1. 30 Jan, 2012 2 commits
  2. 29 Jan, 2012 1 commit
  3. 12 Jan, 2012 1 commit
    • Christophe GISQUET's avatar
      rv34: DC-only inverse transform · 3faa303a
      Christophe GISQUET authored
      
      
      When decoding coefficients, detect whether the block is DC-only, and take
      advantage of this knowledge to perform DC-only inverse transform.
      
      This is achieved by:
      - first, changing the 108x4 element modulo_three_table into a 108 element
        table (kind of base4), and accessing each value using mask and shifts.
      - then, checking low bits for 0 (as they represent the presence of higher
        frequency coefficients)
      
      Also provide x86 SIMD code for the DC-only inverse transform.
      Signed-off-by: default avatarKostya Shishkov <kostya.shishkov@gmail.com>
      3faa303a
  4. 09 Jan, 2012 1 commit
  5. 19 Dec, 2011 1 commit
  6. 14 Dec, 2011 1 commit
  7. 11 Oct, 2011 1 commit
  8. 11 Aug, 2011 1 commit
  9. 03 Jul, 2011 1 commit
  10. 21 Jun, 2011 1 commit
  11. 18 Jun, 2011 1 commit
  12. 05 Jun, 2011 1 commit
  13. 31 May, 2011 1 commit
  14. 21 May, 2011 1 commit
  15. 19 May, 2011 1 commit
  16. 11 May, 2011 1 commit
  17. 12 Mar, 2011 1 commit
  18. 10 Feb, 2011 1 commit
  19. 02 Feb, 2011 1 commit
  20. 17 Sep, 2010 1 commit
  21. 14 Sep, 2010 1 commit
    • Ronald S. Bultje's avatar
      Rename h264_idct_sse2.asm to h264_idct.asm; move inline IDCT asm from · 1d16a1cf
      Ronald S. Bultje authored
      h264dsp_mmx.c to h264_idct.asm (as yasm code). Because the loops are now
      coded in asm instead of C, this is (depending on the function) up to 50%
      faster for cases where gcc didn't do a great job at looping.
      
      Since h264_idct_add8() is now faster than the manual loop setup in h264.c,
      in-asm idct calling can now be enabled for chroma as well (see r16207). For
      MMX, this is 5% faster. For SSE2 (which isn't done for chroma if h264.c does
      the looping), this makes it up to 50% faster. Speed gain overall is ~0.5-1.0%.
      
      Originally committed as revision 25119 to svn://svn.ffmpeg.org/ffmpeg/trunk
      1d16a1cf
  22. 10 Sep, 2010 1 commit
  23. 08 Sep, 2010 1 commit
  24. 03 Sep, 2010 1 commit
  25. 01 Sep, 2010 2 commits
  26. 31 Aug, 2010 1 commit
  27. 30 Aug, 2010 3 commits
  28. 25 Aug, 2010 2 commits
  29. 07 Aug, 2010 1 commit
  30. 05 Aug, 2010 1 commit
  31. 31 Jul, 2010 1 commit
  32. 22 Jul, 2010 1 commit
  33. 11 Jul, 2010 1 commit
  34. 06 Jul, 2010 1 commit
  35. 27 Jun, 2010 1 commit