Age | Commit message (Collapse) | Author | |
---|---|---|---|
2011-05-12 | GIT: Clean up: Suppress SVN tags, now useless | strangerke | |
2011-05-03 | ANDROID: Fix various forbidden symbol clashes | Max Horn | |
2011-04-17 | ALL/GRAPHICS: Remove Surface::bytesPerPixel. | Johannes Schickel | |
2011-04-17 | BACKENDS: Adapt various backends code to set up Surface::format correctly. | Johannes Schickel | |
Note that this change is not tested at all (not even compile wise!). | |||
2011-04-05 | ANDROID: Remove code for paletted textures | dhewg | |
Unused now, because the performance isn't good enough on weak GLES drivers. | |||
2011-04-05 | ANDROID: Use a faked paletted texture for CLUT cursors | dhewg | |
Same change as for the game screen, reduces CPU usage a little | |||
2011-03-25 | ANDROID: Always set the surface properties | dhewg | |
Didn't happen on the shortcut, which led to wrong properties and surface functions scribbling over memory | |||
2011-03-24 | ANDROID: Fix texture double release | dhewg | |
2011-03-24 | ANDROID: Cleanup | dhewg | |
2011-03-14 | ANDROID: Cleanup | dhewg | |
Removed not required checks since we now buffer texture contents | |||
2011-03-14 | ANDROID: Remove clearBuffer() | dhewg | |
Not required. Why did i add that again? | |||
2011-03-14 | ANDROID: Clear fake palette after allocating | dhewg | |
Gets rid of funky gfx artifacts on the overlay | |||
2011-03-13 | ANDROID: Add graphics mode for linear filtering | dhewg | |
2011-03-13 | ANDROID: Fix texture clear color | dhewg | |
2011-03-13 | ANDROID: Some texture convinience functions | dhewg | |
2011-03-13 | ANDROID: Prevent unnecessary reallocs | dhewg | |
2011-03-13 | ANDROID: Buffer 16bit texture contents | dhewg | |
Same issue as in the last commit: glTexSubImage2D is slow, so cache all copyRect*() calls in a buffer, and update the dirty rect once when drawing. Reduces CPU usage on 16bit games significantly. Also, lockScreen() returns now pixel data for non-CLUT8 games instead of asserting. | |||
2011-03-13 | ANDROID: Don't use compressed textures for the game screen | dhewg | |
Some GLES drivers suck so much that uploading data to the GPU takes ages. CLUT8 games now use a faked paletted texture, which internally uses a RGB565 hardware texture (Android's native pixel format). This seems to be the only way to efficiently implement constant changing textures with GLES1 - at the cost of extra buffers. Then again, we can now use glTexSubImage2D to only update the dirty rects, which wasn't possible before because glCompressedTexSubImage2D is only usable on GLES2. This commit does exactly that. Overall, the CPU usage is massively reduced for CLUT8 games. | |||
2011-03-12 | ANDROID: Remove redundant variable | dhewg | |
2011-03-12 | ANDROID: Add more paletted texture types | dhewg | |
2011-03-12 | ANDROID: Cleanup paletted textures | dhewg | |
2011-03-07 | ANDROID: Texture cleanup | dhewg | |
2011-03-05 | ANDROID: Fix warning | dhewg | |
2011-03-05 | ANDROID: Proper fillBuffer() for non CLUT8 colors | dhewg | |
2011-03-05 | ANDROID: Add initial 16bit gfx support | dhewg | |
Supported pixel formats: 565, 5551, 4444 Missing: 555 (doesn't exist on GLES) | |||
2011-03-05 | ANDROID: Reduce mem usage of fillBuffer() | dhewg | |
2011-03-05 | ANDROID: Merge updateTexture() | dhewg | |
2011-03-05 | ANDROID: Remove some vtable overhead on textures | dhewg | |
2011-03-04 | ANDROID: Cleanup | dhewg | |
2011-03-03 | ANDROID: Don't wipe paletted textures on reinit | dhewg | |
The content of a paletted texture is in _surface.pixels. When recreating the surface, don't wipe its data unnecessarily. This fixes gfx garbage on CLUT8 games/cursors when leaving and going back to ScummVM | |||
2011-03-03 | ANDROID: Formatting/whitespaces | dhewg | |
2011-03-02 | ANDROID: Properly release texture resources | dhewg | |
When calling glDeleteTextures() we need a valid surface. | |||
2011-03-02 | ANDROID: cleanup | dhewg | |
2011-02-24 | ANDROID: Let's not do that on the stack | dhewg | |
2011-02-24 | ANDROID: Split code into multiple files | dhewg | |
And get rid of unnecessary JNI calls to get a pointer to g_system |