[Gstreamer-openmax] GSTOMX and N-1 / 1-N IL components

Topi Hukkanen topi.hukkanen at gmail.com
Wed Dec 3 02:59:54 PST 2008


Do you think there should be a gstomx base_mux or base_demux (2 to 1, or 1
to 2)?

At least with GStreamer, I assumed there wasn't a base mux and base demux,
because the AV synch was so complex.  But if GST OMX pushes this
responsibility down to the IL layer, why can't there be a base_mux and
base_demux in GSTOMX?

On Wed, Dec 3, 2008 at 12:24 PM, Felipe Contreras <
felipe.contreras at gmail.com> wrote:

> On Wed, Dec 3, 2008 at 11:28 AM, Topi Hukkanen <topi.hukkanen at gmail.com>
> wrote:
> > Hi,
> >
> > I took a look at the GST OMX code, and it appears that there's only base
> > components for src, filter, and sink.
> >
> > Are there any plans for extending these bases to muxers and demuxers...
> or
> > is this responsibility delegated because "N" implies there can be many
> > ports, and the very nature of muxers and demuxers make them difficult to
> > control.
>
> I don't know of anybody doing that, so no, no plans.
>
> > At least I don't think there's a base mux for GStreamer... doesn't a
> writer
> > of a plugin have to start completely from scratch?
>
> Yes.
>
> > I guess in the OMX world, the clock synching could be a big problem.  Has
> > anyone looked at this yet?  Is there any way I can get involved?
>
> Well, if properly done GStreamer and OpenMAX IL would communicate
> properly and GStreamer would take care of the input synchronization.
>
> The problems come with tunnelling and omx renderers, there are
> proposals on how to solve the synch issues, but no one has implemented
> them yet.
>
> The best way to get involved is to ask questions here, or start coding
> and ask for feedback :)
>
> --
> Felipe Contreras
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/gstreamer-openmax/attachments/20081203/78b823ed/attachment.htm>


More information about the Gstreamer-openmax mailing list