x264 - OpenSource AVC/H.264 Video Codec

x264 r1153 :cool:

Zitat

commit 7b6ce6a005f791f1d82c369d2f1e1c30b9ccbe80 r1153
Author: Loren Merritt <pengvado@akuvian.org>
Date: Tue May 19 02:47:15 2009 +0000

fix a race condition at the end of thread_input

commit fb4e2845497f77e2dae3262e1414de2cdf327ab5 r1152
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Mon May 18 22:40:45 2009 -0400

Various trellis speed optimizations

commit e86a1734047fb3e3af8dbe17a0e29479e839635b r1151
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sat May 16 12:16:34 2009 -0700

Make i686 the default arch on x86_32
Disabling asm will default to a generic arch.
Also fix configure for gcc 4.4.

commit 0e050663d16cd5b095723bd00136873aa461d5cd r1150
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri May 15 20:07:59 2009 -0700

Faster signed golomb coding
3% faster CAVLC RDO and bitstream writing.

commit 3b80f2321937ed4962c6e2fcc679765bbf729723 r1149
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Thu May 14 04:11:15 2009 -0700

Faster spatial direct MV prediction
unroll/tweak col_zero_flag

commit 400740ba1a8a7a6001aa9b01e9be0b3d0905856e r1148
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Mon May 4 04:19:28 2009 -0700

More CABAC and CAVLC optimizations
Simplified function calling for block_residual_write_(cabac|cavlc) and improved sigmap coding.
Tried making 0/1-bit specific versions of CABAC asm, but benefit was minimal under GCC 4.3.
Helped a decent bit under 3.4, but you shouldn't be using such old versions anyways.

commit f526f244cbc21c245064a2a9c5611840dd3b5203 r1147
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Wed Apr 29 22:54:52 2009 -0700

Various optimizations in frametype lookahead

commit ecae94a0ee6689bec9d32d64cce64cab7790a984 r1146
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sun Apr 26 22:13:17 2009 -0700

Some cosmetics/cleanup
Move some macros to x86util.asm that should have been there to begin with.
Fix a typo that didn't cause any issues.

commit d2e1e1c35c43ea9c90c9211be6202143b69b35b9 r1145
Author: Guillaume Poirier <gpoirier@mplayerhq.hu>
Date: Tue Apr 21 21:18:44 2009 +0000

fix "incompatible types in initialization" compilation issues with GCC 4.3 (which is stricter than previous compiler version)

commit 4fc8c03ad568efe3dd2f57db33b0863d29cb63a0 r1144
Author: Guillaume Poirier <gpoirier@mplayerhq.hu>
Date: Tue Apr 21 17:32:21 2009 +0200

fix conversions between vectors with differing element types or numbers of subparts errors

commit 755470584932877c4d5ea1b51c2cc2dbd044b7ca r1143
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sat Apr 18 16:07:53 2009 -0700

Add "coded blocks" stat to output information.
This measures the total percentage of blocks, intra and inter, which have nonzero coefficients.
"y,uvAC,uvDC" refers to luma, chroma DC, and chroma AC blocks.
Note that skip blocks are included in this stat.

commit a908c88f0bbd4250afbd16b6c55b3c9f613af96d r1142
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri Apr 17 23:38:29 2009 -0700

Enable asm predict_8x8_filter
I'm not entirely sure how this snuck its way out of holger's intra pred patch.

commit c74726e1ed0312e2073f0ea8c45804a870512579 r1141
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri Apr 17 06:00:39 2009 -0700

Remove various bits of dead code found by CLANG.

Alles anzeigen

Zitat

libx264 SVN-r1153:

libx264-r1153, MinGW GCC 4.4.0, optimized for Core 2
libx264-r1153, MinGW GCC 4.4.0, optimized for K10
libx264-r1153, MinGW GCC 4.4.0, optimized for Pentium 2 and later

libx264 SVN-r1153 using experimental AutoVAQ patch v0.2 by BugMaster:

libx264-r1153, MinGW GCC 4.4.0, optimized for Core 2
libx264-r1153, MinGW GCC 4.4.0, optimized for K10
libx264-r1153, MinGW GCC 4.4.0, optimized for Pentium 2 and later

These builds will NOT work with Avidemux 2.4.x any longer. Please update to Avidemux 2.5 SVN-r4763 or later now!
The patches used for my builds can be found at this location. More info on BugMaster's patch in this thread!

Zitat

x264 r1153 x64 (unpatched)
gcc 4.4.0 (release) fprofiled build.
-------------------------

x264.1153M.x86.exe / x264.1153M.x64.exe
gcc 4.4.0 (release) fprofiled builds (x86 uses -march=pentium2).

Patches used:

x264-r1134-threaded-slicetype-v14.diff
x264_threaded_slicetype_windows_fix.diff
x264_hrd_pulldown.12_interlace.diff
x264_win_zone_parse_fix_05.diff

Alles anzeigen

Es scheint das letzte Skystrife Build war kaputt. Neue Builds jetzt verfügbar:

Zitat

x264.1153M.x86.exe / x264.1153M.x64.exe
gcc 4.4.0 (release) fprofiled builds (x86 uses -march=pentium2).

Patches used:

x264_hrd_pulldown.12_interlace.diff
x264_win_zone_parse_fix_05.diff

Use this if the slicetype patch is giving you issues. I'll also put up an AutoVAQ build here in a bit, as well as a x64 build without the slicetype patch.

Zitat

x264.1153M.AutoVAQ.x86.exe / x264.1153M.AutoVAQ.x64.exe
gcc 4.4.0 (release) fprofiled builds (x86 uses -march=pentium2).

