I use the ICSP, since you probably have a programmer anyway it
makes it really simple, either put a connector on the board, or for
production I use large pads and spring loaded contacts, the only
extra component to add is a diode to the reset line.  Makes
reprogramming a breeze, with my setup I use either a 16C76 or
16F876, all I change is the header files and rebuild it.

Martin

> That should work in a clean (electrically) environment.  Don't expect it to work
> well on a factory floor, long wires make good antennas.
>
> I would suggest, however, that you use a pic16f877 for prototyping.  There are
> only a few things you need to change to make it work, and you can load the
> program using a bootloader rather than dealing with a programmer and ICSP, or
> even the older than dirt program-plug-test-unplug-erase-etc cycle (it's a pain
> even with two or three chips used in a rotating manner)
>
> Once you've got the program working on the flash part, changing the program for
> the c64 is trivial.
>
> -Adam
>
> Martin McCormick wrote:
> >
> >         I will be using a 16C64 in a project in which there are
> > enough significant issues that I will most likely be needing to
> > pull and reprogram the PIC several times before everything is
> > right.  The project is on a perforated board and the PIC will
> > eventually sit in a wire-wrap socket, but the plan is to build a
> > second little board containing a ZIF socket and two roughly
> > 6-inch or 16-CM pig-tails containing female connectors that will
> > fit down on to the pins of the 40-pin wire-wrap socket.
> >
> >         The pig-tails are each 40-lead ribbons that I salvaged
> > from some old gear but only one 20-pin row can be used on each
> > connector.  I plan on grounding half the conductors so that
> > ribbons will have a signal, ground, signal, ground pattern.
> >
> >         In this case, the 16-meg oscillator is being supplied by
> > a 7404-based circuit I built as I need clocks for other devices
> > on the board.
> >
> >         Does this idea have any hope of working?:-)  I see it as
> > saving a lot of wear and tear on the 40-pin socket which I can
> > wire normally.  When it is good, the test set can come off and
> > the PIC can fit in to the socket in the normal way.
> >
> >         This PIC will not be operated in Microprocessor mode so
> > the fastest switching with the exception of the clock input will
> > be at 4 MHZ or 250 ns per event.
> >
> >         I plan to put decoupling capacitors on both the permanent
> > and test sockets.
> >
> >         I bet one will be able to hear the test setup very well
> > on a HF receiver.
> >
> > --
> > http://www.piclist.com hint: The PICList is archived three different
> > ways.  See http://www.piclist.com/#archives for details.
>
> --
> http://www.piclist.com hint: The PICList is archived three different
> ways.  See http://www.piclist.com/#archives for details.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.