Yeah, this is one part of the CS80 that I would not want to emulate
exactly in a VA/Origin. I'd rather have choices, like on the Oberheim
4/8-Voice, which let's you select different key assignment algorithms,
each of which suits a particular type of sound and/or playing style.
Combine that with Yamaha's brilliant Sustain I/II choices and you can do
about any style of mono or poly playing.
I don't even want to think of what it would take to interface them, but
it could you imagine (visually and sound-wise) a CS80 with an Oberheim
8-SEM cabinet slaved to it! Now there's an inexpensive, simple way to
get the LEDs to tell which voice is playing :^)
David
The Old Crow wrote:
> This is what you get when a team of organ engineers use their key
> assigner logic for a synthesizer. ;) On the EX-2 each manual can sound
> 7 voices each, but together they can only sound 11. They use the same
> ring memory table in the GX-1 and key assigner of the CS 50/60/80.
> There is a somewhat cryptic explanation in the CS service guide, pages
> 11 to 19, describing the encoder for 4 voices. JH references the
> service guide on his CS50 page,
>
> http://home.debitel.net/user/jhaible/jh_cs50.html
>
> If you look closely you'll see the actual electrical operation of the
> key scanner is ternary, not binary, it uses both -6.5v to 0 and 0 to
> +8.5v to derive logic states.
>
> Crow
> /∗∗/
>
> Colin f wrote:
> > The GX has all its voice cards in 80 pin 0.1" card edge sockets, so swapping
> > round a bad card is simple.
> >
> >
> >> I'd be interested to find that out too.
> >> The GX is odd. I assume it keeps a list of free voices in least recently
> >> used order, assigns the voice at the head of the list to a new note and
> >> sends that voice to the end of the list when it is released.
> >> After an initial clear, for a repeated single note, it starts at voice 2,
> >> then cycles thru 3,4,5,6,7,1, back to 2...
> >> If you play an overlapping note before the first cycle thru 7 has complete,
> >> it plays the second note on the next voice.
> >> But once thru the cycle of 7, a second note will always be on voice 8.
> >> That suggests to me that the intial state of the LRU list is
> >> 2,3,4,5,6,7,8,1.
> >> Once you go through the first cycle, it skips voice 8, leaving 8 at the head
> >> of the LRU, where it stays until at least two notes are overlapped.
> >> One particularly odd thing is that if voice 8 has been brought in during a
> >> cycle of single notes, on release it seems to put both voices 8 and 7 at the
> >> end of the LRU list, so voice 7 is not used as soon as you'd expect.
> >> As you might imagine, mimicking this behaviour precisely in software is
> >> going to be a fun job.
> >>
>