generic-library/vpx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Tuning for the more exact quantizer.

Browse Source 
Small changes to the default zero bin and rounding tables.
Though the tables are currently the same for the Y1 and Y2 cases
I have left them as separate tables in case we want to tune this later.

There is now some adjustment of the zbin based on the prediction mode.
Previously this was restricted to an adjustment for gf/arf 0,0 MV.

The exact quantizer now marginal outperforms and is the default.

The overall average gain is about 0.5%

Change-Id: I5e4353f3d5326dde4e86823684b236a1e9ea7f47
This commit is contained in:
Paul Wilkins
2010-11-08 15:28:54 +00:00
parent 458f4fedd2
commit 6adbe09058
5 changed files with 68 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
vp8/encoder/onyx_if.c
View File

@@ -3494,8 +3494,18 @@ static void encode_frame_to_data_rate
cpi->zbin_over_quant = 0;
cpi->zbin_mode_boost = 0;
// Enable mode based tweaking of the zbin
// Enable or disable mode based tweaking of the zbin
// For 2 Pass Only used where GF/ARF prediction quality
// is above a threshold
cpi->zbin_mode_boost = 0;
cpi->zbin_mode_boost_enabled = TRUE;
if (cpi->pass == 2)
{
if ( cpi->gfu_boost <= 400 )
{
cpi->zbin_mode_boost_enabled = FALSE;
}
}
// Current default encoder behaviour for the altref sign bias
if (cpi->source_alt_ref_active)