Robert Noyce: Microchip

By Frederic Golden

From the day he first handled one in college in 1948, Robert Noyce knew the new gadget meant the end of balky, bulky vacuum tubes. But he also realized you couldn't do much with transistors until you could link them together, like fibers in an Oriental rug. To everyone's astonishment, the gifted young man from Grinnell, Iowa--a minister's son--achieved that goal in a decade. His integrated circuit, or microchip, not only helped rename an orchard-filled California valley but also led to a seemingly endless harvest of silicon devices, from PCs to coffeemakers.

Recruited by Shockley himself, Noyce joined the new Shockley Semiconductor Laboratory a few years after getting his Ph.D. from M.I.T. in 1953. But Noyce quickly despaired of Shockley's imperious style, and in 1957 he and seven colleagues--"the traitorous eight," Shockley groused--formed Fairchild Semiconductor.

At Fairchild, Noyce used a new chemical etching method not only to print transistors on silicon wafers but also to lay down tracks between them. Besides eliminating expensive wiring, the new integrated circuits operated much faster. Six months earlier Texas Instruments' Jack Kilby had produced a similar chip, but it was made of germanium, required external wires and was tougher to manufacture. Noyce's chip won the ensuing patent race, but the two friendly rivals were content to regard themselves as co-inventors.

In 1968 Noyce joined with fellow Shockley "traitor" Gordon Moore to found Intel. Under Noyce's shirt-sleeves leadership, it soon produced a landmark memory chip and the so-called computer-on-a-chip, or microprocessor. By 1974 Intel was so successful that Noyce traded day-to-day management for industrywide concerns, like leading a consortium called Sematech to stave off foreign competition. He died in 1990 at 62.

--By Frederic Golden