> Actually I'm not going to bother. Here's why: moving to the Intel part > would require you learning a completely new toolchain, from programming > language, to programmer, to architecture, to programming style. As a > 16F84A user you have already invested a significant amount of time in > learning how PICs operate. In short unless the 196 absolutely does > something the 877 cannot, > it is doubtful that switching will be very productive. > > One last point. Where exactly are you going to go to ask 80C196 > questions? ;-) > > Unless your friend can explain to you in detail why you need a new > hammer, you are much better off using the hammer you already know. The > very fact that you are here asking why you should consider switching > means that he doesn't. I agree with Byron but want to add a few points. I haven't looked up the 80C196, so I don't have a good idea of its capabilities. It is probably a larger chip than the 16F877. As Byron said, if a PIC can do the job and you're already familiar with PICs, then use a PIC. However, I recommend the 18F452 instead of the 16F877. The 18F part is only a little more expensive in small quantities and actually cheaper in large quantities. In either case the difference is tiny compared to other costs of building a robot. The 18F part has roughly twice the code space and twice the compute power, and more than twice the RAM. It is also easier to program. Bank switching issues have been reduced and page issues have been nearly eliminated. Do you really need a 40 pin part to control the robot arm? Remember, these chips have internal RAM and ROM, so most of the pins are I/O lines. The next step down is the 18F252. This is essentially the same chip in a 28 pin instead of 40 pin package. I have done Intel 8051-type microcontroller development, and prefer using PICs by a large margin. This has nothing to do with the programming architecture or their capabilities. The two architectures are very different, but in the end both can usually do the job for about the same price. The Intel family has a higher high end than PICs do, so it can do some tasks PICs can't. The main reason I prefer PICs is because of the great support I get from Microchip. I get the development software, simulator, emulator, and the chips all from one source, and they are reasonably priced. Microchip seems to understand that its worth a few $$ to help me design in their parts once so they can make more $$ selling chips later. When I have a PIC problem I have one place to contact, whether it has to do with the assembler/librarian/linker, IDE, simulator, emulator, the PIC itself, or some interaction between these components. That kind of support for the 8051 family is just not available. That's because Intel (or several others) make the chips, Kyle and a few other make the software tools, and somebody else makes the ICE. Whichever party you call, their first reaction is to blame one of the other parties, this is assuming you can talk to someone intelligent at Intel at all. This isn't hypothetical speculation, by the way. I've been there. It's a pain. The tools also cost a lot more because the people selling you the tools won't make any money on the chips later. They have to get what they can from you up front. For PIC development, the assembler, librarian, linker, IDE, and simulator are totally free and maintained by Microchip. You will need a programmer and something beyond the simulator for testing, especially for complex projects or if you're a beginner. If I remember right, the ICD2 is a few hundred $$ (I use the ICE-2000 and never bought an ICD, so this could be quite wrong), which is the only cash outlay you need to get started. ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads