apple2freak{at}gmail.com wrote:
> On Mar 6, 2:15 pm, "Simon D. Williams"
 wrote:

>> I'm not all that jazzed on the idea of going 32-bit, but that's because my
>> chances of understanding anything beyond 8-bits is practically nil. What I
>> would like to see is an Apple II clone that can run mutliple
"shells"...
>> I'm not talking about multi-tasking or anything like that -- just the
>> ability to switch between virtual machines. Functionally the same as if
>> you had say an 8-way KVM switch hooked up to a stack of Apple IIs.
>>
> I think this is a wonderful idea, sort of extending on what MJM
> brought up earlier.
> 
> I don't see any reason why Alex & Steve's FPGA implementation of the
> Apple //e could not be modified to implement such a system.  The limit
> on the number of //e systems would be the number of gates on the FPGA.

Replicating the logic would provide full speed for the "background"
Apples, but that's probably unnecessary.

Just replicating the state registers, including a "master RAM bank"
register would make switching from one "machine" to another very fast
and simple.  The "foreground" machine (with display focus) could run
an arbitrary factor faster than the "background" machines just by
pausing the round-robin state scheduler.

This is now often referred to as "multithreading", though it's an
old idea that I think was first used in the Honeywell 800, and then
to great advantage in the CDC 6600 peripheral processors:  one set
of processor logic with a "rotating" set of state shift registers.

Of course, the total amount of memory is also multiplied by the number
of state copies, so each "machine" has its own full set of memory.

Interesting things are possible if machines can communicate with
other machines' memories, but Apple II's are not designed to be
SMPs, and Apple II software doesn't support it!  It could be used
as a very fast message-passing capability, though.

>> VGA support is a definite plus since that would allow the use of LCD
>> monitors. I like the idea of it being a much more energy-efficient system
>> than the original. Maybe something that could even run on batteries?
>>
> Good points.  I'd enjoy this too.

Unfortunately, power efficiency (MIPS/Watt) is something that FPGAs
are not particularly good at in comparison to custom designs--but
it would take very high volume to justify a custom design.  ;-(

An FPGA design could certainly be low-power *enough* to satisfy
most portability requirements.  For example, a backlight on an LCD
would be much more demanding.

-michael

******** Note new website URL ********

NadaNet and AppleCrate II 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."
--- SBBSecho 2.12-Win32