Hi all,

First of all, I would like to point out that Jim Robertson told me a couple
of times that he couldn't guarantee that the board would work on my iMac.
He made sure that I understood that before sending me the product, and as
far as everything else goes, his service was excellent.  The product arrived
promptly, he even rang me up to apologise for using a second-hand box
because his supplier of blank boxes was late!  (Which in my book goes far
and above the call of duty).  The warp-13 looks very professional, I got a
serial cable with it, and I have absolutely no complaints against him.

Unfortunately, and due to no fault of Jim's, I'm still having problems
communicating with the warp-13.  I think this is more a problem of macs
being a bit touchy with their protocols than the board being defective.

Here is what I'm up to at the moment:

I tried putting a MAX-232 in between the programmer and the computer, with
the same result as the voltage divider trick.  (ie. I get a serial port i/o
error in MacPIC when
reading or writing code).  I CAN say that the computer is happy that the
programmer is the correct type - when the programmer selected in MacPIC is
a different one to the PICStart plus, it spits an error straightaway.

And blank checks work fine... so long as they don't take too long.

I tried using my PC emulator running MPLab, and it seemed to connect fine.
It told me that there was a newer version of the firmware available -
which means to me that it connected fine.  (By the way, Jim told me that
the firmware was out of date and I told him to ship it anyway).

However, after a few seconds MPLab told me that it couldn't read from the
PICStart +.  MPLab would then reset the programmer, tell me it had reset,
then try for a response again.  Two seconds later it would tell me that it
couldn't read from the PICStart, reset it.... etc.

Here's part of a previous post:

>         As a long-time Mac user, I always get a little upset when I see
> technically oriented people that do not even make an attempt to understand
> what is 'under the hood' of something...
<SNIP!>
>
> Mac (DIN 8)             RS232 (25 Pin DIN)
> 1       (HSKo)          4       (RTS)
> 2       (HSKi)          5       (CTS)
> 3       (TXD-)          2       (Tx)
> 4       (Gnd)           7       (Gnd)
> 5       (RXD-)          3       (Rx)
> 6       (TXD+)  n.c.
> 7       (GPi)           8       (DCD)
> 8       (RXD+) short to Mac pin 4
>                         short 6 (DSR) to 20 (DTR)
>

I use the connections given by apple.  You can have a look at them by
opening up the "Apple Modem Tool" in your extensions folder in ResEdit.
There
should be a PICT in there that shows you the correct connections.  They're
not too different to yours.

It might be worth mentioning here that I have used my cable to program an
intel 80c196 board (using my PC emulator) and also my DonTronics AVR
development board (I wrote the software to program it).  I have also
tested it by surfing the net through a modem that I was testing.  All
worked fine.

When I tried putting the voltage divider circuit in, the computer was a
lot happier - ie I could actually connect to the programmer - but I'm
still having the same problems as written above.  Although one of you
suggested that the "Serial port I/O error" message doesn't mean anything,
I'm not convinced.  If that were the case, then shouldn't the programmer
work with MPLab?  Hmmm...

It seems to me that the computer and programmer are getting slightly out of
synch when they are communicating.  Maybe the crystal is just slightly
wrong?

Could I put a different crystal into the programmer?  Maybe half the
frequency?  hang on, that would mean a different baud rate for the
programmer applications.  Maybe I need to put a variable clock frequency
into it and see which gives the most reliable output..

>         At any rate, if you ignore the standards, actual voltages, etc, and
> use my simplistic approach to construct the cable I described above, you
> may be pleasantly surprised.
>
I would LOVE to be pleasantly surprised at this point, however at the
moment I very much doubt it.

Now as for the USB connection:

>Speaking of PIC and MACs, does anyone know if it's possible to use one
>of the USB->parallel printer "cables" with the cheap parallel port PIC
>programmers (and similar devices for other processors?)  I have no idea
>whether these things allow any individual bit twiddling, but it's at least
>POSSIBLE...

I investigated this option for a different sort of programmer, and it's
difficult.    You have to write your own extension to control the
USB-parallel chip.  I started to do this, invested about 20 hours, and
gave up.    Even when you do this, you still have to write all the other
software to go with it - it's a lot easier to create a chip that takes a
serial input/output and converts it to parallel.
If you're interested, the chip that almost everyone uses inside their
USB-parallel converter is a Lucent chip.  I have the specs at home.
You can bit-twiddle but it means a LOT of stuffing around.

Ben