Innovation Through Exaggeration

Imagine what could have happened with the much-publicized launch of Intel's wireless-enabled Centrino platform.

March 31, 2003

2 Min Read
Network Computing logo

Imagine what could have happened with the much-publicized launch of Intel's wireless-enabled Centrino platform. Intel might have:

• Jump-started the multimode wireless LAN market, providing instant legitimacy to high-speed 802.11a and 802.11g standards.

• Introduced a wireless interface that integrated Wi-Fi with emerging 2.5G cellular data standards like GPRS and CDMA 2000 1XRTT.

• Showcased a breakthrough radio design that overcomes the thirsty power requirements of Wi-Fi interfaces.

None of this happened, of course. Instead we got a cute logo and a catchy theme song. If there's a silver lining, it's that Intel is spending $300 million to market Centrino, which is sure to raise the visibility of Wi-Fi and lend credibility to the hotspot market.The Centrino platform consists of a new processor (Pentium M), a new system chipset (Intel 855) and a mini-PCI WLAN NIC (Pro/Wireless 2100 Network Connection). The processor-chipset combo is touted as significantly more power-efficient. Intel cites a battery life of more than five hours for a 1.6-GHz Pentium M system, compared with 235 minutes for a 2.4-GHz Mobile Pentium IV, when measured by BAPCo's MobileMark 2002 benchmark. So the step-down in clock speed buys you an additional hour and 10 minutes of battery life.

When it comes to the wireless interface, the results are far murkier. Using Webmark 2001, an independent benchmark Intel asserts measures the "performance of a user interacting on the Internet," the Pentium M scores 285 while the faster Mobile Pentium IV scores 253. My hunch: Intel hasn't done much of anything to improve the Wi-Fi interface's power efficiency.

Centrino's lack of multimode support is shocking given Intel promises dating back at least two years. When asked to defend the lack of 11a and/or 11g support, Intel's vice president and manager of mobile platforms, Anand Chandrasekhar, said Intel "made the choice to go with b because wireless LAN vendors were all b-based." That's news to me. I don't know of a single major wireless vendor that doesn't have an 11a offering. Chandrasekhar was quick to point out that Intel will have a dual-mode a/b offering later this quarter. I guess we should conclude that, by then, all wireless LAN vendors won't be b-based. --Dave Molta

Post a comment or question on this story.

SUBSCRIBE TO OUR NEWSLETTER
Stay informed! Sign up to get expert advice and insight delivered direct to your inbox
More Insights