At DDA, we shoot our raw video with full-quality High Definition digital cameras. The picture quality is astounding, but the filesize is… large, to say the least.
When a video file is way too big, it must be compressed. Maybe it needs to stream over the web in a nifty Flash video player. Perhaps we need to run video simultaneously on top of an interactive training simulation, and simply cannot allow it to freeze up while the video loads. Compression is something that helps people working on web video tremendously, but can also be of aid in the creation of content for DVDs and television too.
In essence, compression makes a video file smaller. It does this by reusing data for pixels that don’t change over certain periods of time. For instance, if you have a video of a person standing against a white background and talking, the compression takes all of the space around the person, and everything else that’s not moving in the shot, and makes a keyframe out of it. Then, the file can reuse that keyframe for as long as those pixels are not changing. For better video quality, some compressors use more keyframes. For smaller file size, they can use less.
Ultimately, it comes down to video quality vs. how fast you want your video to load. Generally, with the permeation of broadband internet today, you can get a pretty nice quality video streaming on the web with barely noticeable compression. Indeed, some services (such as Netflix) are now streaming high quality HD video directly to people’s televisions through an Xbox 360, Playstation 3 or other device. Pretty soon we won’t need to compress videos at all; but until then, we’ll continue rendering video to formats that will be compatible with the maximum number of internet connection speeds!
Entry by: Rob