Age | Commit message (Collapse) | Author |
|
svn-id: r13661
|
|
svn-id: r13649
|
|
MSVC6's idiosyncracies.
svn-id: r13644
|
|
I've also made the SaveGameHeader struct packed, which may break savegame
compatibility on some architectures (though not on the Linux and Windows
boxes I've tried it on). But I'm hoping it will guarantee, or at least make
it more likely, that savegames will be portable across architectures.
svn-id: r13634
|
|
cache after it's been closed. (Currently it always is, but ideally I'd like
for BS to work even if resource caching is disabled.)
svn-id: r13610
|
|
didn't make. :-)
(That is, I made the change and then thought better of it, but I forgot to
change one line back to its original form.)
svn-id: r13608
|
|
and I may as well clear it when a resource is read from disk as well.
svn-id: r13606
|
|
resource manager. All new code! All new bugs!
svn-id: r13603
|
|
svn-id: r13582
|
|
svn-id: r13579
|
|
svn-id: r13495
|
|
svn-id: r13494
|
|
svn-id: r13493
|
|
messages to be more like in BS1.)
svn-id: r13491
|
|
svn-id: r13457
|
|
svn-id: r13431
|
|
This removes a bunch of debugging code/commands that either didn't do
anything useful under ScummVM (e.g. "soft" and "hard"), or which did things
that was already easily avaiable elsewhere (e.g. "save" and "restore").
I didn't have the heart to remove the "tony" command, though. :-)
svn-id: r13422
|
|
I think the reason I didn't do this from the start was that BS2 used to
call clearScene(), or whatever the function was called back then, between
every frame. Nowadays, it simply assumes that each frame will cover the
previous one.
Anyway, this change prevents the restart/restore dialog from appearing
briefly between the two intro cutscene animations.
svn-id: r13421
|
|
svn-id: r13410
|
|
svn-id: r13407
|
|
svn-id: r13406
|
|
svn-id: r13405
|
|
One of the changes, I'm not quite sure about: buildDisplay() used to open
and close the _thisScreen.background_layer_id resource for each layer it
processed. In particular, it used to "release the screen resource before
cacheing the sprites".
I have no idea why, because I can't see any trace of a sprite cache, and I
can't think of any harm in keeping the resource open during the whole
render cycle. The resource is probably loaded into memory already anyway,
though its reference counter may be 0.
svn-id: r13401
|
|
be needed when using the -x command-line option to restore a game, though
I'm not quite sure why.
svn-id: r13400
|
|
And a teensy-weensy little cleanup.
svn-id: r13399
|
|
overlay, that needs to be cleared at the beginning of the movie. Otherwise
what was on the screen may show at the end of the movie.
svn-id: r13398
|
|
of how the savegame is loaded. (ScummVM adds two alternative methods: the
-x command-line parameter, and the restart/restore dialog at the beginning
of the game, which is only shown when there are savegames available.)
svn-id: r13386
|
|
svn-id: r13385
|
|
Invalidate the lookup table when the screen changes. (TODO: We also have to
invalidate it if the change happens between cutscenes, don't we?)
Some cleanup, particularly in the BS2 cutscene player. More needed, I
guess...
svn-id: r13377
|
|
svn-id: r13365
|
|
svn-id: r13361
|
|
extremely grateful that the BS1 and BS2 cutscene players are finally
sharing at least some of their code now.)
svn-id: r13360
|
|
svn-id: r13359
|
|
start)
svn-id: r13358
|
|
svn-id: r13357
|
|
svn-id: r13354
|
|
it's Logic::_scriptVars[ID] instead of just ID. Apart from looking cool, it
makes it much easier to tell the difference between variables and constants
when looking at the code.
Of course, this sort of sweeping changes is jolly good for introducing
truly weird regressions, which is why I waited until after 0.6.0.
svn-id: r13331
|
|
svn-id: r13286
|
|
should only check if the music is fading, not in which direction. (Also
made a minor cleanup.)
svn-id: r13235
|
|
svn-id: r13180
|
|
the game starts. (I know it could look prettier, but I don't have much to
work with here...)
svn-id: r13178
|
|
into the chunk of memory handled by the memory manager, not a well-defined
value. Also, it introduced a theoretical memory leak as fnPlaySequence()
wouldn't get to clean up after itself.
The correct place for this check is probably in MoviePlayer::play(), where
you can test the name of the cutscene. On the other hand, the original demo
will gladly play eye.smk if there is one, so why shouldn't we?
svn-id: r13139
|
|
svn-id: r13118
|
|
svn-id: r13107
|
|
Prevent false warnings in demo.
svn-id: r13102
|
|
NewGuiColor to OverlayColor; fixed some calls to error() in the SDL backend
svn-id: r13087
|
|
OSystem API is not yet finished); porters will have to fix their ports to get them to compile again
svn-id: r13036
|
|
svn-id: r12996
|
|
svn-id: r12984
|
|
svn-id: r12974
|