Am I even approaching the problem the right way?
I want to create a new output muxer, in my current case a file via
AVFormatContext
Using the encoder context (AVCodecContext) I am going to push frames and read
packets
Then I want to write those packets to the muxer
If the application layer has to loop over the encoder context with
avcodec_send_frame / avcodec_receive_packet
Why does the muxer care about the encoding? Should he only know about the
packets?
Without a demuxer how to you get stream data into the muxer?
Thanks
cco
On 08/23/2016 09:47 AM, Nicolas George wrote:
Le septidi 7 fructidor, an CCXXIV, Charles a écrit :
4) parameters = video_st->codecpar;
parameters->codec_tag = tag;
parameters->format = AV_PIX_FMT_YUV420P;
parameters->bit_rate = 5e6;
parameters->width = 1024;
parameters->height = 768;
parameters->bit_rate = 5e6;
You are not supposed to fill the stream's parameters with random values, you
are supposed to fill them with the values used by the encoder. These values
will be very similar to the settings you chose for the encoder, but the
encoder is allowed to tweak them.
5) av_encode_codec_ctx = avcodec_alloc_context3( av_encode_codec )
av_encode_codec_ctx->bit_rate = 5e6;
av_encode_codec_ctx->width = 1024; /// \note multiple of 2
av_encode_codec_ctx->height = 768; /// \note multiple of 2
av_encode_codec_ctx->time_base = (AVRational) { 1, 60 }; //* 2;
av_encode_codec_ctx->gop_size = 15; // Intra frames per x P frames
av_encode_codec_ctx->pix_fmt = AV_PIX_FMT_YUV420P; // MUST DO NOT CHANGE
nvenc required
// This appears to make h264_nvenc happy
6) avcodec_parameters_to_context( av_encode_codec_ctx, parameters )
You are still overwriting the parameters of the encoder. It can not work.
The parameters are used for communication from demuxers to decoders and from
encoders to muxers. They are output from the decoders and you are using them
as input.
Regards,
_______________________________________________
Libav-user mailing list
[email protected]
http://ffmpeg.org/mailman/listinfo/libav-user