Patches used:

x264_hrd_pulldown.12_interlace.diff
x264_win_zone_parse_fix_05.diff
x264_AutoVAQ.02.diff (see AutoVAQ thread)

Also note the new x86 build including the lookahead patch (this contains a built-in pthread library, unlike the previous one, my mistake).

Da "x264_hrd_pulldown.12_interlace.diff" scheinbar einen Bug hatte und beim zweiten Durchgang crasht, hier nochmal der Hinweis auf die gefixte Version:

Zitat von Trahald

Ahh. it was some debug code i added and didnt completely delete. Sorry. Still not sure why its not crashing here (it should). here is 13.

Zitat von Skystrife

x264.1153M.02.x86.exe / x264.1153M.02.x64.exe
gcc 4.4.0 (release) fprofiled builds (x86 uses -march=pentium2).

Patches used:

x264_hrd_pulldown.13_interlace.diff
x264_win_zone_parse_fix_05.diff

-------------------------

x264.1153M.AutoVAQ.02.x86.exe / x264.1153M.AutoVAQ.02.x64.exe
gcc 4.4.0 (release) fprofiled builds (x86 uses -march=pentium2).

Patches used:

x264_hrd_pulldown.13_interlace.diff
x264_win_zone_parse_fix_05.diff
x264_AutoVAQ.02.diff

Alles anzeigen

Die x264 Standard-Einstellungen werden sich in Kürze ändern:

Zitat von Dark Shikari

A lot of patches are going to be pushed soon. And among other things, the x264 defaults are changing. We will also be adding presets to the CLI interface (but not the library interface).

New defaults (suggestions welcome):

--subme 7 --bframes 3 --weightb --8x8dct --ref 3 --mixed-refs --trellis 1 --crf 23 --threads auto

Logically, this results in the following changes in commandline options:

--no-mixed-refs replaces --mixed-refs
--no-8x8dct replaces --8x8dct
--no-weightb replaces --weightb

The intention here is that you can simply use x264 with defaults and it will give Very Good Results that nobody can complain about.

Alles anzeigen

http://forum.doom9.org/showthread.php?t=148148

Darüber hinaus wird es einen ganzen Haufen von "Presets" geben:

Zitat von Dark Shikari

Presets will be used as follows:

--preset name

and apply before all other options. In other words, they override the x264 defaults, which, with the default changes, will be:

--subme 7 --bframes 3 --weightb --8x8dct --ref 3 --mixed-refs --trellis 1 --crf 23 --threads auto

As such, all options that you set will act on top of the preset you choose, regardless of the option order.

ultrafast:

--ref 1 --scenecut 0 --nf --no-cabac --bframes 0 --partitions none --no-8x8dct --me dia --subme 0 --aq-mode 0

veryfast:

--partitions i8x8,i4x4 --subme 1 --me dia --ref 1 --trellis 0

fast:

--mixed-refs 0 --ref 2 --subme 5

medium:

<defaults>

slow:

--me umh --subme 8 --ref 5 --b-adapt 2 --direct auto

slower:

--me umh --subme 9 --ref 8 --b-adapt 2 --direct auto --partitions all --trellis 2

placebo:

--me tesa --subme 9 --merange 24 --ref 16 --b-adapt 2 --direct auto --partitions all --no-fast-pskip --trellis 2 --bframes 16

For all encodes, regardless of preset or not, using --pass 1 will automatically trigger "turbo" settings, which are as follows and override all commandline options:

--trellis 0 --no-8x8dct --partitions none --me dia --ref 1 --subme MIN(2, previous subme)

This can be disabled using --slow-firstpass.

For all encodes, if you don't set your own --ref option and instead use the default or a preset, x264 will automatically lower --ref if the --level you set conflicts with it. If you set --ref explicitly, it will not override what you set.

Profiles will be used as follows:

--profile name

The options will be baseline, main, and high, which do what you think they do. Note that these override all other options no matter what, with two exceptions.. Because interlaced and lossless coding are options with very explicit meanings, the encoder will terminate with an error if you attempt to use them with an incompatible profile.

Tunings will be used as follows:

--tune name

The options will be "film" and "animation", and will tweak AQ strength, psy-rd/trellis strength, deblock, refs, and so forth:

film:

--deblock -1:-1 --psy-rd 1.0:0.15

animation:

--ref (previousref==1?1:MIN(previousref*2,16)) --deblock 1:1 --psy-rd 0.4:0 --aq-strength 0.6 --bframes 5

All of this is subject to change before final release, and suggestions are welcome.

If you don't use any of the new presets, the only changes for you will be the new defaults, and the automatic fast first pass.

Alles anzeigen

http://forum.doom9.org/showpost.php?p=1301866

Mal wieder Zeit für'n Update, wie ich das sehe ;D

Zitat

commit 88b35c2d3bd86b42059e27db365752da9f2cd032 r1184
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date: Sat Jul 18 16:30:18 2009 -0700

New AQ algorithm option
"Auto-variance" uses log(var)^2 instead of log(var) and attempts to adapt strength per-frame.
Generates significantly better SSIM; on by default with --tune ssim.
Whether it generates visually better quality is still up for debate.
Available as --aq-mode 2.

commit f21daff3dc11cf5881f1727c3c9d505f0810d20b r1183
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Wed Jul 15 12:43:35 2009 -0700

Cacheline-split SSSE3 chroma MC
~70% faster chroma MC on 32-bit Conroe
Also slightly faster SSSE3 intra_sad_8x8c

commit 6c13403195d42b2c0ee707e9f2a6e9f9cd81afd6 r1182
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sun Jul 12 12:07:01 2009 -0700

Improve documentation of qp/crf options

commit 49bf7673b2f52b2cd8e9c10d8d6e9cbbb5422cf7 r1181
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Thu Jul 9 19:02:57 2009 -0700

