Re: [sdiy] MIDM-CV DACs, Part #3 (last)

[jhaible at debitel.net]

>Assume you have 1 oscillator tuned to 256Hz (2 ^ 8)
[...]
>So we have a beat frequency of 0.148 Hz - that's about one beat every 7
>seconds.

One beat every 7 seconds at a certain frequency means more than
one beat per second if you go 3 octaves higher. Wether this hurts or not
I don't want to decide - just wanted to mention it.


>It wouldn't even do that if the second oscillator is driven from the same DAC 
>(via the S&Hs).

Actually, if the you play sevaral VCOs at the same note (unison), you can
tolerate *more* detuning for a pleasant sound than if you play a chord.
(That's why some polysynths slightly detune their voices if you switch
them to unison.) I don't know why, but apparently the ears work like this.
The same random oscillator detuning that sounds "fat" in unison, sounds
"detuned" with chords. 

Which is strange, because with the equal tempered system we don't have pure chords 
anyway. Can anybody explain this? (Maybe out typical unison detuning is more
than the maximum error in equal tempered scales? Or detuning adds to the error,
just shifting a tolerable error "over the edge"?)

I'm pretty unexperienced with different scales. My Rev. 3 Prophet 5 theoretically
allowed to program different scales, but it was all overshadowed by the inaccuracy
of pitch from one VCO the the next.

I hope I can explore this a little more when the PS3200 is finished in a few months,
which has individual tune and detune pots for each note.

JH.