On Fri, Jun 18, 2010 at 5:03 PM, Barry Gershenfeld wro= te: > Imagine single-stepping with the debugger, and taking a snapshot each time > you hit the step button. =A0The trace is a review of what instructions got > executed, where the next instruction was fetched from, what was in > such-and-such register, maybe even a record of bus signals. =A0I made up = that > feature list since I don't actually have one, but it's based on what we d= id > many years ago with similar equipment. > > The trace memory is where you put all that info, so you can look back 128 > instructions or whatever. =A0And often you can set up a trigger to capture > what's in the trace memory (like a "freeze" option) even though you keep > running. Oh, cool! Thanks for the explanation, that sounds really useful actually. To verify routines I'll sometimes step through a bunch of code while watching various registers. If I'm understanding you correctly, I'd be able to say, set a breakpoint at the end of a section of code then watch what the registers did for the last X number of steps. That seems much quicker! Do you get to choose which registers it follows or does it dump the entire memory? Would that slow down execution (like the old "jog" mode)? Is there anyone with a RealICE that can show me a screenshot of how trace appears? I'd love to see if it will actually do what I think it does. By the way, I've noticed that single stepping on the PK3 seems to be quicker than with the PK2. I'm guessing RealICE should be at the same speed as PK3/ICD3 right? Thanks! Josh -- = A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams -- = http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist