[x265] [PATCH] no-rdo: Use entropy encoder for bit estimation

2013-10-31 Thread deepthidevaki
# HG changeset patch # User Deepthi Devaki deepthidev...@multicorewareinc.com # Date 1383203307 -19800 # Node ID 4b4332d038832ab8812773d618b38329ec75ae4b # Parent ec6b4d35f11053b06d0e1ea46df798ff89a4c127 no-rdo: Use entropy encoder for bit estimation. Instead of me-bit estimation, use entropy

[x265] [PATCH] asm: fix the bug which occured at win32 compile

2013-10-31 Thread yuvaraj
# HG changeset patch # User Yuvaraj Venkatesh yuva...@multicorewareinc.com # Date 1383204505 -19800 # Thu Oct 31 12:58:25 2013 +0530 # Node ID f6e35bfe1fd67668cc3c18bc41260a3f1d71dffc # Parent a406f7c1dd3bcc471c0885ad0720e60fa0007983 asm: fix the bug which occured at win32 compile diff -r

Re: [x265] [PATCH] no-rdo: Use entropy encoder for bit estimation

2013-10-31 Thread Deepthi Nandakumar
Posting the efficiency results FYI. The performance drop will be taken care of once the rd 0/1 refactor is complete. Before BasketballPass_416x240_50 (11.04 fps), 328.96 kb/s, Global PSNR: 35.613 After BasketballPass_416x240_50(10.70 fps), 326.64 kb/s, Global PSNR: 35.609 Before

[x265] [PATCH] asm: reduce large code size in sad_16xN, sad_32xN for better cache performance

2013-10-31 Thread dnyaneshwar
# HG changeset patch # User Dnyaneshwar Gorade dnyanesh...@multicorewareinc.com # Date 1383212434 -19800 # Thu Oct 31 15:10:34 2013 +0530 # Node ID 5d6ed411995acd674b838f989385c61039760780 # Parent a3235963489588dea19415cf7d4c6e6f979dba41 asm: reduce large code size in sad_16xN, sad_32xN for

[x265] [PATCH] compress: cleanup, remove unused data structs

2013-10-31 Thread deepthi
# HG changeset patch # User Deepthi Nandakumar deep...@multicorewareinc.com # Date 1383214067 -19800 # Node ID c92fadfe223a9740ab353f8abc0f7b8723a46287 # Parent ec6b4d35f11053b06d0e1ea46df798ff89a4c127 compress: cleanup, remove unused data structs diff -r ec6b4d35f110 -r c92fadfe223a

[x265] [PATCH] asm: Optimized sad_64xN for better cache performance. Reduced lea instruction by half. Performance gain is average +5x w.r.t. previous asm code

2013-10-31 Thread dnyaneshwar
# HG changeset patch # User Dnyaneshwar Gorade dnyanesh...@multicorewareinc.com # Date 1383216695 -19800 # Thu Oct 31 16:21:35 2013 +0530 # Node ID 86ff1a3ec89720a73325148e8ac01ec1dbdab3c2 # Parent 5d6ed411995acd674b838f989385c61039760780 asm: Optimized sad_64xN for better cache performance.

[x265] [PATCH] asm: Optimized sad_48x64: +5x and sad_24x32: +2x asm routines

2013-10-31 Thread dnyaneshwar
# HG changeset patch # User Dnyaneshwar Gorade dnyanesh...@multicorewareinc.com # Date 1383218218 -19800 # Thu Oct 31 16:46:58 2013 +0530 # Node ID 515b0af5eb805407d40ead87fd29a8c32118d3a2 # Parent 86ff1a3ec89720a73325148e8ac01ec1dbdab3c2 asm: Optimized sad_48x64: +5x and sad_24x32: +2x asm

[x265] [PATCH] aq: set qp, lambda for every CU in the row before processing the CU

2013-10-31 Thread Aarthi Thirumalai
# HG changeset patch # User Aarthi Thirumalai # Date 1383219394 -19800 # Thu Oct 31 17:06:34 2013 +0530 # Node ID 0e321e5dd34869c0054a5c218cd14e5dfd30fd38 # Parent 9a0da4e6d9e363e383eae7243f0c64026a5f6d00 aq: set qp, lambda for every CU in the row before processing the CU enabled bUseDQP