Merge array_non_zero into zigzag_sub
Faster lossless, cleaner code.
SSSE3 version of zigzag_sub_4x4_field, faster lossless interlaced coding.

commit b63f5919e3f5367a0df3dbf218d5a94d2fdba5fb r1180
Author: James Darnley <james.darnley@gmail.com>
Date: Thu Jul 9 11:25:55 2009 -0700

Fix bug in reference frame autoadjustment
For some types of input file, x264 did the adjustment before width/height were known.

commit 96e2229e96d65420d491596affa9aaa068d718d6 r1179
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Tue Jul 7 11:13:39 2009 -0700

Fix fprofile settings to match changes in defaults
Also add b-adapt 2 to fprofile.

commit 3f6713d5c794d4fbfd3131985e33a822a40cb870 r1178
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri Jul 3 02:33:44 2009 -0700

Slightly faster dequant_flat assembly
Eliminate some redundant shifts.

commit af2a4ecd7bcefc97c8aa83913c9a2980206f9cd0 r1177
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Wed Jul 1 21:14:57 2009 -0700

Totally new preset system for x264.c (not libx264), new defaults
Other new features include "tune" and "profile" settings; see --help for more details.
Unlike most other settings, "preset" and "tune" act before all other options.
However, "profile" acts afterwards, overriding all other options.
Our defaults have also changed: new defaults are --subme 7 --bframes 3 --8x8dct --no-psnr --no-ssim --threads auto --ref 3 --mixed-refs --trellis 1 --weightb --crf 23 --progress.
Users will hopefully find these changes to greatly improve usability.

commit 72534d466a6bd99b9cbf32c74e667bea608c6dee r1176
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Wed Jul 1 16:33:12 2009 -0700

Update Gabriel's email address in AUTHORS

commit f4dac817e45b967572f5c4e4af4644dc7d263512 r1175
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Tue Jun 30 15:20:32 2009 -0700

Early termination for chroma encoding
Faster chroma encoding by terminating early if heuristics indicate that the block will be DC-only.
This works because the vast majority of inter chroma blocks have no coefficients at all, and those that do are almost always DC-only.
Add two new helper DSP functions for this: dct_dc_8x8 and var2_8x8. mmx/sse2/ssse3 versions of each.
Early termination is disabled at very low QPs due to it not being useful there.
Performance increase is ~1-2% without trellis, up to 5-6% with trellis=2.
Increase is greater with lower bitrates.

commit 7fd6a9099f18ec028d6c73890258280e6f8a6c02 r1174
Author: David Conrad <lessen42@gmail.com>
Date: Fri Jun 26 13:09:44 2009 -0700

Fix bug in checkasm
frame_init_lowres_core check didn't check the C plane.
However, all x86 and PPC assembly was correct regardless of the unit test being incorrect.

commit f6d31669a2547110b9c1323aa51437296f2f3506 r1173
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Wed Jun 24 14:39:15 2009 -0700

Add subpartition cost for sub-8x8 blocks
Improves sub-p8x8 mode decision.

commit b484fe1bff3cb68b3325a9b77d802789cf77e600 r1172
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Wed Jun 24 13:24:18 2009 -0700

Yet more CABAC and CAVLC optimizations
Also clean up a lot of pointless code duplication in CAVLC MV coding.

commit 2c7cb4c3f111b49d3d961366ed338f38a0555716 r1171
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri Jun 19 18:49:55 2009 -0700

Various CABAC optimizations and cleanups
Faster CABAC CBF context calculation for inter blocks.
Add x264_constant_p(), will probably be useful in the future as well.
Simpler subpartition functions.
Clean up and optimize mvd_cpn a bit more.
Various other minor optimizations.

commit 364d7dff8dd96f71465bed10594b9f1e78fe6139 r1170
Author: David Wolstencroft <wolstencroft@alum.rpi.edu>
Date: Sat Jun 20 21:42:55 2009 +0200

AltiVec version of frame_init_lowres_core. 22.4x faster than C on PPC7450 and 25x on PPC970MP.

commit ab85c9b0ae08a237472bfd14558353d5ecb92b3d r1169
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri Jun 19 16:03:18 2009 -0700

MMX CABAC mvd sum calculation
Faster CABAC mvd coding.

commit 803c9d94641e57544932114f61f523e19bba6b4d r1168
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri Jun 19 16:02:39 2009 -0700

Faster MV prediction
Smaller code size, plus I get to use goto.

commit 6199685d989facfd6105c30b50e30615e784fba3 r1167
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Wed Jun 10 10:37:01 2009 -0700

Fix potential crash in checkasm
ssim_end4_sse2 requires aligned sums

commit b555e3f90b2060542d44bcb1a254d5a7bfc5d23a r1166
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Wed Jun 10 10:11:00 2009 -0700

SSSE3, faster SSE2/MMX integral_init4v
The real reason I wrote this was an excuse to use shufpd.

commit 6841c5e2407e60de92c46bc6c649cd2fc4a13d75 r1165
Author: Mike Frysinger <vapier@gentoo.org>
Date: Thu Jun 11 08:29:27 2009 +0000

configure check for uclinux

commit a9526973579d7d48bb2730b64a547ad10b7ef6ef r1164
Author: Loren Merritt <pengvado@akuvian.org>
Date: Thu Jun 11 08:27:46 2009 +0000

fix a crash on frame width <= 48 pixels

commit 3f56e271ac3d8a0e054b8b18e63886a6070ef05e r1163
Author: Loren Merritt <pengvado@akuvian.org>
Date: Wed May 27 20:47:18 2009 +0000

configure check for cc, rather than reporting lack of compiler as an asm error.
configure check for -mno-cygwin, since it's removed from gcc4.

