Digital control of accelerators - The first ten years

While both accelerators and digital computers have existed for more than forty years, it was not until the mid seventies that they were successfully united to form the computerized accelerators that are standard today. When the first mini-computers became available circa 1965 they were immediately employed at accelerator sites - for experiments. The physicists were very quick to put them to use in spite of their deficiencies: assembler programming, no operating system to speak of, and limited reliability. Controls people were sure that they did not need them and agreed that knobs and dials were the ''natural'' interface between men and machines anyhow. Also, a potentiometer setting was far cheaper and more reliable than a value in core storage plus its associated ADC. Nevertheless, computers started creeping into control systems, took over the tedious tasks such as setting magnet currents and logging them, but always on a two-ported basis via an added connector guarded by that little switch that enabled good old manual control. Of course, computer breakdowns were entered into logbooks besides traditional obstacles such as water leaks or power supply failures. Today we know that computers wouldn't become reliable until people decided to rely on them