Technicalities | Stretching the limits
The computing industry observed the 50th anniversaries of two seminal events this month: the invention of the integrated circuit and a failed foray into supercomputing by IBM.
In September 1958, Jack Kilby of Texas Instruments unveiled the first integrated circuit, using germanium for conduction. About six months later, Robert Noyce of Fairchild Semiconductor showed off his design, developed separately. The integrated circuit replaced vacuum tubes and led to the development of microprocessors.
Kilby later invented the handheld calculator and thermal printer and won the Nobel Prize in Physics for the integrated circuit in 2000, 10 years after Noyce's death. Noyce, whose circuit design was capable of mass-production, started Intel.
While they were going small with integrated circuits, IBM was swinging for the fences with Stretch, an early supercomputer intended to be 100 times faster than any computer at the time. It turned out to be only 40 or 50 times faster. It was soon branded a failure and abandoned.
However, Stretch included innovations that are still in use today, including multiprogramming, protection against unauthorized memory access and the eight-bit byte as a standard unit of data. Its legacy proves the value of visionary failures.
Kevin McCaney is the executive editor of GCN. Follow him on Twitter: @KevinMcCaney.