[MPlayer-dev-eng] Re: Re: [Mplayer-devel] Help needed in Vidix successor!

[Alex Beregszaszi]

<20020602143502.A32625 at localhost.localdomain.hu>; from alex at naxine.org on Sun, Jun 02, 2002 at 02:35:02PM +0200
> Hi,
> 
> > > > we need volunteers who would help us making more drivers available in
> > > > vidix. The drivers should be ported from Xv or any other opensource codes.
> > > >
> > > > If anyone's fearing from Vidix, we're going to make an api redesign, to
> > > > eleminate pitches and make the API more simplier. In the future we will
> > > > able to compile vidix drivers as (linux) kernelmodules too, if one likes
> > > > this method (as by old mga_vid drivers).
> > > 
> > > How about writing some docs on how to transfer code from Xv drivers to vidix
> > > It would give a better starting point.
> > 
> > hmm. not a bad idea!
> > 
> > we should design the new vidix (xivid? :)) api to be as close to Xv API as
> i vote for smart video overlay ;) (svo)
> 
> > possible, but extending it a little bit to do direct rendering instead of sw
> > conversion & memcpy. xv already has control()-like stuff to set extra
> > options, brightness etc. if our api is similar enough, then converting xv
> > drivers can be done really fast.
> 
> and use that ports stuff? no thanks
> 
> my plans are:
>  the driver would have one entrypoint (==control) with the following
>  controls:
>    INIT         - start the driver, find cards
>    CONFIG       - set the dimensions and return the dga address and other stuff
     QUERY_FORMAT
>    QUIT
>    SET_EQ
>    GET_EQ
>    ...
>    SELECT_CARD  - if more cards simultaneously present
>    GET_CARDS    - return the available cards
> 
> No need for more shit.
> 
> Also i prefer to make this as an own project, and write the supporting
> vo_ interface. So it will be independent from mplayer project.
> 
> -
> alex
> _______________________________________________
> MPlayer-dev-eng mailing list
> MPlayer-dev-eng at mplayerhq.hu
> http://mplayerhq.hu/mailman/listinfo/mplayer-dev-eng