Age | Commit message (Collapse) | Author | |
---|---|---|---|
2009-02-13 | GUI: Fixed long-standing bug in calcGradient (gradients look a bit better now) | Max Horn | |
svn-id: r36311 | |||
2009-02-13 | Changed DotMatrix to use pixelformat instead of bitformat/colormasks | Max Horn | |
svn-id: r36310 | |||
2009-02-13 | cleanup | Max Horn | |
svn-id: r36309 | |||
2009-02-13 | cleanup | Max Horn | |
svn-id: r36289 | |||
2009-02-13 | Rewrote VectorRendererSpec & VectorRendererAA to use a ↵ | Max Horn | |
Graphics::PixelFormat, instead of relying on color masks -> should fix bug #2537212, possibly at a slow down on low-end systems (a regression which we certainly could resolve, though). Note: I left the PixelType template param in on purpose, in case we want one day to compile in both 16 and 32 bit pixel support simultaneously svn-id: r36288 | |||
2009-02-12 | GUI renderer: some cleanup; added comment on sqrt implementation (on my ↵ | Max Horn | |
system, plain sqrt is 50x faster -- on highest optimization levels, otherwise it is something like 200x) svn-id: r36277 | |||
2009-02-10 | 1555 is ABGR (not ARGB) | Joost Peters | |
svn-id: r36257 | |||
2009-02-01 | Silenced MSVC warning (MSVC is being silly here...) | Filippos Karapetis | |
svn-id: r36185 | |||
2009-02-01 | Merged r36175 from branch-0-13-0: | Marcus Comstedt | |
Fixed alpha blending in the UI dialogs by adding alpha computation to the blendPixelPtr() function. To minimize impact, pre-multiplied alpha is assumed, meaning that the blending of the R, G and B components is the same both with and without hardware alpha. svn-id: r36177 | |||
2009-01-30 | Got rid of VectorRenderer::surfacePitch() & ↵ | Max Horn | |
VectorRenderer::bytesPerPixel(): the latter was not used, for the former it is silly to go through a virtual method to compute that value; all in all, this code performs too many unnecessary divisions ;) svn-id: r36151 | |||
2009-01-30 | Removed unused area convolution code from VectorRenderer | Max Horn | |
svn-id: r36148 | |||
2009-01-30 | do not include common/debug.h from common/util.h | Max Horn | |
svn-id: r36143 | |||
2009-01-29 | A ton of code formatting fixes; also fixed warnings about single line loops ↵ | Max Horn | |
like 'while(cond);' by inserting newlines svn-id: r36127 | |||
2009-01-27 | Fixed hq2x ASM scaler | Max Horn | |
svn-id: r36115 | |||
2009-01-27 | Re-enabled ASM versions of HQx scalers (oops) | Max Horn | |
svn-id: r36099 | |||
2009-01-27 | Added new interpolate16_1_1_1_1 func, got rid of interpolate32_1_1_1_1 | Max Horn | |
svn-id: r36090 | |||
2009-01-27 | Replaced small lookup table in 2xsai by direct computation | Max Horn | |
svn-id: r36089 | |||
2009-01-27 | Removed interpolate16_2 and interpolate16_3 | Max Horn | |
svn-id: r36088 | |||
2009-01-27 | Speed up HQ2x/HQ3x C++ version a bit (about 20% faster with -O3 on my system) | Max Horn | |
svn-id: r36087 | |||
2009-01-27 | New code for interpolate32_1_1 overflowed in 565 mode, so reverting to the ↵ | Max Horn | |
old code (with some variable name changes to match the other interpolators) svn-id: r36086 | |||
2009-01-26 | Removed use of LUT16to32 in HQx asm versions, replacing some MMX code with ↵ | Max Horn | |
'plain' x86 code. Advantage: got rid of a 256kb table (reduces cache load, so over here the code is about as fast as before; in particular, since the affected interpolators are not used that often, it seems). Moreover, the new code is more accurate than the old ASM code, which actually differed from what our C++ HQx did (sacrificing precision for speed, i.e., cheating ;-) svn-id: r36078 | |||
2009-01-26 | Slightly improved interpolate32_1_1 | Max Horn | |
svn-id: r36075 | |||
2009-01-26 | error out when no renderer was created due to unknown/unsupported PixelFormat | Joost Peters | |
svn-id: r36074 | |||
2009-01-25 | TUCKER: modified flic playback to make use of dirtyrects | Gregory Montoir | |
