• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 99 subscribers
  • Views 208 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.

About DM365 pseudo-multi-pass encoding

Livio Lima
Prodigy 150 points

Hello all,
I'm trying to use the pseudo-multi-pass encoding feature of DM365. I'm running 2 instance of H.264 encoder at the same resolution. The encoder 1 is configured with mvSADoutFlag=1 and metaDataGenerateConsume=1. I verified that it correctly return the FrameInfo_Interface structure with frame data and Motion vector data. Furthermore i verified that the encoder does not return any information about macro blocks and gmv information per row, since this functionality seems not implemented.

Then I configured the encoder 2 with mvSADoutFlag=1 and metaDataGenerateConsume=2. I performed the encoding and I can't see any difference compared to encoder1. Quality and bitrate are the same as well as the encoding time, but I expected to reduce the encoding time of encoder 2 since it should reuse the motion information computed by the encoder 1.

Is this the expected behavior or there is something wrong?

Thanks for any suggestion

LL

  • 0
    TI__Expert 8465 points

    Hi,

    Normal use case of multipass encoder will be 1st encoder runs for lower resolutions and collects the metadata and it passes this info to second encoder with higher resolution.In multipass case you will see the better quality in second encoder and there will not change in speed performance. Encoder uses motion information as extra information to find better match.

    Please run with complex video(scene changes) and codec debug library, observe the log. Compare the single pass encoder output(higher resolution) with multipass second encoder output.

    Thanks,

    Veeranna