Recherche avancée

Sur d’autres sites (7311)

• x264 encoding taking longer when encoding static frames (than

  14 septembre 2015, par Danilo

  ​Hi,
  I’m using x264 for live video streaming and I’ve noticed that the thread
  responsible for encoding uses ​​more cpu (sometimes 50% more with 1920x1080) when the video stream is frozen (i.e. : camera is sending the same frame over an over again) or when I make it encode the same image over and over again.

  This seems somewhat counter intuitive to me, as I would expect x264 to use
  more processing power when encoding complex scenes other then static ones.

  My encoder settings are the following :

  1280x720 fps=25/1 timebase=0/0 bitdepth=8 cabac=0 ref=1 deblock=1:0:0 analyse=0x3:0x113
  
  me=hex subme=2 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1
  
  8x8dct=0 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=0 threads=1 lookahead_threads=0
  
  sliced_threads=0 slice_max_size=1190 nr=60 decimate=1 interlaced=0 bluray_compat=0
  
  constrained_intra=0 bframes=0 weightp=0  keyint=1200 keyint_min=120 scenecut=40
  
  intra_refresh=0 rc_lookahead=0 rc=crf mbtree=0 crf=24.0 qcomp=0.60 qpmin=0 qpmax=69
  
  qpstep=4 vbv_maxrate=1024 vbv_bufsize=350 crf_max=35.0 nal_hrd=none

  I created a github gist based on the example.c encoder bundled in x264’s
  source code and tested encoding times with it. (You can find it here :
  https://gist.github.com/danilogr/ab4976ff4e0831ab274b)

  Average encoding time for the static scene is 38% bigger than for a scene
  with movements. (You can find my test case and also the output from my test
  encoder on the link above).

  ​​
  ​I’ve also noticed that by setting ​​scenecut=0, subme=0, trellis=0 and me=dia I can get rid of this problem​, but with noticeable quality​ decrease.

  ​
  ​Could anyone, please, shed some light on the reasons for this odd behavior ?
  ​Also, what can be done in order to avoid this situation without a major decrease in quality ?​

• x264 encoding taking longer when encoding static frames (than

  14 septembre 2015, par Danilo

  ​Hi,
  I’m using x264 for live video streaming and I’ve noticed that the thread
  responsible for encoding uses ​​more cpu (sometimes 50% more with 1920x1080) when the video stream is frozen (i.e. : camera is sending the same frame over an over again) or when I make it encode the same image over and over again.

  This seems somewhat counter intuitive to me, as I would expect x264 to use
  more processing power when encoding complex scenes other then static ones.

  My encoder settings are the following :

  1280x720 fps=25/1 timebase=0/0 bitdepth=8 cabac=0 ref=1 deblock=1:0:0 analyse=0x3:0x113
  
  me=hex subme=2 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1
  
  8x8dct=0 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=0 threads=1 lookahead_threads=0
  
  sliced_threads=0 slice_max_size=1190 nr=60 decimate=1 interlaced=0 bluray_compat=0
  
  constrained_intra=0 bframes=0 weightp=0  keyint=1200 keyint_min=120 scenecut=40
  
  intra_refresh=0 rc_lookahead=0 rc=crf mbtree=0 crf=24.0 qcomp=0.60 qpmin=0 qpmax=69
  
  qpstep=4 vbv_maxrate=1024 vbv_bufsize=350 crf_max=35.0 nal_hrd=none

  I created a github gist based on the example.c encoder bundled in x264’s
  source code and tested encoding times with it. (You can find it here :
  https://gist.github.com/danilogr/ab4976ff4e0831ab274b)

  Average encoding time for the static scene is 38% bigger than for a scene
  with movements. (You can find my test case and also the output from my test
  encoder on the link above).

  ​​
  ​I’ve also noticed that by setting ​​scenecut=0, subme=0, trellis=0 and me=dia I can get rid of this problem​, but with noticeable quality​ decrease.

  ​
  ​Could anyone, please, shed some light on the reasons for this odd behavior ?
  ​Also, what can be done in order to avoid this situation without a major decrease in quality ?​

• avutil/mathematics : speed up av_gcd by using Stein’s binary GCD algorithm

  11 octobre 2015, par Ganesh Ajjanagadde

  avutil/mathematics : speed up av_gcd by using Stein’s binary GCD algorithm
  This uses Stein’s binary GCD algorithm :
  
  https://en.wikipedia.org/wiki/Binary_GCD_algorithm
  
  to get a roughly 4x speedup over Euclidean GCD on standard architectures
  
  with a compiler intrinsic for ctzll, and a roughly 2x speedup otherwise.
  
  At the moment, the compiler intrinsic is used on GCC and Clang due to
  
  its easy availability.
  Quick note regarding overflow : yes, subtractions on int64_t can, but the
  
  llabs takes care of that. The llabs is also guaranteed to be safe, with
  
  no annoying INT64_MIN business since INT64_MIN being a power of 2, is
  
  shifted down before being sent to llabs.
  The binary GCD needs ff_ctzll, an extension of ff_ctz for long long (int64_t). On
  
  GCC, this is provided by a built-in. On Microsoft, there is a
  
  BitScanForward64 analog of BitScanForward that should work ; but I can’t confirm.
  
  Apparently it is not available on 32 bit builds ; so this may or may not
  
  work correctly. On Intel, per the documentation there is only an
  
  intrinsic for _bit_scan_forward and people have posted on forums
  
  regarding _bit_scan_forward64, but often their documentation is
  
  woeful. Again, I don’t have it, so I can’t test.
  As such, to be safe, for now only the GCC/Clang intrinsic is added, the rest
  
  use a compiled version based on the De-Bruijn method of Leiserson et al :
  
  http://supertech.csail.mit.edu/papers/debruijn.pdf.
  Tested with FATE, sample benchmark (x86-64, GCC 5.2.0, Haswell)
  
  with a START_TIMER and STOP_TIMER in libavutil/rationsl.c, followed by a
  
  make fate.
  aac-am00_88.err :
  
  builtin :
  
  714 decicycles in av_gcd,    4095 runs,      1 skips
  de-bruijn :
  
  1440 decicycles in av_gcd,    4096 runs,      0 skips
  previous :
  
  2889 decicycles in av_gcd,    4096 runs,      0 skips
  Signed-off-by : Ganesh Ajjanagadde <gajjanagadde@gmail.com>
  
  Signed-off-by : Michael Niedermayer <michael@niedermayer.cc>
  

  • [DH] libavutil/intmath.h
  • [DH] libavutil/mathematics.c