News:

--

Main Menu

libx264 GIT builds

Started by LoRd_MuldeR, January 30, 2009, 12:39:23 AM

Previous topic - Next topic
Print
Go Down Pages 1 ... 13 14 15 16
User actions

LoRd_MuldeR

#210
January 29, 2011, 07:43:53 PM
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

LoRd_MuldeR

#211
January 31, 2011, 06:36:23 PM
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

LoRd_MuldeR

#212
February 20, 2011, 08:47:04 PM
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

LoRd_MuldeR

#213
March 25, 2011, 11:31:26 PM
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

LoRd_MuldeR

#214
May 27, 2011, 04:34:24 PM
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

LoRd_MuldeR

#215
June 24, 2011, 02:12:07 PM
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

LoRd_MuldeR

#216
July 11, 2011, 03:27:40 PM
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

LoRd_MuldeR

#217
August 11, 2011, 09:08:56 AM Last Edit: August 11, 2011, 06:08:43 PM by LoRd_MuldeR
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

LoRd_MuldeR

#218
September 05, 2011, 03:05:20 PM Last Edit: September 06, 2011, 12:23:51 AM by LoRd_MuldeR
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

LoRd_MuldeR

#219
September 25, 2011, 09:03:38 PM Last Edit: September 26, 2011, 09:24:57 AM by LoRd_MuldeR
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

dado023

#220
October 27, 2011, 04:31:35 PM Last Edit: November 03, 2011, 02:35:48 AM by dado023
new revision is out :)
Waiting patiently  :o

EDIT:
Got it from here:
http://code.google.com/p/mulder/downloads/detail?name=libx264-118-r2106M-win32.7z&can=2&q=&sort=-uploaded

thank you in the name of all avidemux junkies

PS: what is the minimum avidemux version to use it?

LoRd_MuldeR

#221
December 07, 2011, 11:30:12 PM
libx264 SVN-r2120 library pack:

Changelog:
Code Select
commit 0c7dab9c2a106ce3ee5d6ad7282afb49e1cc3954 [revision 2120]
Author: Jason Garrett-Glaser <jason@x264.com>
Date:   Tue Dec 6 14:39:21 2011 -0800

    Fix regression in r2118
    Broke trellis with i16x16 macroblocks.

commit 0637cd67cb245fce5ba190fa4b9c341319ea2b37 [revision 2119]
Author: Jason Garrett-Glaser <jason@x264.com>
Date:   Wed Nov 30 13:02:12 2011 -0800

    Modify MBAFF chroma deblock functions to handle U/V at the same time
    Allows for more convenient asm implementations.

commit 67f1fdc4d9c030568eac8cf9ab9d0bb249f520db [revision 2118]
Author: Jason Garrett-Glaser <jason@x264.com>
Date:   Thu Nov 10 16:16:13 2011 -0800

    CABAC trellis optimizations: use SIMD quant
    Significant speed increase, minor change in output due to rounding.

commit e047b3c475cd42b6647397a244e239ebfca53bf6 [revision 2117]
Author: Steven Walters <kemuri9@gmail.com>
Date:   Sun Nov 6 09:48:30 2011 -0800

    YUV range detection and support for x264CLI
    Two new options: --input-range and --range.
    --input-range forces the range of the input in case of misdetection; auto by default.
    -- range sets the range of the output; x264cli will convert if necessary, TV by default.
    --fullrange is now removed as a CLI option (but the libx264 API is unchanged).

commit 00df989cc06208050230756525633438d76b5a6a [revision 2116]
Author: Kieran Kunhya <kieran@kunhya.com>
Date:   Fri Nov 4 20:09:13 2011 +0000

    Pass through user data

commit 04a0aeefd2f5b152c5dbca4a1c6569bd27c9f721 [revision 2115]
Author: Jason Garrett-Glaser <jason@x264.com>
Date:   Thu Oct 27 14:05:56 2011 -0700

    Remove unpredictable branch in CABAC dqp

commit 4185ee883b04d9cee57a64fdebd153830b7b27ba [revision 2114]
Author: Loren Merritt <pengvado@akuvian.org>
Date:   Sun Oct 23 23:15:11 2011 +0000

    x86inc: AVX symmetry optimization
    3-arg AVX ops with a memory arg can only have it in src2,
    whereas SSE emulation of 3-arg prefers to have it in src1 (i.e. the move).
    So, if the op is symmetric and the wrong one is memory, swap them.
    Eliminates redundant moves in some cases when using 3-operand without AVX with memory arguments.
    Also fix movss and movsd in some cases, and flag shufps correctly as float.

