How DVDs Store Video Content
How DVDs Store Video Content A standard, single sided DVD is capable of storing approximately 4.7Gb of data. Hover, even with such a large storage capacity, a single DVD could not store a whole feature length movie without ‘compression’.
Compression is the process of condensing a data file into a smaller physical size, while maintaining no, or little noticeable loss in quality when the video is played back.
A typical feature film is condensed using MPEG compression techniques, originally developed by the Moving Picture Experts Group, who are responsible for establishing standards for video compression.
An immediate problem with compression is often that the final compressed file will often display a noticeable degradation in quality when compared to the original master. Over the years, the MPEG group has been responsible for developing various different MPEG standards as technology has moved, data files have become ever larger and compression requirements have changed.
Video is typically stored on DVDs using the MPEG-2 standard. This was specifically designed to allow for maximum compression of the master file, while maintaining excellent video quality.
DVD players are essentially powerful MPEG decoders, able to read the source file and decode the contents before replaying the image to the TV screen.
PAL and NTSC:
Two dominate formats have immerged relating to video formatting. European markets typically recognise a system called PAL while America and some Asian markets use NTSC.
NTSC delivers 30 frames per second to a TV screen using 60 alternating lines called ‘fields’. PAL displays 50 fields a second; however it does so using a higher resolution.
Therefore, the original source file needs to compressed using a MPEG standard which is either compatible for PAL or NTSC DVD players.
An MPEG encoder is used to scan each frame of the source file and encode it using one of several methods. These methods are primarily used to eliminate data that is earthier irrelevant or of no real use, therefore reducing the overall size of the compressed file. Each frame will be encoded as:
- Intraframe: an intraframe contains all of the data from the master frame. That is to say, that the encoder decides that almost all the information in this frame is important and will therefore contact the minimal amount of compression on this frame.
- Predicted: the encoder compares the current frame and the previous frame to see how the current frame has changed. A predicted frame will only contain the data that has changed from the last frame and therefore the compression of this frame can be quite high.
- Bidirectional: here, the encoder uses information from the closest intraframe and predicted frame and uses interpolation to calculate how and where each pixel should be displayed upon the screen.
The encoder will decide which of the above methods to use based on the type of scene which is being encoded. Very fast action scenes are likely to consist largely of intraframes because the information from one frame to the next is likely to be very different.
However, a TV news bulletin will likely be made up more of predicated frames because the background environment in particular is likely to remain fairly static from one frame to the other.
