Re: C64 MMU POC

Den Wed, 10 Jul 2019 20:11:41 +0200 skrev Gerrit Heitsch
<gerrit_at_laosinh.s.bawue.de>:
> On 7/10/19 7:47 PM, Jim Brain wrote:
> > On 7/10/2019 12:26 PM, laughton_at_cyg.net wrote:
> >>
> >> How strong a guarantee do we need?  In the 1980's I built a
> >> computer that uses 65C02 (CMOS) undefined opcodes, so I'm familiar
> >> with those -- and some of them are very ill suited for
> >> cycle-saving branches etc.  I'm *not* familiar with NMOS undefined
> >> opcodes.  But it's reasonable to suppose there may be some which
> >> could serve the MMU function but otherwise are never (or almost
> >> never) used.  It's true this can't be 100% guaranteed.
> >>
> >> As for the missing SYNC pin: one solution consumes PLD resources
> >> to create a state machine.  That seems like (and may indeed be) a
> >> rather expensive solution.  But it would be no surprise if someone
> >> who's willing to engage with the challenge (this is crucial) found
> >> it simpler than initially imagined.
> > The Faks6509 project uses a similar idea, a small state machine to 
> > reverse engineer the SYNC pin function.
> >>
> >> Failing that, I know of a $10 IC that's available in surface-mount 
> >> which exactly matches the cycle-by-cycle behavior of an NMOS 
> >> 6502/6510  ;o)
> > 
> > Hmm, you thinking of the 65C02S?  Because, I think we determined
> > that the CMOS 6502 and the NMOS 6502 are not cycle exact.  I know
> > the Rockwell 65C02 differs in the execution pipeline and thus it
> > fails with the Fake6509 project.
> 
> Hm... could you take an NMOS 6502 and have it run in parallel to a
> 6510, sync'ed at RESET? Plus some logic to tristate the address lines
> and unable to write so it only reads in parallel to the 6510.
> 
> Then you should have a source for the SYNC signal.

Seems better to combine a 6502, a 6522 and some tristate thingies to
emulate a 6510, than using both a 6510 and a 6502.

-- 
(\_/) Copy the bunny to your mails to help
(O.o) him achieve world domination.
(> <) Come join the dark side.
/_|_\ We have cookies.