commit f7bfcfaa36ac3f04447f668f94db73801ca86e4d r1162
Author: Loren Merritt <pengvado@akuvian.org>
Date: Sun May 24 05:01:26 2009 +0000

a better way to keep track of mv candidates.
2-4% faster dia, hex, and umh.

commit 8572d517a52cb46692d2fcc0723fdbd01d1878b6 r1161
Author: Loren Merritt <pengvado@akuvian.org>
Date: Sun May 24 05:01:19 2009 +0000

reorder some motion estimation patterns.
this change is useless on its own, but segregates the bitstream-changing part out of my next optimization.

commit 79377d7309844d4e197ddf043fa9f115015fe286 r1160
Author: Loren Merritt <pengvado@akuvian.org>
Date: Mon May 25 19:16:05 2009 -0400

Fix VBV warning broken in r915
x264 will now correctly warn about maxrate specified without bufsize even when a level is not set.

commit 3da3f957c096ad7885312b58d7ab12a7fab10111 r1159
Author: Loren Merritt <pengvado@akuvian.org>
Date: Mon May 25 07:03:10 2009 +0000

configure check for ssse3-capable binutils

commit a2c01c124e4ee4b9d71a4b2a20ed572d439aea7b r1158
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sun May 24 16:58:08 2009 -0400

Fix 10L in r1155
Broke --me esa/tesa due to forgetting to add handling for x264_cost_mv_fpel.

commit 37241b84054d5d012994c4567270c1b32dd9a038 r1157
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri May 22 21:28:15 2009 -0700

Fix bug where satd was incorrectly used with subme<=1
Faster subme<=1 with i4x4 enabled.

commit be17f0306ec42c2fc6e572a50d16c3717c7991ce r1156
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri May 22 20:40:27 2009 -0700

Remove some pointless error handling code in cabac/cavlc

commit b6470f07f02342d1abf960b1482e3e9e835fbc5d r1155
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri May 22 18:40:12 2009 -0700

Save some memory on mv cost arrays
Have quantizers that use the same lambda share the same cost array.

commit 83d2c126382fb5f012e979b1a80f92fd49f35771 r1154
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri May 22 16:57:33 2009 -0700

Various CABAC and CAVLC optimizations
Backport CAVLC partial-inlining early termination to CABAC (~2-4% faster CABAC residual coding)

Alles anzeigen

Macroblock-Tree Ratecontrol ist da ;D
http://forum.doom9.org/showthread.php?t=148686

Zitat

commit a1ed468f67476fbbe49e1fbfe1a567be0c052d44 r1198
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Thu Aug 6 23:09:46 2009 -0700

Add "veryslow" preset
Apparently some people are actually *using* placebo, so I've added this preset to bridge the gap.

commit bb66c482242a0747823661b212114c1a2f015fe3 r1197
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Tue Aug 4 17:46:33 2009 -0700

Macroblock-tree ratecontrol
On by default; can be turned off with --no-mbtree.
Uses a large lookahead to track temporal propagation of data and weight quality accordingly.
Requires a very large separate statsfile (2 bytes per macroblock) in multi-pass mode.
Doesn't work with b-pyramid yet.
Note that MB-tree inherently measures quality different from the standard qcomp method, so bitrates produced by CRF may change somewhat.
This makes the "medium" preset a bit slower. Accordingly, make "fast" slower as well, and introduce a new preset "faster" between "fast" and "veryfast".
All presets "fast" and above will have MB-tree on.
Add a new option, --rc-lookahead, to control the distance MB tree looks ahead to perform propagation analysis.
Default is 40; larger values will be slower and require more memory but give more accurate results.
This value will be used in the future to control ratecontrol lookahead (VBV).
Add a new option, --no-psy, to disable all psy optimizations that don't improve PSNR or SSIM.
This disables psy-RD/trellis, but also other more subtle internal psy optimizations that can't be controlled directly via external parameters.
Quality improvement from MB-tree is about 2-70% depending on content.
Strength of MB-tree adjustments can be tweaked using qcompress; higher values mean lower MB-tree strength.
Note that MB-tree may perform slightly suboptimally on fades; this will be fixed by weighted prediction, which is coming soon.

commit f21e71a04ba65aff9b5a4bfa8a73fd86c463f4ee r1196
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Mon Aug 3 20:52:30 2009 -0700

Various 1-pass VBV tweaks
Make predictors have an offset in addition to a multiplier.
This primarily fixes issues in sources with lots of extremely static scenes, such as anime and CGI.
We tried linear regressions, but they were very unreliable as predictors.
Also allow VBV to be slightly more aggressive in raising QPs to avoid not having enough bits left in some situations.
Up to 1db improvement on some clips.

commit 5d75a9bd5b942392c4ab64156a266eed64c0793f r1195
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Tue Jul 28 20:41:27 2009 -0700

Fix another 10L in QPRD
An entry in subpel_iterations was missing.
I have no idea how QPRD was working at all without this change.

commit 97ed27054005a85e7c49209c20fd0b280917ac02 r1194
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Tue Jul 28 01:16:23 2009 -0700

Update help and cleanup in ratecontrol.c
Deal with some out-of-date information.

commit 0538c56a95cdc41a42a206079276a57d5d76b5a5 r1193
Author: Loren Merritt <pengvado@akuvian.org>
Date: Tue Jul 28 07:16:31 2009 +0000

15% faster refine_bidir_satd, 10% faster refine_bidir_rd (or less with trellis=2)
re-roll a loop (saves 44KB code size, which is the cause of most of this speed gain)
don't re-mc mvs that haven't changed

commit 306c3ee4b1c3cae804185597305725d2484f21b9 r1192
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Mon Jul 27 21:03:00 2009 -0700

