Lessons from Sematech
By all appearances, the U.S. semiconductor industry was on the ropes in the mid-1980s. After a multiyear effort to become a force in semiconductor memory chips, Japan now led the industry—both in market share and in the quality of its products. That raised fears the U.S. could lose not only a highly innovative industry but the components crucial for everything from computers to weapons systems. Yet by the early 1990s, that decline had reversed and U.S. chip makers regained the lead.
What happened? Among other things, a 1986 trade agreement that gradually reduced Japanese competition, a recession in Japan, a shift by U.S. companies to making more lucrative kinds of chips—and Sematech. Short for Semiconductor Manufacturing Technology, the consortium of 14 American chip makers such as Intel and Texas Instruments began operations in 1988 with an ambitious goal: to revitalize the U.S. semiconductor industry by finding ways to reduce manufacturing costs and product defects.
No one believes Sematech accomplished that all by itself. But before Sematech, it used to take 30 percent more research and development dollars to bring about each new generation of chip miniaturization, says G. Dan Hutcheson, CEO of market researcher VLSI Research. That increase dropped to 12.5 percent shortly after the advent of Sematech and has since fallen to the low single digits. Perhaps just as important, Sematech set a goal in the early 1990s of compressing miniaturization cycles from three years to two. The industry has done just that since the mid-1990s, speeding innovation throughout the electronics industry and, consequently, the entire economy.
Sematech has become a model for how industry and government can work together to restore manufacturing industries—or help jump-start new ones. The National Alliance for Advanced Transportation Battery Cell Manufacture, formed in 2008, was designed on the model of Sematech, for instance. So is the Department of Energy’s new SunShot Initiative, which aims to reduce the cost of solar energy by 75 percent by 2020. “That model is viable for many industries,” says Robert D. Atkinson, CEO of the Information Technology and Innovation Foundation, a Washington think tank.
Like any consortium, Sematech struggled with its members’ competing interests and with wrenching changes in its industry. And not everyone in the industry supported it. Cypress Semiconductor CEO T.J. Rodgers complained that Sematech was an “exclusive country club” of large chip makers that for too long didn’t share technologies with smaller companies. But a number of academics say it nonetheless succeeded because of the way it was conceived and managed, providing lessons for future consortia.
For one, U.S. chip makers took to heart the adage that crisis breeds opportunity. The leaders of the semiconductor industry, who had created it in the 1960s and thus had a personal stake in restoring it, knocked on countless doors in Washington to get support for Sematech. National Semiconductor CEO Charles Sporck, known as the father of Sematech, spent so much time on that effort he nearly let his company go under, says former Sematech CEO William Spencer. Eventually the group won a five-year commitment of $100 million in annual funding from the U.S. Defense Department’s Defense Advanced Research Projects Agency, matched by Sematech’s members.
Once it was under way, Sematech’s leaders worked to translate that sense of urgency into the organization’s culture. A simple imperative they often repeated was “If it’s not competitive, change it.” Sematech’s technologists would identify important goals, such as reducing circuit line widths to reduce chip size, or manufacturing challenges, such as reducing chip defects, and research the best way to solve them. Often, that involved helping and funding U.S. companies that manufactured chip-making equipment.
More than anything, several observers attribute the no-nonsense culture to Robert Noyce, coinventor of the integrated circuit and cofounder of Intel, who left retirement to head Sematech shortly after it started. His egalitarian management style initially caused a certain chaos and frustration among Sematech’s managers, but it also kept bureaucracy at bay by encouraging experimentation and innovative thinking, according to University of Texas professor Larry D. Browning, lead author of a 1995 study on Sematech.
Despite the promising start, by 1990 some critics said Sematech hadn’t yet accomplished much—partly because at first, many companies didn’t send their top talent to work the requisite two-year stints at the organization. After Noyce died unexpectedly in 1990 and longtime Xerox executive Spencer was appointed CEO, he began targeting key people at member companies to join Sematech, cajoling company CEOs into making sure he got them. “By the next wave, we had top people competing for new spots,” recalls Spencer. When those people returned to their companies, they were in a position to push the technologies researched at Sematech into the design and manufacturing process.
The presence of top technical people and the deep involvement of their technically savvy CEOs also ensured that the right technologies, not just those championed by the most powerful companies, got funded while others were abandoned quickly. For example, says Hutcheson, when it became apparent in the 1990s that a particular technology for transferring chip designs onto silicon wafers was too costly, Sematech helped develop the one that became the standard. Says Hutcheson: “What’s innovative about Sematech is that it’s essentially a crowdsourced method of deciding what should be the next technologies to pursue.”
Not least, Sematech proved itself willing to make repeated shifts in strategy as market needs changed. As the U.S. industry recovered in the mid-1990s, for instance, the group decided to stop seeking federal funding, and members stepped up their contributions to partly fill the gap. And in another wrenching shift in strategy, Sematech expanded to include some international companies as it became apparent that a U.S.-only supply chain would no longer work.
Today, Sematech is in the midst of yet another transition. Under CEO Dan Armbrust, it has created and will manage a new Photovoltaic Manufacturing Consortium funded by the industry, the Department of Energy, and the state of New York. With a mission to help develop a new kind of photovoltaic manufacturing technology—one that, ironically, could replace polysilicon in solar panels—the project again places Sematech in the role of revitalizing a key industry.
The inside story of how ChatGPT was built from the people who made it
Exclusive conversations that take us behind the scenes of a cultural phenomenon.
How Rust went from a side project to the world’s most-loved programming language
For decades, coders wrote critical systems in C and C++. Now they turn to Rust.
Design thinking was supposed to fix the world. Where did it go wrong?
An approach that promised to democratize design may have done the opposite.
Sam Altman invested $180 million into a company trying to delay death
Can anti-aging breakthroughs add 10 healthy years to the human life span? The CEO of OpenAI is paying to find out.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.