Recherche avancée

Sur d’autres sites (9189)

• avcodec/jpeg2000 : replace naive pow call with smarter exp2fi

  8 décembre 2015, par Ganesh Ajjanagadde

  avcodec/jpeg2000 : replace naive pow call with smarter exp2fi
  pow is a very wasteful function for this purpose. A low hanging fruit
  
  would be simply to replace with exp2f, and that does yield some speedup.
  
  However, there are 2 drawbacks of this :
  
  1. It does not exploit the integer nature of the argument.
  
  2. (minor) Some platforms lack a proper exp2f routine, making benefits available
  
  only to non broken libm.
  
  3. exp2f does not solve the same issue that plagues pow, namely terrible
  
  worst case performance. This is a fundamental issue known as the
  
  "table-maker’s dilemma" recognized by Prof. Kahan himself and
  
  subsequently elaborated and researched by many others. All this is clear from benchmarks below.
  This exploits the IEEE-754 format to get very good performance even in
  
  the worst case for integer powers of 2. This solves all the issues noted
  
  above. Function tested with clang usan over [-1000, 1000] (beyond range of
  
  relevance for this, which is [-255, 255]), patch itself with FATE.
  Benchmarks obtained on x86-64, Haswell, GNU-Linux via 10^5 iterations of
  
  the pow call, START/STOP, and command ffplay /samples/jpeg2000/chiens_dcinema2K.mxf.
  
  Low number of runs also given to prove the point about worst case :
  pow :
  
   216270 decicycles in pow,       1 runs,      0 skips
  
   110175 decicycles in pow,       2 runs,      0 skips
  
    56085 decicycles in pow,       4 runs,      0 skips
  
    29013 decicycles in pow,       8 runs,      0 skips
  
    15472 decicycles in pow,      16 runs,      0 skips
  
     8689 decicycles in pow,      32 runs,      0 skips
  
     5295 decicycles in pow,      64 runs,      0 skips
  
     3599 decicycles in pow,     128 runs,      0 skips
  
     2748 decicycles in pow,     256 runs,      0 skips
  
     2304 decicycles in pow,     511 runs,      1 skips
  
     2072 decicycles in pow,    1022 runs,      2 skips
  
     1963 decicycles in pow,    2044 runs,      4 skips
  
     1894 decicycles in pow,    4091 runs,      5 skips
  
     1860 decicycles in pow,    8184 runs,      8 skips
  exp2f :
  
   134140 decicycles in pow,       1 runs,      0 skips
  
    68110 decicycles in pow,       2 runs,      0 skips
  
    34530 decicycles in pow,       4 runs,      0 skips
  
    17677 decicycles in pow,       8 runs,      0 skips
  
     9175 decicycles in pow,      16 runs,      0 skips
  
     4931 decicycles in pow,      32 runs,      0 skips
  
     2808 decicycles in pow,      64 runs,      0 skips
  
     1747 decicycles in pow,     128 runs,      0 skips
  
     1208 decicycles in pow,     256 runs,      0 skips
  
      952 decicycles in pow,     512 runs,      0 skips
  
      822 decicycles in pow,    1024 runs,      0 skips
  
      765 decicycles in pow,    2047 runs,      1 skips
  
      722 decicycles in pow,    4094 runs,      2 skips
  
      693 decicycles in pow,    8190 runs,      2 skips
  exp2fi :
  
     2740 decicycles in pow,       1 runs,      0 skips
  
     1530 decicycles in pow,       2 runs,      0 skips
  
      955 decicycles in pow,       4 runs,      0 skips
  
      622 decicycles in pow,       8 runs,      0 skips
  
      477 decicycles in pow,      16 runs,      0 skips
  
      368 decicycles in pow,      32 runs,      0 skips
  
      317 decicycles in pow,      64 runs,      0 skips
  
      291 decicycles in pow,     128 runs,      0 skips
  
      277 decicycles in pow,     256 runs,      0 skips
  
      268 decicycles in pow,     512 runs,      0 skips
  
      265 decicycles in pow,    1024 runs,      0 skips
  
      263 decicycles in pow,    2048 runs,      0 skips
  
      263 decicycles in pow,    4095 runs,      1 skips
  
      260 decicycles in pow,    8191 runs,      1 skips
  Reviewed-by : Michael Niedermayer <michael@niedermayer.cc>
  
  Signed-off-by : Ganesh Ajjanagadde <gajjanagadde@gmail.com>
  

  • [DH] libavcodec/jpeg2000.c

• Problems using Cuda for video transcoding [closed]

  19 août 2022, par Jay Adlard

  I just bought a new pc as my faithful old windows 7 i7 laptop died. Now the laptop had both intel and nvidea gtx660m chips. When I transcoded video with handbrake the intel graphics managed 8fps when I changed over to the nvidia it managed 80fps and the graphics card got nice and hot so was obviously working.

  


  The new machine i bought is a i9 with an nvidea gtx780 running windows 10. When I tried to transcode some video hoping for it to use the cpu and Gpu but no joy,the card stays cool power usage is only a handful of watts higher than it running something simple.I had been using handbrake but I read it doesn’t support Cuda so why the laptop speeded up I don’t know. A friend of mine that is into ffmpeg came round found that ffmpeg reports no Cuda yet the card works fine in every other respect. Unlike the rest of the machine the graphics card wasn’t new but it looked like new…

  


  One thing to note I am not talking about nvenc as the quality is rather poor, no point in capturing uncompressed files cleaning them up and not using an encoder that will do 2 pass and an exhaustive search.
Can anyone recommend a reasonably priced encoder that will use the Cuda cores ?

  


  Has anyone got any idea what the problem is ?

  


• Demo of WebM Running on TI OMAP 4 Processor

  15 octobre 2010, par noreply@blogger.com (John Luther)

  Texas Instruments has made a video of HD-resolution (1080p) VP8 (WebM) video playing on their new TI OMAP™ 4 processor, in both Android and Ubuntu.

  (If you have a WebM-enabled browser and are enrolled in the YouTube HTML5 beta the video will play in WebM HTML5, otherwise it will play in Flash Player.)

  For more info about the OMAP 4 and the IVA 3 video accelerator that enables low-power HD playback of VP8 on the chip, see the TI web site.