Faster bidir_rd plus some bugfixes
Cache chroma MC during refine_bidir_rd and use both the luma and chroma caches to skip MC in macroblock_encode.
Fix incorrect call to rd_cost_part; refine_bidir_rd output was incorrect for i8>0.
Remove some redundant clips.
~12% faster refine_bidir_rd.

commit d6eed014d0af8f87045d6d5daf3376c486efdea7 r1191
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Mon Jul 27 04:45:03 2009 -0700

Add "fastdecode" tune option
It does what it says it does.

commit 43773d27a6dd74c62b6d29d0ae0a80397469bfbf r1190
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sun Jul 26 12:20:09 2009 -0700

Fix two bugs in QPRD
fprofile settings now actually fprofile QPRD.
Don't use i_mbrd before initializing it.

commit 9d0b5e95bbf3f9077806927264add762952f77ad r1189
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sun Jul 26 03:03:12 2009 -0700

Fix 10l in QPRD
Trellis used wrong lambda with trellis=1

commit 4074956df13e058421fb5ba89b872be143742ffd r1188
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sat Jul 25 22:31:06 2009 -0700

Fix a nondeterminism with threads and subme>7
Also add a few more checks to eliminate the need for spel_border.

commit 7733721e410acb96fdf740ca95d2a394b2a2b713 r1187
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Thu Jul 23 12:20:39 2009 -0700

Add QPRD support as subme=10
Refactor trellis lambda selection to be done in analyse_init instead of in trellis.
This will allow for more easy adaption of lambda later on; for now it allows constant lambda across variable QPs.
QPRD is only available with adaptive quantization enabled and generally improves SSIM and visual quality.
Additionally, weight the SSD values from RD based on the relative QP offset for chroma; helps visually at high QPs where chroma has a lower QP than luma.
This fixes some visual artifacts created by QPRD at high QPs.
Note that this generally hurts PSNR and SSIM, and so is only on when psy-RD is on.

commit f5e6980b3eb34ed610f5fc36a4378a0ed4277753 r1186
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Tue Jul 21 19:56:21 2009 -0700

SSSE3 cachesplit workaround for avg2_w16
Palignr-based solution for the most commonly used qpel function.
1-1.5% faster overall on Core 2 chips.

commit 29569051505a78db9dbbc8fda53ab11e7e08b994 r1185
Author: Loren Merritt <pengvado@akuvian.org>
Date: Wed Jul 22 20:20:52 2009 +0000

shut up valgrind warnings in trellis

commit 88b35c2d3bd86b42059e27db365752da9f2cd032 r1184
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date: Sat Jul 18 16:30:18 2009 -0700

New AQ algorithm option
"Auto-variance" uses log(var)^2 instead of log(var) and attempts to adapt strength per-frame.
Generates significantly better SSIM; on by default with --tune ssim.
Whether it generates visually better quality is still up for debate.
Available as --aq-mode 2.

Alles anzeigen

Mal wieder Zeit für ein Update:

Zitat

commit 383621879573013bf5f85ce8d6eeefb18d9625c0 r1281
Author: Loren Merritt <pengvado@akuvian.org>
Date: Wed Sep 30 22:39:13 2009 +0000

rm msvc project files and related ifdefs

commit 5cda2cf81a75bfd70a5529d3b6fb01b0a7897a5e r1280
Author: Holger Lubitz <holger@lubitz.org>
Date: Tue Oct 6 15:17:34 2009 -0700

SSE4 version of 4x4 idct
27->24 clocks on Nehalem.
This is really just an excuse to use "movsd" in a real function.
Add some comments to subsum-related macros in x86util.

commit d971ace393dc49b09b930aa22e8b0c70f8a6f386 r1279
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sun Oct 4 19:15:28 2009 -0700

Constrained intra prediction support
Enable with --constrained-intra. Significantly reduces compression, but required for the base layer of SVC encodes and maybe some other use-cases.

commit sponsored by a media streaming company that wishes to remain anonymous.

commit a88f2d78533af5ff38c01054cf336bdd7f2cf27c r1278
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sun Oct 4 00:48:27 2009 -0700

Slightly improve non-RD p8x8 mode decision
Subpartition costs are effectively zero in CABAC if sub-8x8 search is off.

commit c318d8fa1e83b8807b66d45b225c905f480703ce r1277
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sat Oct 3 00:59:02 2009 -0700

Reorder reference frames optimally on second pass
About +0.1-0.2% compression at normal bitrates, up to +1% at very low bitrates.
Only works if the first pass uses the same number of refs as the second (i.e. not with fast first pass).
Thus, only worthwhile at insanely slow speeds: as such, enable slow-firstpass by default with preset placebo.
Note that this changes the stats file format!

commit 4f3ad6a28e1ac017103a77d135a48aaad114d262 r1276
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Wed Sep 30 12:13:16 2009 -0700

Fix typo in ratecontrol_summary

commit 8be3935a8cef95fe94e0b054f3b6cb904a1f363a r1275
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Tue Sep 29 23:32:07 2009 -0700

Clip log2_max_frame_num
It's still much higher than it needs to be, but that will be fixed with the upcoming MMCO patch.
Also make sure we don't write too large a frame_num or poc in slice header.

commit 83ca0f2cec3909b38d55e664bcd3f5c4b1546de8 r1274
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date: Sat Sep 26 12:44:53 2009 -0700

Fix some issues with 3-pass statsfile handling
The value of i_frame during encoder_close was incorrect.

commit a4334d41ea61b70fcff308ba988193309bb5608a r1273
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date: Sat Sep 26 12:42:46 2009 -0700

Fix ctrl-C termation message with few frames encoded

commit db518eae22e57b37c806625685faba2f2a857f88 r1272
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri Sep 25 16:23:52 2009 -0700

Add support for single-frame VBV, improve compliance
This allows both constant-framesize and capped-framesize encoding.
Literal constant framesize isn't actually supported yet due to the lack of
filler support.
Example with 30fps video: --vbv-bufsize 200 --vbv-maxrate 6000 will ensure that
no frame is ever larger than 200 kilobits.

