| commit | cd489264e1011c367476664b65db589380115c6f | [log] [tgz] |
|---|---|---|
| author | Geza Lore <gezalore@gmail.com> | Fri Jul 08 14:10:53 2016 +0100 |
| committer | Pascal Massimino <pascal.massimino@gmail.com> | Mon Jul 11 18:14:21 2016 +0000 |
| tree | 87b5d5819210a807f17ac4916e7d8849a46149a1 | |
| parent | bfa59b4a5f3f96526b93e758d753ee9b8e389c22 [diff] |
Optimize and cleanup supertx predictor. Use vpx_blend_a64_hmask and vpx_blend_a64_vmask to speed up computing the supertx predictor. Decoder speedup of up to 4% has been observed. Change-Id: I255a5ba4cc24f78dc905d25b6e2f7fbafac13253