)]}'
{
  "commit": "5500ce76f43349d0a6a3ef16de8a6d57077103b9",
  "tree": "eb8fc76be786ae2fedb5d637377400c2bf680058",
  "parents": [
    "9134586fe6d6136143cd1e5456a24d6eac5cc772"
  ],
  "author": {
    "name": "Monty Montgomery",
    "email": "cmontgomery@mozilla.com",
    "time": "Mon Nov 06 21:47:34 2017 -0500"
  },
  "committer": {
    "name": "Christopher Montgomery",
    "email": "cmontgomery@mozilla.com",
    "time": "Sat Nov 11 07:10:59 2017 +0000"
  },
  "message": "Move all of LBD Daala TX to up-4, down-1 shift\n\nNow that tran_low_t is assumed to be 32 bit when Daala TX is active,\nthere\u0027s no reason for multi-stage shifting to fit coefficients into 16\nbits for the inter-tranform transpose matrix. Go to a consistent up by\nfour, down by one shifting scheme for all TX block sizes.\n\n(Note this is for the current AV1 coefficient scaling scheme with\nav1_get_tx_scale and deeper coefficients for higher bitdepth input.\nDaala TX is moving to the long-intended constant-coefficient-depth in\nupcoming patches).\n\nsubset 1:\nmonty-4-1-baseline-s1@2017-11-11T05:57:15.857Z -\u003e\n monty-4-1-test-s1@2017-11-11T05:57:52.983Z\n\n   PSNR | PSNR Cb | PSNR Cr | PSNR HVS |   SSIM | MS SSIM | CIEDE 2000\n-0.0117 | -0.0246 |  0.0530 |   0.0238 | 0.0254 |  0.0447 |    -0.0442\n\nChange-Id: I2214e94ac822542c504d472276723277ed350abf\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "dc0715531641bc27fc5e3ace462301b798fc88b2",
      "old_mode": 33188,
      "old_path": "av1/common/idct.c",
      "new_id": "c3bde4c3967c3e8c20ee5b404e6911036faad6e9",
      "new_mode": 33188,
      "new_path": "av1/common/idct.c"
    },
    {
      "type": "modify",
      "old_id": "6ae3080c8f068b7bf0e3b3fb6add8e99827d5956",
      "old_mode": 33188,
      "old_path": "av1/encoder/dct.c",
      "new_id": "43d059af436041886880fc8775a9504734430920",
      "new_mode": 33188,
      "new_path": "av1/encoder/dct.c"
    }
  ]
}
