For seven years, a small but devoted band of chief executives, investors, and technophiles has gathered in the Indian-summer shadows of this Olympic ski resort to divine the future of the communications industry. Their latest assessment? The market recovery looks real, but a generational change in chip design looms.
"If we don't have any 9/11-type disasters, the economy is poised for a very significant recovery," said Steve Forbes, president and chief executive of Forbes Inc., which organized the Telecosm conference held here last week.
The group of some 150 free-marketeers--buoyed by the recent tax cut, generally lower interest rates, and improved liquidity of investment capital--gave the former presidential candidate a standing ovation after his keynote talk. Their enthusiasm, however, was tempered by the memory of better days.
"Three years ago, this conference drove stock prices. People would leave sessions to go out and buy stocks on their cell phones in the lobby. The mood here today is the opposite of that," said one of the participating investors.
Cellular and 802.11 advocates both pointed to signs of growth and agreed their technologies generally will complement, not compete, with one another.
Paul Jacobs, executive vice president of Qualcomm Inc., San Diego, said Korea's SK Telecom has already deployed its CDMA 1x EV-DO technology to 10% of its subscribers, offering data of hundreds of Kbits per second. Japan's KDDI will deploy the technology by the end of the year, Jacobs added. In the United States, carriers Sprint and Verizon also use the technology.
"The most aggressive people are saying 25% of carrier revenue will be from data in 2005," he said.
Meanwhile, an estimated 5,000 802.11 public hot spots are up and running in the United States, according to Sky Dayton, chief executive of Boingo Wireless, a Santa Monica, Calif., start-up trying to establish the underpinnings of an 802.11 roaming service. That number must double in key areas such as airports, hotels, and conference centers to get Wi-Fi to critical mass, he said.
"We're working with OEMs to create a hot-spot-in-a-box that will have access points and everything you need for well under $200."
Geoff Barrall, chief technology officer for BlueArc Corp., San Jose, said he is seeing signs of an uptick in corporate IT spending. To date, the start-up has sold more than 230 of its high-end file servers for storage-area networks, he said.
Others pointed to recent quarterly reports from Intel Corp. and Network Appliance that suggested those companies are experiencing a significant pickup in spending on laptop computers and enterprise networks, respectively. And one Analog Devices Inc. technical manager said his company has logged several quarters of single-digit growth, in part driven by power management components.
Optical networks don't appear to be participating in the recovery, according to one Telecosm presenter, who noted that current trans-Atlantic optical cable use amounts to a fraction of one provider's lit capacity.
Conference host George Gilder said too much government regulation and too little support for deploying broadband is holding back the U.S. market. "South Korea has 20 times the bandwidth per capita of the U.S. and five times the bandwidth to homes and businesses," he said.
As many as 73% of Korean homes have broadband links, compared with about 20% in the United States, Gilder said. Nevertheless, "as broadband is rolled out in the U.S., it will occasion another nonlinear surge in traffic that will take most experts by surprise and generate another spurt of panic equipment buying."
As the economy slowly rebounds from a three-year downturn, the technical community faces a generational shift.
The 30-year reign of the microprocessor and the treadmill of ever-finer CMOS process advances is coming to an end, said some industry pundits at the conference. In its place, a new generation of programmable logic devices and "good enough" CMOS processes will serve a widening circle of applications.
"It's not going to be just PLDs and FPGAs," said Carver Mead, a pioneer of VLSI design and computer science professor at the California Institute of Technology. "There will be as many types of programmable logic devices as there are styles of microprocessors today."
A new crop of companies will define reconfigurable interconnects, processing elements, and a host of other options, Mead said.
A former IBM PLD researcher, Dan Coffey, said his start-up, NetHarmonix Inc., Burlington, Vt., is already working with Sprint on a kind of software architecture that is re-programmable on the fly.
"It's a very fertile ground for research," Mead said. "The trick is finding the demanding apps that will pay the bills and get the paradigm going."
Today's work in programmable logic represents a new wave in thinking after years of brute-force advances in CMOS and microprocessors, he added.
"The picture we drew in 1971 [of a 150nm transistor] is pretty much the picture of what's being built today," Mead said in closing remarks at the conference. "There's been a lot of problem-solving but no new ideas. Not only have we not had new ideas in the structure of how we build semiconductors, we haven't seen new ideas in the architecture of devices either."
BlueArc, whose file server uses a whopping 13 high-end Altera FPGAs, served at the conference as a poster child for future system architectures that will rely on programmable logic.
"I've seen a number of start-ups moving to this technology because they couldn't do it with ASICs or microprocessors," Barrall said.
"I think we're in for a sea change that's as important as the microprocessor was," said analyst Nick Tredennick, who has written newsletter articles on the trend and is a consultant for QuickSilver Technology, a chip start-up pioneering the field.
90nm costs loom large
The shift comes as many companies are reporting diminishing returns for the latest CMOS processes, Tredennick said, claiming that foundries are seeing a slowing rate of adoption for new technologies such as 0.13-micron processes. Tredennick said designers will increasingly turn to "good enough" trailing-edge processes, adding that 0.25-micron technology offers the best bang for the buck in 2003.
Indeed, Jordan Plofsky, senior vice president of Altera Corp., San Jose, said ASIC technology is hitting a wall, and provided a detailed analysis of the prohibitive costs of designing a 90nm chip. Such a design will cost $30 million and require targeting a diminishing number of multibillion-dollar systems markets, he said.
"Only a few major microprocessors will migrate to 90nm, and that will leave a lot of applications looking for a host," Plofsky said. "For example, Nvidia's latest PC graphics chip reportedly cost $100 million to design. They're at the hairy edge of what makes sense going forward."