We know the way forward, we just haven't had the resources to put into it. I'm hoping post 1.2 that we can do it, but load testing seems to be of higher priority right now. My goal for the CA in post 1.2 is first adding arbitrary encoding outputs (e.g. h264) and second adding back confidence monitoring. The latter to move towards live streaming (could there be a better confidence monitoring?).
Chris On Thu, 24 Mar 2011 22:31:21 +0200 Lovemore Nalube <[email protected]> wrote: > Thanks Chris. Stephen is for the idea of finding alternative methods > of monitoring if confidence monitoring does not work for us. For now > we've de-proritised it - for now :) > > Thanks > > > -- > Lovemore Nalube > OLE Developer (Vula) > University of Cape Town > http://www.cet.uct.ac.za/LovemoreN > > >>> On 3/23/2011 at 9:55 PM, in message > <20110323135543.16917eb1@bobble>, Christopher Brooks > <[email protected]> wrote: > > Hi Lovemore, > > > Thanks for your prompt response. We'd prefer having monitoring on as > > it has proved useful with keeping an eye on remote venues. Version > > 1.0 and 1.0.1 have done this well over the months. If 1.1 is not > able > > to do this then its a bug, regression and perhaps a release blocker. > > Is there a Jira for this? > > No, this was a beta feature of 1.0 that we did not pursue for 1.1 (or > 1.2) and are hoping to address in 1.3. We focused on adding support > for webcams and other video devices instead. There are jiras to > refactor and get confidence monitoring going though. > > > Basically we need this to work. Any idea where it's breaking? > > No, but I can suggest that if you want to hack something special into > the gstreamer command before we record you could potentially do that > with a custom video producer. I would not recommend this because > making sure it works properly on your hardware would take some time, > but you *should* be able to do live confidence monitoring by changing > the gstreamer device line to include a tee similar to the one Armin > did > here: > http://opencast.jira.com/wiki/display/MH/Examples+of+Gstreamer+Piplines+with+NCast+VGA+and+Video+Grabber+Cards > > It would require beefy capture agents hardware wise, or a huge > reduction in framerate, but I think it would be doable. Time > investment would be large. > > Chris > > > -- Christopher Brooks, BSc, MSc ARIES Laboratory, University of Saskatchewan Web: http://www.cs.usask.ca/~cab938 Phone: 1.306.966.1442 Mail: Advanced Research in Intelligent Educational Systems Laboratory Department of Computer Science University of Saskatchewan 176 Thorvaldson Building 110 Science Place Saskatoon, SK S7N 5C9 _______________________________________________ Matterhorn-users mailing list [email protected] http://lists.opencastproject.org/mailman/listinfo/matterhorn-users
