Into the nineties: VLSI trends and education

Abstract

Any consideration of VLSI must take into account two features of IC technology: decreasing line-widths and increasing device count. These are the twin driving forces of research and development programs. However, it is necessary also to consider a third factor: producibility. The history of microelectronics technology is full of good ideas and exciting research projects which never quite became a commercial reality. The important question is therefore not “what can be done?”, but “what will be done?”. This paper looks at some of the ways in which current technology may lead to new devices in the last decade of the twentieth century. Inevitably this will put emphasis on increasing complexity—designers seem never to be satisfied with current capability in this respect. However, a “simple” increase of chip size to achieve multi-mega device circuits seems unlikely. At the very least, it will not be simple. Future development of VLSI (or ULSI) devices seems almost certain to involve new technology, perhaps in three dimensions, or new design concepts such as fault tolerant architectures. Indeed, the ultimate VLSI — wafer scale integration — will certainly require innovation in both technology and design. A review of this kind should also not neglect the human factor. New ideas and new technology will require the development of new engineers and scientists. A revolution is taking place in our educational system to meet the challenges posed by the dramatically changing needs of our industry. This problem will also be considered and ways suggested of tackling it.