gnupic@linuxhacker.org

	A thousand thanks to Scott.  After reading his message to the
list, I got the tar ball, unpacked it, did ./configure --disable-gui
and then make and it worked like a Swiss watch.  I had logged the make
output to a file in case anything unusual happened, but that turned
out to be unnecessary as nothing went wrong at all.  Now, the command
gpsim results directly in a functional text display.

	It now looks like the simulator and the programmer driver
works so I will use gpasm to develop a 12C54-based replacement for a
circuit I was building on the work bench that has a string of 7490
counters feeding off of a crystal oscillator that was to have been
used as a frequency standard.  It's a loooooong story, but the logic
IC's I have in my boxes and what I needed to do to make the generator
select 1 of four frequencies was about to overflow the perforated
board space and I really didn't like the way things were going.  I'll
keep the oscillator because I finally got it on frequency, but I think
a 12c54 will divide 2.4 MHZ down to 50, 60, 100 and 120 Hertz, each
selected separately, and give it to me on a pair of complementary
outputs with a few left over.  I thought of using a 12C509, but I
would run out of leads or come very close to it.  What this will be is
the driver to an audio chopper modulator used to test the speed
accuracy of open-reel tapes and cassettes when they are being dubbed
on to CD's.

	Again, many thanks for the extra effort.  I was perfectly
happy to use the existing version until the problem got corrected
later.  After all, these things happen.

Scott Dattalo writes:
>On Sun, 26 Oct 2003, Scott Dattalo wrote:
>
>> On Sat, 25 Oct 2003, Martin McCormick wrote:
>> 
>> > 	I guess I'm getting tired.  The missing file is 
>> > paraface.h:30: gtk/gtk.h: No such file or directory
>> 
>> 
>> Martin,
>> 
>> This is a bug in gpsim.
>> 
>> The problem is that the --disable-gui option is not carried over to 
>> gpsim's modules. The modules are separate entities like LCD displays that 
>> expand gpsim's simulation capabilities. Give me a couple of hours to fix 
>> this.
>
>Martin,
>
>it would've been too difficult to describe what would need to change to 
>get around this problem. So I fixed the code and created a new tarball. 
>
>http://www.dattalo.com/gnupic/gpsim-0.21.0.tar.gz
>
>You should be able to configure gpsim without the gui and build it now. 
>Note, to get this to work, I had to inhibit certain modules like the LEDs, 
>LCDs, boolean logic, and switches. A better approach would be to go 
>through and implement the textual equivalence of the graphical output.
>
>I guess I could've just put the fix into CVS and pointed you to that. 
>However, I got tired of forcing everyone to go to CVS to get the latest 
>stuff. So, starting now, I'm going to start making releases of gpsim 
>again.
>
>Caveat -- this particular version does have some experimental code with 
>regards to the way simulated time is handled. There's a good chance that 
>a bug may be there. When completed, this change will enable gpsim to 
>better handle multiple time sources. e.g. the time for the processor can 
>be separated from the time of modules. This way, the PIC processor clock 
>can vary (or even stop) while the modules can continue on.
>
>Scott
>
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: gnupic-unsubscribe@linuxhacker.org
>For additional commands, e-mail: gnupic-help@linuxhacker.org
>