Conclusions
It seems clear that our industry is departing from at least some of Gordon Moore's earlier quantitative predictions. One of these is illustrated in Figure 6. Gordon's 1974 tongue-in-cheek but genuine extrapolation of wafer sizes suggested that by the year 2000, we would have 57-inch diameter wafers! Clearly, this is off by about an order of magnitude. Yet simple extrapolations of Gordon's trends does lead to qualitatively correct predictions.
![Figure 7: Extrapolated Year 1999 wafer size[1]](../images/art1fig7.gif)
Figure 7: "Extrapolated" Year 1999 wafer size[1]
Regardless, two trends seem inescapable: everything in the production of semiconductor devices is moving toward more expensive factories, and there is swiftly expanding use of information and knowledge to reduce costs, improve delivery, and improve quality. These two trends need to be linked to try to alleviate the effects of the former by using the latter. At the same time, one must also recognize the emergence of other forces: the need for cleaner, safer, and less energy-consuming manufacturing enterprises, the evolution and indeed revolution of materials and materials' processing, and the change from local politics and culture to global politics and culture. All these trends will result in a significantly greater emphasis being put on manufacturing as a competitive weapon in the 21st century.
Page 6 of 8