Since resolution means different things with regard to digital and analog video, the most basic thing you need to understand is what we mean by analog or digital video. You may have read or been told at some point that digital has something to do with electronics. In fact, nothing could be further from the truth. In reality, digital refers to something that can only have a limited number of states. For example, when I was growing up we had analog FM tuners. When we were trying to tune in a radio station at a frequency of 94.9 we might actually end up tuning to 94.83452 or 95.00002 because there was essentially an infinite variance. A modern tuner, on the other hand, will normally be digital. That means if you tune upward from 94.7 you'll automatically get 94.9. If you tune downward you'll get 94.5.

Digital Logic

To be more specific, digital video specifically refers to using digital logic. Digital logic involves problem solving using a special type of math known as boolean algebra, which changes all problems into math to represent them as 1s and 0s. Video is represented by a series of points in a grid, each of which is described by a binary (Base2) number of a fixed length. The dimensions of the grid are also fixed, and are what most people think of when they hear the word resolution.

Display Resolution

The width and height of the grid of pixels (points) used for each frame (full picture) of video are called display resolution. This is somewhat misleading, since there's no guarantee that all dots will be visible when the video is displayed. Display resolution is really the amount of detail that's stored (digitally) for each frame. This is purely a digital term, and while the resolution used impacts the detail displayed on your analog television, and is also a determining factor in how much detail can be captured from an analog source.

Visual Resolution

While only digital video has a display resolution, all video has a display resolution when played back. Display resolution is the amount of detail that can be distinguished by the human eye. How it's measured is dependent on the display format.

Line Pairs

The most basic type of analog source is film. Since this guide isn't designed to get into the level of detail someone who works with film transfers needs, I won't go into a great deal of detail. What is useful to understand is the concept of Line Pairs. In order to determine the maximum detail can be represented on film a series of black and white lines is used. The lines are divided into pairs - one black followed by one white - because the edges where each pair meets are where the details are. The visual resolution of film is measured in Line Pairs per millimeter (LP/mm), meaning the maximum number of Line Pairs that can be individually distinguished from a single millimeter of a frame.

Analog Video

Unlike film, in analog video (ie video in analog television format) there are discrete lines. This means that a television is made up of a specific number of (horizontal) lines that can be counted. The number of lines depends on whether the video is destined for a PAL or NTSC television, with 483 lines for NTSC or 576 for PAL. These are known as active lines, because a television signal also includes many more lines that aren't visible. Each active line is measured in the amount of time it takes to draw it from beginning to end, known as the active period.

Analog Waves

While digital pictures are made up of points, all analog signals are essentially waves. Waves have a period (how much time it takes to get from the beginning to the end) and frequency (how many waves there are over a period of time). Frequency is measured in Hz (Hertz) and 1Hz is equivalent to 1 complete wave per second. Period is the inverse of frequency, meaning a 100Hz wave has a period of .01s (.01 seconds).

TV Lines

In the case of television, each line is a series of waves, each of which can represent a pair of lines - one white and one black. This means two TV Lines are equivalent to one Line Pair. The number of TV Lines displayed on an analog television for a given signal is dependent primarily on the frequency of the analog source, which could theoretically be as high as 13.5MHz, although when considering digital sources you're generally limited by the limits of the television. You may find the number of TV Lines listed in your television's owner's manual. Remember to cut that number in half to determine the actual horizontal visual resolution.

Kell Factor

One affect of using discrete lines for video is that details may either fall between the lines or across multiple lines. Since each point (on a TV Line) can only represent a single color, details smaller than a line may also disappear completely regardless of position. Consider what would happen if you looked through horizontal blinds at a ladder leaning up against a wall. Depending on the dimensions of the ladder and how far away it was you might be able to see every rung, no rungs, or any amount in between. Scan lines have the same issue. This is represented by a number called the Kell Factor. The Kell Factor for analog television (on a CRT display) is assumed to be 0.7, meaning you can find the vertical visual resolution by multiplying the number of active lines by that number. The Kell factor is a combination of the underlying system (scan lines) and the limitations of display equipment. It actually varies above and below 0.7, but that's the generally accepted average for modern televisions. This may also be different for newer display technologies.


when analog video is converted to digital, the process used is sampling. Sampling is the process of selecting a regular interval of time to determine the value of an analog wave. The samples are then saved, and can be used to reconstruct the waves when it must be returned to analog for display. In this case, the analog waves being sampled are Line Pairs (pairs of TV Lines). The basic requirements for sampling are laid out in the Nyquist-Shannon Sampling Theorem, including what's often referred to as the Nyquist frequency. The Nyquist frequency is twice the frequency of the wave being sampled. This is because at least two samples are required to accurately determine the shape of a wave. Unfortunately the actual technology used for digitizing video isn't accurate enough to perfectly duplicate the original wave, so a better target frequency for sampling would be as many as you can get.

ITU-R BT.601

The International Telecommunications Union has established the guidelines used as a starting point for sampling analog video. This standard, part of a document called ITU-R BT.601, specifies a samplerate of 13.5MHz for both PAL and NTSC. This document also specifies a display resolution of 720 pixels in width for both systems. As we'll discover later this doesn't necessarily mean what you think it means. The major difference is in the number of vertical pixels. PAL's 576 active lines translate to 576 vertical pixels in a digital frame. However, since NTSC's 483 active lines doesn't divide evenly into any multiple of 2 (this is very important in digital compression) it's reduced to 480 vertical pixels. Therefore PAL's full resolution is 720x576 and NTSC's is 720x480.

Table of Contents

  1. 1. Introduction
  2. 2. Definitions
  3. 3. Resolution of Common Formats
  4. 4. Aspect Ratio
Written by: Rich Fiscus