[MEncoder-users] the options for the best x264 encoding with crf

[Grozdan]

On Fri, Jul 2, 2010 at 8:06 PM, Goga777 <goga777 at bk.ru> wrote:
> Hi
>
> in mencoder official documentation there's some settings for different options, for example for the best
> quality the documentation recommend to use the following options (I will keep in the mind crf method,  not
> 2pass)
>
> http://www.mplayerhq.hu/DOCS/HTML/en/menc-feat-x264.html

Those docs are *very* outdated so you shouldn't rely on them

>
>
> Very high quality
> subq=6:partitions=all:8x8dct:me=umh:frameref=5:bframes=3:b_pyramid:weight_b

You can up subq to 9 if you want and also add trellis=2. For motion
estination use umh, as esa/tesa are way slower and you won't notice
the diff between them & umh. b_pyramid needs a parameter (choose
between normal, strict and none -- strict is only for blu-ray
compatibility so use normal)

if you want, just use in mencoder this option: preset=veryslow
(requires a very recent mplayer svn checkout). It'll directly use the
veryslow preset of the libx264 library. You can use any of the
available x264 presents in mencoder, do a x264 --fullhelp to see them
all

>
> speed (in fps)
> 6fps
>
> Relative PSNR loss (in dB)
> 0dB

PSNR is not a reliable metric, especially when psy optimizations are
involved as is the case in x264. Psy optimizations virtually always
decrease PSNR while increase subjective quality

>
>
> are there any additions options for the best quality of encoding which yu can recommend with crf for
> 576i video ?
>
> BTW,  by default currently subq=7, why docs do recommend worse subq=6 ?

See above

>
> Goga
> _______________________________________________
> MEncoder-users mailing list
> MEncoder-users at mplayerhq.hu
> https://lists.mplayerhq.hu/mailman/listinfo/mencoder-users
>