commit cc129adcaaf5604f3d4fea9ebcb289403192a741 [revision 2113]
Author: Anton Mitrofanov <Bugmaster@narod.ru>
Date:   Tue Nov 29 13:45:13 2011 -0800

    checkasm: shut up gcc warnings, fix some naming of functions in results

commit f0ccc98bb747b8ee0fe9329f4205cf382788bb89 [revision 2112]
Author: Mans Rullgard <mans@mansr.com>
Date:   Mon Nov 28 16:29:12 2011 -0800

    checkasm: fix build on ARM
    Because of how ALIGNED_ARRAY_16 is defined on ARM, array initialisers cannot be used here.  Use memset() instead.

commit d8d8e756b1fee72b4771761d6aa4cfb31edc0b67 [revision 2111]
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Sat Nov 12 01:31:49 2011 +0400

    Improve makefile rules
    Remove the need for "make clean" after most reconfigures.

commit e6d33a931c08918e78dcae97e4d80d0c3411bf2c [revision 2110]
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Sat Nov 12 00:47:48 2011 +0400

    Mark some local functions as static, cosmetics

commit e0c11dc6e283569606aaa97767401c6a13c2529d [revision 2109]
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Fri Nov 11 23:19:02 2011 +0400

    Fix crash if timecode file opening fails

commit a14db080c3fdba4cadc38152a292bb1fa216d50e [revision 2108]
Author: Fabian Greffrath <fabian+debian@greffrath.com>
Date:   Fri Nov 11 13:25:43 2011 -0800

    Configure: force PIC for shared build on PARISC and MIPS

commit 6a0bd421bf5fd006012ddcd1be2072a8736b2d27 [revision 2107]
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Sat Oct 22 19:41:07 2011 +0400

    Improve yasm version check
    Previous check allowed certain earlier versions that weren't fully compatible.

commit 07efeb45db224b7757880d4d63bb549fb454f6db [revision 2106]
Author: Jason Garrett-Glaser <jason@x264.com>
Date:   Tue Oct 18 14:30:26 2011 -0700

    Add fenc prefetching to adaptive quant
    Many fewer cache misses, faster adaptive quant.

These builds will NOT work with Avidemux 2.5.4 or older. Please update to Avidemux 2.5.5 r7694 or later now!
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

LoRd_MuldeR

#222
January 01, 2012, 09:58:19 PM
libx264 SVN-r2120 library pack (updated for the Avimdeux 2.5.6 release):
Changelog:
Code Select
commit 0c7dab9c2a106ce3ee5d6ad7282afb49e1cc3954 r2120
Author: Jason Garrett-Glaser <jason@x264.com>
Date:   Tue Dec 6 14:39:21 2011 -0800

    Fix regression in r2118
    Broke trellis with i16x16 macroblocks.

commit 0637cd67cb245fce5ba190fa4b9c341319ea2b37 r2119
Author: Jason Garrett-Glaser <jason@x264.com>
Date:   Wed Nov 30 13:02:12 2011 -0800

    Modify MBAFF chroma deblock functions to handle U/V at the same time
    Allows for more convenient asm implementations.

commit 67f1fdc4d9c030568eac8cf9ab9d0bb249f520db r2118
Author: Jason Garrett-Glaser <jason@x264.com>
Date:   Thu Nov 10 16:16:13 2011 -0800

    CABAC trellis optimizations: use SIMD quant
    Significant speed increase, minor change in output due to rounding.

commit e047b3c475cd42b6647397a244e239ebfca53bf6 r2117
Author: Steven Walters <kemuri9@gmail.com>
Date:   Sun Nov 6 09:48:30 2011 -0800

    YUV range detection and support for x264CLI
    Two new options: --input-range and --range.
    --input-range forces the range of the input in case of misdetection; auto by default.
    -- range sets the range of the output; x264cli will convert if necessary, TV by default.
    --fullrange is now removed as a CLI option (but the libx264 API is unchanged).

commit 00df989cc06208050230756525633438d76b5a6a r2116
Author: Kieran Kunhya <kieran@kunhya.com>
Date:   Fri Nov 4 20:09:13 2011 +0000

    Pass through user data

commit 04a0aeefd2f5b152c5dbca4a1c6569bd27c9f721 r2115
Author: Jason Garrett-Glaser <jason@x264.com>
Date:   Thu Oct 27 14:05:56 2011 -0700

