432 Commits (f122deb45cbd00ef0e1726e79dd3658b553d1eab)

Author SHA1 Message Date
Michael Niedermayer d9fc1cfec1 c optimizations 23 years ago
Michael Niedermayer d022ce5c4f c optimizations (array is faster than pointer) (16bpp variants tested and 2% faster) 23 years ago
Michael Niedermayer 8405b3fdaf minor speedup 23 years ago
Michael Niedermayer 497d4f9932 interpolate chrominance for every second line in the 1:1 vertical scale function 23 years ago
Michael Niedermayer 1faf086709 bugfixes 23 years ago
Nick Kurshev 51da31f16b cosmetic 23 years ago
Michael Niedermayer 541c4eb9ba fixed a warning 23 years ago
Nick Kurshev d9384b8b77 No longer required 23 years ago
Nick Kurshev a3aece931b Finish mmx2, 3dnow optimiz. 15to16 should be tested. Better fix of can't compile 23 years ago
Michael Niedermayer e697a1413b cant compile on non x86 bugfix 23 years ago
Michael Niedermayer 17f715fa64 patch from Martin Decky <deckm1am@ss1000.ms.mff.cuni.cz> applied and unnecassery "memory" removed 23 years ago
Michael Niedermayer 2add307d05 tell the c compiler that the memory changed 23 years ago
Michael Niedermayer 99cefd0bda 3dnow prefetch & sfence 23 years ago
Michael Niedermayer 3885847036 yv12 to yv12 scaler 23 years ago
Nick Kurshev 494a6294bf 32to24: MMX, MMX2, 3DNOW optimization 23 years ago
Nick Kurshev 90f8a479b7 More elegant solution 23 years ago
Nick Kurshev 2f6b90b0f9 preprocessor cleanup 23 years ago
Nick Kurshev 79811694c5 sfence 23 years ago
Nick Kurshev 18224bc9be More accurate prefetching 23 years ago
Nick Kurshev 96b956cc46 mmx, mmx2, 3dnow optimized 24to32 23 years ago
Nick Kurshev 49a0c6eec8 mmx optimized 24to32 23 years ago
Nick Kurshev 607d7d0373 forgot, sorry 23 years ago
Nick Kurshev fde33ab52e Prev ver could work only on x86 23 years ago
Nick Kurshev b238eb2e77 vo_vesa: more rgb2rgb support 23 years ago
Nick Kurshev 59ac5a936c vo_vesa: more rgb2rgb support 23 years ago
Nick Kurshev b234ae818b vo_vesa: rgb2rgb support 23 years ago
Michael Niedermayer 390b20a6c9 yuv2rgb bugfix 23 years ago
Michael Niedermayer 9927c7ee0d faster dering 23 years ago
Michael Niedermayer 2e2126188f dering in c 23 years ago
Pierre Lombard dda87e9fbb a few warning fixes (missing #include's) 23 years ago
Michael Niedermayer e0f8ffae99 3dnow dering 23 years ago
Michael Niedermayer cf76316aa3 just a few "not implemented yet" removed 23 years ago
Michael Niedermayer 70c5ae870b dering in mmx2 23 years ago
Michael Niedermayer 2ff198c11a downscale 23 years ago
Michael Niedermayer e5c30e0692 fixed difference with -vo md5 between doVertDefFilter() C and MMX / MMX2 versions 23 years ago
Michael Niedermayer 4e4dcbc584 much better horizontal filters (transpose & use the vertical ones) :) 23 years ago
Nick Kurshev 44d01eea32 Independed compilation of SUBDIRS 23 years ago
Michael Niedermayer 73d3355400 more speed 23 years ago
Michael Niedermayer 9736722ad7 more speed 23 years ago
Michael Niedermayer ac0b0b2f6d and another +2% speedup 23 years ago
Michael Niedermayer 37da00fcaa optimizations (+2% speedup) 23 years ago
Michael Niedermayer 534a602dca brightness / contrast fix/copy optimizations +2% speedup 23 years ago
Michael Niedermayer 57d04d3f11 negative black bugfix 23 years ago
Michael Niedermayer 911879d15e auto brightness/ contrast bugfix 23 years ago
Michael Niedermayer a525ce8db1 BGR24 bugfix 23 years ago
Michael Niedermayer d604bab901 fixed alignment (static variables where sometimes not 8-byte aligned) 23 years ago
Michael Niedermayer 5486f774bd more logic behavior if the altenative deblock filters are used (turning a alt filter on without turning the deblock filter on uses the alt filter instead of using no filter now) 23 years ago
Michael Niedermayer 84adc106cf first line bug fixed (happend if the image was scaled down or 1:1) 23 years ago
Arpi 311832de1f PP_FUNNY_STRIDE disabled 23 years ago
Michael Niedermayer a0e8aca364 added compiletime option to turn width%8==0 on 23 years ago