Hi,
I'm using libav to decode live audio streams and process them using
custom code. In this process the start timestamp of the stream is quite
important.
Currently, i take the timestamp (using std::chrono) at the end of the
first decoded packet (av_read_frame -> ... -> av_free_packet). Most of
the time this is fine and i observe no or little difference between this
start timestamp and the "reality".
In some cases, i observe huge differences (many seconds).
So, i wonder if i can be more precise in my timestamp choice.
Does libav starts buffering when i call avformat_open_input and keeps
the buffers until i call av_read_frame? When does it really starts to
read the stream?
Thanks for your advices
Florian
_______________________________________________
libav-api mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-api
