Electronic Media 

Published: 11-06-95 



TIME TO BRUSH UP ON YOUR DIGITAL FACTS    



Byline: Joe Fedele 



 Digital. You bump into the word everywhere you turn. In this

industry, digitally compressed video is the wave of the future,

and with it comes a torrent of new and confusing language.

 Here's a cheat sheet to help you keep track of important

digital acronyms and lingo. Many thanks to Sony Corp.'s Hugo

Gaggioni, who wrote a paper on the subject that I consulted.

 To start with, there are two primary categories of compression,

JPEG and MPEG.

 JPEG, which stands for Joint Photographic Experts Group, treats

video like a series of still photos and compresses each frame

one at a time. You may have also heard about DCT (discrete

cosine transform) and wavelet coding. These are also JPEG

schemes.

 MPEG, which stands for Moving Pictures Experts Group, is a

compression that analyzes the motion characteristics of video

and compares frames to each other. It then ``predicts'' what the

difference is and passes that information along instead of the

entire frame.

 MPEG-2 is a new version of MPEG-1. There was talk of creating

an MPEG-3 version for high-definition TV until the standards

committee agreed to include HDTV in MPEG-2.

 Compression standards generally use ratios to express how much

information is ``sampled'' from video. These ratios present the

picture brightness and color content of video in their basic

components.

 A commonly used ratio is 4:2:2, with the digits representing a

sampling of the luminance, R-Y and B-Y, respectively.

 You will also see ratios such as 4:4:4, 4:2:0 and 4:1:1. The

main difference between them all is that equipment using higher

sampling rates will yield better pictures. And you probably

already guessed that picture quality is directly proportional to

equipment cost.

 You might come across the terms 8-bit and 10-bit bandwidth used

in conjunction with sampling ratios. The difference is that

10-bit or full bandwidth video contains four times more picture

information than 8 bit. Consequently, it is better quality at a

premium price.

 The amount of compression used is also reflected in a ratio. I

have seen compression ratios ranging from 2:1 to as high as

160:1. Most compression schemes operate between 2:1 and 15:1.

Just remember that as the ratio increases so does picture

degradation.

 Transmission of digitally compressed video is generally

expressed in megabits per second. Simply put, the higher the

value, the higher the picture quality (and the higher the cost).

 Typical bit rates for 4:2:2 studio component video are 210 Mbps

for 10-bit video and 168 Mbps for 8-bit video. Bit rates of 90

Mbps are used for program origination and post production, while

rates as low as 1.5 Mbps can be used for ``VHS-like'' quality.

 Storage of digital video is expressed in gigabits. The amount

of space occupied by video is directly proportional to the

length of time the video runs and the bit rate it is received

at.

 One hour of 4:2:2 component 10-bit programing will fit into 97

GB of disk space. But if you compress it at a 2:1 ratio you can

squeeze it into about 58 GB of disk space.

 Still confused? Don't fret-you're not alone. Many top-notch

engineers in broadcasting can't keep up either!