Viewpoint

Economic Necessity Demands New Design, Fab Portability, and Manufacturing Strategies Changing times require more flexibility and new technology. by Michael Reinhard

Over the past four years, we have all witnessed a spectacular growth period. All the customer segments have been growing: computer, communications, consumer, and industrial electronics. For many semiconductor companies, keeping up with demand was the major customer-related challenge. Now, however, the environment has changed. Growth in the semiconductor industry is slowing. This year, the book-to-bill ratio dropped below 1, and opinions are mixed as to whether the rest of 1996 will see a continued contraction of the ratio, or if billings will recover with modest growth. Regardless, the "rules of the game" have changed in our business. A different set of economic forces is now at work. Tomorrow's winners will recognize these changes and respond early.

Today, the pendulum is swinging from a state of scarce fab capacity to one of over supply as previous investments in new fabs come to fruition and start production. Simultaneously, semiconductor demand is cooling. Economic laws dictate that increased supply or reduced demand will cause prices to drop. This combination is a lethal attack on a semiconductor supplier's profitability.

With the recent drop in the book-to-bill ratio, foundries have cut prices to maintain capacity. Just as the airlines practice "yield management," discounting ticket prices to maximize revenue per plane, foundries strive to optimize revenue per wafer. When this scenario occurs, semiconductor companies must become even more efficient.

For example, consider two semiconductor companies, A and B, each of which manufactures competing products. If Company A optimizes its IC layouts, then it boosts its manufacturing yields. This translates into lower costs. Company A can pass these savings to its customer by selling at a lower price or use them to boost its profitability. Company B then has a real competitive problem. With many fabless companies' success dependent on only one or two products, being Company B could be a true disaster. Wall Street's trashing of many semiconductor companies' stocks shows the unpleasantness of being in this situation.

However, Company B can chose to improve by becoming more flexible in their ability to optimize layouts for specific manufacturing processes. Why is focusing on IC layout the solution? Layout databases are the manufacturing blue prints for ICs, and their quality is the most significant determinant of manufacturing yield that design and production groups can control.

This is true whether Company B has a fab or not. Company B has to improve yield at its own fab or at the fabs of its suppliers. If Company B is a fabless company, then it has the additional advantage of being able to shop around for the best prices.

Today, there are new layout optimization and conversion technologies available to optimize layouts. These tools automatically convert layouts from one process to another--for example, 0.5-µm to 0.35-µm--and optimize the layouts for yield and electrical performance. Tomorrow's winners will chose the most flexible of these tools.

Business slow-downs provide good motivation for rethinking and changing strategies. When good times return and demand increases, new methodologies and processes that optimize manufacturing become very important for remaining competitive. And for the semiconductor industry, the focus returns to portability and finding additional sources of supply.

Michael Reinhardt is the CEO of Rubicad Corp. (San Jose, CA). Rubicad provides LACE, a layout conversion environment.

To voice an opinion on this or any Integrated System Design article, please e-mail your message to michael@isdmag.com.

integrated system design  November 1996