if your camera is set for more than your upload bandwidth, you will not have favorable results. there is no automatic bandwidth detection feeding back to the camera when using groups. insufficient bandwidth for a group multicast will usually manifest as picture breakup and other undesirable distortion, as well as an increase in latency and jitter.
however, when using 1:1 P2P (peerID/NetStream.DIRECT_CONNECTIONS instead of groupspec) there is feedback to throttle the camera, but the independent (and different priority) queues for video and audio will cause a perceived out-of-sync at the receiver for as long as your bandwidth is constrained.
Thanks for your important answer. I have written incorrect, my fault sorry.
My max. upload-bandwidth is 256 kb/sec, I setup my webcam for 256kb/sec of quality and bitrate.
What about if I use "groupsec" with my maximum bandwidth.
Is it not possible or not good for p2p because I have used with maximum bandwidth.
Or it wont upload if I use maximum bandwidth?
thank very much.
And I hope you understand my drawing
I think, it should be 2. Method??
1 person found this helpful
when doing P2P multicast (NetStream with a groupspec), our experiments indicate that the original publisher will need to have about 4-5x the stream bandwidth in upload capacity, and most subscribers must have at least 1x the stream bandwidth+overhead in upload capacity (and ideally more). so if your stream (plus RTMFP overhead) is 256kb/sec, the original P2P publisher should have at least 1Mbps upload, and most of the subscribers should have at least 256kbps upload, and preferably even more.