Home > encode, x264, x264-bin > x264 test build with opencl lookahead

x264 test build with opencl lookahead

17th May 2012, Thursday Leave a comment Go to comments

The x264 opencl lookahead patch has been tested for months on x264dev’s IRC, but it lacked too many features compared to currect x264, e.g., the weightp was still not supported at the beginning of May. However, it is glad to see the latest patch was released on doom10 yesterday, in which most features has been implemented (but only 8-bit, high bit-depth has not been supported). So a test build should be available here.

Based on tMod+MixAQ branch (rev2197+666+30). Compiled with fprofile and win32thread, and AMD APP v2.6, which may require OpenCL v1.1 runtime to enable opencl lookahead. Lavf/ffms/swscale/interlaced was disabled, and chroma format was set to 4:2:0. Tested on AMD HD6770M and NVidia GTX460, not sure if it can run on Ivy Bridge or not.
Download: MediaFire, NMM-Mirror
See the latest x264 build page for newer builds.

Attention: the patch may not be stable enough for everyday use, so this build should only be used for test purpose. Not only speed but also encoded results are not identical with or without opencl.

Categories: encode, x264, x264-bin
  1. defderdar
    17th May 2012, Thursday at 10,34pm UTC

    hi, do i need some specific option or open cl kernel to use this build? or does it uses open cl internally always. it seems it doesnt using open cl on my system. i tried –vbv-maxrate auto_high s.avs -o s.mkv as command line. thanks

    • 17th May 2012, Thursday at 11,19pm UTC

      Your GPU and driver need to support opencl, and opencl runtime has to be installed on your system ( for example, APP runtime for AMD GPUs ). Then x264 will automatically compile the opencl kernel by default when initialing, otherwise it will not use opencl. If you have the opencl runtime enabled but do not want x264 use opencl, --no-opencl is needed to disable it.

  2. Astez
    19th May 2012, Saturday at 09,40am UTC

    Did a quick comparison on my system (2600k@4.7Ghz, GTX580) with CBR:

    Not sure why your build is faster in the second pass though.

    • 19th May 2012, Saturday at 10,36pm UTC

      vanilla builds by jeeb seem to be built with pthread, and mine are with win32thread. win32thread is sometimes faster than pthread.

  3. Astez
    25th May 2012, Friday at 02,26pm UTC

    r2200 speeds up the first pass considerably:

    Is it possible to create an OpenCL test build with r2200?
    Would love to try that one 🙂

    • 26th May 2012, Saturday at 02,33am UTC

      It is possible, but I would probably not do that. r2200‘s threaded lookahead mainly aims at avoiding bottleneck of lookahead by threading it. But for opencl version, multi-threading those GPU calculations won’t help much. It may only eat up memory and even make the system much unstable on certain drivers. So for test purpose I think r2197 build should be enough.

  4. dE
    28th March 2013, Thursday at 08,50am UTC

    This requires vim installed?

  5. dE
    28th March 2013, Thursday at 08,59am UTC

    What about Intel? I’ve the SDK installed, and it works here, but —

    opencl: no

  1. 2nd August 2012, Thursday at 01,59am UTC

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: