Dictionary of Computers - animation, computer

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Computer-generated graphics that appear to move across the screen. Traditional animation involves a great deal of drudgery in creating the 24 frames per second needed to deceive the human eye into seeing a moving picture on film. In computer-generated animation, while humans still create the key frames that specify the starting and ending points of a particular sequence – a character running through a landscape, for example – computers are faster and more accurate at calculating the in-between positions and generating the frames.

First achievements
The first completely computer-generated character to appear in a major motion picture was the sea-water creature in James Cameron's film The Abyss (1990), developed at the leading special effects shop Industrial Light & Magic. It was quickly followed by the liquid-metal man in Cameron's Terminator 2 (1991). The first entirely computer-animated full-length feature film was Pixar's Toy Story (1995), which was the first film ever to achieve independent motion of characters and backgrounds in the same sequence.

Algorithms and initial image creation
The basis of computer animation is algorithms developed by academic researchers. These are used to develop software routines that handle the complex calculations needed to work out the precise colour of each pixel in each of the finished frames; the process demands exceptionally powerful hardware with massive storage capacity. For the animator, an image begins as an on-screen collection of lines that look much like a wire frame. There are a variety of techniques the animator can use to develop 3-D objects – they can be extruded from a cross section, or ‘swept’, which is the on-screen equivalent of turning a cross-section on a lathe to produce an evenly curved surface. Less symmetrical objects may be defined by a series of Bézier curves.

Adding solidity and colour
The object then has to be rendered, which essentially means making it into an image of a solid object. To do this, the computer needs four types of information. First, the object has to be located in space. Second, it has to be assigned a colour, specified either by levels of red, blue, and green or by levels of hue, saturation, and brightness. Third, the location and focal point of the camera photographing the object have to be specified – these determine how the object appears on screen, in perspective. Fourth, the location and type of light sources must be specified: colour, brightness, and, in the case of spotlighting, the size of the cone-shaped pool of illumination. Ray-tracing, meanwhile, calculates how the light directed at the object reaches it, with what intensity, and in what areas. From all this information, the computer can calculate the colour intensity of each pixel making up the object.

Light reflection
There is another element, too: how the object itself reflects light. Two techniques model this, each named after its creator. If the object's surface can be described as a mosaic of polygons, Gouraud shading works by measuring the colour and brightness at the vertices of the polygons and mixing these to get values for the areas inside them. Phong shading extends this by taking into account the angle of reflection; it is therefore a more accurate technique for creating specular highlights. Gouraud, because it is simpler, is faster, and there is specialized hardware available for it; Phong has to be implemented in software. Both methods produce an object that looks as though it is made of soft, smooth plastic – the smoothness comes from anti-aliasing, a process that removes the jagged edges or stepped effect which mars the edges of diagonal lines on a computer display screen.

Mapping for realism
Mappings are what make the objects look as though they are made of real-world materials. There are four main types of mapping: texture, environment, bump, and transparency. Texture is the actual texture of the material the object is made of: brick, water, wood, and so on. The system essentially wraps the object in the texture the animator chooses. Environment mapping adds the reflections on the object's surface of its surroundings; a shiny, round, metal object rolling down a hill, for example, must show accurate reflections of the trees and other objects it rolls past. Bump mapping takes into account the shape of the object itself and the way this affects reflections and shadings in its surface colour. Transparency mapping defines what can be seen through the object, and the distortion caused by the substance of the object; this was a key element in animating the monster, made of sea water, in The Abyss.

Fog and haze
Finally, fog and haze are important elements of computer animation, particularly for backgrounds, as computerized images tend to look too flat and sharp. The introduction of a little fog hides the sharp edges and makes the scene look more realistic. This is vital for one of the largest growth areas in computer animation for film and video: simulated flyovers, which are impossible in live action and expensive and difficult in model work.

The star, which was added to the flag in 1793, initially had more than five points. The star may represent the Morning Star mentioned in the Koran. Effective date: 5 June 1936. >>