Geza Lore 4c4f04ac11 Optimize and cleanup obmc predictor and rd search. ...

Use vpx_blend_a64_hmask and vpx_blend_a64_vmask to speed up
computing the obmc predictor. Clean up calc_target_weighted_pred.

Encoder speedup: 1.3%
Decoder speedup: 6.5%

Change-Id: I0c774fe53d22399e92a10d1daf3af0010d88d2c5

2016-07-13 16:54:20 +00:00

..

bitreader.h

Add pluggable bitreaders.

2016-03-25 11:02:41 -07:00

decodeframe.c

Optimize and cleanup obmc predictor and rd search.

2016-07-13 16:54:20 +00:00

decodeframe.h

VP9Decoder -> VP10Decoder

2015-08-13 14:43:55 -07:00

decodemv.c

Merge bi-predictive frames to EXT_REFS

2016-06-17 12:43:39 -07:00

decodemv.h

ANS experiment: Use ANS everywhere.

2016-04-01 11:32:31 -07:00

decoder.c

Remove reference frame buffer update for show_exsiting_frame

2016-07-01 09:26:54 -07:00

decoder.h

Merge bi-predictive frames to EXT_REFS

2016-06-17 12:43:39 -07:00

detokenize.c

Various cosmetics on the new_quant experiment

2016-06-29 13:18:52 -07:00

detokenize.h

Store ANS token CDFs in the FRAME_CONTEXT rather than in a global table.

2016-04-19 09:39:48 -07:00

dsubexp.c

ANS experiment: Use ANS everywhere.

2016-04-01 11:32:31 -07:00

dsubexp.h

ANS experiment: Use ANS everywhere.

2016-04-01 11:32:31 -07:00

dthread.c

Merge bi-predictive frames to EXT_REFS

2016-06-17 12:43:39 -07:00

dthread.h

vp10: add extern "C" to headers

2015-09-09 23:15:59 -07:00