplanars driven by a Krell?


jumping in

depending on how the signature matching was done, you wouldn't necessarily have
to play it back through speakers/phones to do the matching.

Not sure how it would have been done in the "good old analogue days" (through
hardware?), but it would be fairly straightfoward to punch out some code to do
it via software nowadays (or using DSP hardware).

Taking a simplistic (naive?) view here, a Fourier Transform of the incoming
signal is one way of determining the signature of a sub.

Or maybe even a Wavelet transform, since it gives the frequency content of the
signal and the point in time at which the frequency content appears (though
wavelet transforms weren't officially defined until the late 80s)