[MPlayer-users] vo issue with 422 h264

[Reimar Döffinger]

On Thu, Jan 23, 2014 at 07:39:52PM +0000, Andy Furniss wrote:
> Reimar Döffinger wrote:
> >On 23.01.2014, at 18:50, Andy Furniss <adf.lists at gmail.com> wrote:
> >>While testing something I made an h264 422 vid with ffmpeg/libx264.
> >>
> >>On playback I notice that the vo is saying YV12, testing with an old mplayer this does not happen and the vo gets fed 422.
> >>
> >>full -v outputs below current then old, the relevant parts -
> >>
> >>Current -
> >>
> >>VDec: vo config request - 1920 x 1080 (preferred colorspace: Planar 422P)
> >>Trying filter chain: vo
> >>VDec: using Planar YV12 as output csp (no 0)
> >>Movie-Aspect is 1.78:1 - prescaling to correct movie aspect.
> >>VO Config (1920x1080->1920x1080,flags=0,'MPlayer',0x32315659)
> >>VO: [gl] 1920x1080 => 1920x1080 Planar YV12
> >>
> >>Old -
> >>
> >>VDec: vo config request - 1920 x 1080 (preferred colorspace: Planar 422P)
> >>Trying filter chain: vo
> >>VDec: using Planar 422P as output csp (no 5)
> >>Movie-Aspect is 1.78:1 - prescaling to correct movie aspect.
> >>VO Config (1920x1080->1920x1080,flags=0,'MPlayer',0x50323234)
> >>VO: [gl] 1920x1080 => 1920x1080 Planar 422P
> >
> >And both _look_ correct?
> >Asking because there is nothing about a scale filter being selected.
> >In fact what it seems to say is that originally the decoder claimed 422 but when MPlayer asked again it changed its mind to actually wanting 420...
> >I can think of one related change, though I can't imagine how it would cause this.
> 
> Both look correct.
> 
> Further testing looks like this is an h264 issue, I can reproduce
> with another sample I have that is not made by me/ffmpeg.
> 
> I also have a mpeg2 422p stream which does not have the issue.
> 
> Raw 422p also works OK (well mplayers raw422P has another issue with
> u/v reversal, but -vc ffraw422p is correct and both result in 422p
> output).

Yeah, it's some very special-case code, though it's mostly chance H.264
is (almost?) the only affected.
Fixed.