Commit Graph

477 Commits

Author SHA1 Message Date
nmlgc 3e7fa3a3d1 [Reduction] #500: over_put_8 2014-09-09 05:54:40 +02:00
nmlgc 2ba86d6887 [Reduction] #499: super_roll_put_tiny 2014-09-09 05:34:40 +02:00
nmlgc 9c94cb9e6e [Reduction] #497-498: getdate and gettime 2014-09-08 21:57:20 +02:00
nmlgc 80fc5c6d82 [Reduction] #496: mem_unassign
Which challenges a lot about what we thought to know about Amusement Makers'
modifications to master.lib, due to the fact that TH02 contains the modified
version of this function, but the original of draw_trapezoid...

And I haven't even begun to research how this removal of conditional branches
could have a positive effect on the game, especially since it's only called
before exiting anyway.
2014-09-08 20:47:09 +02:00
nmlgc 9ff29d3159 [Reduction] #495: localeconv 2014-09-07 22:05:49 +02:00
nmlgc 08092bef2b [Reduction] #493-494: DOS file attribute functions 2014-09-07 19:10:29 +02:00
nmlgc 45f1b0d447 [Reduction] #492: unlink 2014-09-07 19:01:21 +02:00
nmlgc 9ed6e1e93f [Reduction] #488-491: General key input support 2014-09-07 17:01:58 +02:00
nmlgc 3d05fb85c9 [Reduction] #486-487: key_wait_bios and key_sense_bios 2014-09-07 16:21:01 +02:00
nmlgc 0d213f41b5 [Reduction] #484-485: resdata_exist and resdata_create
Looks like the 'pal98 grb' string was merely copy-pasted from the respal
module and isn't actually used by the function, which means that we don't have
to work around a naming collision after all.
2014-09-07 15:47:50 +02:00
nmlgc 61a0a024e2 [Reduction] #483: dos_free
OK. I'm going to spend one day on TH01 alone, and if that doesn't end up
shredding the code significantly, I'm not going to cover that game.
2014-09-07 15:41:48 +02:00
nmlgc 23a3a4116a [Reduction] #479-482: pfgetc 2014-09-07 15:07:36 +02:00
nmlgc 0acdb448d1 [Reduction] #478: grcg_polygon_cx 2014-09-07 02:39:52 +02:00
nmlgc 9bb41bd72b [Reduction] #477: draw_trapezoidx 2014-09-07 02:29:58 +02:00
nmlgc 4f2e3e4944 [Reduction] #476: super_put_1plane 2014-09-07 02:18:27 +02:00
nmlgc 9e405575ea [Reduction] #474-475: super_wave_put 2014-09-07 01:56:57 +02:00
nmlgc 99b60ff9b9 [Reduction] #473: execl
And thus, we've singled out all Borland C++ runtime functions in all games but
TH01.
2014-09-06 19:08:18 +02:00
nmlgc d575a37e1e [Reduction] #470-472: LoadProg 2014-09-06 19:07:54 +02:00
nmlgc fe2a73c888 [Reduction] #468-469: ihypot and isqrt 2014-09-05 20:17:22 +02:00
nmlgc 84b7d0f1ab [Reduction] #467: Far long multiplication
Yeah, the code is identical to the near version, with the only difference
being the PROC directive declaring the function as either near or far. Now, I
could either turn the function body into some kind of macro stored in a
separate file and then instantiate it from both near and far functions... or I
could just copy the original structure. Who cares, anyway.
2014-09-05 10:57:12 +02:00
nmlgc ccc560ab37 [Reduction] #466: searchenv 2014-09-04 20:55:28 +02:00
nmlgc dc9fc37b3f [Reduction] #465: searchstr 2014-09-04 20:55:27 +02:00
nmlgc 97711aac8f [Reduction] #464: mbcjmstojis
"Multi-byte-character-<something>-shift-to-JIS"?
2014-09-04 19:24:14 +02:00
nmlgc c0aa5b8a67 [Reduction] #461-463: fullpath.c 2014-09-04 19:04:39 +02:00
nmlgc af7f0b0ad6 [Reduction] #458-460: Double-byte character set functions 2014-09-03 23:23:25 +02:00
nmlgc bab649b884 [Reduction] #457: getch 2014-09-03 19:30:32 +02:00
nmlgc e54a6ad120 [Reduction] #456: DOSCMD
... I, um, cannot comprehend how the C source code I have for this function
could have been compiled into such an assembly.
2014-09-03 19:13:47 +02:00
nmlgc 92046a8021 [Reduction] #455: getenv 2014-09-03 17:08:02 +02:00
nmlgc bf7fb83197 [Reduction] #454: respal_set_palettes 2014-09-03 16:08:44 +02:00
nmlgc b60d3a0f56 [Reduction] #453: respal_get_palettes 2014-09-03 16:07:27 +02:00
nmlgc 3f7a29acc6 [Reduction] #452: respal_free 2014-09-03 15:45:21 +02:00
nmlgc 61c95ec603 [Reduction] #450-451: respal_exist and respal_create 2014-09-03 15:23:51 +02:00
nmlgc 01a126da71 [Reduction] #449: setvbuf 2014-09-03 14:02:14 +02:00
nmlgc 00e419e9da [Reduction] #448: setblock 2014-09-02 23:38:26 +02:00
nmlgc b77f2cfba0 [Reduction] #447: access 2014-09-02 23:26:19 +02:00
nmlgc 23aa61c002 [Reduction] #446: abort
The one with the single underscore, which is just raise + a wrapper around the
one with two underscores.
2014-09-02 21:45:19 +02:00
nmlgc 9d5aa934d4 [Reduction] #445: flushall 2014-09-02 21:44:35 +02:00
nmlgc 429f134a51 [Reduction] #442-444: fseek and ftell 2014-09-02 21:04:29 +02:00
nmlgc b82a03c3f8 [Reduction] #441: grcg_round_boxfill
No local symbol definitions for variables here, as they are used in both the
normal function and the retfunc.
2014-09-01 21:08:57 +02:00
nmlgc 6250206235 [Reduction] #432-440: xxv.cpp
OK, *that's* the last piece of C++ crud shared across all main executables.
According to the object in the library file though, it seems to include one
more dword named
	__DestructorCountPtr
in the BSS segment. Neither games nor the runtime itself seem to use it, and
as a consequence, it doesn't even seem to be included in the games' BSS
segments, given that they all end with the symbols of xx.cpp...
2014-09-01 13:51:23 +02:00
nmlgc f994832a28 [Reduction] #431: toupper
Neither is this one. Also, interesting how IDA didn't identify the function in
one third of the cases.

