Digital Video Formats

MPEG

MPEG has emerged as the video compression scheme of choice for digital video delivery. It is the scheme most used on DVD's, digital satellite/cable signals as well as multimedia video over the Internet.

Because MPEG was designed for final digital video delivery is not the best format for source video. It is best to use a different video compression scheme for your source video then convert the final product to MPEG.

DV Codec

Most Codecs (Compressor/Decompressor) were created to compress audio/video for multimedia delivery, that is, CD-ROM's and the Web. The DV codec is different from most other codecs, in that it's not intended for multimedia output. Instead, it's designed to be used in two primary ways:

  • For storing original footage, and transferring it to a computer
  • For storing final edited footage, for playback to a television

Lets look at the amount of data generated by an uncompressed video signal.

NTSC video is composed of 30 frames per second. At this rate we will consume disk space at the rate of 1,036,800 x 30 = 31,104,000 bytes per second or 30 Mb/s. At this data rate you could only store a little over 30 seconds of video per gigabyte (Gb) of hard drive space. A 30 minute video show would consume approx. 60 Gb.

To help cope with this massive amount of data and to achieve better performance levels from our hardware we must use compression. There are 2 varieties of compression schemes:

  • lossless
  • lossy

Lossless compression means that when the image is uncompressed again it will be exactly as it was before it was compressed. Lossy compression means that some non-critical information is thrown out and not included in the compressed data. This causes the image to degrade slightly when it is again uncompressed.

DV uses a compression scheme called Discrete Cosine Transform or DCT. It is the underlying scheme behind JPEG images. The DCT compression scheme is a lossy compression scheme.

The uncompressed audio/video stream is around 30 megabytes of data per second and we have to compress it enough to allow our disk drives (or tape drive in the case of a DV camcorder or VCR) to keep up. Algorithms are used to determine which parts of the data can be ignored while still maintaining the original appearance of the picture. Every frame is compressed to about 120,000 bytes and data is written to the DV tape at a rate of 3.6 megabytes per second (including audio). 

Because of the constant compression rate of 3.6 Mb/s you video can fluctuate in quality. If there is lots of action in your video, image quality will be reduced. For this reason it is always better to shot with a tripod. This keeps the background steady from frame to frame and leaves more of the video bandwidth to be put towards the action in each frame.