[x265] [PATCH] assembly code for pixel_sad_x4_12x16

2013-10-31 Thread yuvaraj
# HG changeset patch # User Yuvaraj Venkatesh yuva...@multicorewareinc.com # Date 1383219583 -19800 # Thu Oct 31 17:09:43 2013 +0530 # Node ID 56368c1e4df4d043eadc1352d75542f77c405077 # Parent 1c0f0aa845b1c8a520aa91c9fffc68144effd75a assembly code for pixel_sad_x4_12x16 diff -r 1c0f0aa845b1

[x265] [PATCH] Generating sum ssd values for weightp decision in lookahead

2013-10-31 Thread shazeb
# HG changeset patch # User Shazeb Nawaz Khan sha...@multicorewareinc.com # Date 1383220857 -19800 # Thu Oct 31 17:30:57 2013 +0530 # Node ID c2cba0b6736c52ad177afa6a4974dfe2d35f56ef # Parent 9a0da4e6d9e363e383eae7243f0c64026a5f6d00 Generating sum ssd values for weightp decision in

[x265] [PATCH 3 of 4] asm: fix bug in luma_p2s and active it in encoder

2013-10-31 Thread Min Chen
# HG changeset patch # User Min Chen chenm...@163.com # Date 1383224489 -28800 # Node ID 21dbf988079b0e33265ae48578c26347cc779fbe # Parent a64e813de628856f6916bdafcc069f19d9d81082 asm: fix bug in luma_p2s and active it in encoder diff -r a64e813de628 -r 21dbf988079b

[x265] [PATCH 4 of 4] asm: chroma_p2s to replace ipfilter_p2s

2013-10-31 Thread Min Chen
# HG changeset patch # User Min Chen chenm...@163.com # Date 1383224503 -28800 # Node ID 4a40c4069ad12bc72a1c443b45a91c65d319d35d # Parent 21dbf988079b0e33265ae48578c26347cc779fbe asm: chroma_p2s to replace ipfilter_p2s diff -r 21dbf988079b -r 4a40c4069ad1

Re: [x265] [PATCH] asm: reduce large code size in sad_16xN, sad_32xN for better cache performance

2013-10-31 Thread chen
seems right___ x265-devel mailing list x265-devel@videolan.org https://mailman.videolan.org/listinfo/x265-devel

Re: [x265] [PATCH] asm: reduce large code size in pixel_sad_8x32 for better cache performance

2013-10-31 Thread chen
At 2013-10-31 16:37:24,dnyanesh...@multicorewareinc.com wrote: # HG changeset patch # User Dnyaneshwar Gorade dnyanesh...@multicorewareinc.com # Date 1383208572 -19800 # Thu Oct 31 14:06:12 2013 +0530 # Node ID a3235963489588dea19415cf7d4c6e6f979dba41 # Parent

Re: [x265] [PATCH] asm: Optimized sad_64xN for better cache performance. Reduced lea instruction by half. Performance gain is average +5x w.r.t. previous asm code

2013-10-31 Thread chen
right except pixel_sad_64x32, it is loop 2 times only, I am not sure which is better between loop 4 times and all unroll ___ x265-devel mailing list x265-devel@videolan.org https://mailman.videolan.org/listinfo/x265-devel

[x265] [PATCH Review Only, Need Test bench Intergration] blockcpy_pp all partion sizes

2013-10-31 Thread praveen
# HG changeset patch # User Praveen Tiwari # Date 1383225672 -19800 # Node ID 055109bdb24c60618b9fe1eb98fe801d547726d9 # Parent ec6b4d35f11053b06d0e1ea46df798ff89a4c127 blockcpy_pp all partion sizes diff -r ec6b4d35f110 -r 055109bdb24c source/common/x86/blockcopy8.asm --- /dev/null Thu Jan 01

[x265] [PATCH 3 of 3] Lost in merge: (common files)check need of signed/unsigned int

