Redo the forward 4x4 dct
The new fdct lowers the round trip sum squared error for a 4x4 block ~0.12. or ~0.008/pixel. For reference, the old matrix multiply version has average round trip error 1.46 for a 4x4 block. Thanks to "derf" for his suggestions and references. Change-Id: I5559d1e81d333b319404ab16b336b739f87afc79
Showing
- vp8/encoder/block.h 0 additions, 5 deletionsvp8/encoder/block.h
- vp8/encoder/dct.c 16 additions, 125 deletionsvp8/encoder/dct.c
- vp8/encoder/dct.h 0 additions, 10 deletionsvp8/encoder/dct.h
- vp8/encoder/encodeintra.c 1 addition, 1 deletionvp8/encoder/encodeintra.c
- vp8/encoder/encodemb.c 26 additions, 13 deletionsvp8/encoder/encodemb.c
- vp8/encoder/ethreading.c 0 additions, 3 deletionsvp8/encoder/ethreading.c
- vp8/encoder/generic/csystemdependent.c 2 additions, 2 deletionsvp8/encoder/generic/csystemdependent.c
- vp8/encoder/onyx_if.c 0 additions, 6 deletionsvp8/encoder/onyx_if.c
- vp8/encoder/rdopt.c 2 additions, 2 deletionsvp8/encoder/rdopt.c
- vp8/encoder/x86/csystemdependent.c 20 additions, 6 deletionsvp8/encoder/x86/csystemdependent.c
- vp8/encoder/x86/dct_mmx.asm 33 additions, 359 deletionsvp8/encoder/x86/dct_mmx.asm
- vp8/encoder/x86/dct_x86.h 2 additions, 11 deletionsvp8/encoder/x86/dct_x86.h
- vp8/encoder/x86/x86_csystemdependent.c 16 additions, 18 deletionsvp8/encoder/x86/x86_csystemdependent.c
- vp8/vp8cx.mk 0 additions, 1 deletionvp8/vp8cx.mk
Loading
Please register or sign in to comment