    Remove unpredictable branch in CABAC dqp

commit 4185ee883b04d9cee57a64fdebd153830b7b27ba r2114
Author: Loren Merritt <pengvado@akuvian.org>
Date:   Sun Oct 23 23:15:11 2011 +0000

    x86inc: AVX symmetry optimization
    3-arg AVX ops with a memory arg can only have it in src2,
    whereas SSE emulation of 3-arg prefers to have it in src1 (i.e. the move).
    So, if the op is symmetric and the wrong one is memory, swap them.
    Eliminates redundant moves in some cases when using 3-operand without AVX with memory arguments.
    Also fix movss and movsd in some cases, and flag shufps correctly as float.

commit cc129adcaaf5604f3d4fea9ebcb289403192a741 r2113
Author: Anton Mitrofanov <Bugmaster@narod.ru>
Date:   Tue Nov 29 13:45:13 2011 -0800

    checkasm: shut up gcc warnings, fix some naming of functions in results

commit f0ccc98bb747b8ee0fe9329f4205cf382788bb89 r2112
Author: Mans Rullgard <mans@mansr.com>
Date:   Mon Nov 28 16:29:12 2011 -0800

    checkasm: fix build on ARM
    Because of how ALIGNED_ARRAY_16 is defined on ARM, array initialisers cannot be used here.  Use memset() instead.

commit d8d8e756b1fee72b4771761d6aa4cfb31edc0b67 r2111
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Sat Nov 12 01:31:49 2011 +0400

    Improve makefile rules
    Remove the need for "make clean" after most reconfigures.

commit e6d33a931c08918e78dcae97e4d80d0c3411bf2c r2110
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Sat Nov 12 00:47:48 2011 +0400

    Mark some local functions as static, cosmetics

commit e0c11dc6e283569606aaa97767401c6a13c2529d r2109
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Fri Nov 11 23:19:02 2011 +0400

    Fix crash if timecode file opening fails

commit a14db080c3fdba4cadc38152a292bb1fa216d50e r2108
Author: Fabian Greffrath <fabian+debian@greffrath.com>
Date:   Fri Nov 11 13:25:43 2011 -0800

    Configure: force PIC for shared build on PARISC and MIPS

commit 6a0bd421bf5fd006012ddcd1be2072a8736b2d27 r2107
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Sat Oct 22 19:41:07 2011 +0400

    Improve yasm version check
    Previous check allowed certain earlier versions that weren't fully compatible.

commit 07efeb45db224b7757880d4d63bb549fb454f6db r2106
Author: Jason Garrett-Glaser <jason@x264.com>
Date:   Tue Oct 18 14:30:26 2011 -0700

    Add fenc prefetching to adaptive quant
    Many fewer cache misses, faster adaptive quant.

These builds will NOT work with Avidemux 2.5.4 or older. Please update to Avidemux 2.5.6 r7716 or later now!
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

LoRd_MuldeR

#223
January 21, 2012, 02:24:10 PM Last Edit: March 01, 2012, 11:11:21 PM by LoRd_MuldeR
libx264 SVN-r2146 library pack:
Changelog:
Code Select
commit bcd41dbcaa4430b2118d9f6828c2b9635cf9d58d r2146
Author: Loren Merritt <pengvado@akuvian.org>
Date:   Wed Jan 18 15:47:07 2012 -0800

    Fix trellis 2 + subme >= 8
    Trellis didn't return a boolean value as it was supposed to.
    Regression in r2143-5.

commit 748fe16c1303b89d2a1d0378addd83fb4198f51a r2145
Author: Loren Merritt <pengvado@akuvian.org>
Date:   Fri Jan 6 15:53:29 2012 +0000

    CABAC trellis opts part 4: x86_64 asm
    Another 20% faster.
    18k->12k codesize.
   
