>At work, we joke that all the other database guy has to do is _walk into_
>my cube for me to finally figure out what's wrong with my code, so don't
>worry about it. I think there's something along the lines of that once
>your mind relaxes that little bit b/c it's given the problem to someone
>else, that relaxation allows you to quit buggin' out and actually _find_
>your problem.
Yah, there's a great deal of truth to what Ruffin says here. This, along with
the paradox of the "I didn't change anything!" broken software are legion in
engineering. :)
>Speaking of, all the bugs have been hacked out of MacDAsm, so if you're
>on a Mac (ha!) and have been having trouble, drop back by the pages and
>try again... ;)
Yah, I found that filling unused bytes in the ROM image with 0x00 instead of
0xFF as DASM originally did allows some "weak" Stella emulators with bad
checksumming routines to properly play those hot new titles you're trying to
develop and debug. :)
I include my readme in the MacDASM distro:
November 17, 1998 Update
------------------------
In order to work-around weak checksumming routines in Macintosh Stella (0.7), I
have changed the default fill byte value to 0x00 rather than 0xFF. This value
seems to allow much more code-in-progress to function correctly than the 0xFF
setting. The 0xFF setting was a concession to EPROM programmers who prefer to
leave unused areas "unburned" to faciliate later overpatching if need be. Since
I suspect users of this assembler are primarily Stella emulator aficionados, the
actual value of this isn't important, other than working around weak
checksumming
algorithms.
If someone screams for it, I'll modify DASM to accept a fill byte parametre, but
I think this issue has already been beaten to death. If it matters that much to
you, the #define is in asm.h and is called DEFORGFILL.
=Rob=
--
Archives (includes files) at http://www.biglist.com/lists/stella/archives/
Unsub & more at http://www.biglist.com/lists/stella/
