The start of CMOS technology came in about 1970,” recalls the former CEO of Toshiba Semiconductor, Tsuyoshi Kawanishi, in his book Chip Management, “for Toshiba, which was late in introducing the previous generation NMOS technology, trying to get a head start in CMOS was crucial.”
It was Kawanishi who had gone to see Sharp a few years earlier to ask them what they wanted in the way of ICs. Sharp said they wanted CMOS chips for their portable products which required low power, and Kawanishi went back to Toshiba and started the development.
“We owe a lot to the engineering groups who worked on perfecting the new technology,” continues Kawanishi, “but we cannot forget the many challenges which confronted those of us in production:
1 Yields were only a few per cent.
2 We had only one ion implanter at the time and it was constantly breaking down, and we had to wait to get spare parts from the US.
3. The managers used to come on the line at 11pm to try to solve problems.”
Nonetheless, Kawanishi reckons it was worse for senior management who “had a lot of headaches because we would lose money whenever deliveries were late,” recalls Kawanishi, “but top management stood up against the stress pretty well, and those of us in production ploughed ahead.”
It was a good thing they did. “Toshiba made sufficient progress in CMOS technology that we were able successfully to use it for the 1Megabit DRAM,” writes Kawanishi modestly.
He is being modest because he does not mention that Toshiba got a six months lead on the rest of the world on the 1Mbit DRAM and made windfall profits which sustained it through many subsequent years