MCMs: niche market or mainstream product?

Abstract

Are multichip modules (MCMs) a niche packaging solution or a mainstream market? The answer may be both. There is a market for high density substrates for high-performance applications. Products containing MCMs include high-end computers and military electronics. This is the high value market and accounts for a large share of the MCM market in dollar value, but a smaller shave of the market by unit volumes (thousands per year). The other segment is for multichip packages (MCPs), usually considered lower in cost. These packages can be found in cellular telephones, modems, PC cards, and some automotive applications. The volume in this market segment is high (tens of thousands per month) but the dollar value per module is low. This paper examines the worldwide developments in MCMs and MCPs for applications including computers and telecommunications (with an emphasis on portable products), consumer electronics, and automotive applications. Also included is an assessment of the infrastructure for MCMs and MCPs. Outside of cost, the most often cited issue for halting the implementation of MCMs is the availability of known good die (KGD). From mainframe and workstation computer makers in Japan, to mid-range computer makers in Europe and a variety of electronics companies in the United States, many complain that KGD is an issue. Is this a major problem? If so, what developments are underway to remedy this problem? A discussion of KGD as a potential barrier to increased MCM adoption is included. The paper concludes with an outlook on the future of MCMs.