One example use-case of this is for zero-delay streaming where bandwidth costs
need to be minimized. If every frame is smaller than 200 kilobits and the
client has a 6 megabit connection, every single frame can be instantly sent
to the client and handled without any decoder-side buffer.

Fix a mistake in VBV calculation--this may have caused the VBV to be slightly
non-compliant in some situations without x264 realizing it.
Add primitive prediction handling for rows with quantizers lower than their
reference. This slightly improves VBV in CBR mode.
Various other minor improvements to VBV, mostly to make single-frame VBV work.

commit sponsored by a media streaming company that wishes to remain anonymous.

commit 496d79dfcb90066f0254e07d593471f2c885a153 r1271
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Thu Sep 24 08:40:45 2009 -0700

Fix 10l in API change
frame_num was set to 1, not 0, for the first frame. This broke spec compliance.
Didn't actually seem to cause any problems though except for breaking decoding on Quicktime.

commit 2b4e823809617d605bc058acba681c37e14e0ed0 r1270
Author: Steven Walters <kemuri9@gmail.com>
Date: Wed Sep 23 15:04:02 2009 -0700

Allow user-set FPS for inputs other than YUV

commit e8b9dfe479a6a945d9cd91b4aae8a3cba06b9db9 r1269
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date: Wed Sep 23 12:31:53 2009 -0700

Improve threaded frame handling
Avoid unnecessary cond_wait

commit d5429261cc2a2205d225b0f0c0802d3b565eb7bd r1268
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Tue Sep 22 17:30:19 2009 -0700

Attempt to detect miscompilation due to bug in gcc 4.2
I don't know if this bug still affects latest x264, but it can't hurt to try to detect it.
Accordingly refuse to open the encoder if detected.
Apparently VLC (on Windows) has been distributed for some time with a completely
broken x264 due to the use of a completely broken compiler (gcc 4.2). In
particular, the MV costs seem to be calculated incorrectly on win32 when linking
from an application compiled without -ffast-math to an application with
-ffast-math.
I am not entirely certain why this occurs, but the result is, unsurprisingly,
encoding quality that makes MPEG-2 look good, due to the motion search being
completely broken.

commit c62060d242879d8271c22a486b286b91147d2ebf r1267
Author: Steven Walters <kemuri9@gmail.com>
Date: Tue Sep 22 12:14:23 2009 -0700

Really fix encoder_close crash this time
Not-entirely-fixed in r1253.

commit d166e7f3f8f68a768cf60cb4512f9efe8623b5cb r1266
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sun Sep 20 21:58:08 2009 -0700

Check for 16x16 partitions masquerading as smaller ones
Saves a few bits when using qpel-RD.

commit c49e14f21c7fd9f7b5d1a82f29cb3b6a1c8ecb70 r1265
Author: David Conrad <lessen42@gmail.com>
Date: Sun Sep 20 01:16:51 2009 -0700

Update config.guess/sub; add Snow Leopard support

commit d52d2db898fa4e7790964cd656a1da3a0c005881 r1264
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sat Sep 19 09:50:59 2009 -0700

Fix integer overflow in 2-pass VBV
Bug caused slight undersizing in 2-pass mode in some cases.

commit ca95d6198f5c21ff6904a842b52b893d0fee6aab r1263
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri Sep 18 14:28:31 2009 -0700

Fix bug with various bizarre commandline combinations and mbtree
Second pass would have mbtree on even though the first pass didn't (and thus encoding would immediately fail).

commit 91f3e81cf9e030f860abdfe0e0c0a37c5e2a4ac2 r1262
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Thu Sep 17 13:02:02 2009 -0700

Add intra prediction modes to output stats
Also eliminate some NANs in stat output with intra-only encoding.
Marginal speedup: disable stat calculation if log level is below X264_LOG_INFO.
Various minor cosmetics.

commit c91a3c98a72492a5e167cb6c98c1807fde76c528 r1261
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Wed Sep 16 21:34:48 2009 -0700

Overhaul syntax in muxers.c/matroska.c
The inconsistent syntax in these files has finally come to an end.

commit 031e25d8cc909af2d138f9fedc8252961f34a6ac r1260
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Wed Sep 16 20:00:00 2009 -0700

Major API change: encapsulate NALs within libx264
libx264 now returns NAL units instead of raw data. x264_nal_encode is no longer a public function.
See x264.h for full documentation of changes.
New parameter: b_annexb, on by default. If disabled, startcodes are replaced by sizes as in mp4.
x264's VBV now works on a NAL level, taking into account escape codes.
VBV will also take into account the bit cost of SPS/PPS, but only if b_repeat_headers is set.
Add an overhead tracking system to VBV to better predict the constant overhead of frames (headers, NALU overhead, etc).

http://forum.doom9.org/showthread.php?t=149593

commit dd026f296748ef370f1651fdbddec6d9238c6cc0 r1259
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Mon Sep 14 12:30:38 2009 -0700

Add missing fclose for mbtree input statsfile on second pass
Bug report by VFRmaniac

commit 8cbb85d7682f78af892ea2f5b0362cc1ef583c02 r1258
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Mon Sep 14 11:07:23 2009 -0700

Improve progress indicator behavior
Progress indicator will now indicate based on output frame, not input frame.

commit b5770ea28a83141252be66870cfabc0ceae626b7 r1257
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Mon Sep 14 03:21:14 2009 -0700

Update yasm configure check
lzcnt apparently requires yasm 0.6.2.

commit 935488e22376c33fbdc38909d9a119ac092e8878 r1256
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sun Sep 13 01:02:37 2009 -0700

