
In the summer of 1981 I was asked to consider the possibility of manufacturing a 600,000 transistor microprocessor in 1985. It was clear that the technology would only be capable of manufacturing 100,000-200,000 transistor chips with acceptable yields. The control store ROM occupied...