Age | Commit message (Collapse) | Author |
|
svn-id: r10300
|
|
screwed up)
svn-id: r10299
|
|
svn-id: r10298
|
|
svn-id: r10297
|
|
svn-id: r10296
|
|
svn-id: r10295
|
|
svn-id: r10294
|
|
This slows down the build procedure enormously, but on the other hand it
actually does something now. :-)
Though, for some reason, make always crashes shortly before the end when I
build it from scratch. I have to repeat the command to get it to finish.
Time to check if there's a newer MinGW version to download, I guess...
svn-id: r10293
|
|
work here, I'll add more build rules. Work in progress, a lot ain't work, but the plugin stuff is disabled by default, so no harm should arise
svn-id: r10292
|
|
svn-id: r10291
|
|
svn-id: r10290
|
|
to get rid of all of them, I'm thinking about it :)
svn-id: r10289
|
|
bound to hit 1.0 any year now.
Of course, this is completely untested.
svn-id: r10288
|
|
barebone ScummVM (or maybe I just want to increase our CVS stats? <g>)
svn-id: r10287
|
|
svn-id: r10286
|
|
the lib.a file
svn-id: r10285
|
|
svn-id: r10284
|
|
Scumm engine will do the work from then
svn-id: r10283
|
|
(this removes the need for an ugly hack in the build system, and is also conceptionally cleaner)
svn-id: r10282
|
|
svn-id: r10281
|
|
svn-id: r10280
|
|
svn-id: r10279
|
|
simplify some code; added a global g_sound pointer in bs2, this cuts down on uses of g_sword2 (of course both should be removed on the long run); some other minor tweaks/fixes
svn-id: r10278
|
|
svn-id: r10277
|
|
svn-id: r10276
|
|
svn-id: r10275
|
|
svn-id: r10274
|
|
svn-id: r10273
|
|
Direct_read_file() and Direct_write_file().
This change is completely untested, but that should be ok since Read_file()
(which is now the only function left in tony_gsdk.cpp) only seems to be
used in debug builds.
Perhaps we should remove these files completely?
svn-id: r10272
|
|
have to clean up the header files at some point...
svn-id: r10271
|
|
svn-id: r10270
|
|
svn-id: r10269
|
|
Since this file contains a list of global variables which are accessed
both directly by the engine and - presumably - by the scripts, this could
lead to some nasty regressions. I've triple-checked the list against the
old version, so it *should* be ok, I just thought I'd point it out.
svn-id: r10268
|
|
svn-id: r10267
|
|
svn-id: r10266
|
|
Implemented a parser for Euphony music. No
FM instrument support yet, as the FM chip used
by FM Towns is not being emulated yet. In the
meantime, a stock FM-emulated GM instrument is
being used instead.
This at least makes the Zak Towns kazoo tune
and the Loom Towns distaff audible. Emulation
of the FM Towns synth chip, or suitable
emulation using the OPL2 synth, is still
under investigation.
svn-id: r10265
|
|
svn-id: r10264
|
|
svn-id: r10263
|
|
particularly noticeable in games with PC speaker emulation, but I could
hear faint noises in AdLib music as well.)
svn-id: r10262
|
|
performance by omitting redundancies
Removed the use of the _bank member for iMuse Parts.
The member is still being saved/loaded, and a
warning will display if a non-zero _bank value is
set. If it turns out some musical passage is actually
using the _bank, we'll need to reinstate it.
Otherwise, after sufficient testing we can obsolete
it in the savegames.
svn-id: r10261
|
|
svn-id: r10260
|
|
svn-id: r10259
|
|
svn-id: r10258
|
|
svn-id: r10257
|
|
svn-id: r10256
|
|
svn-id: r10255
|
|
Implemented _cmd_queue save/load. In addition to requiring
_cmd_queue information, this bug arises from a rare assumption
that sound resources are loaded in memory even though they
aren't currently playing. Therefore, a list of sound resources
loaded in memory is included in the savegame, so that all
relevant sound resources are reloaded when the savegame is
loaded. This also fixes an unreported music bug in S&M when
saving a game while outside the Bumpusville mansion.
As a result of savegame format modifications, we are now at
savegame version 23.
svn-id: r10254
|
|
access of the actor position
svn-id: r10253
|
|
bug... ;)
svn-id: r10252
|
|
original, but I think to make it completely right a bigger change to
the adlib backend would be necessary.
svn-id: r10251
|