[Bino-list] bino recognising bt601 vs sRGB & bt709 movies?

[Ben Simons]

Hi,

I'm trying to generate a movie with ffmpeg (and x264, but it doesn't need to be)
which will playback in bino in sRGB or bt.709 colour space.

However, every movie seems to be interpreted by bino as being in bt.601. 
eg. ffmpeg option:  -pix_fmt = yuvj420p  should spec bt.709
but the bino says it's 601, eg:

   bino: [inf] Input:
   bino: [inf]     Video 1920x2400, 1.6:1: 1920x2400-0.8:1-yuv420p-601-mpeg-l

Can someone give an example of a (linux) ffmpeg config which will create a movie
which bino will play in 'sRGB' or 'bt.709' colorspace?

eg.  here's roughly the ffmpeg command (The first 3 shell variables support experimentation)
it's in a bash shell script, but this is basically it in 4 lines. Input image filenames are:
inputImageSeq.0001.exr to inputImageSeq.9999.exr

    CODEC="libx264 -probesize 2G -analyzeduration 2G"
    OPTIONS="-r 25 -profile:v high10 -pix_fmt yuvj420p -tune zerolatency -tune fastdecode"
    TEST_OPTIONS="-ss 32 -t 5"  # eg skip into seq 32sec, duration 5sec

    ffmpeg -y -framerate 25 -s 1920x2400 -i inputImageSeq.%04d.exr -c:v $CODEC  $OPTIONS $TEST_OPTIONS outputMovie_tb.mp4


There appears to be some comments about 'color_space' in:
    src/media_object.cpp
    src/video_output.cpp
and there's a comment about issues with a broken SRGB texture implementation.

Is there an issue with sRGB? Is anyone playing sRGB, and if so, could you please
report the working codec & pixel format? or the ffmpeg command you used?

I have found a way to generate a 10-bit sRGB movie via Nuke (use an "Uncompressed
RGB 10-bit" codec, saved in an .avi container) but i'm still wondering if someone could
perhaps help describe/explain the current situation with sRGB (textures) and bino?

    bino: [inf] Input:
    bino: [inf]     Video 1920x2400, 1.6:1: 1920x2400-0.8:1-bgra32-srgb

Unfortunately, this format is uncompressed, so it doesn't play back at 25fps.
Nuke does not appear to offer another bgra32 pixel format, so it looks like it's
back to ffmpeg itself.. Still, seeing bino report the above is encouraging.

thanks,
ben.

---

UTS CRICOS Provider Code: 00099F DISCLAIMER: This email message and any accompanying attachments may contain confidential information. If you are not the intended recipient, do not read, use, disseminate, distribute or copy this message or attachments. If you have received this message in error, please notify the sender immediately and delete this message. Any views expressed in this message are those of the individual sender, except where the sender expressly, and with authority, states them to be the views of the University of Technology Sydney. Before opening any attachments, please check them for viruses and defects. Think. Green. Do. Please consider the environment before printing this email.