Not Quite Ready to Roll Vitesse’s Lou Tomasetta, struggling to find revenues wherever he can, puts off release of his latest next-generation product. By CARLOS MARTINEZ Staff Reporter Lou Tomasetta thought he was on to something as Vitesse Semiconductor Corp. prepared to roll out its 10-gigabit network processor later this year. Granted, networking equipment makers like Alcatel, Lucent Technologies Inc., IBM Corp., Nortel Networks Inc. and Cisco Systems Inc. make up 90 percent of its sales, so the Camarillo-based company Tomasetta founded and runs was hit as hard as any by the tech downturn last year. In the first quarter of this year, the company posted a $44.4 million loss on revenue of $42 million, compared to an $11.2 million loss on $121.7 million at the same time last year. Still, a new high-end product seemed like it might be what Vitesse needed to fuel a rebound. The new 10-gigabit processor would vastly improve the efficiency and speed of Ethernet networks. Ethernet networks are a series of computers, usually in the same organization and linked together, that allow users to access information from every other computer in the network. Unlike the Internet though, which connects users to servers all over the world, the Ethernet connects workplace or home computers to each other, only at much greater speeds, up to 2 gigabits per second. The speed and efficiency provided by the Ethernet network would cause customers to flock to the new processor, thought Vitesse CEO Louis R. Tomasetta. But that doesn’t appear to be the case yet. Few companies have shown interest in the new processor. In fact, at this point demand is nearly non-existent and earlier this month Tomasetta indefinitely postponed the release of Vitesse’s 10-gigabit network processor. Instead, Vitesse is focusing on developing slower, more affordable, devices for the 2.5 Gbit optical connection market, products that are “less sexy” but more likely to produce revenue, Tomasetta said. Tomasetta insists he’s been through more than one industry downturn since founding the company in 1987, and this recent market decline doesn’t faze him. An engineer who helped develop the gallium arsenide chip more efficient than chips made of silicon Tomasetta has learned to survive sharp market changes. Vitesse started out as a chipmaker, then expanded into communications components when the chip market slipped, and later moved into network processors and fiber optic devices as markets again changed. Tomasetta spoke with Business Journal reporter Carlos Martinez recently about the travails, and opportunities, of running a semiconductor company in 2002. Question: Over the last year and a half, Vitesse has been hit hard by the tech downturn as customers like Alcatel, Cisco Systems and Lucent cut orders drastically. Did you see all this coming? Answer: We felt there was no way this growth would last. We felt we could resize ourselves and plan for even a 50-percent decline in revenue and still stay profitable. But I don’t think I’ve seen a cycle where almost across the board all the communication chip companies went down 70 to 80 percent of revenue from their peak. With that kind of scenario, there’s no way you can plan for (it). Q: Once you realized what was going on, did you come up with a strategy, or did you feel compelled simply to respond to market forces? A: We had a plan. We all expected a downturn, but I don’t think anybody expected to have this sharp of a downturn. When the first decline happened, you have to resize manufacturing because there is no reason to have a manufacturing facility that supports $150 million in revenue when you’re only doing $40 million. So we resized our manufacturing. Q: The tech industry’s problems appear only to have gotten worse in recent months with the collapse of Global Crossing and the telecom sector’s continuing troubles. How do you see things playing out? A: We’re starting to see recovery from the storage connection and the Ethernet backbone, so that’s what you’re starting to see from companies like Cisco. The long haul (telecom) service market, which used to be 80 percent of our business three years ago, still has a lot of overcapacity and a lot of equipment that they don’t need. We don’t see the recovery, at the earliest, until the first half of 2003. Q: Vitesse has had the 10-gigabit processor in development for quite some time. How difficult was it to pull the plug on its release? A: We felt there was a lot more opportunity to build variations of our 2 gigabit chip that would solve problems of one to two and a half gigabits per second than to go to a whole new product that would be very expensive, and where we couldn’t see where the customers would be for that product. Q: When will the market be ready for the 10-gigabit processor? A: That’s probably three or four years away. Q: Why then have other companies decided to go forward with plans to release their 10 gigabit processors when Vitesse feels it makes more sense to hold off? A: It’s a question of where they want to put their resources. The market for network processors is very fragmented. It’s not like the PC market where there is one architecture and all you’re arguing about are price and speed. There is no question that you can build a 10 gigabit processor, but can you invest 40, 50 or 60 engineers to work on a product for 18 months that can take away from other products that are not as sexy but may generate more revenue right now? Q: Vitesse has moved into fiber optics, but that area has been slow in developing too. Do you see any growth there next year? A: We’re seeing that side of the business growing again slowly because it’s replacing (devices) that were built 10 years ago, primarily to connect data centers to data centers and not long-haul (telecom) providers. Q: Heading a company in the middle of the kind of troubles your industry is having must be a test of your management skills. Have you had to reevaluate your management philosophy and, if so, what have you changed? A: We’ve been through these types of cycles before. In the early ’90s, we had a dominant business in gallium arsenide custom chips in high-performance computer chips when communications was just getting started. That business went through a major restructuring where we lost 65 percent of our revenue over a six-month period, so we always reevaluate what we’re doing. We don’t believe there is a secret formula to managing a company. You have to use your common sense a lot because in these down cycles you just don’t have accurate information. Q: How did you come to start Vitesse? A: I came here in ’77 (from MIT). I worked at the Rockwell corporate research center in Thousand Oaks. We were doing a lot of contract research for the U.S. Department of Defense. You could build products with the technology they had there, but it wasn’t part of the culture to do the research and invest money before you got revenue. So I left. We were able to raise some money and get the company started. We never raised much money in those days. Even when we went public we only raised $30 million, which wasn’t a lot. SNAPSHOT: Louis R. Tomasetta Age: 53 Title: President and CEO, Vitesse Semiconductor Corp. Education: B.S., M.S. and Ph.D. in electrical engineering from the Massachusetts Institute of Technology Most admired persons: Vitesse Chairman Pierre R. Lamond; Bob Paluck, co-founder of Convex Computer Corp.; T.J. Rodgers, founder and CEO of Cypress Semiconductor Corp. Personal: Married, two children
