Subject:
Re: Odyssey and Trivial 16FXXX HVP
From:
Detrick Merz ####@####.####
Date:
1 Mar 2005 19:25:48 +0000
Message-Id: <4224C19D.3080207@mci.com>
Byron A Jeff <byron <at> cc.gatech.edu> writes:
>
> On Tue, Feb 22, 2005 at 12:23:08PM -0500, Detrick Merz wrote:
> > If anyone is using Odyssey under Linux to program a 16F628A or
16F872 in
> > a Trivial HVP, could you tell me if the following odyssey config
> > settings look right:
> >
> > [io]
> > driver=directpp # this is easier for me (I said I was impatient right
> > # now) as it doesn't require me to bother with loading a
> > # module, or recompiling a kernel
> > port=0 # there should only be 1 lpt on this box anyway...
> >
> > wdatapin=2
> > clkpin=3
> > pwrpin=4
> > vpppin=-5 # does this even matter for a LVP or HVP, as long as it
> > # has a value set to make Odyssey happy?
> > rdatapin=10
> >
> >
> > Oh, and yes, I have the long cable termination scheme wired in,
although
> > my cable is only about 2' to begin with (maybe 2'6", somewhere around
> > there anyway).
> >
> > thanks,
> >
> > -detrick
>
> Detrick,
>
> I'm the hardware guy so I guess I'll throw in my 2 cents.
>
> The vpppin is critical for the HVP hardware. It's the power pin that may
> or may not matter.
>
> From the schmat you should set vpppin=-4. It's inverted as my release
notes
> point out.
>
> BTW my plan is to officially support pikdev and pkp
(http://pikdev.free.fr)
> for the trivial programmer series. Alain has been quite responsive about
> updates and requests.
>
> Long cable termination is a dicey issue. One that I really need to solve
> effectively. If Odessey or PikDev doesn't seem to work and everything
tests
> out, I would suggest trying with a shorter cable before chalking it up to
> something else.
>
> Hope this helps,
>
> BAJ
Byron,
I've had some more luck with this... after bringing the DMM into work,
and looking at what the pins on the parport cable were doing. Long
story short, it helps if you turn your parport *ON* in the BIOS *runs
and hides behind a tree before he gets flogged for not 'checking the
power cord' first'*. So, as it turns out, once you turn on the parport,
and your parport actually starts showing up under /proc, things work a
little bit better. Minimally, doing a:
odyssey PIC16F872 check
when a p16f872 is in the programmer, odyssey reports back that it sees
it, and likes it (device ID matches what it expects). Of course, when I
put a p16f628a in, odyssey complains that the device ID is wrong, but
then again, odyssey doesn't come out and say they support the p16f628a,
only the p16f628. So, to "solve" that problem (I'm a complete PIC newb,
and my 'tutorials' are for a 16f628, so that's what I really wanna
program for now), I'm emerge'ing pikdev as I write this.
Now, as far as the pwrpin and vpppin settings go, odyssey doesn't seem
to care which one is -4 and which one is -5 for the 'check'...
presently, I've followed your advice, and set vpppin to -4 (and pwrpin
to -5, but we think this doesn't matter much for the HVP). All this
effort, for software I don't plan to use provided I can get pikdev up
and running. At least I've solved some issues that would be common to
either application, and I can see my linux box talking at some level to
the PIC. That makes me happy. Now, if I could only figure out why
google finds my originally message, relayed through some news group, but
it never was sent back to me by the ezmlm...
-detrick
