[MPlayer-users] Re: [-] Re: [-] Crop before deinterlacing OK?

[Rich Felker]

On Fri, Jul 22, 2005 at 10:31:53PM -0700, RC wrote:
> On Fri, 22 Jul 2005 22:28:28 -0400
> Rich Felker <dalias at aerifal.cx> wrote:
> 
> > If you follow inverse telecine by a deinterlacer, output will be
> > almost as bad as if you just used a deinterlacer. 
> 
> I'm getting tired of arguing this point over and over again, and on
> the same thread...  

:)

> We've already established that he isn't going to be able to guess what
> the content will be before he captures it, and isn't going to be
> capturing interlaced and re-encoding, so a combination of filters that
> can handle both is necessary.

Yes, I understand this.

> You've already said that detc should do very little if any damage to
> pure interlaced content.

No, I have never said this!!!!!!!!!!!!!!!
I said that all of my inverse telecine filters will do no damage to 30
FPS PROGRESSIVE CONTENT! They will all butcher interlaced content in
random ways. The only inverse telecine filter 'safe' to use on
interlaced content is filmdint, which was not written by me.

> Deinterlacers like kerndeint or pp=lb should
> also do little/no damage to the content if the detc filter has already
> converted it.  So I'm not sure why you're still arguing.

This is true. pp=md however will do serious damage.

> I have also never known any of mplayer's inverse telecine filters to be
> perfect...  So not including a deinterlacer often leaves videos with
> mismatched fields, which of course looks very bad.  

With the defaults pullup tends to skip lots of fields rather than
output mismatching ones. This looks bad too, of course, but it
shouldn't happen with clean input, which you may or may not have from
a TV signal.

Rich