In 1983, Siemens’ top management announced that developing and producing megabit memory chips would now be a strategic company goal – with the aim of closing the gap to the market-leading Japanese producers by 1989. The company went to work in cooperation with Dutch and Japanese partners, and by the end of 1987 the first 1-megabit DRAM chips went into series production in Regensburg. The following summer, Siemens was one of the world’s first manufacturers to release customer samples of a 4-megabit DRAM device. The company had joined the ranks of the world’s leading semiconductor manufacturers.
Japan and the USA in the lead – Siemens aims to catch up
Early in the 1970s, the transition from analog to digital technology had accelerated vastly in every branch of electrical engineering. As digitalization advanced, more and more memory capacity was needed to process programs and data. Both supply and demand exploded – chip capacity quadrupled every three years, as new, more powerful, yet also cheaper chip generations reached the burgeoning market. By 1984, chip technology had advanced to being able to make the first megabit chips – chips with more than a million binary memory cells, enough capacity to store around 64 pages of text.
In those days, the European microelectronics industry lagged more than two years behind its Japanese and American rivals. But all that changed with an extremely ambitious venture named the MEGA Project. Joining forces together and girded with government support, Siemens and Philips aimed to catch up with the world leaders within five years. Siemens’ goal: to produce 4-megabit dynamic random access memory (DRAM) chips, able to store four million bits of data.
To achieve its goal, Siemens adopted a paradigm change. For the first time, it left the path of serial planning – research, development, production – and planned a factory to make products that had not even gone into development yet. This parallel networked planning was essential to keep pace with the extremely dynamic market, rife with both price wars and enormous risk. For even as research expenditures soared, prices for each chip generation dropped by 90 percent within just two years.
Even short delays could cause immense financial losses. Which was also why Helmut Lohr, head of Standard Elektrik Lorenz (SEL) AG, warned companies against getting into the fierce price war between American and Japanese makers; instead, he advised buying memory chips as cheaply as possible on the world market.
Microelectronics as a field for the future – And a door-opener for processor technology
Siemens saw things differently. The company had already transformed from an energy to an electronics company back in the 1970s, under President and CEO Bernhard Plettner. His successor, Karlheinz Kaske, vigorously advanced the company’s drive into microelectronics from 1980 onward. This approach was based on the strategic theory that only a company with the best chips could market the best higher-level products.
On top of that, the company viewed memory chip development simply as a door-opener to the development of far more interesting processor technology, which the company expected to retain and develop as an in-house core competence. In that spirit, the Managing Board unanimously approved the MEGA Project on February 6, 1984. Key elements of the project included concentrating development work at two development halls for one- and four-megabit chips, to be built at the Munich-Perlach research center; building up production capacity in Regensburg; and hiring over 100 additional engineers. All in all, the company planned to invest DM 1.4 billion within five years.
Competition toughens – But Siemens reaches goal ahead of schedule
The project celebrated its first milestone in 1987 – working with Toshiba, Siemens was the first company in the western world to start series production of 1-megabit chips. The new semiconductors made at the Regensburg plant sold extremely well. The cooperation was not free from controversy; after all, the Europeans had originally wanted to advance development on their own. But the year before, prices in the microelectronics market had plunged, making the market even tougher. The race had sped up.
After a wide range of difficulties, the company finally reached its goal in the summer of 1988 – ahead of its own schedule – making laboratory samples of the 4-megabit chip. The Siemens Mitteilungen employee magazine proudly reported: "In December, some 400 employees of the MEGA Project gathered for Bavarian snacks at the Munich-Perlach location. […] They were celebrating the successful completion of the MEGA Project and thus the company’s accession to the ranks of global competitors in microelectronic technology." Series production, though, was able to start only at the end of 1989, a year behind the Japanese competition.
Research, development and production under one roof – A role model for further company development
All the same, before the year was out, Hans Günter Danielmeyer, the company’s Chief Research Officer at the time, was calling the MEGA Project a net gain. It marked the start of an entirely new type of cooperation between Siemens research, development and production. Driven by the power of the market and ever-shorter production cycles, collaboration between employees would become increasingly vital. Technical expertise would no longer be enough. Only close integration would allow the company to react quickly to fast-changing customer needs and, at the same time, to cut costs. Thus, quite apart from its technological advances, the MEGA Project offered a role model for Siemens’ further development.
Dr. Franz Hebestreit