Works okay over my 3mbps down/512k up DSL connection in Southern Ontario, Canada. The audio dropped out a couple times, but that sounded like a silence level thing. There wasn't any easily noticeable artifacting, however every second to 2 seconds it looked like it was dropping frames.
Jason Interalab wrote: > I have updated to the trunk revision 1932 which does not contain the > "pending videos" code. > > We're re-launched the live camera (pointed at the road again) and will > leave it up all night. Please take a look and report back on quality if > you have a chance. > > http://www.magicoast.com/media_center/embedded_player.html > > Thanks to everyone who helped test this today. > > I guess this was a day for testing . . . the video conference and the > multiple streaming tests seemed to go well too! > > Bill > > Al Melquist wrote: > >> Video stream stopped >> >> >> >> -----Original Message----- >> *From:* [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] *On >> Behalf Of *Lenny Sorey >> *Sent:* Wednesday, April 25, 2007 1:18 PM >> *To:* [email protected] >> *Subject:* Re: [Red5] live webcam test >> >> >> >> Joachim, >> >> >> >> I made your suggested change bellow and recompiled. >> >> >> >> Now the streaming side seems to be back working. >> >> >> >> Exactly what impact does this change make? >> >> >> >> Thanks. >> >> >> >> >> >> Lenny >> >> >> >> >> >> >> >> On 4/25/07, *Joachim Bauch* <[EMAIL PROTECTED] >> <mailto:[EMAIL PROTECTED]>> wrote: >> >> Hi Klaus, >> >> Klaus schrieb: >> >>> Check conf/red5-common.xml >>> >> [...] >> >>> Remove the bean with org.red5.server.stream.SimpleBWControlService and >>> remove comments around the bean with DummyBWControlService. >>> >> this will only change the way Red5 uses the bandwidth settings defined >> in serverside code. >> >> Another thing you could test is changing line 1780 in >> org.red5.server.stream.PlaylistSubscriberStream from >> >> if (pendingVideos > 1 || writeDelta[0] > writeDelta[1]) { >> >> to >> >> if (pendingVideos > 1) { >> >> Recompile and restart Red5 afterwards. This will change the handling >> of slow client connections to drop less frames. However without this >> check, I had issues with clients from low-latency, low-bandwidth >> connections. You seem to have high-latency, high-bandwidth so maybe >> we need different checks to handle both cases. >> >> Please let me know if this helped with your problems. >> >> Joachim >> >> _______________________________________________ >> Red5 mailing list >> [email protected] <mailto:[email protected]> >> http://osflash.org/mailman/listinfo/red5_osflash.org >> >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Red5 mailing list >> [email protected] >> http://osflash.org/mailman/listinfo/red5_osflash.org >> >> > > _______________________________________________ > Red5 mailing list > [email protected] > http://osflash.org/mailman/listinfo/red5_osflash.org > _______________________________________________ Red5 mailing list [email protected] http://osflash.org/mailman/listinfo/red5_osflash.org
