[sdiy] Prom generating TOS like MK50240 bits..

[nvawter at media.mit.edu]

heya.   great question!  I considered it a little bit once...

the size would be pretty large..

and it depends one how accurate it is.

the real MK50240 uses dividers in the 200-300 range iirc.

That essentially means you must multiply all 12 values together to get  
the Least Common Multiple and thus the minimum PROM size...

that's (very roughly) 256^12 = 512^6=1024^3 = 79 * 10^27  = 2^96;

if you were to use my set of dividers, the smallest set usable with  
error <=2%, then you're looking at multiplying:
16*17*18*19 * 20*21*23*24 * 25*27*28*30

16*17*18*19  * 20*21*23*24 * 25*27*28*30 = 12228309918720000
l(.)/l(2)= 53.44107454018484483288
=2^53

also potentially problematic is that these numbers are not even powers  
of two, so you'd have to fudge something somewhere or look for an even  
larger PROM.

excuse me if these calcs are off, I'm kind of loopy today :)

p.s.  I designed a chip that does my dividers..  you can see  
simulation pix up at:   
http://noahvawter.blogspot.com/2012/04/first-simulation-of-chip.html



Quoting Jean-Pierre Desrochers <jpdesroc at oricom.ca>:

>
> Suppose you read all the 12 generated notes from
> a MK50240 TOS with a signal analizer, after a while is there a
> 'sequence' in the 12 generated square waves that 'wraps around' ?
> I mean all the tones repeat all together a pattern
> that could be burned into a  PROM then scanned with
> a binary counter that would wrap at a precise address
> to make all the tones seam running continuously.. ?
> JP
>
>
>
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>