OK, I am back from my tampering with the 950908 prototype. I think I've gotten as much as I can from this unless somebody knows a way to get one of those DOSBox builds to work, or else if they know something I could do in Linux with a tool like socat.
The first piece of information I have is that the EXE
apparently ends at 120090h and strangely, deleting everything beyond this point doesn't destroy the .EXE but it does yield a different error message, instead complaining about GRAPHICS.DAT which is just... unknown... I'm not a programmer and I definitely don't do assembly.
On the parallel port I can only capture a few pins and it is entirely possible the dongle used the pins differently to most devices. I can't get a straight capture because no tool will capture it and the monitors I can get don't log - which is silly, because the port is much faster than the screen refresh rate.We can, however, look at the signals from D4 D2 D1 and D0 with pretty precise timing which might shed some light as to what is happening, though I doubt it.
pposc_000.png (676bytes)
Number of downloads: 18
I have enclosed the captured file and the utility here, it was the only thing I could log
at all with, everything else just displays on the screen in real time - you know, because you can totally see what's happening on a device at 2500000 times a second, on a display that updates only 72 times per second no less.
Here are the files (Open O1.DAT unless you want to be hardcore and read O2.ASC in Notepad) for looking at this partial log, you'll need to use DOSBox to run the program. It doesn't require a parallel port or any configuration just to look at existing captures;
PPO.zip (91.46K)
Number of downloads: 2
This is the utility I used;
http://spiro.trikaliotis.net/pposc
I doubt this is of any use at all, but without using expensive scopes and such, which I don't have, I'm not sure there's much more that can be done aside from hoping a skilled cracker is bored some day and decides to try breaking it.
I certainly won't be touching it again any time soon because my head hurts from all the time I spent in Linux, damn I forgot what a chore that whole thing was and how slowly it likes to do things. Not a shred of consistency, everything is a damn command line with warping text and everything wants compiling, requiring hunting down specific versions of libraries from years ago of course. Bloody thing. Oh, also, You can actually see a similar waveform if you capture audio in DOSBox from the Covox emulation... Which makes sense, given the Covox was just a DAC plugged into the parallel port in the real world. I also wanted to say to Mark. that I did wonder about that, I wouldn't be surprised if there were cheap dongles that were barely more than loopback devices. I tried setting random pins to High/Low but this affected nothing that i could tell.