    This patch series may have a large impact on encoding speed.
    For example, 24% faster at --preset slower --crf 23 with 720p parkjoy.
    Overall speed increase is proportional to the cost of trellis (which is proportional to bitrate, and much more with --trellis 2).

commit cfdb36ece729209631f7213506685ae733d7f5d4 r2144
Author: Loren Merritt <pengvado@akuvian.org>
Date:   Fri Jan 6 15:53:04 2012 +0000

    CABAC trellis opts part 3: make some arrays non-static

commit 65bd12ae875a768a06b67ec6297dec18323e0768 r2143
Author: Loren Merritt <pengvado@akuvian.org>
Date:   Thu Dec 22 17:56:06 2011 +0000

    CABAC trellis opts part 2: C optimizations
   
    Hoist the branch on coef value out of the loop over node contexts.
    Special cases for each possible coef value (0,1,n).
    Special case for dc-only blocks.
    Template the main loop for two common subsets of nodes, to avoid a bunch of branches about which nodes are live.
    Use the nonupdating version of cabac_size_decision in more cases, and omit those bins from the node struct.
    CABAC offsets are now compile-time constants.
    Change TRELLIS_SCORE_MAX from a specific constant to anything negative, which is cheaper to test.
    Remove dct_weight2_zigzag[], since trellis has to lookup zigzag[] anyway.
   
    60% faster on x86_64.
    25k->18k codesize.

commit e176619d010fc32c970c7ab7a769bbfbe2665f61 r2142
Author: Loren Merritt <pengvado@akuvian.org>
Date:   Thu Dec 22 17:55:06 2011 +0000

    CABAC trellis opts part 1: minor change in output
    Due to different tie-break order.

commit 4e87f36a0e1a78242f04db611e06f80b6b38d900 r2141
Author: Henrik Gramner <hengar-6@student.ltu.se>
Date:   Sun Jan 8 04:14:10 2012 +0100

    x86inc improvements for 64-bit
   
    Add support for all x86-64 registers
    Prefer caller-saved register over callee-saved on WIN64
    Support up to 15 function arguments

commit 84a06e611aff1267a720bf9552b3bcf263bd83b5 r2140
Author: Ilia Valiakhmetov <zakne0ne@gmail.com>
Date:   Sun Jan 15 04:47:58 2012 -0600

    High bit depth SSE2/AVX add8x8_idct8 and add16x16_idct8
    From Google Code-In.

commit c605e3174410ba5c7d1d0a777082e2397734d637 r2139
Author: Edward Wang <edward.c.wang@compdigitec.com>
Date:   Wed Jan 4 15:35:54 2012 -0800

    MMX/SSE2/AVX predict_8x16_p, high bit depth fdct8
    From Google Code-In.

commit 6b06f6d3f7f800dca1a4ea154f54427d5b3cea2b r2138
Author: Jason Garrett-Glaser <jason@x264.com>
Date:   Thu Dec 22 14:03:15 2011 -0800

    XOP 8-bit fDCT
    Use integer MAC for one of the SUMSUB passes.  About a dozen cycles faster for 16x16.

commit c4b54c83629bb92af6c4836a8859e9432dc7333a r2137
Author: Cristian Militaru <cristipiticul@yahoo.com>
Date:   Wed Jan 4 12:38:08 2012 -0800

    High bit depth intra_sad_x3_4x4
    From Google Code-In.

commit c032fbaa3801fb4cf8dd1dd95a6479ca5bd262e2 r2136
Author: Jason Garrett-Glaser <jason@x264.com>
Date:   Thu Dec 8 13:45:41 2011 -0800

    Use a large LUT for CAVLC zero-run bit codes
    Helps the most with trellis and RD, but also helps with bitstream writing.
    Seems at worst neutral even in the extreme case of a CPU with small L2 cache (e.g. ARM Cortex A8).

commit ebb1429e2d24f57aa4ea75284386a15f2eab553e r2135
Author: Matt Habel <habelinc@gmail.com>
Date:   Fri Dec 16 23:16:09 2011 -0800

    High bit depth intra_sad_x3_8x8, intra_satd_x3_4x4/8x8c/16x16
    Also add an ACCUM macro to handle accumulator-induced add-or-swap more concisely.

commit 6d921c5bdefae1a733a3a4c29d88ea15fcece76e r2134
Author: Shitiz Garg <mail@dragooon.net>
Date:   Sat Dec 3 15:34:57 2011 -0800

    MMX 10-bit predict_8x8c_h and predict_8x16c_h
    From Google Code-In.

commit 47cdaa9c3d8197d4deb711d9bcc4af869ef8a426 r2133
Author: Aaron Schmitz <me@aaronschmitz.com>
Date:   Wed Nov 30 00:15:45 2011 -0600

    Some MBAFF x86 assembly functions.
    deblock_chroma_420_mbaff, plus 422/422_intra_mbaff implemented using existing functions.
    From Google Code-In.

