wow what a find?

[MKTSteiner at aol.com]

In the rec.music.makers.synth newsgroup I found a description about a
'soundcard' for the Apple II. Perhaps it's interesting for you, and I hope
it's o.k. to quote this posting. It come from Rodger Raino, eMail:
raino at ix.netcom.com

Malte

<<
I don't have an Alpha Syntauri, but I do have it's close cousin the 
Sounchaser which was made by Passport (yup they did hardware long ago).
The 2 products used the same sound board, insalled in an Apple ][, which 
was made by Mountain Computer (I think).  All of this pre-dates midi, Macs
and commerically available FM synthisis.  The date on my manual is 1981.
The keyboards were used to generate control info to be used by the sound
boards.  The Alpha Syntauri was volocity sensitive and the Passport was
not (and hence much more affordable when I got it).  The sound boards 
were the real heart of both systems.  You could also access and control
the boards directly via hardware registers.  That's where the real fun was.

Here's how the boards (it was a pair of interconnected boards) worked.  They
had 16 "digital oscillators", with 8 bit resolution and used a fixed length
wavetable.  While simple, this design was fairly elegant & flexible.  
Especially for the price (don't ask I don't recall, but reasonable) and the
times.  All of the control registers could be changed at any time, not just
at the start of a note.  In fact the boards didn't know anything about
"notes", just pitch, amplitude & waveforms; from which notes could be 
constructed.  Basically it was 16 continious channels of sound that was only
silent if the amplitude was 0, or the wavetable a flatline. 

For each of the 16 voices:
- The way you controlled pitch was to load a register pair on the 
boards that told them how fast to step thru the table.  Since you could
use any numbers, this allowed creating exotic tunings and the ability
to continiously change the pitch (vibrato, pitch bend, whatever).  
- Envelopes were generated by loading a register with a value for the
amplitude.  So by loading a series of these an envelope could be generated.
The numbers were discreet and did not need to be continious (no ADSR here) so
you could create an envelope of whatever complexity desired.  Didn't even 
need to look like an envelope, could be random points or a wavetable.
- Finally there was a register that pointed wavetable for the osc. to use.
Once again you could change this whenever, so you could use one wave for
the attack  and another for the rest; or cycle thru a series of wave tables.
The wavetables could have anything in them.  The system came with a bunch
of typical waves (sines, tri, saws etc.).  With some programing (or tedious
data entry) you could build waves for yourself.  Due to expense, sampling
wasn't particularily viable way to get waves.

So how did it sound?  Well bad and good.  When the DX7 hit with it's clear
sounds
and then sampling that kinda re-set the rules on what "good sound" was.  But
things
are comming full circle with lo-fi sound, analog all the rage, and people 
purposfully lowering their sample rates.  In reality there is a place for
both
ends of the spectrum.  I think a lot of people would hear these today and
think
they were some funky old analog synth.  They don't *really* sound analog, but
they
don't sound like digital as we think of it either.  You can create some
really unique
sounds, mostly due to the flexibility of changing anything on the fly-
especially
envelopes.  I'd say they sound weird and cool.

I've tried running mine in order to sample it.  Unfortunantly I can't get 
either of the old Apples in the closet to function (funny they worked when
they went in), the keyboards (ascii) seem dead on both.  If anyone in the 
SF bay area is interested in working on this drop me a line.

thanks
rodger raino
raino at ix.netcom.com>>