Computer animation gives to film making the advantages of computing, i.e. only one cycle of a repetitive action need be programed; one program generates a whole family of films; and successful programs build into a library for subsequent use. Moreover, the scientist can now communicate directly through the film medium in his natural language of mathematics. This suggests the possibility of Moreoverimaginative educational films. It is recognised that computer animation has certain limitations. For example, cartooning by computer would probably be unprofitable. Four computer-made films which illustrate these points are available on loan.
COMPUTER ANIMATION is a new tool of exciting possibilities. If nothing else, it helps solve a current critical problem - how to absorb the flood of numbers that a high-speed digital computer can produce. By translating the numbers into pictures, and then viewing the pictures in sequence as a motion picture, one can often enormously increase the efficiency of communication from computers tO human beings.
The use of computer animation in this way is important. But an equally important aspect of computer animation is its use as an educational tool, or, in other words, as a way for people to communicate to people. In science and technology especially, in the author's opinion, computer animation has great potential, both for scientists communicating with other scientists and for scientists communicating with the nonspecialist.
First of all, how does a computer make motion pictures? The process is illustrated in Figs. 1 and 2; which are themselves computer-made. First, one writes a program that computes the picture to be drawn. This is fed into a digital computer, usually by means of punched cards as shown in Fig. 1. The computer translates the program into a series of commands for the electron beam in a cathode-ray tube and the film-advance mechanism in a camera. These commands are read onto magnetic tape.
Next the magnetic tape is read into a device consisting of a cathode-ray tube to display the computed picture, which is recorded on the film of the camera. When the picture is complete, the film advance command on the tape goes to the camera, causing the film to advance to the next frame (the camera shutter is always open). The next picture on the tape is then traced out on the face of the cathode-ray tube and recorded on film, the film is advanced, and so on, frame after frame until the movie is complete.
The form of the computer program varies depending on the particular computer and the computer language that is used. However, all programs have certain features in common - the features that make computers such powerful tools. Namely, the program for any systematically repeating action need be written only once as part of a loop, regardless of the number of repetions; the program can be written in terms of symbols whose specific numerical values are assigned at running time, so that a single program produces not a single film but a whole family of films; and successful programs can be incorporatcd as a subprograms into other films, to accumulate into a library of very flexible building blocks that can be used over and over again.
However, another feature of computer animation is possibly of still greater importance, especially with regard to science and technology. Standard scientific computing languages, such as Fortran, allow the animation to be coucned at the outset in the universal language of the physical sciences, i.e. mathematics. As a result the scientist can produce animation with a minimum of dependence on other specialities. His position vis-a-vis film production is now akin to his position vis-a-vis book production. To produce a film, he needs only pencil and paper to write a program, just as to produce a book he needs only pencil and paper to write a manuscript. Although in both cases technicians and machinery must process his original document, their work from the author's viewpoint is routine. Psychologically, he feels that he is master of the final output - if it is a one-man product bearing his stamp. He need not at the outset enter into partnership and share the creative responsibility with representatives of unfamiliar technologies.
Furthermore, computer animation of the sort the scientist typically wants to do is cheap. For example, the total cost of a minute of computer animation for line drawing animation may be as little as $50. This means that ideas can be easily tested, or, to borrow some fashionable jargon, computer animation allows of easy feedback.
This direct access to the film medium, plus easy testing, may be the most significant feature of computer animation. In fact, the author believes that, because of this feature, the scientist's contribution to animated educational films will soon be appreciably more creative and imaginative than what we have generally seen heretofore.
Paradoxically, this feature - the possibility for the scientist to be his own filmmaker - should make professional filmmakers very happy. If our experience is typical, the scientist soon discovers that he is a verv amateurish filmmaker. The virtuoso pianist's greatest admirers are amateur pianists; likewise, given a chance, via the computer, to be a filmmaker, the scientist gets a new appreciation of the filmmaker's art and a new awareness of its requirements Thus, the analogy between computer animation and book writing is superficial. With computer animation, the partnership between scientist and filmmaker is still needed, but the partnership s strengthened rather than weakened.
