[MEncoder-users] encoding technical question

Corey Hickey bugfood-ml at fatooh.org
Thu May 15 08:37:59 CEST 2008


Philippe MONROUX wrote:
>> 2) Unless I am very much  mistaken, CQ will not give you predictable
>> file sizes, since it will  use whatever bitrate is needed to provide
>> the quality specified., so the  basis of your question is flawed. In
>> my experience, the only way to achieve a fixed final file size is to
>> do two pass encoding specifying bitrate, not CQ or CR.
> 
> Sure but, read in doc :
> 
> resolution x CQ = (bitrate x 1000)/25 (if 25 fps)

Where is this written? At a constant quantizer, bitrate (and, hence,
file size), are the result of many other factors.
- what format is used (MPEG-4 ASP, h.264, flv, etc.)
- what implementation is used (lavc, XviD, DivX, etc.)
- what options are used (mbd=2, trell, B-frames, etc.)
- how compressible the video is (how much motion, detail, etc.)
- ...whatever I'm forgetting...



$ mencoder input.avi -o test1.avi -ovc lavc -frames 2400 -lavcopts \
vqscale=2
$ mencoder input.avi -o test2.avi -ovc lavc -frames 2400 -lavcopts \
vqscale=2:mbd=2
$ mencoder input.avi -o test3.avi -ovc lavc -frames 2400 -lavcopts \
vqscale=2 -ss 100
$ ls -l test1.avi test2.avi test3.avi
-rw-r--r-- 1 bugfood bugfood 22878790 2008-05-14 23:24:35 test1.avi
-rw-r--r-- 1 bugfood bugfood 21281050 2008-05-14 23:25:11 test2.avi
-rw-r--r-- 1 bugfood bugfood 20190498 2008-05-14 23:27:43 test3.avi

test2.avi is smaller than test1.avi because the mbd=2 option makes lavc
more efficient at compression; test3.avi is even smaller because the
second 100 seconds of the video are more easily compressed than the first.

-Corey



More information about the MEncoder-users mailing list