archive.worldofdragon.org

For the rom, maybe we should follow the msx route, and create new system roms from scratch, compatible with the old ones. That, if anyone cares, which I think is not the case (neither Dragon Data or Eurohard)... Okay... Would be really nice, a OpenSource ROM that's compatible with the original... B...

I have been thinking of sharing a list of comments that I have collected, which people can combine with their own disassemblies. Maybe with scripts to combine and decompose again. I think that would be safely legal. That is a idea! EDIT: My collected comments are here: https://github.com/jedie/Drag...

I have been thinking of sharing a list of comments that I have collected, which people can combine with their own disassemblies. Maybe with scripts to combine and decompose again. I think that would be safely legal. That is a idea! EDIT: My collected comments are here: https://github.com/jedie/Drag...

e.g.:
I created a startup_trace.txt with only run PRINT "A", then i start filtering with PRINT 6 and this is the result: https://gist.github.com/jedie/ab730ed530c67c96218f

The startup trace is ~165MB big. The filtered file is only ~200KB small...

I commit a real nice feature: Live trace filtering ;) (see commit https://github.com/jedie/PyDragon32/commit/7dc9312dcf3ef180f19155cb426f04f80773b8aa ) How it works: Create a startup trace: $ xroar -trace > startup_trace.txt Let xroar start the machine e.g.: until the prompt is blicking. Then quit x...

ROMDIFF.TXT is the result of a binary comparison between the D32 ROMs and the D64 32K mode ROMs. For each address where there is a difference the file shows the D32 byte and the D64 byte stored at that address. I produced the file to see how different the two ROM sets were. The answer is 'not very'...

It's my error... I used virtualbox... With may daily used LinuxMint Debain Edition i have no running opengl... But then i found a missed packages related to OpenGL... After install, XRoar and the test program glxgears worked, but only in fullscreen :( With a second test box with a fresh Linux Mint 1...

I tried xroar_0.32+33SNAPSHOT20140623081438+0100-1~testing_i386.deb and get this: Module init: GTK+-2 UI Module init: GTK+-2 file requester Module init: GtkGLExt video (xroar:11662): GdkGLExt-WARNING **: Window system doesn't support OpenGL. Module shutdown: GTK+-2 file requester Module shutdown: GT...

OK, it's now usable: $ python filter_xroar_trace.py --help usage: filter_xroar_trace.py [-h] [--display [MAX]] [--filter [MAX]] infile [outfile] Filter Xroar traces positional arguments: infile Xroar trace file. outfile If given: write output in a new file else: Display it. optional arguments: -h, -...

I started a simple python script to filter Xroar trace files... The idea is simple: Skip lines with addresses that are calles very often, e.g.: * The "how many RAM is installed" ROM routine * The BASIC Interpreter idle loop So you can see the really interesting parts of the trace... I put the script...