The Frame Buffer

Our CM-2 is in no sense a production machine. There is no way, for example, to get data in and out of the device in a hurry. Also, it is quite small in size (two 4K partitions), and has almost antedeluvian memory (256Kbit DRAM). Nevertheless it is almost an ideal configuration for algorithmic and code development. One reason for this is the frame buffer. This is a means of transferring data on the fly from the core of the machine to a graphics monitor with a simple subroutine call. If, for example, this call is in a loop over time, and if the image represents some field variable, then the graphics monitor will present a movie of the field as it changes from moment to moment. The frame buffer turns the CM-2 into a very interactive machine that I find invaluable for code development. For example, I have had up to twenty turn-arounds in a single session.