2013-10-31 Thread kavitha
# HG changeset patch # User Kavitha Sampath kavi...@multicorewareinc.com # Date 1383226415 -19800 # Thu Oct 31 19:03:35 2013 +0530 # Node ID 0a2f6547b5b6213f6bb9ebe1374f70659aef0540 # Parent 2cdef1dd17b2d66dc5a84f2e40ae3130a3f9e325 Lost in merge: (common files)check need of signed/unsigned

Re: [x265] [PATCH] asm: Optimized sad_64xN for better cache performance. Reduced lea instruction by half. Performance gain is average +5x w.r.t. previous asm code

2013-10-31 Thread Steve Borho
On Thu, Oct 31, 2013 at 5:53 AM, dnyanesh...@multicorewareinc.com wrote: # HG changeset patch # User Dnyaneshwar Gorade dnyanesh...@multicorewareinc.com # Date 1383216695 -19800 # Thu Oct 31 16:21:35 2013 +0530 # Node ID 86ff1a3ec89720a73325148e8ac01ec1dbdab3c2 # Parent

Re: [x265] [PATCH] Assembly routines for vertical luma filter for all block sizes

2013-10-31 Thread Steve Borho
On Thu, Oct 31, 2013 at 10:50 AM, naba...@multicorewareinc.com wrote: # HG changeset patch # User Nabajit Deka # Date 1383234610 -19800 # Thu Oct 31 21:20:10 2013 +0530 # Node ID e11e3328ff72aabab96ee2e6dac802a1284fe2d1 # Parent 9a0da4e6d9e363e383eae7243f0c64026a5f6d00 Assembly

[x265] [PATCH] Assembly routines for vertical luma filter for all block sizes

2013-10-31 Thread nabajit
# HG changeset patch # User Nabajit Deka # Date 1383235449 -19800 # Thu Oct 31 21:34:09 2013 +0530 # Node ID b3a3131e1c32e7cd4cef833455ca8c541307f5e1 # Parent 5584339605f0c3d1abb1e8ac3a1998a55a7cbba5 Assembly routines for vertical luma filter for all block sizes diff -r 5584339605f0 -r

[x265] [PATCH] Set up the function pointers for luma vertical filters

2013-10-31 Thread nabajit
# HG changeset patch # User Nabajit Deka # Date 1383235660 -19800 # Thu Oct 31 21:37:40 2013 +0530 # Node ID 9ae563379b0fb57e5701fde2932277dddea4598a # Parent b3a3131e1c32e7cd4cef833455ca8c541307f5e1 Set up the function pointers for luma vertical filters. diff -r b3a3131e1c32 -r

Re: [x265] [PATCH 4 of 4] asm: chroma_p2s to replace ipfilter_p2s

2013-10-31 Thread Steve Borho
On Thu, Oct 31, 2013 at 8:03 AM, Min Chen chenm...@163.com wrote: # HG changeset patch # User Min Chen chenm...@163.com # Date 1383224503 -28800 # Node ID 4a40c4069ad12bc72a1c443b45a91c65d319d35d # Parent 21dbf988079b0e33265ae48578c26347cc779fbe asm: chroma_p2s to replace ipfilter_p2s

Re: [x265] [PATCH] refactor: (common files)check need of signed/unsigned int

2013-10-31 Thread Steve Borho
snipped diff -r 0666d56aaa42 -r b7f2aac2b3fe source/x265.h --- a/source/x265.h Mon Oct 28 16:13:05 2013 +0530 +++ b/source/x265.h Thu Oct 31 14:33:52 2013 +0530 @@ -130,7 +130,7 @@ typedef struct { void* planes[3]; -int stride[3]; +intptr_t stride[3];

[x265] [PATCH] api: add zero-latency tune target

2013-10-31 Thread Steve Borho
# HG changeset patch # User Steve Borho st...@borho.org # Date 1383254053 18000 # Thu Oct 31 16:14:13 2013 -0500 # Node ID 8afb161419df87f7c1b088e057e7ce6dab654322 # Parent 885e41fac72615b270e558cf85273d1b8425a2c8 api: add zero-latency tune target This just disables lookahead and B frames