svn-id: r36060 | |||
2009-01-25 | Fixed compilation | Filippos Karapetis | |
svn-id: r36054 | |||
2009-01-25 | Renamed lowBits -> kLowBitsMask and highBits -> kHighBitsMask | Max Horn | |
svn-id: r36049 | |||
2009-01-24 | HQX: Removed some more branches (like in my previous commit) from ↵ | Max Horn | |
interpolation macros (they are meant to speed up things, but in reality cause a slowdown -- branches are expensive on modern CPUs) svn-id: r36047 | |||
2009-01-24 | Turned two vars of the HQ2x/HQ3x ASM implementation into global vars, to ↵ | Max Horn | |
make it possible to adjust them for 555 vs. 565 mode (555 mode is still a bit buggy, due to the interpolation code they use) svn-id: r36046 | |||
2009-01-24 | Moved 565 mode specific bits of HQ2x/HQ3x asm to 'variables' | Max Horn | |
svn-id: r36034 | |||
2009-01-23 | Changed Graphics::ImageDecoder to allow custom PixelFormats | Max Horn | |
svn-id: r36026 | |||
2009-01-23 | Remark that the HQnx ASM code only works in 565 mode | Max Horn | |
svn-id: r36025 | |||
2009-01-23 | Added some comments to hq2x/hq3x asm code | Max Horn | |
svn-id: r36024 | |||
2009-01-23 | Renamed Archive::openFile to createReadStreamForMember | Max Horn | |
svn-id: r36021 | |||
2009-01-23 | Made InitLUT use a PixelFormat instead of a ColorMask | Max Horn | |
svn-id: r36008 | |||
2009-01-22 | Added an '!=' operator to PixelFormat. | Johannes Schickel | |
svn-id: r35998 | |||
2009-01-22 | Speed up HQ2x/HQ3x C versions by reordering tests in diffYUV: most likeliest ↵ | Max Horn | |
to be taken now is first, least likely is last svn-id: r35996 | |||
2009-01-22 | Got rid of several occurrences of gBitFormat | Max Horn | |
svn-id: r35995 | |||
2009-01-22 | Got rid of gBitFormat in thumbnail scaler code (and even simplified it in ↵ | Max Horn | |
the process) svn-id: r35994 | |||
2009-01-22 | Moved Graphics::PixelFormat into its own header file; turned RGBToColor etc. ↵ | Max Horn | |
into methods, and added an operator== svn-id: r35993 | |||
2009-01-20 | Moved Surface::blit to backends/vkeybd/virtual-keyboard-gui.cpp | Max Horn | |
svn-id: r35965 | |||
2009-01-20 | merge SurfaceKeyColored into Surface | Andre Heider | |
svn-id: r35951 | |||
2009-01-11 | Stop recalculating the x and y coordinates for every frame in playVideo() | Filippos Karapetis | |
svn-id: r35819 | |||
2009-01-11 | Oops... *really* fixed compilation | Filippos Karapetis | |
svn-id: r35818 | |||
2009-01-11 | Fixed compilation | Filippos Karapetis | |
svn-id: r35817 | |||
2009-01-11 | Committed a modified version of wjp's patch for the video player: | Filippos Karapetis | |
- Split the video player from the video decoders. It's now possible to have one video player for multiple decoders - Added the palette weight calculation from the BS1 engine into VideoPlayer::setPalette. It's now possible to find the values of the white and black colors via getWhite() and getBlack() (useful for subtitle overlays) - Adapted FTA2's movie playing code to the new changes to video player - Fixed a slight bug in the DXA decoder (_videoinfo.startTime was not initialized) svn-id: r35816 | |||
2009-01-11 | Merge in Virtual Keybpard & KeyMapper branch, | Eugene Sandulenko | |
- Merge is perfromed in order to not let rotting the code - Makefile modifications were avoided Stuff to resolve: - Circular dependency of common/vkeybd from graphics/ - Make it compilable (?) - Add some keyboards - Decide on the key bindings svn-id: r35813 | |||
2009-01-07 | Reduced header deps | Max Horn | |
svn-id: r35781 | |||
2009-01-07 | videoIsLoaded() -> isVideoLoaded() | Filippos Karapetis | |
svn-id: r35777 | |||
2009-01-07 | Clear the screen to black before showing a video with playVideo() (it's ↵ | Filippos Karapetis | |
meant to play full screen videos) svn-id: r35776 | |||
2009-01-07 | Initialize DXA buffers to 0 when they're allocated | Filippos Karapetis | |
svn-id: r35775 |