[MPlayer-dev-eng] Hardware acceleration with fbdev

Diego Biurrun diego at biurrun.de
Thu May 5 15:15:02 CEST 2011


On Thu, May 05, 2011 at 03:07:07PM +0200, David Kuehling wrote:
> 
> >> > VIDIX is legacy code in a general state of disrepair.
> >> 
> >> Hopefully you're not going to remove it?
> 
> > The threat of removing XYZ seems to be the only thing that drives
> > people using XYZ out of the woodworks...
> 
> So far so good, but now that you drove us out of the woodworks, what
> alternative API would you like us to migrate to?

Have you tried the external VIDIX?

> >> There a some embedded systems around which are not capable to run X,
> >> or won't have an accelerated X driver available. VIDIX makes it
> >> easily possible to support these devices, without having to code a
> >> full 'vo' driver. MPlayer is probably the most leightweight OSS video
> >> player around, making it a good match for embedded systems.
> >> 
> >> (BTW I'm the developer of a VIDIX driver for the Ingenic jz47xx SOC
> >> for use with the NanoNote -- http://sf.net/projects/mplayer-jz47xx/).
> 
> > Why have you created a sourceforge project for this, but not submitted
> > it here?  Or are you using the external vidix.sf.net version?
> 
> This is still work in progress, was planning on submitting it, once
> everything stabilized.  But for now, it's much easier to develop it
> outside mplayer's svn (where i have no commit access anyways).

You should have a look at git / git-svn if you have trouble developing
without svn commit access.  Bazaar and Mercurial also support detached
operation with a Subversion backend.

> It's just a vidix driver, plus a patch to mplayer to load it from
> a shared object (so we can package/update it independently from
> mplayer).

That part is unlikely to be accepted upstream.

Diego


More information about the MPlayer-dev-eng mailing list