[Don Pearce[_3_]]

On Tue, 27 Oct 2009 01:12:01 -0700, Beta_Carotene
wrote:

David Nebenzahl wrote:
On 10/25/2009 1:59 PM Bag of Peanuts spake thus:

David Nebenzahl wrote:

The only way this could work would be if you could somehow get an
*absolutely perfect* copy of the noise to be removed from the
signal+noise recording. But of course that is impossible.

When it comes to sound, there's no such thing as "perfect anything" [...]

That's just the *point*, you nimrod.

I wouldn't want to do any "subtracting" when XOR works so much better.

Please explain how XOR would work "so much better" than subtraction
using the formula I gave:

[S + N] - N = S

where [S + N] is the signal plus noise, S is the signal only, and N is
the noise.

If you really think that 1+1 gives you the same result that 1 OR 1 gives
you, then no amount of arguing will convince you that S-N is any
different than S XOR N.

How would you use XOR to remove noise (N) from [S + N]?

I can't wait to hear your explanation. This oughta be good.

Last question: Are you a troll, as some have accused you of being? Or
are you posting this stuff in earnest?

The real problem here is that there is no N. What you have is two
terms, N1 and N2. So the equation now goes

[S + N1] - N2 = ?

And the answer is?

If it were actually possible to record the identical N from two
places, the equation would work quite nicely. It isn't

d