[MEncoder-users] broken mencoder options (was: High quality encoding of anime and broken mencoder options)

Jorge Peixoto de Morais Neto please.no.spam.here at gmail.com
Sun Mar 9 19:32:39 CET 2008


On Fri, 07 Mar 2008 08:57:57 -0800
RC <cooleyr at gmail.com> wrote:

> On Thu, 6 Mar 2008 17:51:24 -0300
> Jorge Peixoto de Morais Neto <please.no.spam.here at gmail.com> wrote:
> 
> > In my case, the mbcmp=2 option is helpful, 
> 
> mbcmp=<0-2000>
> Sets the comparison function for the macroblock  decision, *only used if
> mbd=0.*
> 

I am reasonably satisfied with my current settings for encoding
anime*, so now my worry is about the broken mencoder options.
Specifically:

1) The man page says that mbcmp only has an effect if mbd=0. This is
false. mbcmp=2 increases the PSNR for a given bitrate. I tested this
on mencoder compiled from SVN on 13/02/2008 and also on mencoder
compiled from SVN on 08/03/2008.

2) the option vrc_override is broken (tested on mencoder from SVN
13/02/2008). I have found other people on the internet complaining
about this.

3) In my case, setting the options cmp=2 or subcmp=2 actually
diminishes the PSNR.
3.1) If this means that they actually harm the video quality, then
perhaps they are not adequate for anime encoding (but perhaps this only
happens on a few specific videos, including mine...)
3.2) If they actually increase the video quality, but diminish PSNR
(which is possible but unlikely), it should be noted in the
documentation

Should I file a bug report?

* My current options are:

#High quality encoding of anime:
[anime0]
lavcopts=vcodec=mpeg4:keyint=300:vmax_b_frames=2:vb_strategy=2:brd_scale=2:mbd=2:trell=yes:v4mv=yes:last_pred=3:predia=3:dia=3:mbcmp=2:cbp=yes:preme=2:vqcomp=0.6

#Even higher quality encoding of anime (but slower):
[anime1]
lavcopts=vcodec=mpeg4:keyint=300:vmax_b_frames=2:vb_strategy=2:brd_scale=2:mbd=2:trell=yes:v4mv=yes:last_pred=3:predia=3:dia=3:mbcmp=2:cbp=yes:preme=2:vqcomp=0.6:mv0=yes

-- 
Software is like sex: it is better when it is free. --Linus Torvalds



More information about the MEncoder-users mailing list