Jim Nitchals wrote:
> A 6502/816 doing single-step by placing instructions into a temporary
> holding area then executing them can run at about 7% of real-time. If
> doing address computation, masking of the 2600's high order address
> bits, etc you'd be lucky to get to 2% of realtime (times 20 for your
> fast '816, for a total of 40% of the speed of a real 2600.)
40% might not be all that bad... possibly worth doing. But point
taken about the overhead. I suppose it is faster letting the 65xx
execute the instructions like you said (that is what I was thinking)
then emulating each instruction like a non 65xx would have to do...
> I wrote a 6502-based package for doing fast code "emulation" for the
> purpose of debugging games while working at Electronic Arts. It was
> too slow to bother with for any but the *worst, intractable* bugs.
Cool, working at EA, wow.
> And that was presuming hand-tuned assembly language. Go to C, not the
> best language for the '816 (stack addressing modes are *SLOW*) and you're
> looking at a worthlessly slow solution.
Yup, I was only wanting the C source as a bit of a guide. I'd
definitely
be writing it in assembly, I'm an awful lot better in it than C. Thanks
for the comments.
Robin Harbron
macbeth@xxxxxxxxxxx
--
Archives available at http://www.biglist.com/lists/stella/archives/
E-mail UNSUBSCRIBE in the body to stella-request@xxxxxxxxxxx to be removed.
