October 30, 2001 6:20 PM PST
How Intel plans to survive the downturn
The apparent paradox--declining industry revenues during a period of increasing customer use--was the central theme in Intel's biannual financial Webcast on Tuesday. Barrett and other executives said that all of the markets in which the company participates struggled this year.
"For those of us who've been in the business for 30 years, this is the biggest slowdown that our industry has ever seen. This is twice the slowdown that we saw in 1995," Barrett said. "The Internet is just starting its early stages of buildout...The massive volume of traffic is just beginning."
To capitalize on that rebound, Intel is tweaking its overall strategy to better target growth areas. The company, unlike in recent years, will now almost exclusively focus on its core chip markets.
Intel will place increasing emphasis on emerging PC markets, such as China and India. This strategy will require a renewed focus on costs, the company said.
"It requires us to be very scrupulous about costs," said Paul Otellini, executive vice president and general manager of the Intel Architecture Group. "In Latin America, parts of Eastern Europe and Russia, we are seeing growth in excess of 100 percent."
Intel will also increase contacts with dealers and regional manufacturers--the main conduits for technology in these regions, Otellini said.
In established markets, the company will work on "influencing the upgrade cycle," Otellini said, by working with software developers to come up with more glamorous applications that, in turn, require faster processors to run effectively.
Server strategy will follow a different approach. Intel plans to come out with a host of Xeon and Itanium processors in 2002 for servers that will compete against products from Sun Microsystems.
In addition, Intel will devise "solution blueprints" with partners that will take some of the pain out of installing complex back-end systems. One blueprint, for example, combines Compaq Computer hardware, software from a number of companies, and consulting how-to from Ernst & Young.
"We're creating the shrink-wrap equivalent for the e-business conversion," Otellini said. "We are absolutely, deadly focused on increasing our market share."
In cell phones and mobile devices, Intel will work with software developers and carriers to better incorporate its chips with upcoming 2.5G and 3G services. The more advanced wireless networks will help the company, as consumers will need to upgrade their current handsets to accommodate the technology, according to Ron Smith, general manager of Intel's Wireless Communications and Computing Group.
"The installed base is aging, and data will be coming that will force an upgrade cycle. The DoCoMo iMode service has 26 million subscribers, and Japan is not an isolated test case," Smith said.
Intel also plans to push through an integrated cell phone chip that will combine a digital signal processor, a microprocessor and flash memory, he added.
Company executives detailed a number of product plans during the call:
For desktops, the Pentium 4 processor will hit 3GHz in 2002. In addition, a chipset with integrated graphics for low-priced Pentium 4 computers will debut in the second quarter.
By mid-2002, the first Celerons using the NetBurst architecture, the underlying architecture of the Pentium 4, will be released. Celerons now are based on the older Pentium III architecture.
Prestonia, a dual-processor Xeon chip for servers based on the Pentium 4 architecture, comes out in the first quarter. Gallatin, a multiprocessor Xeon for servers, will be unveiled in the fourth quarter.
Commercial systems containing McKinley, the next version of Itanium, will come out mid-2002. Pilot production on McKinley will begin this quarter.
In the first quarter of 2002, consumers can expect the first Pentium 4 for mobile computers. Meanwhile, in 2003, Intel will release Banias, a processor specifically designed for notebooks. The chip will perform 25 percent better than other Intel chips at that time, Otellini said, and will provide 25 percent better battery life.
By the end of 2002, six factories will be manufacturing chips on the more advanced 0.13-micron manufacturing process, and two of these factories will process 300-millimeter-diameter wafers. Shifting to the 0.13-micron process will boost speeds and lower costs, while 300-millimeter wafers will lower costs.