The Man Behind the Microchip Read online

  There was nothing graceful or subtle in the price-cutting approach; it was, instead, an attempt to develop the market and reduce costs by means of brute force. This brazen play for market share leapfrogged Fairchild to the top position in number of circuits sold each year. At the same time, the entire market for integrated circuits took off. In early 1964, industry-wide integrated circuit sales for 1966 were projected to reach $58 million, or 8 million units at an average cost of $7.25 a unit. Within a year of the Fairchild-triggered price drop, estimates had been upped more than 150 percent, to $157 million annually—remarkable growth, given that the average price per unit had dropped.35

  LESS THAN A YEAR after the dramatic price cuts, the market had so expanded that Fairchild received a single order (for half-a-million circuits) that was equivalent to 20 percent of the entire industry’s output of circuits for the previous year. One year later, in 1966, computer manufacturer Burroughs placed an order with Fairchild for 20 million integrated circuits. By this time, the circuits had begun pulling in radio stations on stereo receivers and amplifying sound in state-of-the-art hearing aids.36

  The integrated circuit had also come to play a key role in the burgeoning commercial market for computers. In 1964, IBM introduced the revolutionary “System/360 family” of six different-sized computers. To research, develop, build, and market the System/360 series, IBM spent $5 billion, more than twice what it cost the United States government to develop the atomic bomb. The System/360 machines, designed to share software and peripherals such as printers and tape drives, were meant to be “the only computers anyone would need”—and this standardized system ushered in the rapid-fire proliferation of computers in business and government offices around the world. “Electronic Girl Fridays” (as one contemporary characterized computers) calculated payrolls, figured mortgage payments for loan companies, kept track of inventory and utility uses, and processed billing records. By 1967, some 95 percent of all banks in the United States would use computers to handle their checking accounts.37

  Many of these new machines were “minicomputers,” smaller machines that used integrated circuit technology to offer mainframe-competitive computing power at a fraction the size and cost of mainframes. The most popular of the minicomputers was an offering from DEC called a PDP-8, which was the size of several refrigerators and cost about $18,000. In 1966, there were 3,600 minicomputers in use worldwide, all of them dependent, to no small degree, on integrated circuit technology.38

  As the demand for integrated circuits grew, the lawyers for Fairchild and Texas Instruments continued their fight in the patent courts. In 1964, the patent office interference board split its decision on who owned the patent to the integrated circuit, awarding 4 of 5 claims to Texas Instruments’ inventor Jack Kilby. Both sides appealed the ruling, but they also recognized that it meant each company needed a license from the other in order to manufacture integrated circuits. In the summer of 1966, Noyce, Hodgson, and Fairchild counsel Borovoy met with Texas Instruments president Mark Shepherd and counsel Matt Mims. The group agreed that each company would grant the other licenses. They further decided that any other company that wanted to build integrated circuits would need to negotiate separate licenses from Texas Instruments and Fairchild. With these agreements, each side acknowledged the other’s claim to some part of the invention of the integrated circuit.

  In October, Noyce and Kilby shared the prestigious Ballantine Medal of the Franklin Institute, presented to them “for their significant and essential contributions to the development of integrated circuits.” The citation noted, “The full extent of the [integrated circuit] revolution is not yet in sight: for a dream, the dream of doing things with electronic motion alone, has come true, and the coming to fruition of dreams is a rare event and full of implications difficult to comprehend completely.”39

  Many at Fairchild objected that Kilby’s device was not a monolithic integrated circuit at all. In Texas, meanwhile, there was grumbling about Noyce having done nothing more than adapt Kilby’s idea for industrial use. A few people who had helped bring the integrated circuit to life at Fairchild wondered how it had happened that once again, Noyce seemed to get all the credit from their company. Why was Noyce’s interconnection patent any more important than Hoerni’s planar process or Last’s isolation work or Isy Haas’s patient attempts to build the ideal transistor for use in integrated circuits? Absent any of these, the integrated circuit would never have emerged from the Fairchild lab. Moreover, Noyce had almost nothing to do with building the device, since he moved to the general manager’s position even before the patent was filed. His designation as the lone inventor of the integrated circuit from Fairchild struck several of his early laboratory co-workers as arbitrary. Noyce himself freely admitted that his insight “was a question of having these rather vague concepts of insulators, of isolation, of interconnection … so that you drew on your bag of tricks to combine these elements to make the integrated circuit. There was no huge lightbulb flashing.”40

  On the most simplistic level, Texas Instruments, not Noyce or Fairchild, was responsible for Noyce, and not someone else at Fairchild, being named co-inventor of the integrated circuit with Kilby. It was Texas Instruments, after all, that claimed that Noyce’s patent for interconnecting devices infringed upon Jack Kilby’s application for a patent that covered putting more than one device on a single substrate. Once this interference was filed, the Fairchild claim to any part of a potentially enormously valuable product rested on Noyce and his January 1959, lab notebook entry. Noyce did not lobby to be the Fairchild representative in the integrated circuit debacle with Texas Instruments. The opposition drafted Noyce, and then his own organization devoted huge amounts of time and money to ensuring he prevailed.

