)]}'
{
  "commit": "b49c6aea692236ce1939cefc8eca3c74f4b4e1d3",
  "tree": "1a41fda443d0c90aacc0500791b523e4c9c52ea4",
  "parents": [
    "730c8054fbb542a102a44edc534ce39254cfbe30"
  ],
  "author": {
    "name": "Jingning Han",
    "email": "jingning@google.com",
    "time": "Mon Nov 27 18:14:05 2017 -0800"
  },
  "committer": {
    "name": "Jingning Han",
    "email": "jingning@google.com",
    "time": "Mon Dec 04 23:17:05 2017 +0000"
  },
  "message": "Add the speed feature structure for codec dev\n\nThis commit re-structures the speed feature setup for the codec\ndevelopment purpose. Instead of progressively reducing encoder\ncomplexity at the expense of incremental coding loss, we allow a\nseparate set of speed features, each corresponds to a certain\ncategory of coding units:\n\n1 \u003c\u003c 0: transform coding\n1 \u003c\u003c 1: inter prediction\n1 \u003c\u003c 2: intra prediction\n1 \u003c\u003c 3: block partition\n1 \u003c\u003c 4: loop filters\n1 \u003c\u003c 5: rd early skip\n\n[6 - 7] are left open for next adjustment.\n\nIt is constructed to facilitate the codec development purpose.\nWhen working on a coding functions, one could choose to turn on\none or more less related coding units to speed up the evaluation\nprocess. For example, to test a transform related experiment, one\ncould set\n--dev-sf\u003d2, 6, or 22\nwhich corresponds to turning on:\n2 - inter prediction speed features,\n6 - both inter / intra speed features,\n22 - inter / intra, and loop filter features.\n\nThe goal is to allow faster experimental verification during the\ndevelopment process. With the experiment in a stable state, we\ncan evaluate its performance in speed 0 at higher confidence level.\n\nChange-Id: Ib46c7dea2d2a60204c399dc01f10262c976adf0d\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "2f9bafd9816028c45ec86d0ef901b0926de8c92f",
      "old_mode": 33188,
      "old_path": "aom/aomcx.h",
      "new_id": "f8c7168888526e94be112329cb1452ea4d5c1fd0",
      "new_mode": 33188,
      "new_path": "aom/aomcx.h"
    },
    {
      "type": "modify",
      "old_id": "6071e38948ef98aa6a8899ab221726bf6eae646f",
      "old_mode": 33188,
      "old_path": "aomenc.c",
      "new_id": "9438023e92cd056d19b0c85cb99c84a557573d6c",
      "new_mode": 33188,
      "new_path": "aomenc.c"
    },
    {
      "type": "modify",
      "old_id": "e55625deddc4dd99720484da6b3d27a202ea6177",
      "old_mode": 33188,
      "old_path": "av1/av1_cx_iface.c",
      "new_id": "09dd3a298ffd18f3ac3228ebf6d11542a2b4c695",
      "new_mode": 33188,
      "new_path": "av1/av1_cx_iface.c"
    },
    {
      "type": "modify",
      "old_id": "e1ddd7366161afde4835fc6827551f03f6191698",
      "old_mode": 33188,
      "old_path": "av1/encoder/encoder.h",
      "new_id": "84dcb274432a433eb41c3a082096f3eb83ad95e7",
      "new_mode": 33188,
      "new_path": "av1/encoder/encoder.h"
    },
    {
      "type": "modify",
      "old_id": "8833a7ba3b4de27c5a1e18fb657fc46a8d36fa1d",
      "old_mode": 33188,
      "old_path": "av1/encoder/speed_features.c",
      "new_id": "e81732cc6374513484e15506d16d4cd735ec14a8",
      "new_mode": 33188,
      "new_path": "av1/encoder/speed_features.c"
    },
    {
      "type": "modify",
      "old_id": "fedefaa03b903267b42b727a97292dc93007159b",
      "old_mode": 33188,
      "old_path": "av1/encoder/speed_features.h",
      "new_id": "9f0a7a1b97877e704989bc6b3bc50ea6e8fcef0c",
      "new_mode": 33188,
      "new_path": "av1/encoder/speed_features.h"
    }
  ]
}