[MPlayer-users] mbd, vqscale, and vpass mechanisms

Daniel Moreno comac2k at teleline.es
Sat Jan 3 22:57:11 CET 2004

Hash: SHA1

Elmo wrote:

| I don't see why people use 2-pass encoding with vqscale, basically.  If
| you're encoding at a constant quality, how does getting information
| about the bitrates help the encoding quality?  Should I be doing this
| differently to get a better overall movie, or am I misunderstanding the
| way that vqscale works with vpass?  Also, mbd has three options, and I
| don't really know which is the "best" in terms of quality.

I don't know the internals, but I decided to use vqscale=2 (and -o
/dev/null) in the first pass after I found it worked better under some

I was trying to compress a pretty long movie (about 7560 seconds, which
is... about 2h) in 1 CD. At a given point in the movie, there was a
scene with lots of movement (water flowing rapidly just in front of the
camera for a while) followed by a prety quiet one. (two men sitting in
sofas, talking with the camera fixed)

When using vbitrate=whatever at first step, the quiet scene would look
bad at first and take quite long to "recover". (with this I mean over 30
seconds of prety big quantizers after leaving the high motion scene)
Using vqscale=2 at first pass solved the problem and the quiet scene
looked allright from the begining. Don't ask me why. I just know the facts.

On the other hand... that was the only time I was able to notice a

Version: GnuPG v1.2.3 (GNU/Linux)


More information about the MPlayer-users mailing list