First-Hand:Raster Scanned Display

Submitted by A. Michael Noll

September 18, 2025

Dr. Peter B. Denes at Bell Telephone Laboratories, Incorporated (Bell Labs) in Murray Hill, New Jersey, had a Honeywell DDP-224 computer for use as a dedicated laboratory computer, mostly for interactive speech research. It originally was a Digital Equipment Corporation DDP-124 machine, and later it was upgraded to the DDP-224 machine. The computer had a cathode ray tube (CRT) display for vector graphics output. Denes suggested that we make a scanned display system for the interactive computer system, and I was given this task.

With vector graphics, the electron beam was moved from position to position, but this took time and not that much information could be displayed without flicker, which was particularly a problem when displaying text. The solution was to use a raster scan approach, like used in television. However, the information in the computer was mostly X-Y coordinates, and it would need to be converted to the data needed to turn the electron beam on-and-off for a scanned raster display. The scan data would be stored temporarily in digital memory in an output display array – what today is know as a frame buffer.

Others had done scan display earlier, for example at Brookhaven National Laboratory, which performed the scan conversion with hardware. The system we designed at Bell Labs in the late 1960’s performed the scan conversion using software. The initial scanned display system was compatible with the lower resolution of the Picturephone standard of about 252 scan lines per frame. Later the system was upgraded to work at the US TV standard of 525 scan lines per frame.

The initial system was just black and white display, as was the Picturephone. Later color was displayed using a color monitor, with 3 or more bits used to specify the color. Color had been initially contemplated, but I questioned the need for color. Broadcast television was black and white, although color might make it more entertaining. As an engineer, I was taught to avoid color in graphs since it would not copy well – instead I used dashes and other ways to indicate separate graphs of data.

The patent application filed by Bell Telephone Laboratories (Bell Labs) in 1970, listing me as the inventor, disclosed the use of software. The use of software created controversy with the Patent Office, since software was considered a mathematical algorithm and not patentable back then. After years of appeals by both sides, the case was to appear before the US Supreme Court, but AT&T ordered Bell Labs to abandon it, since AT&T was going to argue its case about Bell System antitrust and did not want to overwhelm the Court with too many Bell cases. I recall when the Bell Labs patent attorney told me (almost in tears) that he had been ordered by AT&T to abandon the case. I guess he looked forward to arguing a case before the Supreme Court. Today. Software is patentable.

A number of creative people made the scanned system a reality. Martha J. Southern wrote software for generating scan characters; John J. Dubnowski designed the hardware; D. Jack MacLean worked on the color displays; and Peter Denes championed the entire project. Later in the early 1970s, resident visitor Laure Spiegel used the color scan system for computer art and wrote an early paint software package for the color scan system. Peter became a good friend in the later years of his life, but earlier was criticized for operating a computer center, although we now know that some very innovative graphics work was done with his interactive laboratory computer.

In the 1960s, the development and promotion of the Picturephone was a huge project at Bell Labs and the Bell System. The use of a Picturephone as a scanned display terminal for text and graphical computer information was innovative, and important. However, the Picturephone was premature by decades and failed. The home computers that ultimately came used raster scan displays, and scan displays with frame buffers were used for animation in Hollywood.

The graphics device used at Bell Telephone Laboratories, Inc. in the 1960s with the mainframe computer was the Stromberg Carlson SC-4020 microfilm plotter. My guess is that Dr. Bela Julesz used it to make his random-dot stereogram images, possibly doing some sort of software scan conversion, or even using an actual raster scan device with a camera. But this was not an interactive display, like the DDP machine. Some of this early history is not that well preserved and documented.