[sdiy] Prophet VS a phase accumulator design

[karl dalen]

> ? So the ESQ1 are fixed and the PV-S the same then?
> I tought we hade come to the conclushion a phase acc
> would be seen as variable sample rate despite the fixed
> rate computation? No?
>
> Ahh, had no idea the PPG was fixed rate osc. ;)

>I think almost every sane (i.e. not Fairlight/Synclavier)
>polyphonic implementation is fixed rate, just some are
>high frequency. And as I said before, if you increment
>on integer divisions of the (high) clock frequency then
>the two cases become equivalent despite "fixed" clock
>rate, just with less pitch resolution. But this is too
>fast to permit a multiplexed DAC.

In other words your not entirely sure anymore? :)

>Where I'm less clear is the Emulator/Emax stuff. Older
>variable rate samplers like the Fairlight required a
>separate sample RAM per voice, but the Emulator
>apparently kept the fixed sample rate, using a single
>shared bank of RAM and apparently some sort of DMA and
>small RAM buffer per voice. I haven't seen any 
>documentation, obviously, but this should put the Emax
>in a uniquely desirable position (cheap, variable rate, 
>analog processing).
 		 	
From what i have read was the E1,2,and 3 DMA variable
sample rate (depending on nowadays the definition of
variable).

The CMI11 and CM111 are entirely different the CMI III who
uses a pole of mem variable voice cards in a multiplexed
fashion and each contain a DAC.