Thank you for the opportunity to ask a question and for your very quick
response
First, we do not support the model whereby a data source 'pushes' media
into a RTSP server (something that's excessively complex, and
non-standard). That's why the "ANNOUNCE" command did not work.
Unexpectedly, it seems ffplay, VLC (for RTP), rtsp-simple-server do it
simply.
So the scheme in the documentation is not correct
http://live555.com/proxyServer/
[back-end RTSP/RTP stream] --> [LIVE555 Proxy Server]... BUT [back-end
RTSP/RTP SERVER] --> [LIVE555 Proxy Server]...
Then it is easier to use forwarding in the firewall, it works much
faster.
You should use our "testOnDemandRTSPServer" demo application - in the
"testProgs" directory - as a model. You should first make sure that
this server can stream a pre-encoded H.264 video file (that you've
generated from your camera using 'ffmpeg'), named "test.h264",
properly. Then, you extend the code to use your own
('ffmpeg'-generated) video source, rather than a pre-recorded video
file.
Did I understand you correctly that I will need to copy the buffer to
memory (test.h264) in a single thread for your server to work and this
will lead to a decrease in live stream fps on my highly loaded H3 SoC
from Allwinner (about 75% CPU for transcoding) ?
_______________________________________________
live-devel mailing list
live-devel@lists.live555.com
http://lists.live555.com/mailman/listinfo/live-devel
