[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Are there any 65816 card schematics?
Michael J. Mahon wrote:
> Tristan Mumford wrote:
>
> <snip>
>
>> Dare I mention my SMP idea? I see no reason why 2x 65816s cannot be used
>> in a multiprocessing environment. They are designed to be coprocessor
>> friendly anyway. I know it's a silly idea but I like the thought of a
>> dual processor apple ][. Time to drag out my concurrent systems textbook
>> from wherever I packed it.
>
> The 6502 line is a natural for a dual-processor system, since the
> processor only accesses the memory during half the 1MHz cycle.
>
> The bad news is that the "easy" way is used up in the Apple II by
> the "video processor", but I suppose you could always slice the
> clock finer during the CPU half-cycle to accomodate different slots
> for more processors (and use SRAM or fast DRAM to accomodate it).
Bear with me. Long day.
Two processors. Running at arbitrary speed , sharing the clock (one high one
low) for memory > 64k. Not hard so far from a hardware POV.
Now, say one CPU wants to write to RAM < 64k (ie original system).
Simple enough. Latch it and tristate the latches until the right part of the
clock comes around. To make it easier stop the CPU until a single shot is
triggered at the end of the clock half.
Now, during the time of all this happening lets say that the second cpu
wants to use memory < 64k.
The line for < 64k (created by using demuxers I assume) on the first cpu
would still be active. AND that with the base system's clock and the single
shot.
Use the output of this to do the same latch and tristate as on the other
CPU.
So when CPU1 is reactivated, CPU2 is waiting for the next system cycle to
roll around.
If CPU1 needs to access <64k again it's forced to wait, like the CPU2 was.
Please note. I believe this would only work with the 65816 CPU anyway
because it can be stopped anytime.
>
>> Wow, I get a Palm, an apple ][, and a mockingboard for my //e. That is
>> what I call a good week, and a great way to start the new year.
>
> That's a *very* good start! ;-)
I'll say! Although it turned out to be a rarish, well-populated clone of a
][, which I consider to be better.
And I may (fingers crossed) get an Acorn Archimedes and a IIgs.
>
>> Speaking of which (1 hour premature)...
>>
>> Happy new year everyone. Let's make 2007 our best year yet!
>
> Right on!
<:D Yeah! I like this year a lot!
--
-----> http://members.dodo.com.au/~izabellion1/tristan/index.html <-----
===== It's not pretty, it's not great, but it is mine. =====