[Binary change] Order of 2 relocations in TH03's MAINL.EXE, TH04's MAIN.EXE
and MAINE.EXE, and TH05's MAINE.EXE.
2014-09-01 12:01:35 +02:00
nmlgc 49d85b2c0b [Reduction] #430: tolower
... huh? That is certainly not multi-byte safe.
2014-09-01 12:01:34 +02:00
nmlgc 4e16a92b07 [Reduction] #429: ctype 2014-09-01 12:01:32 +02:00
nmlgc b4dd0d452d [Reduction] #428: text_puts 2014-09-01 08:58:15 +02:00
nmlgc ce245eae9f [Reduction] #427: grcg_trapezoid 2014-09-01 08:45:00 +02:00
nmlgc c90b76bff9 [Reduction] #426: graph_200line
Yup. TH03's graphics layer is actually set to a resolution of 640x200 and
upscaled by the hardware, which gives the game its distinctive low-res look.
2014-09-01 08:41:20 +02:00
nmlgc 1276b727ff [Reduction] #425: grcg_settile_1line 2014-09-01 07:39:11 +02:00
nmlgc a27c177235 [Reduction] #424: grcg_triangle 2014-09-01 07:30:58 +02:00
nmlgc 5b91dd3b42 [Reduction] #423: grcg_polygon_c 2014-09-01 06:52:28 +02:00
nmlgc b2c9ba67bc [Reduction] #422: draw_trapezoid
Yup. From TH03 on, ZUN uses a different version of master.lib, containing more
features than the last official version 0.23 from May 1995. Given the relative
insignificance of the features in question, I presume that Amusement Makers
must have kept their own, improved fork of master.lib, rather than ZUN having
hacked master.lib on his own.

Anyway, that doesn't matter. In the end, we still have to reverse-engineer each
and every one of these changes, and some of the more complicated functions
won't even be reduced before the actual reverse-engineering step of this
project.

[Binary change] Order of 3 relocations in TH05's OP.EXE.
2014-08-31 18:08:50 +02:00