Tobias wrote:
I can not get D. Taits software to work.

and Antti replied:
D. Taits soft usually works. However you are not the first one saying
he cant get it working.

I always find such statements a bit confusing; that is because I
provide software in three different forms: Turbo-C source (PP.C),
QBasic source (PP.BAS) and an executable (PP.EXE).  Which of these
don't work?  All of them?

Tobias:

If the hardware works with PICSER then I expect my QBasic program will
work too.  Did you try that?  If you tried it but it didn't work, fit
a 10k pullup on the ACK pin and have another go.  My PCs don't seem to
need one but apparently some do.  Antti tells me his software makes
special provision for the lack of a pullup.  With PP.BAS working you
can then take Antti's advice to program a PIC using your separate
program and data files and use PICSER to dump a merged file.  After
that you can throw my software in the bin and use PICSER from then on.

I want to point out that there is nothing wrong with PP.C either.  The
same can't be said of PP.EXE though.  In my FAQ I mentioned that I
thought my compiler (Turbo-C V2.0) had a bug in the routine used to
determine the vital programming delay.  I can now confirm that.

I compiled and ran this simple program on several different machines:

#include <stdio.h>
#include <dos.h>

void main()
{
     int i;

     printf("start\n");
     for ( i=0; i<2000; ++i)
        delay(10);
     printf("finish\n");
}

This should print "start" then wait 20 seconds before printing
"finish".  This works as expected on my old 386DX-20.  On a 486DX-33
the waiting period dropped to 9 seconds and on a 486DX4-75 it was only
3 seconds.  Compiling and running the same program using a more recent
version of Turbo-C (actually Turbo-C++ now) gave a uniform 20 seconds
delay on all PCs.

If you don't have access to a C compiler just let me know and I'll
send you a new PP.EXE as uuencoded mail (that also goes for anybody
else bitten by this problem); make sure you tell me whether you are
using inverting (7406) or non-inverting (7407) buffers.  I will
replace the archived copies of PP.EXE in the next few weeks.

David
--
david.tait@man.ac.uk