1. 27 Mar, 2014 2 commits
    • Paul Wilkins's avatar
      Experimental change to speed 1 settings. · 32c4fdec
      Paul Wilkins authored
      Allow tx search for ARF and GF helps quality but a little slower.
      Setting subpel_iters_per_step to 1 improves encode speed.
      Setting sf->mode_skip_start = 10 improves speed.
      
      Initial local results suggest overall impact on quality is neutral
      but encode is up to 15% faster.
      
      Change-Id: Ibde02cae6626a44c10a1da0cefe888afbb51f037
      32c4fdec
    • Marco Paniconi's avatar
      Move aq_mode=2 (complexity_aq) to separate file. · 2b06bf20
      Marco Paniconi authored
      Change-Id: Iffa45b9b04196c1ded6037622a8644a2500a62de
      2b06bf20
  2. 26 Mar, 2014 1 commit
  3. 25 Mar, 2014 3 commits
    • Yaowu Xu's avatar
      Added a speed feature "allow_skip_recode" · 6eb2cacf
      Yaowu Xu authored
      This commit added a speed feature to make the logic of calculating
      skip_recode on a block level more explicit. This also enable the
      feature to be enabled at speed 5 where the previous logic is too
      conservative, help gain back the lost speed for --rt(-5).
      
      Change-Id: Ieb37ca3e85c2e7bda343486edf13d5f5395f2233
      6eb2cacf
    • Yaowu Xu's avatar
      Removed redundant settings · de8fd6f4
      Yaowu Xu authored
      The recode_loop feature is disabled for one-pass always.
      
      Change-Id: I6de8820add4a6f97196874af83a970f9898f842d
      de8fd6f4
    • Dmitry Kovalev's avatar
      Moving above_context to VP9_COMMON. · ed39c40a
      Dmitry Kovalev authored
      Change-Id: I713af99d1e17e05a20eab20df51d74ebfd1a68d2
      ed39c40a
  4. 24 Mar, 2014 4 commits
  5. 21 Mar, 2014 4 commits
  6. 20 Mar, 2014 1 commit
  7. 19 Mar, 2014 2 commits
    • Jingning Han's avatar
      Enable variable block size test in non-RD mode decision · 60f9ebc3
      Jingning Han authored
      This is an initial attempt to allow variable block size partition
      in non-RD coding flow. It tests 8x8, 16x16 and 32x32 block size per
      64x64 block, all using non-RD mode decision and the associated rate
      distortion costs from modeling, then selects the best block size to
      encode the entire 64x64 block. Such operations are triggered every
      other 3 frames. The blocks of intermediate frames will reuse the
      collocated block's partition type.
      
      It improves the compression performance by 13.2%. Note that the gains
      are not evenly distributed. For many hard clips, the compression
      performance is improved by 20% to 28%. Local speed test shows that
      it will also increase runtime by 50%, as compared to speed -7. It is
      now enabled in speed -6 setting.
      
      Change-Id: Ib4fb8659d21621c9075b3c369ddaa9ecb0a4b204
      60f9ebc3
    • Minghai Shang's avatar
      [svc] Finalize spatial svc first pass rate control · d2053350
      Minghai Shang authored
            1. Save stats for each spatial layer
            2. Add frame buffer management for svc first pass rc
            3. Set default spatial layer to 1
            4. Flush encoder at the end of stream in test app
      This only supports spatial svc.
      Change-Id: Ia89cfa87bb6394e6c0405b921d86c426d0a0c9ae
      d2053350
  8. 18 Mar, 2014 2 commits
  9. 17 Mar, 2014 2 commits
  10. 13 Mar, 2014 3 commits
  11. 12 Mar, 2014 2 commits
    • Deb Mukherjee's avatar
      Preliminary code for variance based paritioning · e39ecfaa
      Deb Mukherjee authored
      Brings back most of Jim's previous patch for choosing
      partitioning based on variance while making it compatible
      with the current state of the code. Also adds a
      nonrd_use_partition() function to recursively encode  for any
      arbitrary sb_type decisions within a 64x64 block; and
      includes some refactoring.
      
      Currently, when the VAR_BASED_PARTITIONING mode is turned on
      for speed 7, there is a 10+% speed-up observed.
      
      Experiments/improvements with this new partitioning method
      will be conducted subsequently.
      
      Change-Id: Ie6f43bfbde30583e941f450bf07c3b48828c9571
      e39ecfaa
    • Yaowu Xu's avatar
      Increase redo_frequency to 4 and cleanups · 6f5a1e03
      Yaowu Xu authored
      Clean-ups include
      a. redundant code in rt -5 speed feature settings
      b. code that guarantees square block availability in
      rd_auto_partition_range()
      
      Change-Id: Ic7b04d45b6dc15c461e0edbbb4e78aec20348291
      6f5a1e03
  12. 11 Mar, 2014 2 commits
  13. 10 Mar, 2014 1 commit
    • Jingning Han's avatar
      Fix use of uninitialized bsize in non-RD mode · 29af074f
      Jingning Han authored
      The block size used for non-RD mode decision in FIXED_PARTITION
      setting was uninitialized. This commit fixes it by setting block
      size to be BLOCK_16X16.
      
      Change-Id: Ief04c9f1ab668de69297d9ab3dc15e2fa0bc4e95
      29af074f
  14. 08 Mar, 2014 1 commit
    • Deb Mukherjee's avatar
      Support for a fast diamond search · bead2e8f
      Deb Mukherjee authored
      Adds a fast diamond search which is about 5% faster than FAST_HEX
      with only a 0.1% drop in psnr when turned on for both speeds 5 and 7.
      This search is turned on for speed 7.
      
      Change-Id: I497630aa88a5148926086bb3038e7975e5f4eb98
      bead2e8f
  15. 07 Mar, 2014 1 commit
  16. 06 Mar, 2014 1 commit
  17. 05 Mar, 2014 2 commits
  18. 04 Mar, 2014 2 commits
    • Alex Converse's avatar
      Prune RT mode decisions for BLOCK_32x32 and up · bbc8c9d2
      Alex Converse authored
      * Remove all non-DC intra modes for BLOCK_32x32 and up
      * Remove all intra modes for blocks bigger than BLOCK_32x32
      * Remove ZEROMV for BLOCK_32x32 and up
      * Only consider NEARESTMV for blocks bigger than BLOCK_32x32
      
      Change-Id: Ia18351a238213e2f072f9e481d622949346a245f
      bbc8c9d2
    • Dmitry Kovalev's avatar
      Removing VP9_PTR. · fe7b1d0a
      Dmitry Kovalev authored
      Change-Id: Ib49d8dbc67c590f22a1a70251ff607c9f38febd7
      fe7b1d0a
  19. 03 Mar, 2014 2 commits
    • Jim Bankoski's avatar
      cost_coefs approximation speed up · 9c63e887
      Jim Bankoski authored
      This patch adds a new speed feature which doesn't do the rather
      expensive entropy context lookup or save to the table,  while
      doing costing.
      
      The speed up on desktop36p.y4m is around 10% other clips much less.
      
      On the RTC test set this was + 1% in overall datarate.
      
      Change-Id: Ia5144bbf45270671e7be9c8e4055369909e2f738
      9c63e887
    • Yunqing Wang's avatar
      Reset search_method for non-rd pick mode · 5b2190b2
      Yunqing Wang authored
      Reset FAST_HEX search_method, which was lost because of code merging.
      
      Change-Id: Ic2b2cb9ec2f1751e96f7db0f9226310f605fcc83
      5b2190b2
  20. 02 Mar, 2014 2 commits