Ben wrote:
> I could be doing something stupid, but I am having
> trouble simulating a simple isr that is supposed to
> happen every 208 cycles. When using the stopwatch, I
> was getting 209. I simplified the problem in a test
> project, and the symptoms are the same, TMR0 is losing
> track of time.

Actually, it sounds like Microchip has finally fixed a long-standing bug =
in
the simulator...

There is actually a 3-cycle fudge factor when adding a value to TMR0: on
the cycle of the add itself, the pre-incremented timer value is read, and=

the result of the addition overwrites the incremented value, so effective=
ly
no increment happens that cycle.  This is followed by two cycles in which=

no increment happens, as documented.  I think that if you were to try you=
r
program in an actual PIC, you'd find that the simulator results now match=

reality.
        Jason Harper

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.