> I think a swishing sound may have too much content for a PIC. I'd
> imagine you'd want to start with white noise and then run over it
> with a band-pass filter (like a wah-wah pedal). It would be far
> simpler to let h/w do the work. For example, from the Nat Semi
> book -

I think this could be done easily in a PIC using the method you described.
There aren't a lot of cycles there.  A band pass filter is just a low pass
and a high pass.  A low pass filter requires a subtract, shift, and add.  A
high pass filter is the original signal minus the low pass.  The high pass
part of the band pass would need to be continuously adjustable, so the shift
becomes a multiply.  That one multiply is the only part of this algorithm
that will take an appreciable number of cycles.

Let's say you update the PWM output at 10KHz rate, which gives you up to
5KHz bandwidth.  With a 20MHz 16-family PIC, that's 500 instructions per
sample.  This would allow software random number generation, but you could
instead just store a list of pre-generated random numbers.  You'd probably
want that anyway to eliminate low frequencies.  All in all, 500 instructions
is a lot for this task.


********************************************************************
Olin Lathrop, embedded systems consultant in Littleton Massachusetts
(978) 742-9014, olin@embedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads