1. 07 Jan, 2011 3 commits
  2. 06 Jan, 2011 2 commits
    • Johann's avatar
      x86 sse2 temporal_filter_apply · 8b0cf5f7
      Johann authored
      count can be reduced to short because the max number of filtered frames
      is set to 15. the max value for any frame is 32 (modifier = 16,
      filter_weight = 2). 15*32 = 480 which requires 9 bits
      
      this function goes from about 7000 us / 1000 iterations for the C code
      to < 275 us / 1000 iterations for sse2 for block_size = 16 and from
      about 1800 us / 1000 iters to < 100 us / 1000 iters for block_size = 8
      
      Change-Id: I64a32607f58a2d33c39286f468b04ccd457d9e6e
      8b0cf5f7
    • Paul Wilkins's avatar
      Disable some features for first pass. · 431dac08
      Paul Wilkins authored
      The following features don't make sense for the first
      pass in its current form and have a significant impact on its
      speed (up to 50%).
      
      Slow quantizer, slow dct and trellis optimization.
      
      Change-Id: Id9943f6765ffbd71fc0084ec7dfbc9d376fd6fcd
      431dac08
  3. 04 Jan, 2011 1 commit
    • Paul Wilkins's avatar
      Adjustment to boost calculation in two pass. · b095d9df
      Paul Wilkins authored
      Calculate a minimum intra value to be used in determining the
      IIratio scores used in two pass, second pass.
      
      This is to make sure sections that are low complexity" in the
      intra domain are still boosted appropriately for KF/GF/ARF.
      
      For now I have commented out the Q based adjustment of
      KF boost.
      
      Change-Id: I15deb09c5bd9b53180a2ddd3e5f575b2aba244b3
      b095d9df
  4. 29 Dec, 2010 2 commits
    • Scott LaVarnway's avatar
      Fixed encoder crash when mult-threading is enabled. · de4e8185
      Scott LaVarnway authored
      Happens in real-time mode.  Will happen in good quality, speed 1.
      
      Change-Id: I3e5b68827b1a5798d0431b088a709256d1ce2c95
      de4e8185
    • Yunqing Wang's avatar
      Always update last_frame_type · a864678c
      Yunqing Wang authored
      Scott pointed out that last_frame_type only gets updated while
      loopfilter exists. Since last_frame_type is also needed in
      motion search now, it needs to be updated every frame.
      
      Change-Id: I9203532fd67361588d4024628d9ddb8e391ad912
      a864678c
  5. 28 Dec, 2010 5 commits
  6. 26 Dec, 2010 1 commit
  7. 24 Dec, 2010 1 commit
    • Yaowu Xu's avatar
      adjusted sad_per_bit to correlate with quantizer · 0f5264b5
      Yaowu Xu authored
      Re-calibrated sad_per_bit16 and sad_per_bit4 tables to linearly
      correlated to quantizer values, these two variables are used in
      motion search for costing motion vectors. This change has an small
      positive effect on compression.
      
      Change-Id: Ic9b5ea6fb8d5078ef663ba4899db019cc51f4166
      0f5264b5
  8. 23 Dec, 2010 4 commits
  9. 22 Dec, 2010 5 commits
    • John Koleszar's avatar
      bd9b383d
    • John Koleszar's avatar
      make yasm generate cv8 debug data on win32 · 30830d5a
      John Koleszar authored
      Native Windows targets should use CV8 format debugging symbols, not
      DWARF.
      
      Change-Id: I9489163fcd9d749b72f6c70ecbce67a6f0790802
      30830d5a
    • Johann's avatar
      improve integer version of filter · 20b855c3
      Johann authored
      the lookup table is based on floating point calculations (see source)
      
      by moving the *3 before the downshift and adding the rounding bit, the
      delta (LUT - integer) goes from:
      ______________________________________
      __ 1__ 1______________________________
      __ 1__ 1______________________________
      ____ 1______ 1________________________
      ____ 1 2__ 2 1________________________
      ______ 1 1 2__ 2__ 2__ 2 1 1__________
      ________ 1 1 2 2__ 1 2 3 1 2__ 2__ 2__
      to:
      __-1__-1______________________________
      ______________________________________
      ____-1______-1________________________
      ______________________________________
      ________-1______________-1____________
      ______________________________________
      
      it's important to be able to use the integer version because the LUT
      more or less precludes SIMD optimizations
      
