Olin Lathrop wrote:

> In the end, I usually use a combination of initial test with the simulator,

For what do you guys use the simulator? The only time I used it was when I
tried to get familiar with the PIC architecture. That's a while ago now.
Since then, I never looked back at it... it doesn't seem to be able to do
me any good. As a consequence, of course, I'm not at all familiar with it,
and may miss out on something :)

For me, it works like this: I do simulate, but only hardware; unusual
constructs or when I'm not sure where the limits are. For the PIC itself,
the simple framework that sets up all the basic functions usually is done
quickly. (C, and past projects help here.) Never before used peripherals I
often get to work by thoroughly reading the manual, or they don't work well
with the simulator anyway (like the CAN bus interface) and may need some
real-world experimenting. After that, almost all the problems that pop up
have to do with interactions with the process environment, which is usually
not easy to simulate with MPLAB. Something like Proteus may help here, but
by the time you have successfully simulated your circuit and the process
environment, you probably have developed the product without the simulation
:)  (And note that I'm a fan of process simulation. I even studied it in
university. It's a lot of fun, but it's also a lot of work :)

So I kind of never found the right place for MPLAB-style firmware
simulation. Where is it?

One thing that just came to my mind is test and optimization of special bit
wiggling routines that work in the microseconds and have critical timing.
That's probably something the simulator does quite well -- it's just
something I don't do that often.

Thanks,
Gerhard

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist