by Marcus Weise

What Does It Mean To Digitize?

Television was originally developed as part of the analog or real world. An analog system has the ease and advantage of fitting into the natural physical system. It also carries with it, however, all the interference and noise problems that are part of the analog world.

Noise is analog information, just as video and audio are, and so getting rid of it is not easy. Analog information is also not easily manipulated, making creative use of the medium difficult. Storage of analog information is also complicated, delicate, and somewhat space hungry. 

Digitizing refers to converting analog information to numerical values. This accomplishes several things at the same time. As digital information, the signals are not subject to real world analog interference. The picture and sound information become very manipulable. Real world physical problems have no effect on the television signals when they are digitized. Digital signals are absolute and do not change over time. 

Sampling Rates

Digitizing is the process of measuring the original signals and assigning a number to that value. The more frequently the measurement is taken, the more accurate the digital reproduction of the changes will be. For example, in a child’s connect-the-dots puzzle, the more dots there are to connect, the closer the curves and outlines will reproduce the picture. The frequency of the readings taken of the signal is referred to as the "sampling rate".

The specific sampling rate for digital television is a little over fourteen million times a second for the luminance or "light" information, and a little over seven million times a second for each of the color signals. The measured value taken each time is recorded as a numerical value on the recording medium (tape, floppy disk, hard disk). What is recorded is not the real world analog signal, but a series of numbers representing the measurements of the video and audio signals at each instant in time.

Early Digital Systems

Early computers were a series of switches that were either on or off representing a zero or a one and taken to mean "yes" or "no". This system allows the processing of information by having the computer go through its memory and extract data based on yes or no answers in response to programmed commands. In human terms, for example, suppose someone wanted to find out all about you; but you could only answer yes or no (represented by a 0 or a 1). To find out your name, they would go through the alphabet and you would answer yes or no to each letter until they got the correct letters of your name. Doing the same with your address, phone number, parent’s names and so on, they could find out everything about you. The process is slow, but accurate. This is essentially what a computer is doing as it goes through a program and its memory. The faster it goes through the yes or no questions, the faster it can process information. The rate at which the computer processes information is measured in cycles per second or Hertz. The speed is so many millions of times per second or Megahertz or billions of times a second or Gigahertz. The higher the rate, the faster the computer can process information.

Binary vs. Decimal

A digital or "binary" bit is a zero or a one. It is called binary because there are only two numbers. The number of binary or digital bits the computer can read at once is known as the "word size." There are 8 bit, 16 bit, 32 bit processors and larger. The bigger the size of the word the computer can handle, the faster it can process information. 

In the decimal system, the system of counting in tens, the far right column represents ones. The next column to the left represents tens. The next column represents hundreds, then thousands and so on. Each column has a value from zero to nine. After nine you go to the next column to the left and continue counting. For example, the number 198 represents eight "ones," nine "tens" and one "hundred." After 198 comes 199, then 200. A 200 means no "ones", no "tens" and two "hundreds." 

In the binary system, a computer does the same type of math but its columns only have values of 0 and 1. The first column represents "ones," the second column is "twos," the third column is "fours," the fourth column is "eights" and so on. 

In an eight bit number system the highest value represented by a one in each column is 255. The number 256 is the start of a ninth column. An eight bit computer processor can read eight bits or one "byte" at a time. For information that requires a larger word, the computer reads in groups of eight bits, putting the eight bit words together.

In the ASCII system (American Standard for Computer Information Interchange) there are 256 numbers, letters, symbols etc., each with its own binary code. When inputting information into a computer each letter or symbol that you include is represented by an eight digit binary code inside the computer.

Converting Analog To Digital

In processing a video or audio signal digitally, the analog information is measured as to frequency, amplitude, voltage etc. These numerical values are recorded as binary or digital bits on the recording media. These numbers can also be used to alter video or audio signals for creative reasons. They can be added, subtracted, reversed or manipulated in an almost unlimited number of ways. This is one of the strong reasons for working in the digital realm; the ability to manipulate the information for creative purposes. 

Digitizing also gives the equipment the ability to detect, correct, and conceal errors far beyond anything available in the analog realm. But there are certain precautions to be aware of in working with digital and analog signals. If analog signals are converted to digital, it is best to remain digital. This is because the digitizing of analog information yields an approximation of the original not an exact copy. To go back and forth between analog and digital invites serious degradation of the information. Something is lost every time the signal is translated from one domain to the other. However, once digitized, if the information remains digital it will yield an exact digital duplicate each time it is copied. Digital copies are exactly like the digital original in every respect.

Recording Digital Signals

There are several ways of recording digital video and audio signals including magnetic media such as tape, floppy and hard disk drives and non-magnetic media such as solid state "chips," optical disks, laser disks, CD’s and DVD’s. CD’s are probably the most common form of non-magnetic digital recording. Audio CD’s, computer CD-ROM’s (Compact Disk-Read Only Memory) and DVD’s (Digital Versatile Disks) are good examples of these non-magnetic media. Compact Disk and DVD recordings are almost immune to damage with reasonable handling. Scratches and surface damage usually do not affect the quality of the playback and the disks are light, small, allow very rapid cueing and repeated use without any wear or loss of quality.

Storage Methods

In computers and computer based products such as non-linear editing systems, hard disk drives are the recording media. In a non-linear editing system, all the program material is "downloaded" or recorded onto the system’s hard drives for the editor to work with. This can be done while the production is going on, or after the fact from whatever media the program was originally recorded on. Like a floppy drive, the hard disk is made to hold magnetic energy which is the way the binary information is stored.

Hard drives are extremely fast, can hold large amounts of information, and allow true random access to the data that is stored there. Because of this random access capability video can be edited in the same fashion as film. Scenes and segments can be moved around and replaced any number of times and ways. When you are watching, the computer is playing back the material you asked to see off the hard drives. If you want to change the order of playback all you do is change the numbers. Since nothing is recorded in any specific order, and the computer is accessing information based on your selection, the order of playback can be changed at any time.

Large capacity hard drives are bringing central servers into the market. A central server is a large hard disk storage system or data bank allowing the compiling of thousands of hours of video and audio in one location, or several locations interconnected by a network. Therefore the material can be accessed by any number of people simultaneously, allowing viewing on demand and also the sale of program material to any place in the world at any moment.

A central server system can allow a program to be produced in one location, but edited in another without anyone having to get out of their chair and without having to wait for the material to be sent anyplace. 

The New Digital Environment

We are moving toward a digital television world. We will create in a digital environment, broadcast digital signals, and watch digital wide screen pictures at home. Digital television will mean more programming, excellent quality pictures and sound, and almost unlimited picture size. It will also allow the world to watch each other’s programs.

Digital television is changing the way we all do our jobs, and also the way we enjoy the results of our work. It is opening more markets and creating a demand for more material. It is allowing more people to create that material without the need for a lot of very expensive, highly technical equipment.

Video Theory and Operations Copyright © 1999 Rev. 2004
by Marcus Weise
Reprinted by permission of the publisher
Weynand Training Publications