[FFmpeg-devel] [PATCH] Warn about PAFF & Spatial

[Michael Niedermayer]

On Wed, Jul 23, 2008 at 06:48:10PM -0700, Serguei Miridonov wrote:
> On Wednesday 23 July 2008, Mark Buechler wrote:
> > Hi
> >
> > On Wed, Jul 23, 2008 at 3:22 PM, Michael Niedermayer
> > <michaelni at gmx.at>
> >
> > wrote:
> > > On Wed, Jul 23, 2008 at 11:44:26AM -0700, Serguei Miridonov wrote:
> > > > On Wednesday 23 July 2008, Michael Niedermayer wrote:
> > > > > On Wed, Jul 23, 2008 at 10:45:16AM -0700, Serguei Miridonov 
> wrote:
> > > > > > On Wednesday 23 July 2008, Michael Niedermayer wrote:
> > > > > > > On Wed, Jul 23, 2008 at 08:02:01AM -0700, Serguei
> > > > > > > Miridonov
> > > >
> > > > wrote:
> > > > > > > > On Wednesday 23 July 2008, Michael Niedermayer wrote:
> > > > > > > > > If i knew where the problem is exactly i would fix
> > > > > > > > > it. Besides i honestly dont care about PAFF & MBAFF
> > > > > > > > > at all.
> > > > > > > >
> > > > > > > > Michael, I do not ask you why just because this is a
> > > > > > > > developing list. But could you please post a few links
> > > > > > > > which you think best explain pro and contra of 24/p
> > > > > > > > video? For me 60/i video looks much much smoother than
> > > > > > > > 24/p. On TV or computer monitor -- does not matter.
> > > > > > > > So,if 24/p is so good, what I do wrong with 24/p video
> > > > > > > > captured by my camcorder?
> > > > > > >
> > > > > > > then use 60/p
> > > > > >
> > > > > > Camcorder does not support that. So?
> > > > >
> > > > > Complain to the company that made it ...
> > > >
> > > > You are so kind. Thank you.
> > >
> > > What did you expect?
> > > Something like:
> > > "Ill spend 1+ weeks and fix the bugs that i dont care about at
> > > all" ?
> > >
> > >
> > > Why dont you fix the bug? Or help cleaning up h264.c if first is
> > > too hard?
> > >
> > > And seriously, if enough people complain about the lack of proper
> > > progressive support in camcorders, maybe the industry will
> > > eventually listen.
> > > And its a good idea even with all the *AFF code working.
> > >
> > > [...]
> >
> > You don't give yourself enough credit. I have every confidence you
> > could have it fixed in 6 days if you tried.
> 
> Mark, please, do not start the flame war. I'm sure Michael is the best 
> ffmpeg developer who knows H.264 better than anyone else here. I 
> respect what he has done so far and I'm glad we have at least 
> progressive stuff really working. My only complain was his view on 
> the interlaced stuff (see above). I have asked only info how to make 
> 24/p footage better looking, as I do not like jerky video at 24 fps. 
> That's it.
> 
> Michael, please, do not take it too seriously. Just give us a clue if 
> you can. Complaining to big companies is not a good idea, you know. 
> Especially for the stuff like 60/p. I'm not even sure if this format 
> will be standard soon because it requires not only much bps but also 
> more sensitive image sensors which always is a problem for small 
> consumer products. 

If you have XxY interlaced at 60 fields per second then
sqrt(X)xsqrt(Y) progressive at 60 frames per second has the same number
of pixels per second and it is much simpler to handle as progressive is
simpler than interlaced.


> So, if you know how to make 24/p stuff to look 
> good, please, post a link.

I dont

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Complexity theory is the science of finding the exact solution to an
approximation. Benchmarking OTOH is finding an approximation of the exact
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20080724/889a1def/attachment.pgp>