AW: Re[2]: Pitch -> CV conversion

[Don Tillman]

   From: chordman at flash.net (Synthaholic AKA sPEW)
   Date: Tue, 01 Apr 1997 18:47:36 GMT

   I am assuming that the major objection to a PLL is that the filter
   adds a 'portamento' to the derived CV that might not be desirable.

There are numerous problems with a PLL, most of the stemming from the
"P".  By that I mean that a phase locked loop's primary objective is
to lock onto *phase*, that's what it does for a living, but in this
application we don't care about phase at all.

For example, say there's exactly one little glitch, an extra
zero-crossing in our input signal.  How does a PLL react to that?  The
right thing would be to ignore it.  But a PLL will do an instant pitch
up correction followed by a longer pitch down correction, and the sum
total will have included the extra cycle.

Compare that to a non-PLL system, one that just times the zero
crossings.  It would output a single cycle glitch CV pulse and the
rest of the time it would be accurate.  If you didn't apply
portamento, that CV pulse would cause a single cycle of noise (the VCO
running a much higher pitch) in a VCO.  Not bad.

So using a PLL for this sort of thing will result in a lot of quirky
behavior.  It might be just the sound you're looking for, and that
would be great, but it won't do the best job at pitch to CV when the
waveform is a real live instrument like a guitar.

  -- Don