[MPlayer-users] pp=hb/vb/dr doesn't work while encoding?

[D Richard Felker III]

On Sun, Mar 14, 2004 at 12:01:44AM +0200, Ville Saari wrote:
> On Fri, Mar 12, 2004 at 01:21:40AM -0500, D Richard Felker III wrote:
> 
> > pp should always be the FIRST filter. Otherwise it will not work
> > correctly.
> 
> Is there any other reasons for this besides the alignment issues?
> I have used pp-deinterlacers successfully after cropping when my
> vertical cropping has been divisible by 4. I also have used deblocker
> after cropping by multiples of 16 and been subjectively satisfied with
> the results.

No, it has not been successful. It's only partially worked, and
actually done the wrong thing. The filter is missing the most
important data, the exported quantizer table. Deblock and dering
filters will NEVER work after cropping when the offset is not a
multiple of 16, and even supporting that case would require changes to
the crop filter which are not likely to be added. So ALWAYS apply pp
first.

> > BTW, I suspect deinterlacing is incorrect. Very few DVDs are actually
> > interlaced, and deinterlacing something that's not actually interlaced
> > will ruin the quality.
> 
> PAL DVDs frequently have the frames-out-of-phase-problem I described in
> my earlier posting and the only effective method to fix it using the
> mplayer's standard filter arsenal is to deinterlace. I use my home-brewn
> phase shift filter.

You should also try filmdint with the right options, or pullup. Either
of these filters should repair it properly. Deinterlacing will ruin
the video, especially with a bad deinterlacer like the lavc kernel
one. Look for ghost outlines in the deinterlaced frames to see what I
mean.

> I'm still searching for a way to fix it. Median deinterlacing should
> theoretically work, but the field that pp=md wants to keep is just
> the wrong field for the material I'm trying to fix and my attempts to
> invert the field order have resulted either colorspace incompatibility
> problems or crashed mplayers :-(

Try vf_flip to invert the field parity.

Rich