It is difficult to illustrate a film technique by verbal descriptions or by still photographs. It is advisable for the interested reader, to see for himself by viewing some computer-made films. At present, four such films are available on loan from: Technical Information Library, Bell Telephone Laboratories Inc, Murray Hill, N.J. 07971. Brief descriptions of these films follow:
This is an example of the type mentioned above where animation helps people absorb the vast amount of data a computer can produce. The author wanted to see the results of a computer study of satellite motions rather than read the results as pages and pages of numbers, and to that end programed the computer to feed the numbers it would normally print out into a sub-program which computed a perspective picture of the satellite, represented by a box. This movie was the outcome.
The scenes in the film are from two different vantage points. One vantage point is in a reference frame fixed with respect to the stars. The viewer sees the satellite orbiting a rotating earth. Figure 3 is the superposition of every fifth frame taken from the first orbit of one sequence (the earth's rotation has been stopped). The upper-right figure is an orbital clock whose minute hand goes around once an orbit and whose hour hand counts orbits.
The second vantage point is in an orbiting reference frame. The viewer is traveling in orbit directly behind the satellite ; he sees only the satellite. Figure 4 is an illustrative frame.
The fact that the same program generated both types of scenes shows the flexibility of computing. To change from the scenes of Fig. 3 to those of Fig. 4, only a few data cards were needed to omit the earth and to change the vantage point.
This is an educational film illustrating Newton's law of motion in two dimensions. Orbits are shown not only for masses which interact according to the inverse square law found in nature but for the inverse cube, direct cube and other interactions as well. The latter are very difficult to create experimentally, but easy to simulate by digital computer. Figures 5 and 6 are frames from this film.
This film combines two paradoxes found recently by psycholo-gists. One, found in 1958 by the British psychologists Penrose and Penrose, is a staircase that carries one ever upward and at the same time stays at the same level. The second, found in 1964 by R. Shepard of the, Bell Telephone Laboratories, is a tone which goes ever upward and at the same time stays in the middle of the scale. This film iilustrates the possibility of using a photographic slide in conjunction with a cathode-ray tube to provide a fixed background. It is also the first pure computer film. Both the sound and animation are done by computer.
This film illustrates BEFLIX, a technique for computer animation invented by K. C. Knowlton. The first three films mentioned above were programed using Fortran, a standard scientific programming language. Coordinates of the end points of straight-line segments were computed and the segments drawn. A drawback of this technique is the need for a certain amount of mathematical background in order to do or to supervise the programming.
In contrast, BEFLIX imitates the scanning technique of television. Because Knowlton did not have available a variable beam intensity on the cathode-ray tube, he used different aplabetical characters to create different shades of gray. For example, an area of B's gives a dark shade, an area of commas a light shade.
Knowlton has also coupled the scanning technique to a special language which is quite different from Fortran. In his language one dirextly varies the position and blackness of rectangular areas of the screen by means of instructions such as PAINT (paint the area a specified shad of gray), ZOOM (zoom in on the picture), DISOLV (dissolve), etc.
Because every point of the screen is computed, motion pictures made in BEFLIX are generally more expensive than line0drawn films programmed in Fortran. On the other hand, BEFLIX has the advantage that little or no mathematics is required for programming. It makes computer animation avalilable to those who have no scientific or engineering training. Figures 1 and 2 are frames from Knowlton's movie.
Computer animation is a powerful new tool because it (1) helps the computer communicate results; (2) allows the scientist or engineer to communicate directly and cheaply, with the film medium and (3) uses features that have made computing itself powerful. Furthermore, with a special-purpose language like BEFLIX one can do computer animation with no special computing knowledge.
Like anything else, computers have limitations. We believe that progress will be best made by exploiting the advantages od computer animation, not by trying to bend it into uses for which it is not suited. For example, it is unlikely that it would be profitable to try to do cartooning as we now know it by computer. The pictorial processing involved in cartooning has evolved along certain lines over the years. We suspect that it would bevery inefficient to imitate this processing or to duplicate its final product by computer. On the other hand, animators turned loose on computer animation, especially since we have BEFLIX, would find a wealth of new and interesting effects, which would supplement and augment conventional cartooning rather than replace it. This combination might result in a new form of cartooning of wider range and scope than what is now available.
1. K. C. Knowlton, A computer techniques for producing animated movies, Amer. Fed of Information Processing Societies, Conf. Proceedings, Vol 25, pp57-87, 1964.
2. E. E. Zajac, Computer-made perspective movies as a scientific and communications tool, Communications of the ACM, Vol 7, pp169-170, Mar 1964.