Re: Hardware emulation of 6509 using 6502?

From: Dr Jefyll <laughton_at_cyg.net>
Date: Wed, 28 Feb 2018 22:45:12 -0700 (MST)
Message-ID: <1519883112579-0.post@n4.nabble.com>
Yes, great work, Jim! *Very* pleased to hear this news!

Jim Brain wrote
>   * I don't know that anything special is needed on the PHI2 signal.

Well, early indications are you're right!  And, y'know, if a modern WDC
65C02 were to be used then the delay from clock input to clock output is
gonna be pretty darn small anyway.  These are *really* fast parts, after
all.

> * I simplified the Verilog (and Dr. Jefyll's schematic by performing
>     the combinatorial logic on the DATA and SYNC inputs before storing
>     into the flip/flop.  This reduces the number of flops (macrocells)
>     needed by 7.  NOt terribly useful, since there's plenty of space,
>     but it's more compact.

 Again, this is probably acceptable.  My original schematic purposely does
things "the hard way" in order to allow more headroom for speed.  You wanna
take pains not to increase the data-hold time required at the end of each
cycle.  And the way to do that is to capture the byte first, then decode it
later (after the cycle has ended).

You changed it around so the decoding happens before the cycle ends and
/then/  the result is captured -- and apparently the CPLD is fast enough to
make that option satisfactory (for that particular computer).

Who know, though -- it's fun to speculate.  Maybe someday someone will take
things to a whole other level and create a 20 Mhz 6509!  In that case it'll
be worth considering a switch back to my original logic -- capture first,
decode later I mean.

Congratulations again on your efforts -- it's great to see the rubber hit
the road at last, and kudos to you for getting serious and taking up the
challenge!

 -- Jeff





--
Sent from: http://cbm-hackers.2304266.n4.nabble.com/

       Message was sent through the cbm-hackers mailing list
Received on 2018-03-01 07:00:02

Archive generated by hypermail 2.2.0.