Date   : Wed, 04 Apr 2001 16:08:50 +0100 (BST)
From   : jim <jim@...>
Subject: Re: 6502 Technical Term

On Wed, 4 Apr 2001, Thomas Harte wrote:

> > Incidentally, very detailed timing information on the 6502 can be found in
> > the most excellent "64doc" which can be found at
> > http://www.cs.cmu.edu/~dsladic/vice/doc/64doc.txt or just type 64doc into
> > Google ... it was written for the Commodore 64 but is mostly accurate for
> > the Beeb also.
> 
> a) "mostly accurate for the Beeb also" - are any instructions timed
> differently, or does that just mean all that muttering about zero page
> registers and such like, and

Perhaps that was an unfortunate turn of phrase. I didn't mean that 64doc
is inaccurate for the beeb, just that it's written from a Commodore 64
point of view so parts of it aren't relevant.

> b) does anyone know the timing for IRQ and NMI? "like BRK" isn't really very
> helpful . . . .

"Like BRK" is all I got, I'm afraid.

I did some work a while ago on a 6502 emulator using the cycle rather than
the instruction as the basic execution unit, which tries to emulate the
exact steps the 6502 goes through (from 64doc) for each instruction, right
down to a memory access every cycle and instruction prefetch on the last
cycle of an ALU op. It worked reasonably well, but was a bit slow. And
yes, I frankly fudged interrupts. I didn't think they were that vital
timing-wise, being a once-in-a-blue-moon occurrence.

jim
-- 
http://madeira.physiol.ucl.ac.uk/people/jim/
"... I naturally gravitated to London, that great cesspool into which all the
 loungers and idlers of the Empire are irresistibly drained."
                                - Sir Arthur Conan Doyle, "A Study in Scarlet"