icd issues

[h]

hello,
anyone had a project that will not work when downloaded to the hardware after working great through the debugger? Stranger still. When the code is compiled with ICD=TRUE it works in the hardware. compile with ICD=false and it wont run.
Any thoughts.

Thanks

[Testicky]

a few days ago, using a 12F629, a small adaptor board from microchip ( wich plug instead of the 12F629), MPlab 6.4, I could run the program with the ICD-2, but not program the same chip in-circuit with the same ICD-2.
Eventually, after a few uninstall, and re-install, the conclusion was that my ICD-2 was faulty!!!! The ICD-2 could run the small board to emulate, but when trying to use it in programming mode, ..no it would not do it.
I am not sure if this is related to your problem...but...seems not far similar.
_________________
Testicky

[schmobol]

Hi,

working with CCS 3.182 or 3.184 (same symptoms) under MPLAB 6.4

I have the same kind of problem. When compiling with ICD=TRUE the program runs, when I take it off it doesn't work anymore !

What should we do ?

[Guest]

Check your fuses both with and without ICD=TRUE. Perhaps one of the default setting with ICD=TRUE is allowing you code to run.

Also check out the Release notes for 3.184 CCS has fixed a few ICD related fuse issues.

Later

[Guest]

I'm using microchips icd2 debugger. I set all fuses off in code and ensured that the icd wasn't compiling for icd. I did notice that when switching the icd to program it didn't automatically change the program range to the full range of memory on the part (16f877A). anyway I got it working. But I'm still uneasy. Afraid there is some hidden bug that I'm not aware of. Eventually (soon I hope) this product will be for sale. I don't need alot of irate customers. hopefully our field testing will uncover any insidious bugs I can't find in the lab.

thanks

[schmobol]

It works now ! I just had to add a fuse setting NOLVP (it was automatically set by ICD=TRUE) to solve my problem. Tricky this "ICD=TRUE" !!!