Home' Technology Review : October 2005 Contents Reviews
PHOTOGRAPH COURTESY OF INTEL
Robert Noyce dreamed up the
microchip in a 1959 notebook entry.
BY ROGER LOWENSTEIN
T of the electronics age arguably
came in January 1959, when Robert Noyce, an engineer
and a founder of Fairchild Semiconductor, scrawled in
his notebook the words "Methods of Isolating Multiple De-
vices." Under that obscure heading, Noyce went on to write, "In
many applications now it would be desirable to make multiple
devices on a single piece of silicon in order to be able to make in-
terconnections between devices as part of the
manufacturing process, and thus re-
duce size, weight, etc., as well
as cost per active element."
Although the word for it
did not yet exist, Noyce was
describing the microchip.
A former protégé of William
Shockley, the coinventor of
the transistor, Noyce under-
stood the transformative
potential of new technol-
ogy as well as anyone alive.
His halting follow-up on his
initial idea therefore casts
light not just on the history of
computers but on the often be-
fogged pathways that lead to
scienti c advancement.
As Leslie Berlin, a visiting
scholar at Stanford University, re-
lates in her new biography, The
Man behind the Microchip: Robert
Noyce and the Invention of Silicon
Valley, "After noting his ideas in his
lab notebook, Noyce did...nothing."
Fairchild was a new company, and,
as Noyce later recalled, he was
preoccupied with selling tran-
sistors, not with inventions "that
might make you some money
somewhere down the road."
Noyce did not "invent" the chip
to create something new but to
solve an existing problem in an
The problem was that cir-
cuits consisted of numerous
discrete components (tran-
sistors, resistors, and so
forth) requiring thousands of interconnections. Electronics us-
ers con gured their own circuits by attaching these components
to each other one at a time, "a process fraught," Berlin tells us,
"with errors and failures." As the number of interconnections
rose, so did the odds of system failure. By the late 1950s, a score
of companies were looking for a solution.
Two months after Noyce's notebook entry, Texas Instruments
announced that one of its engineers, Jack Kilby, had invented a
crude integrated circuit. This may have been the spark that in-
spired Noyce to return to his notebook. In July, ve months after
Kilby, Noyce led a patent on an integrated circuit. Though Kilby
was rst, he merely placed all the components on a single slab of
germanium and wired them together the standard way---by hand.
Noyce's design was easier to mass-produce. His integrated circuit
connected components in a single cir-
cuit on a chip of silicon that was small
enough, as Berlin writes, to be "car-
ried o by an ant."
Berlin's rigorously factual account
portrays the scienti c process in all its
grittiness. Not only were the events
that led to the Fairchild integrated cir-
cuit "murky" (Noyce was inspired by
the work of one of his colleagues, Jean
Hoerni), but after the fact, the engi-
neers failed to realize what they had wrought. Some executives
within Fairchild were opposed to investing in the commercial de-
velopment of integrated circuits on the grounds that they were
prohibitively expensive and threatened transistor sales.
But Fairchild didn't quite give up. In 1961, it did launch a primi-
tive integrated circuit dubbed the Micrologic, though the $100
price tag limited demand. Finally, in 1964, Noyce made a bold de-
cision: to cut the price of the circuit below what it was costing
Fairchild's customers to buy and then solder the individual com-
ponents themselves. Once the chip became economical to pur-
chase, sales took o . Fellow Fairchild founder Gordon Moore
later said the decision to cut prices was as important as the inven-
tion itself. It established a patter n for Silicon Valley that still en-
dures. As Moore put it, "Whenever there's a problem, you lower
the price." By 1965, Noyce could see the future. He told a group of
nancial analysts to get ready for portable telephones, personal
paging systems, and palm-sized televisions.
In 1968, Noyce and Moore bolted from Fairchild and founded
Intel. There, Noyce rather sadly became a front man and eventu-
ally a gurehead. Berlin does not spare us the depiction of Noyce's
shortcomings, including the details of his troubled rst marriage.
After Intel, he became a lobbyist for the semiconductor industry---
not the nale one envisions for a legend, but in keeping with
Noyce's modest self-appraisal.
He was often asked when he would win the Nobel Prize.
"They don't give Nobel Prizes for engineering," he would say
with a smile. Noyce died in 1990. Had he lived, he undoubtedly
would have shared the stage with Kilby, who in 2000 did indeed
win a Nobel in physics for ushering in the age of computers.
Roger Lowenstein is the author of When Genius Failed and
Origins of the Crash.
The Man behind the
Microchip: Robert Noyce
and the Invention of
By Leslie Berlin
Oxford University Press, 2005, $30.00
Links Archive December 2005 January 2006 September 2005 Navigation Previous Page Next Page