| commit | 6ae7b3d9fbdc095e30435cacf2738aea018a1a4a | [log] [tgz] |
|---|---|---|
| author | Jingning Han <jingning@google.com> | Sun Jun 11 07:58:10 2017 -0700 |
| committer | Jingning Han <jingning@google.com> | Mon Jun 12 05:35:59 2017 +0000 |
| tree | c4004d9fefb7f12d9fc9d97d80d3e21969525048 | |
| parent | 32d26bc0af45681fe93c3c9a6f39078569cf707f [diff] |
Fix high bit-depth quantization process Scale the rounding factor according to the scaling factor applied to the quantization step size. This resolves a compression performance regression in 32x32 and above transform size. BUG=aomedia:599 Change-Id: Id3fc9a46c4a8843ff5d77ccaa59ee3112b12d7f4