Re: 65020 CPU? Look this...

["Michael J. Mahon" <mjmahon@aol.com>]

Simon Williams wrote:
> Michael J. Mahon <mjmahon@aol.com> wrote:
>
>
> > Simon Williams wrote:
> >
> > > Michael J. Mahon <mjmahon@aol.com> wrote:
> > >
> > >
> > >
> > > > (BTW,
> > > > I'm working on a synthesizer version of DAC522--one that could be
> > > > an eight-voice synthesizer on an AppleCrate!  More soon.)
> > >
> > >
> > > !!!!!!!
> > >
> > > That would be exactly the incentive needed to turn my closetful of IIes
> > > into a NadaNet... 
> >
> > Ha!  I knew that would hook you, Simon!  ;-)
> >
> > Actually, you made me think about how I would adapt DAC522 to the
> > AppleCrate, and another e-friend, Patrick, got me back on track
> > thinking about the details last week.
>
>
> Okay, that's my good deed for the day!
>
>
> > I found that if I'm willing to use more RAM for the DAC, I can
> > create a synthesizer that plays a sample at any pitch, and with
> > an arbitrary envelope (table-defined).  If the sample/envelope
> > information is sufficiently compact, more than one "voice" can
> > co-reside in memory.
>
>
> That creates some _very_ interesting possibilities...
>
>
> > Only one voice per machine can sound at a time, but that's where
> > the 8-processor AppleCrate comes in.  ;-)
>
>
> What about audio output? Will this make use of the cassette out port, or
> do you have something more exotic in mind?

Right now it outputs to the speaker--which I route to an amplified
speaker.  On the eight-processor AppleCrate, I think I'll connect
four to the left channel and four to the right.  ;-)

I don't use the cassette output, because it's too low-level
to drive my amplified speaker(s).  But, as you know, it isn't
hard to patch all the $C030 references to $C020...

The software DAC, DAC522S, is now 32 discrete pulse generators,
each page-aligned, so it's spread out over 8KB.  I'll put all
the rest of the synthesizer code into the "holes" (which are
about 3/4 of the total space).

> > I've also done an experiment that demonstrates that the eight
> > AppleCrate machines, with their separate clock oscillators, stay
> > synchronized within 22 parts per million, or, put another way,
> > if all eight machines are started in sync, after 46 seconds, they
> > are still within 1 millisecond of being in sync.
>
>
> Close enough for Rock & Roll! (and then some)

Agreed.  Then only issue is making sure that the synthesizer code
accounts properly for all conditional paths, so that no software
dispersion is introduced.

> > Allowing for the usual variations, and assuming 50 ppm, that means
> > that a system that can re-sync every second or so, would only have
> > 50 cycles of dispersion coming from clock variations.  This means
> > that re-synchronization could be guaranteed with only a 50-cycle
> > delay!  (Of course, if they are running different programs, then
> > the conditionals in the program will create additional timing
> > dispersion, unless limited by programming techniques.)
> >
> > -michael
> >
> > New Applesoft BASIC interface for NadaNet networking!
> > Home page:  http://members.aol.com/MJMahon/
>
>
> Anything else we should know about... perpetual motion machine? cold
> fusion Apple II power supply? ;-)

I've been thinking about a power supply that would run on
spam--that ought to come pretty close to perpetual motion.  ;-)

-michael

New Applesoft BASIC interface for NadaNet networking!
Home page:  http://members.aol.com/MJMahon/