[vlc-devel] [RFC] video_output: handle a second picture to display for Frame Sequential 3D sources

[Steve Lhomme]

On 21/08/2018 16:05, Rémi Denis-Courmont wrote:
> 	Hi,
>
> I don't like to made a special case stereoscopy - for comparison, there are no
> special cases for anisotropy.
>
> But more pratically, I fear that will make things worse rather than better for
> capable video outputs - as they might need to assume that the underlying
> picture buffers are tied ?

This is frame sequential. The pictures *have* to be tied. Other forms of 
3D sources won't be passed that way.

Also this patch doesn't have checks on the picture date but there should 
be some so we don't end up displaying mismatching pictures.