Tegra Multimedia API

Hello! I am developping a multimedia application on the TX1 platform.
Because of some problems, will use Tegra Multimedia API directly instead of GStreamer framework.
When examining its sample codes for video encoding (samples/01_video_encode/*.cpp), some questions
have occured. That sample code accepts many coding options, for example ‘-br’, ‘-p’, ‘-rc’ and so on.
Some options have been understood but others not. Especially ‘-vbs’ and ‘-nrf’ options.
Although I read the API documents related to the options, it didn’t help.
What purposes do the two options exist for ? Please help…

Hi DNA01,

vbs ->“virtual buffer size” and default =0.
nrf -> number of reference frames.

Thank you for the reply.
But the explanations are already known in the sample code.
In case of ‘-vbs’ option, if the default value is zero error occurs as the followings in the sample code.

else if (!strcmp(arg, “-vbs”))
{
argp++;
CHECK_OPTION_VALUE(argp);
ctx->virtual_buffer_size = (uint32_t) atoi(*argp);
CSV_PARSE_CHECK_ERROR(ctx->virtual_buffer_size == 0, “Virtual buffer size should be > 0”);
}

In case of ‘-nrf’, it maybe relates to the Multframe Reference to be used on Inter-Prediction
of P/B Slices and indicate the max number of available reference frames?

I will expect any replies.

Hi DNA01,

You can take a look at Video buffering verifier for info about virtual buffer here.

nrf is just what you said. It is for inter-frame prediction.