On 7/25/15 6:02 AM, Moritz Barsnick wrote:
On Sat, Jul 25, 2015 at 11:52:56 +0000, Carl Eugen Hoyos wrote:
But you could test without "--enable-librtmp".
Is it possible to choose different protocol handlers at runtime, as it
is with decoders (analog to the choice of "mp3" vs. "mp3float" using
"-c:v" as inpt option)? Or does using librtmp totally disable ffmpeg's
native implementation?
(Indeed, my librtmp doesn't seem to like this URL, shown by trying with
rtmpdump. But I'm not the original poster. ;-))
Moritz
Well, my rtmpdump _does_ work. It claims to be v2.4 and I was able to
get the stream and pipe it to ffmpeg. Inelegant, but it works. Sadly,
the audio/video is out of sync by about 75ms. Just enough to be annoying.
What can I do to help track down the problem? Is the interest only in
the native solution (try without librtmp) or ??
Side question: With the sync problem, I've used -itsoffset (and mapping
of course) in the past to fix sync issues, but that obviously won't help
if I don't have the streams split _before_ it gets to ffmpeg. Any solution?
Thanks
Deron
_______________________________________________
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user
