I'm glad you got a laugh out of that, and I do hope its not accurate. Part of the reason for posting it was to get out in the open things that I've heard others speculate while complaining. I don't think you really responded to that but your offer to help is appreciated and I'm amazed that none of the simulator lovers on the list have jumped at it. MPLAB is apparently a great product (I'll be finding out soon as I now have an ICD and am scrambling for the time to compare it to the SX Key I know and love). A lot of people really like it and its nice that your efforts for big ticket development wash over into the hobby and small professional users. I am concerned that Microchip will have a hard time justifying the ICD hardware to big ticket manufacturers. I don't see as much usage of the ICD systems on the PICList as I would have expected. Most hobbyists seem to be programming F84s with old tait or other programmers and relying on classic (insert a port toggle for an LED) debugging techniques. Most professionals seem to be using the EMU's their bosses lay out the bucks for. PICList development method survey: A) How do you debug your code? B) Have you ever used an ICD? Now I don't want to get on a rant but: I'm interested in this because simulators are IMH(N)O evil and emulators are only a bit better. ICD is THE ONE TRUE WAY. Religious fervor aside, I think simulators and emulators are only valuable when you don't have access to an ICD. Why else would you want to watch what a PC thinks the uC is going to do or what an emulator (no matter how close to a production chip it is) is doing when you could be watching what the actual uC IS DOING, in the actual circuit. Simulations are doomed to success. Most can't afford an EMU. Why not have something better than an EMU for a tenth the cost that fully "simulates" the entire device? Go to a seminar and get one for what? $70, $80 over the cost of the seminar? I've seen that again and again. "my program just wouldn't work in the simulator, but I went ahead and programmed a chip anyway and it worked perfectly" "it works in the simulator just fine but I can't get it up and running in the F84." "We couldn't get it running and spent a month with a '508 and then finally got a few minutes on a borrowed EMU since we couldn't afford one ourselves and I found a register initialization problem right away". Everyone of these would have been prevented if people had developed on an ICD system even if they then had to port the code to a different target chip. I think there are fewer gottchas porting from and ICD capable chip to another target processor than there are in developing directly on the target. At least that is true if you don't have an EMU for the target, and even then, I'd start on the ICD and switch to the EMU at the end to work out the device differences. My experience with this comes from a former life where I was developing on an early version of the Z8 (an embedded Z80 basically) as a part of a 5 person team. I was software. The hardware guys (analog, analog specialist 1 and 2, and digital) got a circuit together and I wrote a mother big (for a Z8) program that did a lot of complex stuff. Anyway, it didn't work and the hardware guys traced it back to the software not producing an output. As I young kid, I started sweating and going through the code trying to trace back and eliminate reasons why that output wouldn't get made. All of this with out an emulator, simulator, logic analyzer etc... Eventually, management called the local mfg.'s rep and arranged for an emulator. Suddenly, on the EMU, everything worked perfect. Now I had a little credibility, but the hardware team did all kinds of tests to verify that the signal levels getting to the uC where in tolerance for both the EMU and the production chip (slightly different tolerances) and they verified that the same signals were getting to both setups. They pointed the finger back at me saying that I was changing code between the EMU and production chips since some changes were required and didn't really know what I was doing. I wasn't totally professional and things really started to go down hill (my fault mostly, I really didn't know what I was doing completely but wasn't willing to admit it ). Eventually, one of the hardware guys found that the production setup worked for a fraction of a second after a new chip was inserted and that the new chip got hot for a second on initial power-up then cooled back down, but never worked right again. Turned out, the signal driving the uC on that pin was burning out the bondout wire on the production chip before the uC switched the pin to input on startup but not on the EMU which used (apparently) a larger wire. They were seeing a perfect signal at that pin but the uC never got that input. On an ICD I would have had that nailed in no time. If I was a more experienced tech I also would have nailed it in more time, but experience costs more than even an EMU, doesn't it? There isn't anything about the ICD capability that actually requires an ICD module. I bet at '877 and some passives combined with a PC program can make a "free" open source ICD. We already know most of what it would take. If we publish a super low cost ICD replacement for the '877, wouldn't that be a better use of our time than a new simulator? Wouldn't more people be able to write better code faster with that? Your comments appreciated. --- James Newton mailto:jamesnewton@geocities.com 1-619-652-0593 http://techref.massmind.org NEW! FINALLY A REAL NAME! Members can add private/public comments/pages ($0 TANSTAAFL web hosting) -----Original Message----- From: pic microcontroller discussion list [mailto:PICLIST@MITVMA.MIT.EDU]On Behalf Of Darrel Johansen Sent: Tuesday, March 07, 2000 17:50 To: PICLIST@MITVMA.MIT.EDU Subject: Re: PIC18CXXX Assembler? & MPSIM James --I really enjoyed that script, but it's not accurate. The bottom line is that MPLAB is a product that is provided free by Microchip. It supports the simulator, a number of emulators, the ICD, quite a few different language products, and is actively supported and maintained. It is offered as a product to enhance productivity. It is doubtful that simulation will ever be able to compete with emulator capabilities, and you can expect to see more low-cost solutions like the ICD because those functions are enabled by flash technology. My offer stands to provide information for those interested PICLISTer's or others, to take a thread of the simulators and run with it. Darrel Johansen tempe, az