DaWei

Here's the deal. Media files are representative of analog signals, which are continuous. (I won't go into the theory that analog signals are micro-digital signals).

One represents these signals by assigning discrete values to some range of continuous values. Suppose that you have a range of 100 volts, but only 8 bits to represent that range. That's 256 actual values. One-hundred volts divided by 256 is a tad uner a half volt per representation. In other words, a signal of 50 volts will be represented the same as a signal of 47.75 volts or a signal of 50.25 volts.

This removes some information. The audience of a punk band might never detect the difference, whereas the audience of a classical ensemble might very well run off into the grass and puke on their shoes.

When this "lumping into discrete categories" occurs, it is called aliasing. One can reduce aliasing by providing a larger range into which to divide the digital representation. One may go to 16, or 32, or 56, or whatever, divisions for a specific period of time. This has nothing to do with format. It has to do with busyness.

If you want to join two signals together, even if they are the same format (a macro property) then they must have the same digitally-expressed range, or you must convert the range to the area of interest.

Depending upon the way your moving, this might require a guess at some intermediate range to assign to a signal, with no prior information about what direction to slide it, or it might merely require what is equivalent to a flooring or rounding or ceiling of information.

First, you have to establish a common denominator. Then you can splice to your heart's content. You cannot add quality (unless you're Godlike), but you can always detract from quality.

__________________
Abstraction doesn't make it impossible to write bad code; it makes it possible to write superior code.
Contributor's Corner: Grumpy on C++ Exceptions DaWei on Pointers

cloud-

to join the video files together they must be the same size (frame), fps, etc, if you are using vdub, as in the guide posted previously, so it's likely that the videos differ somehow? I didn't check but if that's the case they would need to be re-encoded. alternatively you could use avisynth, it's a "scripting" interface for dealing with video and audio, it allows you to "edit" the video before processing it with the main application (vdub)

an example of joining 2 videos together, with different fps/size etc.


Though I'm not sure if this would cause audio skew (if fps was the case), you'd need to know how the clips differed, you could also just remove the audio like in the example and add that separately.
edit: using avs would need a re-encode too, so it's all the same in the end, i guess, just some time saved..

Ghost

http://www.esfiles.com/cplusplus.wmv

There is everything linked together. Sorry it took me so long to post, kind of forgot I was doing you a favor. (Had a large project released and took up most of my time)

lectricpharaoh

I think this depends on the codec. I'm pretty sure MPEG-4 streams can be joined by simply appending one to another, though the stream data will obviously need to be taken from the container (AVI, or whatever).

@Matt: What kind of videos are these? Just plain MPEG-4 (like Xvid/DivX) inside an AVI container?

__________________
A man's knowledge is like an expanding sphere, the surface corresponding to the boundary between the known and the unknown. As the sphere grows, so does its surface; the more a man learns, the more he realizes how much he does not know. Hence, the most ignorant man thinks he knows it all. - L. Sprague de Camp