  Moreover, if nearly any invention is examined closely enough, it almost immediately becomes apparent that the innovation was not the product of a single mind, even if it is attributed to one. Invention is best understood as a team effort, with the person ultimately called “inventor” occupying much the same space as the pitcher who has just had a perfect game. The outfielders might have caught a dozen fly balls, the first baseman might have nearly broken his neck to step on the bag an instant before the runner, the catcher might have called for pitches perfectly calibrated to each batter’s weakness, but the record books note only that the pitcher threw a perfect game.

  So it had been at Fairchild all those years before. The lab team played beautifully, and Noyce never hesitated to admit that his ideas about integrated circuits relied heavily on ideas that were “in the air” in 1958 and 1959. Without Hoerni, without Moore, without Kurt Lehovec at Sprague, Noyce never would have imagined the integrated circuit in the way he did. Without Last or the development team that decamped to start Signetics, Noyce’s ideas would never have become marketable products. People all around the country were experimenting with integrated circuits in 1958 and 1959, but no one could offer a practical way to build the devices. That is what the ideas Noyce scribbled in his patent notebook in January 1959, provided. In the same way that the great beauty of Noyce’s integrated circuit was its simplified interconnections between existing components, so too did the elegance of his insight lie in its interconnection of ideas already in play. His notebook entry was but one piece of the puzzle, but it was an essential piece, the one that comes at the critical time and slides easily into place—the one that reveals how all the other parts of the picture fit together.

  BY THE MIDDLE OF THE 1960s, Fairchild was one of the fastest-growing companies in the United States. In the first ten months of 1965, Camera and Instrument’s share price ballooned 447 percent, shooting from 27 to 144, with a 50-point growth in the month of October alone. This was the fastest rise of any stock listed on the New York Stock Exchange at the time. Sales and profits hit another record high. IBM bought the rights to the planar process in a lucrative cross-license deal so important to Noyce that on the first day of the negotiation he blocked his calendar with a giant “IBM�
�� that he wrote, circled, and underlined in black, felt-tip pen. (Nearly every other entry is in pencil.) By year’s end, only established industry giants Texas Instruments and Motorola manufactured more semiconductor devices per year than did Fairchild. Fairchild facilities ran throughout the night. Salesmen were encouraged not to move to their territories so their families would not distract them from the business of selling.41

  For Noyce, the energy and growth were incredibly seductive. Piloting Fairchild through its acceleration, Noyce told a friend, was a bit like riding a fast horse—that same combination of exhilaration and fear and teetering on the edge of losing control but never quite doing so. He was a success because the company was a success, and it had triumphed because he had triumphed.42

  AS THE SECOND HALF of the 1960s opened, however, shadows began to form within Fairchild Semiconductor. A two-tiered market for semiconductors had emerged: at one end were mature products available at very low prices, manufactured in very high quantities, and with very low margins. These devices were often used in the entertainment market—as parts for radios, for example—and sold for less than a nickel apiece. On the other end were new devices, technically superior to the old, which were manufactured in smaller runs and sold at higher profits. These devices were used in military applications and also in the burgeoning computer market. Most firms in the young industry, including Semiconductor, served both ends of the market. As a result, the companies needed to function simultaneously as sophisticated research organizations and as mass manufacturer. In 1965, for example, Fairchild sold its integrated circuits of a sort called complementary-transistor logic (CTL) in quantities of a half-million units or more.43

  At the same time that Fairchild was shipping hundreds of thousands of devices each week, it maintained essentially the same organizational structure it had developed as a tiny startup operation. All manufacturing reported to a manufacturing manager, all engineering to an engineering manager, all design to a design manager. As the gap between development and manufacturing revealed, the different divisions of the company did not readily coordinate operations. There were no product managers to ensure that a product moved efficiently from conception through shipping. The only person with responsibility for more than one division was Noyce himself, and his managerial style did not lend itself to the sort of detailed coordination the company needed.44

  Adding to the troubles was Noyce’s promotion to vice president of Camera and Instrument, an assignment that extended his authority to a new instrumentation division, headquartered in Clifton, New Jersey, which sold semiconductor test equipment. Noyce tapped Sporck to become Semiconductor’s new general manager and asked marketing head Tom Bay to run the instrumentation division. With this move, the two men, whose skills had complemented each other’s and Noyce’s so well, were separated, leaving a vacuum within day-to-day management at Semiconductor.

  Between the monthly managers’ meetings in Syosset and the requirements of the instrumentation division, Noyce began to split his time in half between the Fairchild Semiconductor offices and Camera and Instrument headquarters in New York. Betty Noyce declared this the perfect reason to move back east. The state of California had begun building a new freeway near their house. Already several fields of mustard and apricot had been plowed under, and the construction noise was drowning out the sound of a brook Betty used to hear from the yard. She dreaded the prospect of seeing and hearing traffic every minute of her life. Bob offered to enclose their property behind a wall.45

  The freeway, in any case, was simply one in a string of justifications Betty offered for leaving the state. Everything Bob liked about California—the fast pace, the constant change, the opportunity to become wealthy without a family name or pedigree, the relentless drive to perform—she hated. She preferred a life where change was incremental, where people lived in homes that had belonged to their families for generations, and where everyone knew his or her proper place. Bob refused to move, despite his promise, sworn when they first arrived in California, that they would leave if Betty wanted.

