Video container question

Orwell · Post by **Orwell** » Wed Oct 25, 2006 3:58 am

Is there a way to have multiple videos in a single container be selectable, and have one play over the other at the same time? Picture in Picture determined by the viewer, basically. Is that only possible with text? Will any container that supports titles support multipe ones, and allow you to them all over the place, as compared something more standard such as lines in fansubs? What program would you use for that?

BauziOLD · Post by **BauziOLD** » Wed Oct 25, 2006 6:44 am

mp4 can work with it. I´ve found it out, when I wanted to put out the new file, over the old one. When I´ve opened the final mp4 there where two windows with the video at the same time opened.
So: Yes there is a way with mp4, but I don´t know how to fix it with more "style". I mean... Rendering over a file that already exist looks a bit silly to me

For better understanding:
Putted out a video called: video.mp4
-watched the video and was not pleased with quality. So I made another output with the same name "video.mp4" and the "programm" didn´t asked me if I want to write over the file that already exist with the same name. So he writed over this video file... Yeah and after that he opens two windows with the same video in it, when I open the file.

Uhm... does it make sense to you? This is really not easy to explain for me...

Is that only possible with text?

In my example it is not like text. It is more a second preview window opened at the same time, with a different video stream.

Purge · Post by **Purge** » Wed Oct 25, 2006 7:05 am

I dont know a program with such capability but maybe just open up the file in multiple instances of wMP classic and "hey presto" you can have picture in picture.

Orwell · Post by **Orwell** » Wed Oct 25, 2006 7:29 am

Purge wrote:I dont know a program with such capability but maybe just open up the file in multiple instances of wMP classic and "hey presto" you can have picture in picture.

Technically it works, but then it defeats the whole purpose of what I'm trying to model. I suppose I could just confine it to text, however, I'm having doubts about a subtitle maker being able to replicate what you can do with text in photoshop, or even premiere. Theoritically one could make all the text in photoshop, but again, that would require an overlay of another video track on the same one in a single container, unless it were hardsubbed, which... bleh.

Qyot27 · Post by **Qyot27** » Wed Oct 25, 2006 6:34 pm

Sub Station Alpha has the ability to use more than just text (it can use pictures, I don't know about video), but I've never gotten it to work with anything but text, so I can't really say.

Video-on-video overlay is probably something that is a designed or hypothetical feature in MP4 or Matroska, respectively, but I've never heard of there actually being a working implementation. Both containers support more than one video track, but it's the same deal as having multiple audio tracks - you hear/see one or the other, not both (although in older versions of Gabest's Matroska splitter, it would make WMP 6.4 launch additional compact display windows if the MKV had more than one video track - Haali's splitter doesn't display that behaviour, and actually doesn't seem to like dual video MP4s, at least from the file I just tested anyway).

Orwell · Post by **Orwell** » Wed Oct 25, 2006 7:13 pm

Hmm, so its not likely to work except with text. If I wanted to stay with the video idea, would it be possible to create the video, and have it be plain without the text/extra visuals, then create a second video track that references the first to conserve space?

I'll have to look into sub station Alpha, though I have a feeling I'll just be making two videos. Where are those hypothetical features when you need them?