[Telepathy] how to specify codec and encoding/decoding pipeline for empathy?

[Sjoerd Simons]

On Mon, 2010-09-06 at 14:47 +0800, Zhao, Halley wrote:
> I want to specify a dedicated v4l2src and encoder for video
> conference, where should it happen? Inside farsight or inside empathy?
> 
> Since my v4l2src works in shared memory mode, and encoder is hw
> accelerated; my encoding pipeline differ from usual ones.
> Is there any detailed doc to understand empathy? Is there any doc on
> work division between farsight and empathy?

The source is the documentation. The raw division as it stands currently
is that Empathy is responsible for capturing raw video and passing that
on the farsight. Farsight handles the encoder selection (based on the
capabilities of the remote side).

If you have a hardware encoder just ensuring that your gstreamer encoder
element has the right caps and rank should cause farsight to
automatically pick it up and use it. 

If your encoder and webcam are combined in some way then some more work
might be needed, but that's hard to say without knowing more precisely
what your hardware looks like and what you're trying to achieve

> I try to grep 'fsrtpconference' in empathy, no result found, how does
> empathy use the libraries what farsight provide?

Telepathy-farsight handles the creation of the fsrtpconference and a lot
of the media signalling parts on behalf of Empathy
-- 
Sjoerd Simons <sjoerd.simons at collabora.co.uk>
Collabora Ltd.