Wasn't the JP80x0 modelled on the Jupiter 8 (including filters, etc.), hence the JP8xxx title? Would this have contributed to the supersaw's fatness?
Also, I bet there's more to it than the actual individual saw waves that make up the supersaw each being mathematically-correct sawtooths (using fourier analysis), each identical, other than being detuned and panned?...
...meaning, I guess if we could isolate each saw wave, they wouldn't be just simple, "static" saw waves, just stacked in unison and detuned?
Could the Supersaw have also incorporated some kind of waveform-modulation/sync between the individual sawtooths to create an even more rich sound, seperate to detuning? (Check out this thread for examples of waveform-sync amongst saw waves, something I've never been able to easily achieve with the Virus, which could otherwise add a whole new level of fatness, completely seperate to simple detuning by normal methods:- )
[
Forum thread: Quick pulse-width question]
Basically it appears to take one single sawtooth oscillator, break it into two (by duplicating it, then halving the amplitude for each copy), and then modulating the phase between the two to give a "doubling in octave/harmonic" type effect (more apparent when you modulate the phase with an LFO, creating a gentle "rocking" pitching motion between one octave to the next).
...And that's using just one single oscillator. When you mix just
two of these oscillators together, and detune them, the effect is devastating, without even using any Unison or third/sub oscillator.
An audio example taken from that thread:
Timo