The Charactron (sometimes branded a Typotron), a special-purpose electron tube, was a peculiar device with a brief life, but appears to have been an important fulcrum of sorts in the history of computer graphics and art. This Note rather rambles; there's simply too much untold story in here, so I've settled on nailing down some facts and connections.
In the 1950's, before Moore's Law made silicon dominate everything, all sorts of seemingly-odd technologies were pressed into use as information displays for computer and radar systems. My favorites are ones based upon lots of wacko physics, especially those using multiple technologies. The most commonly exploited trickery was the well-developed physics of electron beams: cathode-ray tubes (TV tubes) and good old electron tubes.
The Charactron (Stromberg-Carlson, General Dynamics) or Typotron (Hughes Aircraft Company) is a great example; it is a cathode-ray tube that directly forms alphanumeric characters onto a phosphor screen, either for direct viewing (radar) or to be recorded onto film.
A brief summary of its functioning follows:
(In fact all this bending of electron beams necessitated much complexity; there are all sorts of convergence and collimation controls to control image distortion.)
Provided with many hundreds of pounds of electronic support, the Charactron can form a line or small rectangle of text, at Very High Speeds; approx. 20,000 to 25,000 characters per second. This was a Big Deal back when computers executed that many instructions per second (viz. today's millions/second.
The charactron was used mainly for high-speed printing, by outputting to sensitized paper, or film such as microfilm recorders. (One of my sources(1) has one sentence accompanying one image showing a Charactron-annotated radar image; but how this composite was assembled is left a mystery.) It could also be viewed directly by eye.
Though I have no technical specs for any charactron model, assuming standard CRT spot size and deflection characteristics implies that it was probably much better at small (1/2") rectangles of very high resolution images than large (2") directly-viewable images; perfect for outputting to small film formats, such as archival microfilm.
...and as soon as someone comes up with nifty hardware, some hacker types come by and do perverse things with it -- jump to the Bell Laboratories, 1960's, where pioneering work is being done on computers and perception, sound and images.
Ken Knowlton developed a system called BELFLIX in 1963, a platform for producing animated images with an IBM 7094 transistorized computer and a Stromberg-Carlson 4020 microfilm recorder (allegedly the first computer-output film recorder(5), 1959). Filmmakers Stan VanDerBeek, Noll, Lillian Schwartz used BELFLIX to produce highly artistic movies [WHERE ON EARTH CAN I SEE THESE! --tomj] with this funky hardware. Almost certainly the 4020 used the charactron, since Stromberg-Carlson made the tubes and the film recorder. (I think General Dynamics bought the rights to the charactron in the early 1960's, but don't quote me on this.)
Knowlton also did much work on (static) image processing and perception; though done at Control Data Corp., the famous character-mosaic MONA LISA must have been known to Knowlton.
Knowlton later worked with Lillian Schwartz, a filmmaker/artist.
BELFLIX appears to be a system of hardware and software; other enabling (software) technology by Knowlton are TARPS, EXPLOR, and L6(2).
Only tangentially related: Some people are just terrifying whirlwinds of creativity and breadth: Harry Huskey, part of the ENIAC team who later spent time in early British computing circles, is listed in (2) as a source for EXPLOR 'in "medium sized" FORTRAN', out of UCSC; I assume because he had some involvement or interest in computer art.
This is certainly not the first use of computers or electronics in art, or computers to film; see More on early computers and art can be found at