I didn't know you had to reload the driver to change the settings in the Hauppauge cards. I don't know the technical difficulties for exposing the h264 encoder using the v4l2 specification, but there should be room in such specification for incorporating a device representing the encoder, with all their properties accesible.
Of course using a Gstreamer plugin is another option (anyway, the plugin would need to use a custom method to access the card encoder), but if it is exposed as a v4l device, the solution will be much more "universal", so to speak. Re. your initial question: I believe that any Matterhorn-compatible hardware is always welcome, and such cards are a very interesting product which many adopters could be very interested in (the University of Vigo being one of them). Best regards Rubén 2011/10/13 Curtis Hall <[email protected]> > Greg is right. Technically you could probably use gstreamer without > much, if any, modification. However gstreamer would only be accessing > the MJPEG encoder, not the H.264 compressed stream. The H.264 > compressed stream also gives benefits since it does deinterlacing and > provides a better image quality then the MJPEG encoder. > > Thanks > > 2011/10/13 Rubén Pérez <[email protected]>: > > I'm confused: > > aren't Bluecherry cards already being used with Gstreamer? If the v4l2 > > driver exposes the card capabilities correctly, is there a real need for > an > > specific gstreamer plugin? Wouldn't be better to dedicate efforts to > improve > > the driver conformance to the v4l2 standards and document their use (and > > maybe providing some software to automatize the operations to fine-tune > the > > cards?). > > Best regards > > Rubén > > > > 2011/10/11 Christopher Brooks <[email protected]> > >> > >> Hi, > >> > >> Just to chime in. One of the particularly nice things about using a > >> dedicated hardware encoder for incoming video streams is that the > >> presentation stream (hdmi/vga/etc.) could then have more access to the > >> CPU for encoding. Really nice when you want to squeak more frames out > >> of it at a higher bitrate. > >> > >> The H.264 compression cards are pretty reasonably priced, I think we > >> pay $150 for the non hardware encoding cards right now. And that's for > >> a card that only does one port at 30 fps, and each additional feed > >> would drop the fps in half. That BC-H04120A can do 30 fps > >> simultaneously on all ports. > >> > >> Network streaming would be great for finishing off confidence > >> monitoring. A feature lacking in Matterhorn (though some is coming up > >> in 1.3) in part because of the need to encode multiple streams... > >> > >> Chris > >> > >> On Tue, 11 Oct 2011 > >> 15:40:02 -0500 Curtis Hall <[email protected]> wrote: > >> > >> > Hi, > >> > > >> > I'm Curtis with Bluecherry. We provide software compression cards > >> > that some Opencast members are using (PV-143, PV-149, PV-981 etc). > >> > We also produce H.264 hardware compression cards which does the > >> > compression on the card itself. We also provide a Video4Linux2 > >> > driver for these cards ( https://github.com/bluecherrydvr/solo6x10). > >> > > >> > I've talked to Chris Brooks a couple times about these cards and I've > >> > gained some decent information on how these cards can benefit the > >> > Opencast application by compressing the video on the cards instead of > >> > on the processor. I've attached a couple emails between Chris and our > >> > developers about what the gstreamer plugin would need to support. > >> > > >> > My question comes down to the amount of people that would potentially > >> > be interested in these cards. We would incur a expense in hiring a > >> > contractor to complete the gstreamer plugin, so I'm doing the math to > >> > see if it would be cost effective for us to start on the gstreamer > >> > plugin. > >> > > >> > The following cards are available: > >> > > >> > BC-H04120A PCIe ($219.95) - 4 port > >> > BC-H04120A Mini-PCI ($199.95) - 4 port > >> > BC-H16480A PCIe ($299.95 - 16 port > >> > > >> > > >> > > http://store.bluecherry.net/search.php?search_query=bc-h&button.x=0&button.y=0&button=Submit > >> > > >> > Thanks! > >> > > >> > > >> > >> > >> > >> -- > >> 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 > > > > > > _______________________________________________ > > Matterhorn-users mailing list > > [email protected] > > http://lists.opencastproject.org/mailman/listinfo/matterhorn-users > > > > > > > > -- > Curtis Hall ([email protected]) > Bluecherry - www.bluecherrydvr.com / store.bluecherry.net > (877) 418-3391 x 283 > (573) 642-6161 x 283 > _______________________________________________ > Matterhorn-users mailing list > [email protected] > http://lists.opencastproject.org/mailman/listinfo/matterhorn-users >
_______________________________________________ Matterhorn-users mailing list [email protected] http://lists.opencastproject.org/mailman/listinfo/matterhorn-users
