VSP-330: Update

I'm working hard to redevelop the voicing architecture of the VSP-330.  In beta there were some phase-cancellations that I noticed frequently when playing octaves.  I didn't know what was going on, but a guardian angel suggested where to look so that I could define the problem.  Turns out, it's nothing that phase-locked loop oscillators can't handle like on the original instrument.  The desired result will sound much more in keeping to the original VP-330.

A little technical digression.  The Phase locked loop oscillator scheme will also result in being Fully Polyphonic.  But you may know the original was Paraphonic.  Paraphonic has a master oscillator divide-down circuitry where all keys are triggering a single shared envelope.  If no notes are being held while the retriggering happens the envelope resets.  To me this was an undesirable limitation of the VP-330 that resulted in sustaining notes getting robbed when the envelope gets retriggered.  A much more desirable result would be like the Logan String Melody II in this regard, which uses a separate Attack-Release envelope for each individual key resulting in Full Polyphony.  Why didn't Roland use the Full Polyphony design in the VP-330?  I am guessing because Logan and Hohner already had the patent by then....if anyone knows for sure I would love to know!

Flash forward back to today.  I've successfully modeled the phase-locked loop oscillators, and now face the daunting task of implementing them into the voicing architecture of my plugin: Upper/Lower and 8'/4' scheme.   If all goes well launch is set again for Tuesday.  Much to do until then! -g