commit 027b05e0a22421e477847506a205a49b151ae5bf r2132
Author: George Stephanos <gaf.stephanos@gmail.com>
Date:   Thu Dec 1 16:53:45 2011 -0800

    More ARM NEON assembly functions
    predict_8x8_v, predict_4x4_dc_top, predict_8x8_ddl, predict_8x8_ddr, predict_8x8_vl, predict_8x8_vr, predict_8x8_hd, predict_8x8_hu.
    From Google Code-In.

commit 658a3585b74f77fd8f78588f3f39e0abefb104c4 r2131
Author: Ilia <zakne0ne@gmail.com>
Date:   Mon Nov 28 05:20:09 2011 -0800

    More 4:2:2 asm functions
    High bit depth version of deblock_h_chroma_422.
    Regular and high bit depth versions of deblock_h_chroma_intra_422.
    High bit depth pixel_vsad.
    SSE2 high bit depth and MMX 8-bit predict_8x8_vl.
    Our first GCI patch this year!

commit 978abe065737089913feccffece483bc69a9e5b0 r2130
Author: Henrik Gramner <hengar-6@student.ltu.se>
Date:   Thu Dec 8 16:14:35 2011 +0100

    SSE2 and SSSE3 versions of sub8x16_dct_dc
    Also slightly faster sub8x8_dct_dc

commit 61a78a1b417595c4b5d7ef6831692904a243a9fc r2129
Author: Steven Walters <kemuri9@gmail.com>
Date:   Mon Dec 5 08:46:34 2011 -0500

    Resize filter updates
    Use AVPixFmtDescriptors to pick the most compatible x264 csp for any pixel format.
    Fix deprecated use of av_set_int.
    Now requires libavutil >= 51.19.0

commit bc6c98cf4f76c779c8c07f43aa97ac29b1150bc0 r2128
Author: Oka Motofumi <chikuzen.mo@gmail.com>
Date:   Thu Jan 5 14:23:50 2012 -0800

    Add out-of-tree build support

commit f33c8cb0f8fff7a83100b3e9d15baba53c6f6a35 r2127
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Fri Dec 16 18:17:00 2011 +0400

    Limit SSIM to 100db
    Avoids floating point error for infinite SSIM (lossless).

commit c0d698859c36be611d465f968762f042853be817 r2126
Author: Reynaldo H. Verdejo Pinochet <reynaldo@collabora.com>
Date:   Wed Jan 4 13:16:12 2012 -0300

    Fix wrong conditional inclusion of inttypes.h
    inttypes.h is required by encoder/ratecontrol.c for SCNxxx macros, and HAVE_STDINT_H does not imply having inttypes.h.
    stdint.h is a subset of inttypes.h, but this isn't enough for x264.
    This change fixes building x264 with Android's toolchain.

commit b081d179e741ceffee2217f6fda06779693dce56 r2125
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Wed Dec 21 11:08:56 2011 +0400

    Fix crash with sliced threads and input height <= 112

commit 64da5f9df46ac33a5a6b56ca1510d2082e6fbb62 r2124
Author: Phillip Blucas <pblucas@gmail.com>
Date:   Mon Dec 19 17:43:41 2011 -0600

    Fix loading custom 8x8 chroma quant matrices in 4:4:4

commit 4c08e42504af81cdbe5789a309e868ca8eda2c1f r2123
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Fri Dec 16 01:48:07 2011 +0400

    Fix PCM cost overflow

commit 489a9b2d04c4828877930d2a9104ce93dde8cb85 r2122
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Fri Dec 9 01:54:22 2011 +0400

    Fix overflow in 8-bit x86 vsad asm function

commit c291a9d09263708e9d9f02e28f8442fdbe46bb06 r2121
Author: Anton Mitrofanov <BugMaster@narod.ru>
Date:   Wed Dec 7 19:14:52 2011 +0400

    Fix crash in --fullhelp when compiled against recent ffmpeg
    Don't assume all pixel formats have a description.

These builds will NOT work with Avidemux 2.5.5 or older. Please update to Avidemux 2.5.6 r7716 or later now!
You asked me once, what was in Room 101. I told you that you knew the answer already.
Everyone knows it. The thing that is in Room 101 is the worst thing in the world.

MuldeR's OpenSource stuff: http://muldersoft.com/

dado023

#224
March 16, 2012, 04:51:20 PM
hi mulder,

can we expect new build any time soon  :-\
Print
Go Up Pages 1 ... 13 14 15 16
User actions