[FFmpeg-devel] GSoC 2017

wm4 nfxjfg at googlemail.com
Thu Feb 9 21:22:50 EET 2017


On Thu, 9 Feb 2017 14:01:59 -0500
"Ronald S. Bultje" <rsbultje at gmail.com> wrote:

> Hi,
> 
> On Wed, Feb 8, 2017 at 8:21 PM, Michael Niedermayer <michael at niedermayer.cc>
> wrote:
> 
> > On Thu, Feb 09, 2017 at 01:03:06AM +0100, Michael Niedermayer wrote:  
> > > On Wed, Feb 08, 2017 at 09:07:09PM -0200, Pedro Arthur wrote:  
> > > > 2017-01-27 1:01 GMT-02:00 Michael Niedermayer <michaelni at gmx.at>:
> > > >  
> > > > > we need more (backup) mentors, so yes, i think its needed
> > > > >
> > > > > also if you want to be mentor for a swscale related project i would
> > > > > support that idea  
> > > >
> > > >
> > > > Hi, sorry for the late reply.
> > > > I think  implementing a cascade-less swscale could be a good project.
> > > > If you think  it is doable for a GSOC project, I'm fine with it.  
> > >
> > > It depends on the student, it doesnt sound too complex but students
> > > abilities varies alot and its quite possible there will be noone
> > > interrested in any specific project
> > >
> > >  
> > > > I just doesn't know a good qualification task.  
> > >
> > > Something which shows that the student can work with the swscale
> > > codebase, not something that is just copy and paste work
> > >
> > > Either way, this can be added later  
> >
> > also in absence of any other ideas, YCoCg support may be a
> > qualification task. Maybe not the best choice but certainly usefull
> > on its own  
> 
> 
> Before we get into adding new things to swscale: I will veto anything that
> doesn't use the provided API in libavutil. So no new SWS_CS_*.
> 
> And yes I'm serious about this.

That should be self-evident and not an issue. Did anyone argue for it?

(Also wondering if I should revive my ancient libswscale + AVFrame
patch.)


More information about the ffmpeg-devel mailing list