Make MV costs global instead of static
Fixes some extremely rare threading race conditions and makes the code cleaner.
Downside: slightly higher memory usage when calling multiple encoders from the same application.

commit 4f6a00704c3e36648a32404aa729a4beff438b38 r1255
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Fri Sep 11 17:30:14 2009 -0700

Don't print scenecut message multiple times in verbose mode
Occurred mostly with b-adapt 2.

commit 411ee507328890fb1fde96b37b6eef854d27101a r1254
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Thu Sep 10 02:55:21 2009 -0700

Optimize rounding of luma and chroma DC coefficients
Reduce bitrate mostly-losslessly at low quantizers.
In some rare cases, bitrate reduction may be as high as 10%.
Luma rounding optimization (helps much less than chroma) requires trellis.

commit 08c83969dd45a6e550aaa3be11c69bb74d7a478c r1253
Author: Steven Walters <kemuri9@gmail.com>
Date: Wed Sep 9 12:19:40 2009 -0700

Fix crash if encoder_close is called before delayed frames are flushed
Also no longer flush frames when ctrl-Cing x264, so x264 will close faster.

commit b9b7c12eed6eb69b9df7ccd1289df5ce9c130a81 r1252
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Sun Sep 6 14:55:48 2009 -0700

Improve x264 help
Now has three help options: --help, --longhelp, and --fullhelp.
--help only shows the most basic options; most users should not need more than these.
Add usage examples.
Fix typo in a comment.

Alles anzeigen

Zitat

x264 r1281 64bit unpatched:
download ; hash

• built on Oct 7 2009, gcc: 4.3.4 20090220 (prerelease) (x64.generic.Komisar)
• fprofiled, otherwise defaults

________________________________________________________________________________

x264 r1281 32bit
download ; release notes

• built on Oct 7 2009, gcc: 4.3.4 20090220 (prerelease) (x32.generic.Komisar)
• fprofiled, -march=i686

x264 r1281 64bit
download ; release notes

• built on Oct 7 2009, gcc: 4.3.4 20090220 (prerelease) (x64.generic.Komisar)
• fprofiled, -march=core2

patched with:

• x264_win_zone_parse_fix_06.diff
• x264_hrd_pd_interlace.16_r1281.diff

Zitat

libx264 SVN-r1278:

• libx264-r1278, MinGW GCC 4.3.4, optimized for Core 2
• libx264-r1278, MinGW GCC 4.3.4, optimized for K10
• libx264-r1278, MinGW GCC 4.3.4, optimized for Pentium III
• libx264-r1278, MinGW GCC 4.3.4, without ASM (slow!)

libx264 SVN-r1278 using four slices:

• libx264-r1278, MinGW GCC 4.3.4, optimized for Core 2
• libx264-r1278, MinGW GCC 4.3.4, optimized for K10
• libx264-r1278, MinGW GCC 4.3.4, optimized for Pentium III

libx264 SVN-r1278 with AutoVAQ enabled:

• libx264-r1278, MinGW GCC 4.3.4, optimized for Core 2
• libx264-r1278, MinGW GCC 4.3.4, optimized for K10
• libx264-r1278, MinGW GCC 4.3.4, optimized for Pentium III

libx264 SVN-r1278 with MB-Tree disabled:

• libx264-r1278, MinGW GCC 4.3.4, optimized for Core 2
• libx264-r1278, MinGW GCC 4.3.4, optimized for K10
• libx264-r1278, MinGW GCC 4.3.4, optimized for Pentium III

These builds will NOT work with Avidemux 2.5.0 or older. Please update to Avidemux 2.5.1 r5369 or later now!

The patches used for my builds can be found at this location. More info on AutoVAQ (now officially committed) can be found in this thread.
Information about the new "Macroblock Tree" ratecontrol, which redefines the CRF scale, can be found in this thread.

Alles anzeigen

Achtung:

Es gab wohl ein Problem mit x264 und dem GCC 4.4.1 und 4.3.4, das zu fehlerhaften Builds geführt hat:
http://forum.doom9.org/showthread.php?p=1333035#post1333035

Ein Workaround für x264, welches das Problem beheben soll, wurde hier als Patch bereit gestellt, ist aber noch nicht committet:
http://forum.doom9.org/showpost.php?p=1333460&postcount=2443

Aktualisierte Builds wird es wohl in kürze von den üblichen verdächtigen geben. Frische libx264 Builds schon mal hier:
http://avidemux.org/admForum/viewtopic.php?pid=38530#p38530

Das dürfte gegessen sein.
Jason hat vor 20min committet
>> http://git.videolan.org/gitweb.cgi?p=x264.git;a=summary

Zitat von Pittiplatsch

Das dürfte gegessen sein.
Jason hat vor 20min committet
>> http://git.videolan.org/gitweb.cgi?p=x264.git;a=summary

Negativ. Die VBV Bugfixes sind nochmal ne andere Geschichte

Siehe dazu:
http://forum.doom9.org/showthread.php?p=1333680#post1333680

Der Workaround für den GCC 4.4.x scheint (noch) nicht committet. Da ist wohl noch mehr im argen:
http://forum.doom9.org/showpost.php?p=1333860&postcount=2458

--[EDIT]--

So, jetzt aber ;D
http://git.videolan.org/gitweb.cgi?p=x…b9e06867c9ab91d

Und noch mehr Bugfixes:

Zitat

commit e381f6d0093504a545fe4d90e3a6f178c6877cc7 r1292
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Mon Oct 12 18:40:28 2009 -0700

Fix glitches with slow-firstpass + weightb + multiref + 2pass
Bug in r1277

commit 9045fbed2d696e04d402fc8b545d761ec6c41d34 r1291
Author: Henrik Gramner <hengar-6@student.ltu.se>
Date: Mon Oct 12 15:44:13 2009 -0700

