Digital and High Definition TV

by Marcus Weise

What Is Digital and High Definition Video?

Digital Video and High Definition Video are two separate entities. Digitizing an image is the process of converting analog information to numerical or digital values. High Definition is a wider screen image with more detail than the present standard analog television image. Currently there are approximately 32 different possible standards being considered for digital and high definition television, each with its own advantages and disadvantages. The original high definition systems proposed primarily by the Japanese manufacturers were analog. They had several different variations that included line standards up through 1125 lines per frame. Some of the American manufacturers developed alternate systems such as ACT or Advanced Compatible Television which would allow a wide screen image (16x9 aspect ratio) to be broadcast using the current technology and still be compatible with the existing NTSC system (4x3 aspect ratio). While these systems did address the high definition and wide screen image issues they were never accepted into the market so were never developed.

The Present System

The computer industry has always insisted on a much more detailed and higher quality image than has been used in the television industry. Computer images usually require very fine detail and very often a great deal of text that would be unreadable on a standard video screen. Present video images do not require the same degree of detail as computer information. However, as more and more video is being viewed through computers and on computer monitors, there is a need to bridge the two industries and create a compatible standard.

The present NTSC television system creates the television image using 525 horizontal lines per frame at 30 frames per second. The single 525 line frame is split into two images or fields. Each field is displayed as one image consisting of two hundred sixty-two and a half lines using alternate lines from the whole frame.. The remaining two hundred sixty-two and a half lines are displayed as a second image interlaced with the first.. The interlacing of the two images creates one whole frame of video at the rate of sixty of these half images or fields per second to create thirty frames per second of television pictures. This interlace scanning process conserves broadcast spectrum space and helps eliminate flicker that might be perceived in the picture. The NTSC standard of 525 lines per frame and interlaced fields at the rate of sixty per second was created in the 1940's and has not changed since, except to allow for color. The addition of color changed the frame and field rate to 29.97/59.94 from the original 30/60.

The Progressive Scanning Process

Where television uses an interlaced line scanning process; the computer uses a scanning technique called progressive scan whereby the image is displayed on the monitor as a series of pixels or picture elements from top to bottom as one full frame. Pixels are the individual building blocks that make up a picture on the monitor. The computer industry has variable standards with respect to the number of pixels used horizontally and vertically to create an image. This variable rate is necessary because computer standards have been continually upgraded. Computers have become more powerful, faster, and capable of handling far larger quantities of data in a relatively short period of time. Older model computers cannot handle as much data and require different standards, thus the need for variable pixel rates and image quality.

Computer Images vs. Television Images

This difference between interlace and progressive scanning is one of the main elements to be considered in deciding on a new standard. Each has its own benefits and drawbacks. For broadcast purposes, interlace scanning allows more information to be transmitted in the limited spectrum space allotted for the transmission of analog signals. As spectrum space is not a factor in computer image transmission, progressive scanning or sending the entire image as one frame is possible. The limiting factor in a computer system is the data transmission rate, or the number of bits per second that can be sent and received. This is referred to as bandwidth and is limited by the power of the computer and the compression and transmission techniques employed as well as the type of physical transmission used. These limitations are continually being overcome as the power and speed of computers are constantly increasing and the types of transmission systems are being improved to allow more data at higher rates of speed. While physical spectrum space is finite, the speed of computer systems is continually being expanded.

Frame Rates

The current NTSC system is thirty frames or sixty interlaced fields per second.. Thus, there are compatibility problems between television and film in countries using the NTSC system.. Film has a frame rate of twenty-four frames per second in North America. Techniques were developed many years ago to allow the two mediums to work together but these techniques were a compromise and created a new set of problems in their application. If the television frame rate were 24 frames per second, film and video would be compatible in those countries that use the NTSC system. The Europeans have never had this problem because their television and film are set at the same rate of 25 frames per second. European and American systems thus also need to find a common frame rate if we are to agree on a world standard.

The New Standards

The NTSC or National Television Systems Committee designed the present North American system. The ATSC or Advanced Television Systems Committee is helping to design the new system for digital and high definition. The committee had outlined eighteen possible video standards that were being considered with the possibility that more than one standard could be used simultaneously. The number of possible standards has been reduced as the marketplace has been operating and deciding on what is best. Converter boxes either added on or built into a receiver or monitor could handle the different standards making the differences transparent for the end user or viewer and would also allow analog television sets to receive digital signals that would be converted to the current NTSC format.

