Re: 68008 co-processor board complete working design available ...

[Jim Gilbert <brokerjim@gmail.com>]

On Nov 9, 7:20 pm, Jim Gilbert <broker...@gmail.com> wrote:
> On Nov 9, 7:00 pm, "Michael J. Mahon" <mjma...@aol.com> wrote:
>
>
>
>
>
>
>
>
>
> > On 11/9/2010 3:43 PM, Jim Gilbert wrote:
>
> > > On Nov 8, 7:57 pm, "Michael J. Mahon"<mjma...@aol.com>  wrote:
> > >> On 11/8/2010 2:32 PM, Jim Gilbert wrote:
>
> > >>> Back in 1985-1987 I implemented a 68008 wirewrapped co-processor board
> > >>> that ran on both my Apple II+ and my Apple IIGs.  (I never tested in
> > >>> on a IIe.)
>
> > >>> This design was based on a Motorola design note and featured some
> > >>> considerable extensions:
>
> > >>> 8 Mhz 68008
> > >>> 128 K DRAM (Intel 64x4 FPM DRAMs)
> > >>> 2732 EPROM
> > >>> Communication with Apple II back plane via a dedicated read port and a
> > >>> dedicated write port, each attached to a Mostek 4501
> > >>> 256 x 8 FIFO.  These FIFOs enabled both processors to speak to the
> > >>> other with relative impunity since they handled timing issues well.
> > >>> 68901 MFP chip serving as interrupt controller, parallel ports, serial
> > >>> port (complete with full modem control), and timers for 68008.
> > >>> Hardwired reset pushbutton on top edge of card.
> > >>> Software:
> > >>> (1)  PROM Monitor on 68008 initialized everything, maintained software
> > >>> refresh of DRAM, and implemented bidirectional communications protocol
> > >>> with host 6502 (read/write a byte/word/long/block, run code beginning
> > >>> at x, reset).
> > >>> (2) Applesoft debugging monitor on 6502 which helped load and debug
> > >>> code on the 68k
> > >>> (3) RAM Disk software used to test 68008 where the co-processor was
> > >>> just used as RAM drive for the 6502
> > >>> (4) Print spooler for the 68008 using its memory to spool for the 6502
> > >>> (not sure I still have this)
>
> > >>> I have a working wirewrapped board, block diagrams, schematics, wire
> > >>> lists, theory of operation, construction and debugging guides, timing
> > >>> diagrams, the works...
>
> > >>> I built this out of parts that were mainly free samples from Motorola,
> > >>> Intel, Mostek etc. back when I worked for a system company.
>
> > >>> I used to actually sell the kit on Usenet but unfortunately the key
> > >>> single component that became unavailable was Hollywood Hardware's
> > >>> wonderful full-length Apple II prototyping board which I used.  To run
> > >>> the card in an Apple II GS I had to use a bus extender because the GS
> > >>> case is a little shorter front to back than the IIe or II+ ones
> > >>> were.
>
> > >>> I am seriously thinking of scanning and loading all this stuff into a
> > >>> web site and making it public domain if anyone is interested.
> > >>> As far as I can tell the 68008 and the 68901 and the 4501s are still
> > >>> all available.  The rest of it is all 74 series TTL except for one 555
> > >>> timer chip.
>
> > >>> My design, was heavily influenced and inspired by the Motorola design
> > >>> note but is clearly significantly different and original.
>
> > >>> The main problem with building another one or more would be finding a
> > >>> viable Apple II prototyping board that is big enough.
>
> > >>> The 68008 software is assembly language in the AT&T syntax that was
> > >>> accepted by the BSD UNIX system at work and the public domain 68K
> > >>> assembler for CP/M I used on my Z/80 coprocessor (Softcard, then
> > >>> Applicard).  Yes, the 68008 co-existed with the Z/80 cards...
>
> > >>> I no longer have an Apple II.  In order to build or play with one of
> > >>> these a bus extender is very valuable because you'll wear out your
> > >>> backplane slot with too many insertions and extractions otherwise...
>
> > >> Jim, I'd be very interested in your design.  It is considerably more
> > >> capable than the 68008 card I acquired several years ago!  That card
> > >> had a single 8-bit port, with handshaking--pretty slow communication.
>
> > >> I think it would be a great idea to scan in the documentation and make
> > >> it available on the web.
>
> > >http://jgg-apple2-68008.110mb.com
>
> > > Here's the web site.  Currently there is a photo of the board on the
> > > home page, a link to a page with the conceptual block diagram, a link
> > > to a page with the Programmer's Reference Manual, and a link to a page
> > > with the Conceptual Theory of Operations.
>
> > > This is just the beginning of the technical information available.
>
> > Wow--I see you really *used* that long proto board!  ;-)
>
> > -michael
>
> > NadaNet 3.1 for Apple II parallel computing!
> > Home page:  http://home.comcast.net/~mjmahon/
>
> > "The wastebasket is our most important design
> > tool--and it's seriously underused."
>
> I have now uploaded the Motorola Application Note AN896 by Geoff Brown
> which inspired this design. This material is copyright 1984, Motorola.

Actually the FIFOs were each 512x9, not 256x8.  There is provision in
the design for the 68008 to interrupt the 6502 in-band (i.e. when the
6502 decides to allow itself to be interrupted. There is also
provision for the 6502 to interrupt the 68008 in or out of band.

Debugging RAM-based code on machines which are running software
refresh is kind of tricky.  If you stop the processor all the evidence
evaporates!