Simplify some code in b-adapt 2's trellis

commit b47aca5ed2797b8b46a0bd83d58ffdb60dffc1ef r1290
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Mon Oct 12 15:38:51 2009 -0700

Fix a very rare integer overflow in slicetype analysis
Caused an assert failure when it occurred.
Bug is as old as adaptive B-frames.

commit 2ceb2cc6e6753328d959e42823b2e6059ed007a7 r1289
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Mon Oct 12 13:14:19 2009 -0700

Reduce the aggressiveness of 2-pass VBV
Now that B-frames are properly covered, we don't have to be as aggressive.
This eliminates some issues with skyrocketing QPs in B-frames in 2-pass VBV.

commit a65d36142b765516dd80a6bed5bff7444b24f4aa r1288
Author: Jason Garrett-Glaser <darkshikari@gmail.com>
Date: Mon Oct 12 11:29:23 2009 -0700

Fix regression: disable flash detection without B-frames

commit 287641a7c54d525e0eb74925431541ffb5f9471d r1287
Author: Loren Merritt <pengvado@akuvian.org>
Date: Sat Oct 10 04:43:00 2009 +0000

change all dct arrays to 1d.
the C standard doesn't allow you to iterate 1-dimensionally over 2d arrays, and nothing other than the dsp functions themselves cares about the 2dness of dct.
this fixes a miscompilation in x264_mb_optimize_chroma_dc.

Alles anzeigen

r1330 (November 09, 2009)

Zitat

Changelog:

http://mirror01.x264.nl/x264/changelog.txt

r1332 (November 10, 2009)

Zitat

Fix one (of possibly many) miscompilations in weightp
Use NOINLINE and some emms calls to fix emms reordering issues.
This issue occurred with some GCC versions if threads > 1 and the phase of the moon was right.
Also a cosmetic in x264.c.

r1336 (November 13, 2009)

Zitat

r1336
Actually add -Wno-strict-aliasing to configure

r1335
Various weightp fixes
Make weightp results match in threaded vs non-threaded mode.
Fix two-pass with slow-firstpass.

r1334
Fix all aliasing violations
New type-punning macros perform write/read-combining without aliasing violations per the second-to-last part of 6.5.7 in the C99 specification.
GCC 4.4, however, doesn't seem to have read this part of the spec and still warns about the violations.
Regardless, it seems to fix all known aliasing miscompilations, so perhaps the GCC warning generator is just broken.
As such, add -Wno-strict-aliasing to CFLAGS.

r1333
Fix 10l in weightp on ARM

Alles anzeigen

r1339 (November 15, 2009)

Zitat

r1339
100l in deadlock fix

r1338
FLV muxing support

r1337
Fix rare deadlock introduced in weightp

r1342 (November 16, 2009)

Zitat

r1342
Fix two issues in weightp
If analysis decided on an offset of -128, x264 would create non-compliant streams.
Fix some cases with nearly all intra blocks where analysis could pick very weird weights.
Also add some asserts to check compliancy.

r1341
Allow compilation with non-Apple GCC on OS X

r1340
Use __attribute__((may_alias)) for type-punning
GCC thinks pointer casts to unions aren't valid with strict aliasing.
See http://gcc.gnu.org/onlinedocs/gcc…ype_002dpunning.
Also use M32() in y4m.c.
Enable -Wstrict-aliasing again since all such warnings are fixed.

Alles anzeigen

r1347 (November 23, 2009)

Zitat

r1347
Give more meaningful error if 1st/2nd pass resolution differ

r1346
Fix extremely rare deadlock with sync-lookahead
Patch partially by Anton Mitrofanov.

r1345
Only print weightp stats if there were P-frames

r1344
Faster lookahead with subme=1
If it hasn't been clear already, don't use subme=1 as a "fast first pass" option.
Use subme=2 instead; 1 and below now enable a fast (lower quality) lookahead mode.

r1343
Faster weightp analysis
Modify pixel_var slightly to return the necessary information and use it for weight analysis instead of sad/ssd.
Various minor cosmetics.

Alles anzeigen

r1352 (November 27, 2009)

Zitat

r1352
Much faster weightp
Move sum/ssd calculation out of lookahead and do it only once per frame.
Also various minor optimizations, cosmetics, and cleanups.

r1351
Fix bugs in fps/timestamp handling in FLV muxer

r1350
Fix bug in weightp analysis
Weights weren't reset upon early terminations, so old (wrong) weights could stick around.
Small compression improvement.

r1349
Minor deblocking optimization, update comments

r1348
Fix weightb with delta_poc_bottom
Has no effect yet, but will be required once we add TFF/BFF signalling support in interlaced mode.
Gives 0.5-0.7% better compression with proper TFF/BFF signalling.

Alles anzeigen

r1354 (November 30, 2009)

Zitat

10L in r1353
Broke mp4 output.

r1360 (December 05, 2009)

Zitat

r1360
Actually do r1356
Somehow commit r1356 got lost in the ether.I'm not sure how, but now it's fixed.

r1359
Remove some unused code from x264.c

r1358
SSSE3 version of zigzag_8x8_field
Slightly faster interlaced encoding with 8x8dct.
Helps most on Nehalem, somewhat disappointing on Conroe/Penryn.

r1357
Fix crash in interlaced with >8 refs
Crash introduced in weightp.

r1356
Significantly faster qpel-RD
Cache the results of MC, like in bidir-RD.
Slightly changes output due to the necessary reordering of satd/RD calls.
5-10% faster qpel-RD.

r1355
Add x264 prefix to functions with ffmpeg equivalents
Not important now, but will be when we add libav* input support.

Alles anzeigen