| Age | Commit message (Collapse) | Author |
|---|
|
svn-id: r49095
|
|
svn-id: r48843
|
|
svn-id: r48764
|
|
svn-id: r48287
|
|
svn-id: r47556
|
|
svn-id: r47541
|
|
This makes sense as a default for CLUT8 modes, but not really
for anything else. As part of the gsoc2009-16bit merge, the
default was changed to "all ones", with extra code in the SDL
backend to truncate this to the depth of the mode. However,
"all ones" (white) still isn't a very useful default for RGB modes.
So rather than jumping through hoops to provide a bad default,
it's better to remove the default altogether. Engines which relied
on the old default of 255 have been updated to specify it explicitly.
svn-id: r47118
|
|
svn-id: r44264
|
|
Amiga version of the game.
svn-id: r44263
|
|
highlight selection is still wrong though.)
svn-id: r44231
|
|
svn-id: r44229
|
|
svn-id: r44224
|
|
Future Wars.
svn-id: r44222
|
|
"OS: Assert starting demo (regression)".
svn-id: r43743
|
|
- Now the doors actually do revolve when before they didn't.
- The fix was to alias the 9th background with the collision
page directly when loading collision page data. This way
changes written to the 9th background go to the collision
page and vice versa (And there were changes in the labyrinth).
- Also converted _bgTable from a pure array to a Common::Array
svn-id: r41453
|
|
Also fix some bugs that came up in testing of Cine::Palette.
svn-id: r41386
|
|
svn-id: r35648
|
|
refreshPalette-function compliant with disassembly.
svn-id: r35118
|
|
svn-id: r34840
|
|
Operation Stealth.
svn-id: r34838
|
|
svn-id: r34146
|
|
What little testing I did, this seems to work, but more testing
would be nice.
svn-id: r34132
|
|
Fixes bug #2058539: OS: Assert starting demo (regression).
May possibly also fix bug #2019344: FW: crash with Amiga Italian version (photocopy room),
but not sure about that because I couldn't reproduce the bug myself.
svn-id: r34065
|
|
(Learned this from trying to implement convertPI1_2 i.e. gfxConvertSpriteToRaw myself to see if that was the problem with the Operation Stealth's labyrinth arcade sequence).
svn-id: r33976
|
|
type 0 overlays (i.e. color sprites). Objects with negative frame values are supposed to be jumped over (Verified with disassembly).
svn-id: r33967
|
|
conflicted with taking screenshots using Alt-s. Great. Hopefully F11 doesn't conflict with anything useful.
svn-id: r33877
|
|
isn't used for anything else so one might as well use it for this.
svn-id: r33871
|
|
and checked the original for how the oxygen gauge during the first arcade sequence looks like. They're some kind of sprites most likely and not just simply filled rectangles).
svn-id: r33826
|
|
occasionally). Now using the Common::Rect for clipping, yay! It's good.
svn-id: r33811
|
|
shadows a previous local.
svn-id: r33808
|
|
FWRenderer::setCommand use a Common::String. Hopefully this might help with the command buffer overflow stuff, although this isn't a fix for the problem behind it, just a bandaid.
svn-id: r33793
|
|
by the opcode 0x49 'o1_setDefaultMenuBgColor'. Should fix the 'text hard to read' problems.
svn-id: r33792
|
|
(Negative colors are used for timed text boxes that are totally transparent, only the text is drawn).
svn-id: r33790
|
|
as that's what it is (No need for using malloc & free anymore). Previously we would've tried to free an array that wasn't heap-allocated in freePoldatDat (Freeing fontParamTable_standard or fontParamTable_alt), that's fixed.
svn-id: r33786
|
|
to catch out of bounds access errors which may cause memory corruption.
svn-id: r33726
|
|
catch out of bounds access errors that may cause memory corruption.
svn-id: r33725
|
|
better when it reaches zero this way although in its fully charged state it still does look a bit off. Disassembly seems to agree with not flipping the overlays so I'm going with this one for now).
svn-id: r33722
|
|
- Makes oxygen gauge during first arcade sequence visible as it gave drawPlainBox a height of zero.
Made type 21 overlays be drawn as type 22 (Filled rectangle).
- For an example the oxygen gauge during the first arcade sequence is a type 21 overlay.
Flipped type 21 & 22 overlays as it looks correct for the oxygen gauge.
These features are quite WIP, they need testing to see if they're correct.
svn-id: r33721
|
|
added with opcode 0xA4 (o2_loadMask22) and removed with opcode 0xA5 (o2_unloadMask22)). Things might be wrong so needs testing! WIP!
svn-id: r33713
|
|
function's type 20 overlay case (Previously var5 wasn't updated anywhere!). Also added a lower bound for var5's value into a comparison (Previously only the upper bound was tested for).
svn-id: r33712
|
|
Stealth to avoid possible memory corruptions.
svn-id: r33688
|
|
This may help with some memory corruption issues when for an example trying to draw the player's command string out of screen.
svn-id: r33687
|
|
Operation Stealth:
- Made waitForPlayerClick updating more like in the original.
- Moved removeMessages to after the frame drawing in main loop hoping to be more like the original.
- Added an additional test to Operation Stealth's implementation of overlay type 2 drawing.
- Added an additional parameter incrementing and testing to Operation Stealth's removeMessages.
Hopefully this won't cause any regressions in Future Wars!
svn-id: r33686
|
|
svn-id: r33669
|
|
This way the arcade sequence is completable now.
This hack should be removed later and the first
arcade sequence be made properly playable!
svn-id: r33667
|
|
- Music related settings and adBgVar0 & adBgVar1 aren't loaded currently.
Modified resetEngine to also reset more of the Operation Stealth specific variables.
Added getter for background scrolling value.
Changed additional background indices 1 & 2 from byte to uint16.
Made savegame loading functions return !in.ioFailed() as return value instead of true as previously.
svn-id: r33379
|
|
default, needs more work still. WIP!).
Added backgrounds' name saving (8 names in Operation Stealth instead of just 1 like in Future Wars).
Added 256 color palette saving and restoring (One of the palettes isn't properly handled yet though).
svn-id: r33349
|
|
- Changed savegame loading related functions to use SeekableReadStream
rather than InSaveFile so MemoryReadStream can be used transparently.
- Fixed loadResourcesFromSave to load multiframe animations correctly
and to load 0.11.0/0.11.1 Future Wars savegames which used a slightly
different format.
- Added a savegame format detector that tries to detect between the old
Future Wars savegame format, the new one and a broken revision of the
new one.
- Changed makeLoad to first load the savegame fully into memory and only
then handle it (If the savegame's packed then it's unpacked first). If
the packed savegame can't tell its unpacked size (i.e. it's using zlib
format) then we'll try to load up to 256kB of the savegame data.
Thanks to wjp for his help with nailing this release critical bug.
svn-id: r33192
|
|
svn-id: r33087
|
|
svn-id: r32951
|
