NOVA Artists in the Lab

TELEVISION APPEARANCE

TITLE: NOVA Episode - Artists in the Lab
PRODUCTION: WGBH - Public Television Broadcast
FIRST AIR DATE: November 15, 1981

TRANSCRIPT

         What fascinates me most about computer graphics is that, while I've always been interested in the visual arts, with a background in mathematics and engineering until recently I haven't been able to express those interests. Computer animation gives me the most direct route from conceptualization to visualization and lets me bypass the use of my hands and traditional artist's materials.
         I'd like to show you some of the techniques I've used in a recent production for the new opening for the NOVA show. One of the first jobs we had was to produce the image of the galaxy which you see up here. The first step in that process was to find an image of the galaxy. We used this one here. The next step in that process is to put that image in a form that the computer can understand.
         The first stage of this process is to convert the photograph into an analogue signal. To do this, we use a standard television camera that scans the photograph from left to right, top to bottom. It converts the information into an electronic signal which is analogous to the brightness of each point that it scans. Hence, the term analogue. This signal is sent downstairs to a special computer that measures the signal ten million times a second. These measurements are then converted into numbers, or digits, which are stored in the computer's memory. Once the photograph is in this form, the computer can understand it and manipulate it.
         This board is full of microchips. Each chip contains sixteen thousand switches. It is the process of turning these switches on and off that all computers have in common. Each switch represents a single bit, or digit, and hence the process is referred to as a digital process. This particular device is a frame buffer. It's a device that makes a computer graphics facility different from other computer facilities. At one end, the frame buffer looks to the computer as a standard computer memory. But at the other end, it looks to a video monitor or a tape recorder as a video signal. It allows us to translate the mathematical representation of the picture into a visual one.
         What we're looking at here are individual pixels blown way up. When we go to videotape, they're actually much smaller, as there are a quarter of a million on the screen at any given time. The representation we have of the galaxy now is two-dimensional -- that is, flat -- just like the photograph we started out with. The effect we wanted in the opening, though, was a three-dimensional trip through the galaxy. And to do that, we had to describe to the computer the position, size, and brightness of 40,000 stars.
         To accomplish that, I wrote a program that looked ar each and every pixel in the photograph. If it saw a bright pixel, that indicated that the photograph was bright, and there were a lot of stars in the original galaxy. So for that section of the galaxy, we made a lot of stars; we made them bright; and, we spread them apart. If the pixel was dim, we made fewer stars, made them dimmer and clumped them closer together. Once we have the three-dimensional description, we can ask the computer to draw the galaxy from any position in space. And, if we change that position smoothly, from frame to frame, we've created a piece of animation.

For more information on Mr. Geshwind's animation, click here.

David M. Geshwind - DIGITAL MEDIA GROUP, LTD.
111 Fourth Avenue - New York, NY 10003
212/388-0122 (Voice)