Below is a table showing the current video formats being considered:

Format Pixels/Lines Image Aspect Ratio Picture rate p (Progressive) i (Interlace)

Current NTSC

440x480

4x3

60i

Fields per second
Proposed

-----------------------------------------------------------------------------------------------------------------------

1920x1080 16x9 24, 30p

60i

Frames per second
Fields
1280x720 16x9 24, 30, 60p
Frames
704x480 16x9 24, 30, 60p

60i

Frames
Fields

(Note: 24, 30 and 60 frames per second and 60 fields per second can denote 23.97, 24.97 and 59.94 frame and field rates. The actual rates will be decided.)

The Format column is translated as the number of pixels or picture elements per scanned line by the number of lines per frame. Thus 1920x1080 is one thousand nine hundred twenty pixels per line for each of the one thousand eighty lines per frame. This is different than computer notation where the numbers refer to the number of pixels horizontally and vertically. A computer image noted as 800x600 would be for the number of horizontal and vertical pixels only, as there are no scan lines. The present NTSC television image aspect ratio is 4x3 meaning that the image is four units wide by three units high. The wide screen image used for high definition is sixteen units wide by nine units high (16x9). The possible frame and field rates are listed for both progressive and interlace scanning.

The choice depends on what is required of the image. For example, 1080i at twenty four frames per second is considered best for film viewing on television, but interlacing creates some very apparent and annoying visual artifacts with certain types of video images. A progressively scanned image has about fifty percent greater apparent resolution or clarity of detail than an equivalent interlaced format.

Image Resolution

The resolving capability or resolution of motion pictures using 35mm film far exceeded the original t television standards. The resolution capability of a visual image is the ability to show detail in the reproduced image. This is commonly measured using black and white lines either vertically or horizontally. The lines are paired, one black, one white and the number of lines that can be reproduced accurately is measured. Thirty-five millimeter motion picture film is capable of resolving approximately fifteen hundred line pairs or about three thousand individual vertical lines across the face of the image going from one side to the other. It can reproduce approximately eleven hundred line pairs or twenty-two hundred individual lines going from top to bottom. The resolving capability is dependent on the type of film and how it is used.

In contrast, the present NTSC television system can reproduce approximately two hundred twenty-five line pairs or four hundred fifty individual lines from side to side under laboratory conditions. The average TV set can reproduce about one hundred twenty five line pairs or two hundred fifty individual lines. Top to bottom the system can reproduce about four hundred eighty individual lines. That is the number of viewable scan lines on the picture tube. Though the system consists of a total of five hundred twenty-five lines per frame, only four hundred eighty of them are viewable. Once again the resolution capability depends on the quality of the receiver or monitor being used as well as the bandwidth available.

A comparison within the table will show ample reason for the desire to switch to both a digital and high definition system. The number of pixels in the video image is approximately the number of resolution lines that could be reproduced. In a digital environment, measurements are also given as the number of megapixels per second that the system can transmit and reproduce. The ability to transmit the data depends on the bandwidth available and the transmission medium used. Consequently, if the video frame rate were increased to sixty frames per second in a progressive scan, there would be a sizeable increase in the amount of data needed to be transmitted and the resultant detail information available.

Audio

Sound adds additional variables. Tthe possible number of standards increases as audio is factored in. There are several audio standards being considered. The audio standard being developed as the FCC standard in the United States ( referred to asATSC A53/AC-3) includes not only left and fight stereo channels but also a center channel, left and right surround sound channels and a Low Frequency Enhancement channel. This is referred to as 5.1 channels. There is work underway on creating a 7.1 channel system as well.

The sampling rate for audio is presently set at 48KHz. The sampling rate is the number of times per second that a sample or reading is taken of the signal and a digital value assigned and recorded. Thus, audio sampled at 48KHz is sampled forty eight thousand times a second. The sampling rate for audio is set at a number a little over twice the highest frequency that humans can hear. Video sampling was set at approximately 14 million times a second.

The frame rate for audio need not be the same as video frame rate. For example, if the rate for video is set at twenty-four frames per second, the audio could be recorded at thirty frames per second. There would be no synchronizing problems as long as everyone conformed to the 24/30 frame rate standard. The audio will synch if it is reproduced at the same rate it was recorded. For example multi-camera film television shows are shot at twenty-four frames per second on film while the audio is recorded at 30 frames per second on tape. The difference in rates does not matter as long as both are reproduced at their original rate. The requirements for audio are different than video in terms of what is needed for quality recording and reproduction. Audio requires considerably less bandwidth for transmission and reproduction, and therefore makes fewer demands on a system.

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