• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 100 subscribers
  • Views 357 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TDA4VH-Q1: Dual device nodes failed to execute encoding simultaneously

li quan
Intellectual 955 points

Part Number: TDA4VH-Q1

Hi  TI expert

Our SDK version ： 8.6

Our board model：j784s4_evm

On our custom board we want to encode simultaneously on two device nodes. Our reserved cma memory is 1GB. The resolution encoded on /dev/video0 is 3840 * 3072. The resolution encoded on /dev/video1 is 3840 * 3240. But when encoding node video0, video1 cannot be encoded. We found an error in the following locations：

linux\drivers\media\platform\chips-media\wave5\wave5-hw.c  

As shown in the red box in the figure, I added debugging printing, and the results are as follows：

Does it not support such a large resolution? If supported, how to solve it？ Thank you very much！

Best regards !

Alex

  • 0
    TI__Expert 3840 points

    Hi Alex,

    The device is able to handle a maximum resolution of 8192x8192 per encoder instance. However, is important to mention that the max resolution that is guaranteed real time performance is 4K60 per instance. Your encoding instances exceed this resolution so real time is NOT guaranteed.

    Encoder should still be achieved successfully. Could you share with me your use case? Are these camera inputs, or just using raw video files? If just raw video files, could you share these files with me?

    How are you encoding these videos? Gstreamer or another application? Lastly, which codec are you using for encoding (264 or 265)?

    Thanks

    Brandon

  • 0 in reply to Brandon Brnich
    Intellectual 955 points

    HI Brandon

    We use encoder is 264. The demo we use is as follows:

    videnc-test-app.zip

    When using, you need to modify the dev name, input file name, and output file name.

    We use NV12 format video files as input. However, the file is too large to upload. You can use any NV12 format file instead.

    If there is any progress on this issue, please contact us.Thank you very much！

    Best regards !

    Alex