Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Formerly About Applied Materials -- Ignore unavailable to you. Want to Upgrade?

To: Proud_Infidel who wrote (48223)	6/20/2001 8:34:59 AM
From: Proud_Infidel	Read Replies (1) \| Respond to of 70976

Downturn Accelerates Move Towards 0.13micron Technology United Microelectronics Corp (UMC) in Taiwan is one of the largest IC fabs in the world. As part of a growing trend for Taiwanese companies to set up fabs in other countries see also "Taiwan Fabs Expand Horizons", UMC has recently established a new 300mm fab in Japan. Like all other semiconductor-related companies, UMC has been affected by the slowdown in IC demand. Peter Chang, president and CEO of UMC since 2000, talks to NEA about the impact of the slowdown on UMC. He also discusses the company's preparations for the economic turnaround, and the increasing tendency for customers to move towards lower (0.18micron, 0.13micron) process geometries. NEA: How has the slowdown in IC demand affected your operations? What product segments and geometries have seen the most downturn in demand? What has remained strong? Chang: Of course the slowdown in demand has affected our operations. Everybody has been affected. Our current orders are about only 70% of our potential production capacity. I expect the second quarter of this year to be worse. For most of last year, our orders were running over 100% of our production capacity. I look for a recovery in the electronics industry to occur in about the second quarter of 2002. The move towards the lower process geometries is still relatively healthy and will only become more so. 300mm wafer production will also become more important. It took us about nine months to get our 300mm fab in Japan to be really productive. It was a struggle for those first nine months but it is now doing quite well. NEA: TSMC recently announced a US$1.6 billion decrease in their capital expenditure budget for 2001, from US$3.8 to US$2.2 billion. Do you have similar plans? If so, about how much, in terms of dollars and percentage of your original capital expenditure budget? Chang: We decreased our capital expenditure budget for this year from US$2.9 billion to US$1.4 billion. That's a reduction of US$1.5 billion, or about 50% of the original capital expenditure budget. Our original capital expenditure budget was based on full capacity. Right now there might be too much 8-inch capacity in the world. It's the end of the cycle for that wafer size. NEA: During the Asian economic crisis, Taiwanese foundries aggressively increased their capital equipment procurement, while the Japanese and Koreans cut back. This was one of the things that helped propel Taiwan into the first tier of IC fabrication. Why cut back this time? Chang: Because the Internet bubble has, for now, burst. There has been a slowdown in telecom. We had expected a slowdown in the PC industry. It's now a mature industry and that's just part of the business cycle. The slowdown in telecom was a surprise to us, however. Only the Japanese have been quick adapters to cell phone access to the Internet. There are a few reasons for the slow growth in the telecom sector. In Japan, most people travel to and from work via train. So rather than read a book on the train, they can use a cell phone as a portal to the Internet. In the United States, most people drive a car. That is not a mode of transport currently suited for using a cell phone to access the Internet. In Europe, there are high fees for the carrier licenses. 2.5G cell phones have also been slow to come to the market. All these add up to a slowdown in the worldwide acceptance of cell phone Internet use. Also the slowdown in Internet growth has affected sales of backbone equipment. A tremendous market for servers and the necessary backbone high-bandwidth telecom equipment needed to support a large amount of wireless Internet was expected. That trend has stalled somewhat. NEA: How do you balance the short-term slowdown in demand and resultant decrease in revenue and profits with the need to plan and be ready for the future? Chang: Financially we are doing just fine. Last year we had record profits of about US$1.2 billion. We currently have NT$50 billion in cash in the bank. When the turnaround comes, we will be ready to move quickly. NEA: How has the downturn affected your customers' move towards 0.18micron production? Has it affected your plans to move to 0.13micron? Chang: Actually the downturn has made the move to 0.18micron and below accelerate. Some of our customers are going straight to 0.13micron. And next year, 0.13micron will be a big revenue provider for us. Our quick move to 300mm wafer production will also help us. We already have one 300mm fab in operation in Japan. By the third quarter of 2002 we will have another 300mm fab in Singapore. The combination of 0.13micron and 300mm wafers will really help drive the market. NEA: What's the current status of your use of CU++? Do you have a low dielectric now in production? What's its k dielectric constant value and what process are you using? If you don't have the technology yet, when will you have it? Chang: We are now in production using CU++ with a 2.5k value dielectric process. I can't tell you more since we are using this technology in partnership with IBM. CU++ use will become more important for wafer fabrication. At 0.13micron, CU++ is very important. At the smaller line widths CU++ is essential in order to maintain frequency speed. NEA: When do you anticipate a recovery? Which product areas will rebound first? Chang: The telecom market will become very important. In about two years, I expect a great surge in telecom popularity. In two years, the 2.5G market will really take off. Also, the telecom industry is moving to the idea of selling products that fit your particular needs. Some services and features are just not compatible at a product design level. Most people don't need a very fast CPU so that's not a big issue. The trade-off between screen size and keypad size is one example of where a trade-off must be made. If you use your cell phone for playing video games or video conferencing then you want a larger screen. If you are using it primarily to send typed e-mails, then you need a larger keypad. PDAs might become a major product. I use mine all the time. The screen is large enough for comfortable viewing of the Internet, and you can get an attachable keyboard that makes it a small but useful total package. I find that notebooks, even though they are lighter now, are just too heavy to be constantly carrying them around. Maybe I'm not in as good as shape as I used to be! The number of PCs will increase over the coming years but at a lower revenue per unit. The growth in the United States and Europe will be flat. The growth will be in countries such as China where an inexpensive PC makes the product attractive and open to the potential for very large sales. by David Baldwin, Taipei (June 2001 Issue, Nikkei Electronics Asia)