[MPlayer-dev-eng] [LONG] PSNR comparison, new source and results

[Romain Dolbeau]

Romain Dolbeau wrote:

> New results from my attempt at finding the "best" lavc options.
> I used a very different file this time ; instead of the slow-motion
> scene from "CT,HD", I used a very fast-moving from The Matrix.
> 5000 frames from the subway fight, when Smith beats up Neo.
> I also used 2-pass encoding, only looking at the PSNR result
> of the second pass.
> 
> filters: "-vf crop=720:422:0:78,scale=576:238"
> (yes, not multiple-of-16 height, sorry, but it uses the same AR than
> the properly cropped source and I started the runs before Rich told
> me it was bad)
> 
> common options: "vcodec=mpeg4:vqmin=2:vqmax=31:keyint=125:vpass=X:"
> 
> I also forced the following options in my tool:
> "vme=4:mbd=2:qpel:trell"
> 
> The various tested option were:
> 
> {pre,,sub}cmp=2, 3 or 6
> {pre,}dia=1 or -2
> vmax_b_frames=0 or 1
> mv0 or not
> qprd or not
> v4mv or not
> cbp or not (we always have trell)
> vlelim=-4 or not
> vcelim=7 or not
> vbitrate=500, 650 or 800

I've added some options, after forcing:
*cmp=2:mv0:v4mv

The various tested options were:

vmax_b_frames=0 or 1
{pre,}dia=1 or -2
qprd or not
cbp or not (we always have trell)
qns=1 or not
tcplx_mask=0.1 or not
scplx_mask=0.1 or not
p_mask=0.1 or not
naq or not (doesn't do anything w/o a least one *_mask
vbitrate=500, 650 or 800

that was 3 * 512 results

* qprd was mostly good, more so at lower bitrates
* cbp was mostly good, more so at lower bitrates
* qns=1 was usually bad at 500 kbps with low absolute
   values, and was bigger and better at 650 and 800 kbps
* *_mask was mostly bad, with naq (partially) compensating,
   more efficently at higher bitrates
* dia=-2 was still a winner over *dia=1, in most case

-- 
Romain Dolbeau