[MEncoder-users] a-v synchronization problems of some NTSC DVDs with -oac copy

David Soukal david.soukal at binghamton.edu
Sun Apr 24 19:29:00 CEST 2005


Hello,

I have recently ran into A-V synchronization problems while encoding
some NTSC DVDs. I looked at the mailing list archive and found very
similar symptoms in a thread by Corey Hickey on March 15th ("a-v sync
problems with pullup, softskip (only on some DVDs)"). I have studied the
answers but I was unable to fix the problem. It seems that the problem
was a bug in the code that has been fixed. But I didn't find a reference
to the version of MPlayer that has the bug fixed. I'm using version
1.0pre6 compiled from sources on a Gentoo system (For Gentoo users: I
had to recompile MPlayer recently with the dts flag to get FFMPEG
support again.)

To reiterate the problems. I have noticed a short sync delay (of order
of ms) with some DVDs that I converted while there was no sync problem
with other DVDs. All the DVDs are in the NTSC format. The original DVDs
show no sync problems, of course.

The problematic DVDs are stored progressive with exception of a couple
of short scenes with telecined content. I used the following command
line for the encoding:

mencoder -vf pullup=2:2:4:4,softskip,crop=704:368:8:60 -ofps 23.976 -oac
 copy -ovc xvid -xvidencopts pass=1:${xvidencopts} -o movie.avi

where $xvidencopts is

xvidencopts="bitrate=1800:me_quality=6:nointerlacing:noqpel:gmc:trellis:quant_type=mpeg:
vhq=4:autoaspect:chroma_opt:hq_ac:min_iquant=1:min_pquant=1:min_bquant=1".

I use the same commands (with exception of the cropping parameters and
bitrate) for all DVDs, with some it works with other I get the problem.
So it seems that the problem must be in the DVDs themselves.

Some observations that I made:

1) the sync problem would be there even if I omitted pullup and softskip
filters. One DVD that I have has first telecined content in the middle
of the movie, however the sync problems would be there from the very
beginning (with or without pullup,softskip)

2) there would be no sync problem if I compressed the audio stream with
lame instead of just copying it.

3) MEncoder issues "1 duplicte frame(s) found" warning now and then. A
small fragment of the log:

Pos:   0.0s      2f ( 0%)   0fps Trem:   0min   0mb  A-V:0.000 [0:0]
1 duplicate frame(s)!
ODML: vprp aspect is 16384:7220.
ODML: vprp aspect is 16384:7220. Trem:   0min   0mb  A-V:0.000 [0:0]

1 duplicate frame(s)!
Pos:   0.3s     10f ( 0%)   0fps Trem:   0min   0mb  A-V:0.000 [0:0]
demux_mpg: 24fps progressive NTSC content detected, switching framerate.
Pos:  17.4s    421f ( 0%)  53fps Trem:  48min 2133mb  A-V:0.000 [2801:0]
1 duplicate frame(s)!
Pos: 230.8s   5539f ( 4%)  45fps Trem:  44min 1007mb  A-V:0.000 [1631:0]
1 duplicate frame(s)!
Pos: 255.5s   6131f ( 4%)  45fps Trem:  43min 964mb  A-V:0.000 [1554:0]]
1 duplicate frame(s)!
Pos: 268.3s   6437f ( 5%)  45fps Trem:  43min 944mb  A-V:0.000 [1518:0]
1 duplicate frame(s)!
Pos: 269.2s   6459f ( 5%)  45fps Trem:  43min 943mb  A-V:0.000 [1515:0]
1 duplicate frame(s)!
Pos: 270.8s   6497f ( 5%)  45fps Trem:  43min 940mb  A-V:0.000 [1509:0]
1 duplicate frame(s)!
Pos: 275.4s   6607f ( 5%)  45fps Trem:  43min 933mb  A-V:0.000 [1497:0]
...

Is the message "duplicate frame(s)" a good or a bad thing? (I wasn't
able to make out the right answer in the responses in Corey's thread)
Why does it appear in progressive content?

4) I use Windows Smartripper to backup the DVDs and MPlayer seems to
have wrong information about the length of the movie (checked on the OSD
display). However the result is always OK (as far as length is
concerned). I'm including this observation here only for completeness.
(BTW: I have a very small Linux partition (3GB) so I have to grab the
DVD on a Windows partition)

5) the sync problems exist only for some mixed-content DVDs not all of them.


I'm using MEncoder only for a couple of months now. So I'm no expert. If
I overlooked something obvious, I apologize in advance!

Thank you for your thoughts/comments.

David




More information about the MEncoder-users mailing list