1. 27 Feb, 2012 5 commits
  2. 23 Feb, 2012 2 commits
  3. 22 Feb, 2012 1 commit
    • Marco Paniconi's avatar
      Remove the frame rate factor for key frame size. · 507ee87e
      Marco Paniconi authored
      When temporal layers is used (i.e., number_of_layers > 1),
      we don't use the frame rate boost for setting the key
      frame target size. The factor was forcing the target size to be
      always at its minimum (2* per_frame_bandwidth) for low frame rates
      (i.e., base layer frame rate).
      Generally we should modify or remove this frame rate factor;
      for now we turn if off for number_of_layers > 1.
      Change-Id: Ia5acf406c9b2f634d30ac2473adc7b9bf2e7e6c6
  4. 21 Feb, 2012 4 commits
  5. 17 Feb, 2012 7 commits
  6. 16 Feb, 2012 14 commits
    • Johann's avatar
      Clarify 'max_sad' usage · 6b151d43
      Johann authored
      Depending on implementation the optimized SAD functions may return early
      when the calculated SAD exceeds max_sad.
      Change-Id: I05ce5b2d34e6d45fb3ec2a450aa99c4f3343bf3a
    • Johann's avatar
      OS X shell is incompatible with echo -n · 5f0b303c
      Johann authored
      Built in echo in 'sh' on OS X does not support -n (exclude trailing
      newline). It's not necessary so just leave it off. Fixes issue 390.
      Build include guard using 'symbol' so that it is more likely to be
      Change-Id: I4bc6aa1fc5e02228f71c200214b5ee4a16d56b83
    • Fritz Koenig's avatar
      Include path fix for building against Android NDK. · 3653fb47
      Fritz Koenig authored
      cpu-features.h is not in the common paths, add
      to the cflags for Android.
      Change-Id: Icbafc7600d72f6b59ffb030f6ab80ee6860332bb
    • John Koleszar's avatar
      vpxenc: initial implementation of multistream support · 9e50ed7f
      John Koleszar authored
      Add the ability to specify multiple output streams on the command line.
      Streams are delimited by --, and most parameters inherit from previous
      In this implementation, resizing streams is still not supported. It
      does not make use of the new multistream support in the encoder either.
      Two pass support runs all streams independently, though it's
      theoretically possible that we could combine firstpass runs in the
      future. The logic required for this is too tricky to do as part of this
      initial implementation. This is mostly an effort to get the parameter
      passing and independent streams working from the application's
      perspective, and a later commit will add the rescaling and
      multiresolution support.
      Change-Id: Ibf18c2355f54189fc91952c734c899e5c072b3e0
    • John Koleszar's avatar
      vpxenc: factor out input open/close · 732cb9a6
      John Koleszar authored
      Simplify some of the file I/O for later commits which will add multistream
      Change-Id: Idf1a05f3a29c95331d0c4a6ea5960904e4897fd4
    • John Koleszar's avatar
      vpxenc: add warning()/fatal() helpers · c535025c
      John Koleszar authored
      Cosmetic. Allows exiting with an error message without opening a new
      Change-Id: If227b29b825f0241acea79dd38f19e524552ee18
    • John Koleszar's avatar
      decoder: reset segmentation map on keyframes · e8223bd2
      John Koleszar authored
      Refactoring some of the mode decoding logic introduced a bug where
      the segmentation maps would not be properly reset on keyframes.
      The text of the bug is somewhat misleading as I initially read it to
      imply the bug was present in v0.9.7-p1 (Cayuga), but note the text
      "master", which indicates this was something subsequent. This issue
      bisects back to v0.9.7-p1-84-ga99c20c0, so unfortunately it was broken
      during the Duclair release.
      Thanks to Alexei Leonenko for investigating the root cause.
      Change-Id: I9713c9f070eb37b31b3b029d9ef96be9b6ea2def
    • Makoto Kato's avatar
      Support Android x86 NDK build · 7989bb7f
      Makoto Kato authored
      On Android NDK, rand() is inlined function.  But, on our SSE optimization,
      we need symbol for rand()
      Change-Id: I42ab00e3255208ba95d7f9b9a8a3605ff58da8e1
    • Scott LaVarnway's avatar
      Simplify mb_to_x_edge calculation during mode decoding · 6776bd62
      Scott LaVarnway authored
      Change-Id: Ibcb35c32bf24c1d241090e24c5e2320e4d3ba901
    • Scott LaVarnway's avatar
      Merge "decodemv cleanup/improvements" · a5879f7c
      Scott LaVarnway authored
    • Scott LaVarnway's avatar
      decodemv cleanup/improvements · 12ee845e
      Scott LaVarnway authored
      Removed unnecessary variables, unrolled functions, eliminated
      unnecessary mv bounds checks and branches.
      Change-Id: I02d034c70cd97b65025d59dd67c695e1db529f0b
    • Attila Nagy's avatar
      Consolidate C version of token packing functions · d02e74a0
      Attila Nagy authored
      Replace inner loops of pack_mb_row_tokens_c and
      pack_tokens_into_partitions_c with a call to pack_tokens_c.
      Change-Id: I0341554fb154a14a5dadb63f8fc78010724c2c33
    • Attila Nagy's avatar
      Multithreaded encoder, late sync loopfilter · 78071b3b
      Attila Nagy authored
      Second shot at this...
      Sync with loopfilter thread as late as possible, usually just at the
      beginning of next frame encoding. This returns control to application
      faster and allows a better multicore scaling.
      When PSNR packets are generated the final filtered frame is needed
      imediatly so we cannot delay the sync. Same has to be done when
      internal frame is previewed.
      Change-Id: I64e110c8b224dd967faefffd9c93dd8dbad4a5b5
    • John Koleszar's avatar
      vpxenc: factor out global config options · efd54f8f
      John Koleszar authored
      This is a first step towards specifying multiple output streams
      with one command line.
      Change-Id: Iac784d3911bf553694d024bbd0c3d547261e914b
  7. 15 Feb, 2012 1 commit
  8. 11 Feb, 2012 1 commit
  9. 10 Feb, 2012 3 commits
  10. 09 Feb, 2012 1 commit
    • Johann's avatar
      Fix variance overflow · fea3556e
      Johann authored
      In the variance calculations the difference is summed and later squared.
      When the sum exceeds sqrt(2^31) the value is treated as a negative when
      it is shifted which gives incorrect results.
      To fix this we cast the result of the multiplication as unsigned.
      The alternative fix is to shift sum down by 4 before multiplying.
      However that will reduce precision.
      For 16x16 blocks the maximum sum is 65280 and sqrt(2^31) is 46340 (and
      PPC change is untested.
      Change-Id: I1bad27ea0720067def6d71a6da5f789508cec265
  11. 08 Feb, 2012 1 commit