At 05:12 PM 9/12/96 GMT+2, you wrote:
>------- Forwarded Message Follows -------
>From:          Self <FS7/WTERREB>
>To:            PICLIST
>Subject:       Using the Picstart Plus to program other micro's?
>Date:          Thu, 12 Sep 1996 17:07:40 GMT+2
>
>I bought myself one of the new Picstart Plus programmers and is very
>impressed with.   This is almost a universal programmer in the true
>sense of the word.  Or is it?
>
>What I specifically like about the programmer is that they put DG411
>analog switches on almost all the pins, making it possible to cater
>for almost any kind of pin configuration.  The control is done by a
>PIC17C42 (I think) which is connected via the RS232 serial port to
>your computer.
>
>This brings me to an interesting thought...   It really looks as if
>this hardware should make it possible to also program other devices
>on this programmer if one just replace the existing controlling micro
>with another specially programmed one.  I would particularly like it if I can
> program
>the Atmel 89C20xx and 89Cxx devices.
>
>Has anyone else out there ever tried to do this?  Is the schematics
>for the Microchip Picstart Plus available anyware?  It shouldn't be
>too difficult to reverse engineer it, but I really don't want to do
>that if someone else has already done this and I can just buy a
>pre-programmed 17C42 to replace my existing one and wolla - My
>picstart can now also program the Atmel devices!
>
>It would be interesting to hear what the programmer experts like Jim
>and Antti think of this....   <hint> <hint>   :)
>
>Rgds
>
>Werner
>
Werner,

(I hope you meant me when you said "Jim the programmer expert." I'm not
being presumptuous am I?)  :-)

I have not seen a PS + so I can only make general comments. First, I would
be surprised if the firmware chip is a 17C42 as it would take a bit of effort
to fit all the code into just 2k. Going by previous PS 16 efforts I doubt
the Uchip employees would put this effort in. It is certainly possible to fit
the the required code in to a 17C42 but somehow I suspect the chip is a 17C43.

Again, I am guessing, but I think you might be exaggrating when you say there
are DG411 switches on "almost all the pins." Exactly how many are there and
how many are connected to *unique* pins. Given that you need two i/o lines
for each switch to mean anything, there would have to be plenty of additional
i/o support, a 40-pin PIC could not possibly handle it all. Exactly how much
additional i/o support is there?

I think if you look harder at how and where the DG411s are connected, you might
be disillusioned somewhat, Right?

Never-the-less, it is possible to change the firmware and get the PS+ to program
lots of different devices. However, as for doing it onboard, I have very
strong doubts. I just don't think that you will be able to program much more
than PIC
parts and maybe a few other small devices like  SEEPROMS, and KEELOC parts etc.

I realize the PS+ programs the 17C4x devices and these devices require 25
meaningful i/o lines. This would be just be enough to program some smaller
parallel eproms in theory. However, when you factor in the additional
multiplexing signals required, this is becoming very complicated indeed.
After all,
the 25 i/o lines aren't all in the right places are they?

Now, to Antti's comments.

I have had a discussion with Antti about his pin driver idea. He loves it, I
don't.
At the very best, a universal programmer based on a pin driver concept would
take at least 5 time longer to program than specifically targeted code.
Also, as
Robert Lunn correctly pointed out, there is still no guarantee that you can
cover every possibility. In practice, "final firmware" is a dangerous claim
indeed.

The microchip Promate allows new firmware to be downloaded. This is a far
better
way to guarantee future suitability.

Anyway, hope I have given you have enough food for thought on the matter.

Regards

Jim