Why is it so friggin’ hard to make trivial changes to a video?

By | July 29, 2009

I’ve got a video in WMV format that was recorded from a GoToMeeting session that I hosted.

I want to chop a little off the front and a little off the end and otherwise leave the video unchanged.  I don’t want to change the format, the resolution, the compression, or anything — I just want to produce a video which is the same as the original but with a bit chopped off the beginning and end.

I’ve spent two days for a tool that will do this easily (or at all!) on either Windows or Linux.  I started out with free tools, and when none of them panned out moved onto the ones that cost money.  The two closest possibilities I’ve found are Pinnacle VideoSpin and Adobe Premiere Elements.  VideoSpin refuses to export the edited video at the same resolution as the original — it insists on shrinking it down, and I can’t figure out how to convince it not to.  Furthermore, the video produced by VideoSpin is about twice as big as the original, so it’s obviously not preserving encoding or compression.

Premiere Elements offers many different output formats but offers no guidance about which one to choose.  Furthermore, all of the combinations of export settings I chose seem to produce videos that are humongous compared to the original.

So, can anybody offer any useful suggestions for how to do what I’m trying to do?

Print Friendly, PDF & Email

6 thoughts on “Why is it so friggin’ hard to make trivial changes to a video?

  1. Pingback: AVS Video Converter to the rescue « Something better to do

  2. jik Post author

    My mencoder / ffmpeg plan failed. One of them — not sure which — is causing the video and audio to get out of sync with each other. Damn.

  3. abbasegal

    VirtualDub with “Direct Stream Copy” enabled theoretically should do this (at least according to http://www.doom9.org/index.html?/virtualdub_procedures.htm). The one time I tried it it still wanted to re-encode, but I think I may have forgotten to tell it direct stream copy for audio as well as video.

    It does seem like this shouldn’t be so difficult to do!

  4. jik Post author

    I tried using mencoder as you indicated and encountered two different problems:

    1. With the input file format I’ve got, -oac copy -ovc copy produces a file that can’t be viewed by the players that were able to view the original file.

    2. With the input file format I’ve got, -ss doesn’t work properly. The time specified to -ss isn’t the time that mencoder actually starts at — it gets blown up somehow to a much larger delay, although I wasn’t able to figure out exactly the relationship between the value specified to -ss and the delay that mencoder actually uses.

    What I’m trying now is using mencoder without -ss or -endpos to produce a really big AVI file that ffmpeg is capable of reading, and then using ffmpeg with -ss and -t to produce a QuickTime movie. Yes, this means the file format is changed, but least the resolution is preserved.

  5. Greg

    I agree, everything having to do with video encoding is unnecessarily complex. All the tools are written either by windows programmers who think we just want a big lickable button and don’t understand anything or by pimply kids who want to prove how smart they are by making sure their documentation doesn’t explain anything and their tools are as baroque and complex as possible.

    As it happens I did just this earlier today and this is what I came up with:

    mencoder -ss 5:45 -endpos 5:25 -oac copy -ovc copy -o /tmp/output.mpeg input.mpeg

    That starts at 5:45 in and keeps going until 11:05, ie, the resulting video is 5:25. Actually it doesn’t work perfectly because mencoder can only seek into to a key frame. Even then the first second or so was a bit corrupted because it wasn’t the right kind of key frame? Not sure.


Leave a Reply

Your email address will not be published. Required fields are marked *