DENVER, United States-The third-generation (3G) hype is much less intense than a year ago, as executives and engineers get down to the nitty-gritty of making the new technology-and their business plans-work. It is a time of industry transition and stealth-mode research and development, along with few consistencies in the short-term projections and technical expectations for 3G handsets.
Motorola and Nokia remain confident the volume 3G market will emerge during 2002, creating handset demand. However, several major handset chip makers and industry analysts suggest volume demand for 3G services in general is not expected outside of Japan until 2003 or perhaps even beyond.
“Most operators will delay (until) sometime [in] the third quarter or fourth quarter (of 2002), so we think we’ll be comfortably there to meet their needs and have not slowed down our investment one little bit versus the original plan of operators starting in early 2002,” said John Thode, Motorola vice president and general manager of 3G products. “… We believe that clearly by the third and into the fourth quarter of 2002, you’ll start seeing very large-scale W-CDMA deployments.”
Nokia spokesman Keith Nowak reiterated Nokia’s plans to offer its first W-CDMA/GSM phones in third-quarter 2002 and to be shipping in the millions during the fourth quarter. “We foresee the volume 3G market to start in 2002,” he said.
Ericsson, however, seems more cautious. Before its joint-venture announcement with Sony, Ericsson said it would have 3G phones for the Japanese market at the end of this year and for the rest of the world in the first half of 2002. The company, however, expects take-up of volume 3G services to occur during 2003, according to Ericsson spokesman Peter Bodor.
Gartner Dataquest forecasts 1.7 million W-CDMA terminals will be sold worldwide in 2002, with 90 percent to 95 percent of those in Japan.
Leading handset chip makers are also playing it conservative. Executives from Philips Semiconductors and Texas Instruments said the 3G mass market will begin in 2003.
Tony Sica, director of marketing for Intel’s wireless communications and computing group, said he does not believe 3G will really begin to take off outside Japan until the second half of the decade.
“From all the carriers that I’ve talked to, they still have to figure out how to make money with their existing architecture,” said Sica. “… The service providers are balking a little bit, and some of them don’t have the money to invest. You don’t just move from GPRS (General Packet Radio Service) to 3G after a couple of years without a solid business model. … Unless they get money from GPRS, 3G will be dead in the water.”
Though the top-tier handset manufacturers have their own chip development teams, the influence of large semiconductor companies has been significant. For example, two-thirds of today’s 2G handsets worldwide are shipped with Texas Instruments’ digital signal processors, according to Frank Viquez, wireless analyst for U.S.-based Allied Business Intelligence (ABI). TI reports its customers include Ericsson and Nokia.
Stan Bruederle, semiconductor industry analyst for Gartner Dataquest, said the differing views of these large chip players and handset makers regarding time frames is not going to affect what is going to happen.
“I think everybody’s working very hard to have this all fall into place,” he said. “You’ve just got two different points of view as to what the timeframe’s going to be. And I guess both stand to gain if it happens sooner.”
Added to the lack of a united vision about 3G market development are the inherent difficulties in deploying a new technology like W-CDMA and the industry’s precedent for arriving late with handsets for new technologies, most recently for Wireless Application Protocol (WAP) and GPRS.
A W-CDMA network is very complicated and complex, noted Bryan Prohm, wireless analyst for Gartner Dataquest in the United States. “It’s new spectrum, it’s new technology, it’s new hardware, new software. Everything is new. … So it makes sense that there will be some slip-ups along the way.
“At this point, it’s hard to see how we won’t run into another situation of the networks being deployed and arguably ready to go with very few if any mobile phones required for operators.”
There is a push with 3G to try to make handset availability coincide with network deployment, said Oliver Weigelt, 3G marketing manager for Philips Semiconductors. “What we see today is that handset manufacturers and infrastructure suppliers … are working much closer together because they don’t want to repeat the (scenario that happened) with GPRS,” he said.
Wrapped up in this whole issue of timing is being able to deliver the technology in a package that consumers want and will pay for.
“There’s a lot of uncertainty with what requirements will be needed for 3G,” said Andy Fuertes, semiconductor industry analyst for ABI. “The operators don’t know what the customers want. The customers don’t know what they want themselves. So it’s very difficult for the operators to specify to the handset vendors … what they’re looking for … (and in turn specify) to chip makers.
“They don’t really know what the sweet spot is or the amount of processing power to run the applications. And the memory-it’s clear you need more, but how much more is the issue.”
Motorola’s Thode said his company has spent “oodles” of time and effort during the past three years on 3G consumer research to determine what customers want. He asserted Motorola now has a “pretty good handle” on what is going to sell.
Thode said that research has translated directly into determining technical needs for 3G handsets. “If you want video or multimedia messaging … or whatever the case happens to be, we know exactly one for one how that translates into what kind of capabilities and chip size and memory and everything else we need on the device,” he said.
“In terms of process power and memory, we have some good ideas, because you know Philips is also in the consumer market,” added Weigelt. “We’re doing CD players, MP-3 players, etc., so the requirements from the customer end point of view will be pretty similar to what the consumer market delivers already today. The only thing we have to do is do this with less power (requirements) and (with) less [expense].”
Chip firms also are moving to software-based approaches to prepare for undefined future needs and help speed time to market.
TI, for example, offers a DSP-based architecture, called Open Multimedia Applications Platform (OMAP), which delivers an open software environment for delivering new applications.
“Without knowing what’s going to happen three years down the road, we’ve tried to design our products based on the OMAP platform to basically have a lot of [speed and power] available to sort of cover our bases,” said Tom Pollard, TI director of marketing and business development. “Our only solution is to make sure that we have a flexible and pretty powerful multimedia platform and also third-party software support to basically let the market dictate what’s going to win without making us place too many hardware-centric bets.”
