BACKGROUND: "THE STING" [1973] DVD [2005] has a picture glitch that's
fixed by using frame 2184 to replace the 5 frames that follow it. The
'easy' way to do that is via 'splits' and 'interleave'. Unfortunately,
in 'splits' and 'interleave', FFmpeg changes timebases and PTSes and
ruins things -- see "'split'+'interleave' causes FFmpeg to use the wrong
'tbn'", posted yesterday. As a result, I had to replace the 5 frames the
long way around: by making separate, 1 frame files.
MY ASSERTION: Note the five durations that are marked. I believe that
investigating why 'showinfo' gets them wrong is a simpler task, but one
that seems pretty strongly related to my previous post. That assertion
is based on my experiences between yesterday and today -- it's been a
hard slog.
The TARGET is unformly 24/1.001 fps with a 24k tbn, now, but getting it
there was not what anyone would call easy.
I will keep all the files, scripts, and documentation for a few days in
case anyone is interested in investigating -- feel free to contact me
off-list if you prefer. I think this icicle may be the tip of an iceberg.
I made 10 files:
THE STING [1973].mp4 <= SOURCE (correct PTSes)
THE STING [1973]v..91.091.mp4 <= time=0..91.091 (pts=0...)
+------------------------------------------------------------+
¦ THE STING [1973]v91.132.mp4 <= copy of 91.091 (pts=984) ¦ These are
¦ THE STING [1973]v91.174.mp4 <= copy of 91.091 (pts=1992) ¦ copies of
¦ THE STING [1973]v91.216.mp4 <= copy of 91.091 (pts=3000) ¦ 91.091
¦ THE STING [1973]v91.257.mp4 <= copy of 91.091 (pts=3984) ¦
¦ THE STING [1973]v91.299.mp4 <= copy of 91.091 (pts=4992) ¦
+------------------------------------------------------------+
THE STING [1973]v91.341...mp4 <= time=91.341..7736.812 (pts=123120...)
THE STING [1973]v_concat.txt
THE STING [1973]v.mp4 <= TARGET (correct PTSes)
In the TARGET:
n pts duration duration_time iskey type
0 0 1001 0.0417083 1 I
:
2182 2184182 =>3003<= 0.125125 0 B
2183 2185183 1001 0.0417083 0 P
2184 2186184 1001 0.0417083 1 I
+----------------------------------------------------+
¦ 2185 2187185 1001 0.0417083 1 I ¦ These are
¦ 2186 2188186 1001 0.0417083 1 I ¦ from the
¦ 2187 2189187 1001 0.0417083 1 I ¦ copies of
¦ 2188 2190188 => 1<= 0.0000416667 1 I ¦ 91.091
¦ 2189 2191189 => 1<= 0.0000416667 1 I ¦
+----------------------------------------------------+
2190 2192190 1001 0.0417083 0 B
2191 2193191 1001 0.0417083 0 B
2192 2194192 1001 0.0417083 0 B
2193 2195193 1001 0.0417083 0 B
2194 2196194 1001 0.0417083 0 B
2195 2197195 => 999<= 0.041625 0 P
:
185498 185683498 1001 0.0417083 0 P
_______________________________________________
ffmpeg-user mailing list -- ffmpeg-user@ffmpeg.org
To unsubscribe send an email to ffmpeg-user-le...@ffmpeg.org
