| April 20, 2010 | ||
|---|---|---|
 |
View ba87f08
Remove explicit filename from Doxygen @file commands.Diego Biurrun authored on 2010/04/20 23:45:34 |
|
| March 16, 2010 | ||
|---|---|---|
 |
View 4693b03
Move H264 dsputil functions into their own structMåns Rullgård authored on 2010/03/16 10:17:00 |
|
| March 6, 2010 | ||
|---|---|---|
|
View 84dc2d8
Remove DECLARE_ALIGNED_{8,16} macrosMåns Rullgård authored on 2010/03/06 23:24:59 |
|
| February 19, 2010 | ||
|---|---|---|
|
View 19769ec
H264: use alias-safe macrosMåns Rullgård authored on 2010/02/19 01:24:31 |
|
| February 18, 2010 | ||
|---|---|---|
|
View 40d1122
Use LOCAL_ALIGNED macro for local arraysMåns Rullgård authored on 2010/02/18 05:36:20 |
|
| February 14, 2010 | ||
|---|---|---|
 |
View 78998bf
h264: Remove unused variables.Alexander Strange authored on 2010/02/14 06:09:38 |
|
| February 7, 2010 | ||
|---|---|---|
 |
View 9873ae0
Fix CAVLC+8x8DCT+MBAFF loopfiltering. Fixes issue1250Michael Niedermayer authored on 2010/02/07 11:00:00 |
|
| January 31, 2010 | ||
|---|---|---|
|
View 37b2b0d
Get rid of a check in one direction that cant be true in it in that part of the code. No meassureable speed change.Michael Niedermayer authored on 2010/01/31 11:05:26 |
|
|
View 2646814
Split first reference list comparission from mv comparission. about 0.5% faster MBAFF loop filteringMichael Niedermayer authored on 2010/01/31 05:07:37 |
|
| January 30, 2010 | ||
|---|---|---|
|
View 4e99279
Replace h->left_type[0] by the local variable for it we have. No meassureable speed effect.Michael Niedermayer authored on 2010/01/30 23:33:25 |
|
|
View 012dbcc
slightly faster bit trickery.Michael Niedermayer authored on 2010/01/30 23:10:06 |
|
|
View 77821e1
Replace ?: by branchless code. about 0.5% faster loop filteringMichael Niedermayer authored on 2010/01/30 22:40:20 |
|
| January 29, 2010 | ||
|---|---|---|
|
View 34032e2
factorize first filter call out, this makes the code somewhat smaller without any speed loss.Michael Niedermayer authored on 2010/01/29 04:44:13 |
|
| January 28, 2010 | ||
|---|---|---|
|
View 592e03a
Change wraper functions to always inline, they are faster now that way. 1% faster MBAFF decoding overall, maybe ~0.1% faster for the cathedral sample.Michael Niedermayer authored on 2010/01/28 20:37:35 |
|
|
View 5364db2
indentMichael Niedermayer authored on 2010/01/28 20:18:06 |
|