  As their tenth anniversary approached, it was apparent that Bob and Betty Noyce were fundamentally incompatible. Bob tended to like most people he met; Betty had different standards. Bob always rushed to buy any new technology, any new bit of stereo equipment or electronic gadgetry. Betty collected antiques and enjoyed the old-fashioned craft of quilting. Betty barreled her way through obstacles with sarcasm and intimidation; Bob’s approach, as described by Gordon Moore, was to “spend two hours trying to convince someone about the way something ought to be done, rather than [taking] the two minutes in which he could tell them [to do it].” Betty’s depiction of a party at Charlie Sporck’s house, which she sent to her son Bill in 1963, highlights a few of the differences between the couple: “Da [Bob] went swimming, as did most of the men but very few of the ladies, and did some nice diving and joined in a game of water polo. After he got out and got dressed again, three of the men threw him back in the pool, considering it a very funny joke. They in turn were thrown in by others, which made me want very much to come home, but Da put his clothes in the dryer and was good as new in half an hour, and appeared not to be angry. He’s a very good sport!”46

  Bob and Betty did still occasionally enjoy each other’s company. They tiled a wall of their kitchen in a pattern they designed together. They took the children camping in the Sierras and made matching jackets for the entire family. They tried to top each other’s Tom Swifties. (Betty: “‘My birthday is today,’ Tom said presently.” Bob: “‘We’re all out of pineapple juice,’ Tom said dolefully. Betty: “‘This is our very best ground meat,’ Tom said off-handedly.”) Every Christmas, they invited all the children they knew to join in an informal orchestra that Bob conducted at a party in their home. When Bob broke a leg skiing and a friend sent him a collection of single socks to wear, he and Betty spent hours sewing sock puppets and naming them. The puppet with a laurel-leaf crown was Sockrates. The tie-dyed sock with string hair was Sockadelic. They made Socko and Vanzetti, Sam the Sockeye Salmon, Sockajawea, and a “sockreligious” puppet from a black stocking to which Betty attached a white clerical collar. To add to the amusement, Betty labeled the socks and had the children hide them in the neighbor’s house.47

  Such fun was too rare. Even when he was not at work or traveling for his job, Bob spent as much time as he could away from home. Every Wednesday, the 12-voice madrigal group that he directed met for rehearsals. He helped Charlie Sporck build his barbeque grill. He took Billy to help tune the carillon at Stanford and to fly model airplanes in a field. He scheduled weekend meetings.

  He pushed Betty to buy a ski cottage at Lake Tahoe. “A thousand times he picked us up from a snowbank, dusted us off, and led us up the mountain, the four of us in our matching orange jackets,” recalled his daughter Penny. “He did this until we could share his exhilaration.” At the end of a day of skiing, he would have the children wait for him while he took one last run, at top speed, just for himself. Betty, meanwhile, would spend the day in the cabin. She liked the mountains more than she liked skiing, and she always felt like an “innkeeper” at Tahoe, biding her time until the family returned and she fed them hot cocoa and dinner.48

  Bob assembled collections of Fairchild equipment to send to Grant Gale for use in his classes. He further extended his commitment to Grinnell College in 1963, when at the age of 35, he joined the board of trustees. The trusteeship took him to his alma mater for a few days every four months. His fellow trustees fondly called their youngest member “Boy Noyce” when they thought he could not hear them.

  Even when he was at home, Noyce sequestered himself in the basement. Here he built an organ, a harpsichord, many model airplanes, a microwave oven. The basement was his refuge, an underground lab for a scientist who no longer did much science, a place where, as his son put it, Noyce could “work on something that doesn’t fight back.” Betty suspected he was avoiding her.49

  He was. Betty Noyce’s standards of
behavior were so high, and her methods of reinforcement so verbally combative—she yelled at and even taunted the children—that Bob worried she was damaging the children’s psyches. The children’s memories of their “Da,” by contrast, are generally tender—his coming to check on them at night, the glow of his cigarette scarcely visible in the dark; his placing their cold mittened hands under his shirt and jacket as they rode the ski lift. He encouraged, and occasionally re-directed, their sense of adventure. When 13-year-old Penny wanted to skydive, for example, he introduced her to a man swathed head to toe in bandages. “This skydiver had a bit of trouble with his parachute,” Noyce explained. Then he offered to pay for hang-gliding lessons. “It worked,” Penny said later. “I was diverted”—but still soaring through the sky. To be sure, Noyce tried to imbue the children with his own competitive nature. He never understood why they did not want to race for the swim team at the country club or to show their horses competitively. He made elaborate offers related to grades and money, proposing for example, “I’ll pay you 40 cents for each A, an A–is neutral, and you pay me $1 for every B”—and then encouraged the children to counter with offers geared to their anticipated report cards.50

  “He wanted children who were empowered,” recalled one of them. “She wanted children who behaved properly.”51