      Change-Id: I45a81127dc7b72a06fba951649135d9d918386c0
      20b855c3
    • Johann's avatar
      temporal filter naming changes · 4b6219cb
      Johann authored
      be more consistant with the naming pattern, especially wrt rtcd
      
      Change-Id: I3df50686a09f1dab0a9620b5adbb8a1577b40f2f
      4b6219cb
    • Johann's avatar
      abstract apply_temporal_filter · 092b5bef
      Johann authored
      allow for optimized versions of apply_temporal_filter
      (now vp8_apply_temporal_filter_c)
      
      the function was previously declared as static and appears to have been
      inlined. with this change, that's no longer possible. performance takes
      a small hit.
      
      the declaration for vp8_cx_temp_filter_c was moved to onyx_if.c because
      of a circular dependency. for rtcd, temporal_filter.h holds the
      definition for the rtcd table, so it needs to be included by onyx_int.h.
      however, onyx_int.h holds the definition for VP8_COMP which is needed
      for the function prototype. blah.
      
      Change-Id: I499c055fdc652ac4659c21c5a55fe10ceb7e95e3
      092b5bef
  10. 20 Dec, 2010 1 commit
  11. 17 Dec, 2010 4 commits
    • John Koleszar's avatar
      propagate user private data on decode · c49f49b1
      John Koleszar authored
      The pointer passed in the user_priv argument to vpx_codec_decode()
      should be propagated through to the corresponding output frame and
      made available in the image's user_priv member. Fixes issue #252
      
      Change-Id: I182746a6882c8549fb146b4a4fdb64f1789eb750
      c49f49b1
    • John Koleszar's avatar
    • John Koleszar's avatar
      Add psnr/ssim tuning option · b0da9b39
      John Koleszar authored
      Add a new encoder control, VP8E_SET_TUNING, to allow the application
      to inform the encoder that the material will benefit from certain
      tuning. Expose this control as the --tune option to vpxenc. The args
      helper is expanded to support enumerated arguments by name or value.
      
      Two tunings are provided by this patch, PSNR (default) and SSIM.
      Activity masking is made dependent on setting --tune=ssim, as the
      current implementation hurts speed (10%) and PSNR (2.7% avg,
      10% peak) too much for it to be a default yet.
      
      Change-Id: I110d969381c4805347ff5a0ffaf1a14ca1965257
      b0da9b39
    • Henrik Lundin's avatar
      Inform caller of decoder about updated references · 2a87491f
      Henrik Lundin authored
      Inform the caller of the decoder if a decoded frame updated last,
      golden, or altref frames, required for realtime communication
      proposed in document VP8 RTP payload format.
      
      Added a new vpx_codec_control called VP8D_GET_LAST_REF_UPDATES, to be
      called after vpx_codec_decode. The control will indicate which of the
      reference frames that were updated by setting the 3 LSBs in the input
      int (pointer).
      
      Change-Id: Iac9db60dac414356c7ffa0b0fede88cb91e11bd7
      2a87491f
  12. 16 Dec, 2010 2 commits
    • Scott LaVarnway's avatar
      Changed segmentation check order · 64baa8df
      Scott LaVarnway authored
      In SPLITMV, the 8x8 segment will be checked first.  If the 8x8 rd
      is better than the best, we check the other segments.  Otherwise
      bail.  Adjustments to the thresh_mult were necessary to make
      up for the initial quality loss.
      The performance improved by 20% (average) for good quality,
      speed 0 and speed 1, while the overall quality remained the same.
      
      Change-Id: I717aef401323c8a254fba3e9777d2a316c774cc3
      64baa8df
    • Scott LaVarnway's avatar
      Adjusted breakout RD for SPLITMV · 81cdeb71
      Scott LaVarnway authored
      vp8_rd_pick_best_mbsegmentation looks at y only.  The new
      breakout does not include the frame cost, the prob_skip_false
      cost, or the uv rate.  Performance improved by a few percent
      and the quality remained the same.
      
      Change-Id: I94ff013998ac51e8ecce7130870f7b6600758e15
      81cdeb71
  13. 15 Dec, 2010 1 commit
  14. 14 Dec, 2010 6 commits
  15. 13 Dec, 2010 2 commits