Home Movie Editing

As a general rule of thumb, when editing video on a PC, you should always keep the quality of your source material as high as possible until the point when you publish, whether you are publishing to DVD, back to tape, or to the web. This guide aims to help you understand the basic differences in the compression techniques used in the most common video formats.

Ultimately, the highest quality of video is full, uncompressed video. However, even with the relatively low cost of data storage these days, uncompressed video takes up a hell of a lot of space.

Types of Compression

Intraframe Compression

This means the data within each individual frame is compressed. An example of this type of compression is DV Video, which uses type of JPEG compression, and the resulting file size is about five times smaller than uncompressed video.

The important point here is that each frame is still a whole frame, it's just been made a bit smaller in file size. This makes editing this type of video very easy for your computer, as a cut between frames can take place instantly with little or no processing required.

Interframe Compression

This is the basis of most modern formats (MPEG2/DVD, MPEG4 for example). It works by analysing the difference in motion between frames, and the subsequent frames are compressed by only including the parts of the image which have changed. The first frame is called the 'I-frame' and it is the only complete frame until you get to the next 'I-frame'.

Each set of frames is called a Group Of Pictures, and the number of frames between each 'I-frame' is determined by the software which creates the MPEG file. The diagram below may help to explain things visually.

Video encoded using Interframe compression takes more processing power to edit than that which uses Intraframe compression.

When you are editing footage, most software will not alter the original footage, unless you add a title or a cross fade, or create a cut during a scene. When this happens, the footage must be recompressed, as technically a new picture is being created. This is relatively easy for the software to do with DV, as it can cut the footage after a complete frame, and apply effects to each complete frame individually.

When editing MPEG footage, if you want to make a cut or add effects to any frame other than the I-frame, your software will have to 'build' the incomplete frames between the I-frames, and then recompress again after, which reduces quality, and takes additional processing power.

Some software will not allow you to cut an MPEG file between I-frames which can make editing a bit clunky. Also, you will find that editing MPEG files is slower than editing DV, whatever software you use.

Below is a chart which gives some information about various standard definition digital video formats and their file sizes.