VLSI Device Fundamentals

Abstract

That electronics technology will soon enter a new era, that of Very Large Scale Integration, is clear. In this era the “VLSIC’s” will be integrated systems rather than integrated circuits. Although this differentiation may seem trivial, examination of the economics and the capabilities, and the limitations, of VLSI chips results in the conclusion that profound changes in fabrication, design, and product definition must occur before VLSI chips can become pervasive. Fig. 1 shows the well-known integrated circuit complexity increase with time, known by some as “Moore’s Law” [1,2]. The historical data in this figure, which gives the envelope of integrated circuit complexity growth, shows a doubling of complexity every year up to the present time, with a projected decrease of the rate of growth to a doubling every two years. This decrease in rate of growth is projected as being due to the practical elimination of wasted, or non-functional, space on the silicon chip at or about the 1980 time [1], with no further progress to be made in this area. The projection seems somewhat conservative. More important than the decrease in the complexity growth rate envelope due to this effect, are the possible limitations on the available product spectrum due to a crisis in product definition coupled with the exponentially growing cost of design. The question is just this: will the VLSI chip set consist only of ever-larger memories, and longer-word-length microprocessors, with other functional blocks at the LSI level or below? The answer to this question, and in fact the driving force for VLSI, is implicit in the future systems requirements which will drive the semiconductor industry.