>> One good possibility is the 18F458. >> It should be pin compatible, just don't enable the CAN port. >> I am using these successfully at 39.3216 Mhz. Haven't tried 40 yet. >I haven't checked, but don't these have the same bug as the 18Fxx2 chips? I have not seen any problems like this in my own work. Admittedly, I'm still in the prototyping stages, but I do have what I would consider to be asynch high priority interrupts, along with a few low priority interrupts, and so far all the problems have been directly attributed to my lack of mental organization. 8^) I am using the upper interrupts to time stamp IR transitions, and so far the data has been quite repeatable. I checked the datasheets to be sure, and there is no mention of interrupt related bugs in the 18Fxx8 chips. I am also clocking at 39.3216 Mhz, and haven't had any trouble so far. I even have switched back and forth between HS and HSPLL on successive flashes to verify my programs ability to read the config bits and scale timers and BRGs accordingly, and the switch seems quite transparent. Minor point of info for other users: MPLAB does not "emulate" reading of the CONFIG registers, so I had to test that code in real hardware. I plan on making all of my boards for this project with 9.8304 Mhz crystals, then adding an HSPLL config only on boards that require more processing power. Kinda cool, I think. I'm not yet good enough at PICs to be an authority. My word on this is not to be taken as final and definite proof of anything. Anyway, I've just offered what works for me. Lyle -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.