Mencoder -edl with Comskip and ShowAnalyzer

[gnutech]

I have discovered that recent releases of mencoder do not like the edl files generated by Comskip and ShowAnalyzer. Mencoder will skip only the first segment listed in either edl file when trying to transcode the video.

Comskip Generated EDL:

0.03 78.51 0
648.81 795.43 0
1622.25 1771.74 0
1979.01 1979.04 0

ShowAnalyzer Generated EDL:

0 78.3784 0
648.615 795.329 0
1622.09 1722.52 0
1979.01 1979.01 0

However, when I manually create an edl file using mplayer with the -edlout option it generates the following edl that Mencoder successfully transcodes each delete segment listed.

MPlayer Generated EDL:

71836.203125 71914.710938 0
72484.578125 72631.664063 0
73458.453125 73558.757813 0
73587.757813 73815.257813 0

There is an obvious difference between these edl files even though they are for the same video and nearly the same delete points.

Why the difference between the Comskip/ShowAnalyzer edl's and the manually created Mencoder edl file? Is it possible for Comskip/ShowAnalyzer to output the same type of timing as the manually generated MPalyer edl?

Thank you for your assistance.

Respectfully,


Gary

[erik]

Do I understand correctly that each comskip/sa generated .txt line corresponds to one mplayer created line?

Looks like they changed the source, lets have a look.

[erik]

Other question.

Do you see this offset on every recording this way? Or is it depending on the actual recording?

it looks like there is a constant offset of 71836.203125 added to the cutpoints.

Any problem with the PTS in the recording?

[erik]

MMmm.

I checked the source of mplayer.
No visible changes in the edl part of the software.

[wtg]

Gary,

Have you tried this with more than just the one file? I wonder if there's something screwy in the recording.

I'm using the most recent release of mencoder (an Athlon-specific build) and I haven't run into any trouble at all. The -edl processing works just fine.

That's strange.
Tim

[pBS]

i wonder if that's the libav cpu bug again....only found in specific P4 builds, the sse2 versions...
try a non=sse2 version compile of mencoder and see if that doesn't fixi it...

i believe it's a timestamp issue tho...make sure you use -forceidx to try and reset the timestamps inside..

[gnutech]

Erik,

Yes... Each comskip/sa generated .txt line corresponds to one mplayer generated line and it appears to depend on the particular a/v file, but only the manually created mplayer edl file will work correctly with mencoder.

Another Comskip Generated EDL:

0.03 52.29 0
570.57 766.23 0
1149.38 1269.87 0
1519.32 1670.37 0
1979.18 1979.21 0

And the MPlayer Generated EDL:

44800.500000 44907.242188 0
45363.730469 45566.667969 0
45934.734375 46070.335938 0
46319.953125 46470.601563 0

I am using RC1 of the Windows MPlayer/Mencoder binary.

MPlayer 1.0rc1-3.4.2 © 2000-2006 MPlayer Team
CPU: Intel® Pentium® 4 CPU 2.00GHz (Family: 15, Model: 2, Stepping: 4)
CPUflags: MMX: 1 MMX2: 1 3DNow: 0 3DNow2: 0 SSE: 0 SSE2: 0
Compiled with runtime CPU detection.
c:/windows/fonts/arial.ttf doesn't look like a bitmap font description, ignoring
.
Cannot load bitmap font: c:/windows/fonts/arial.ttf

Playing South Park - 512- Here Comes the Neighborhood.mpg.
Read 4 EDL actions.
MPEG-PS file format detected.
VIDEO: MPEG2 720x480 (aspect 2) 29.970 fps 3500.0 kbps (437.5 kbyte/s)
SUB: Could not determine file format
Cannot load subtitles: ./South Park - 512- Here Comes the Neighborhood.txt
==========================================================================
Opening video decoder: [mpegpes] MPEG 1/2 Video passthrough
VDec: vo config request - 720 x 480 (preferred colorspace: Mpeg PES)
Could not find matching colorspace - retrying with -vf scale...
Opening video filter: [scale]
The selected video_out device is incompatible with this codec.
Try adding the scale filter, e.g. -vf spp,scale instead of -vf spp.
VDecoder init failed
Opening video decoder: [libmpeg2] MPEG 1/2 Video decoder libmpeg2-v0.4.0b
Selected video codec: [mpeg12] vfm: libmpeg2 (MPEG-1 or 2 (libmpeg2))
==========================================================================
==========================================================================
Opening audio decoder: [mp3lib] MPEG layer-2, layer-3
AUDIO: 48000 Hz, 2 ch, s16le, 224.0 kbit/14.58% (ratio: 28000->192000)
Selected audio codec: [mp3] afm: mp3lib (mp3lib MPEG layer-2, layer-3)
==========================================================================
AO: [dsound] 48000Hz 2ch s16le (2 bytes per sample)
Starting playback...
VDec: vo config request - 720 x 480 (preferred colorspace: Planar YV12)
VDec: using Planar YV12 as output csp (no 0)
Movie-Aspect is 1.33:1 - prescaling to correct movie aspect.
VO: [directx] 720x480 => 720x540 Planar YV12

Exiting... (Quit)

I'm not sure if it has anything to do with it, but I am using Maui-III video capture cards.

Thank you for your assistance.

Respectfully,


Gary

[erik]

Could you run comskip with verbose=11 and mail me the first megabyte of the log file (the total file will be HUGE)

[gnutech]

Erik,

I changed the comskip.ini verbose option to 11 and reran as requested.

It didn't seem to effect the log size, so I've included the complete file.

I did try transcoding with mencoder using the -idx option with no success.

Thank you for the assistance.

Respectfully,


Gary

[erik]

PTS seems to be OK.

I tried the new mplayer on a mpeg file and it created
4.133333 8.466667 0
12.933333 15.600000 0
21.566668 25.433334 0
which seems to be normal

I have no idea what to do next.....
However,

Could you upload the first part of the mpeg file to my ftp server?
PM me for the details.