Hi, some days ago I have posted to know if I am using the ffmpeg encoder parameters in the right way but nobody has answered. That's means that: 1 - I am in the right way, 2 - the question is so stupid that there is no way for me :( 3 - nobody knows the answer? That's the question:

I have some YUV files grabbed from a camera and I need to create an m4v file to stream via RTSP. Right now I am using these parameters to encode but I would like to know if I could improve the quality or reduce the size of the m4v produced:

codec_context->bit_rate One of these : 128*1024 -- 512*1024 -- 1024*1024 Is too much?

codec_context->bit_rate_tolerance = codec_context->bit_rate * 20/100 codec_context->pix_fmt = PIX_FMT_YUV420P; codec_context->gop_size = gop; codec_context->frame_rate=fps; codec_context->frame_rate_base=spf; codec_context->flags |= CODEC_FLAG_AC_PRED; //AC Prediction codec_context->flags |= CODEC_FLAG_4MV; codec_context->flags |= CODEC_FLAG_PART; codec_context->profile = 9; //Advanced Real Time codec_context->level = 4;

Are these correct? Can I add some parameters? Am I missing something? Have I to change something? I know that one improvement would to be add the b-frames but what else?

I am asking this because in the images of the stream I produce I can see very well the macroboclks (are those the macroblocks?) square expacially during motion...