Intel, AMD battle over claims on battery life for laptops

SAN JOSE, Calif. — Benny Villanueva recalls being in a Starbucks rushing to finish a college paper that was due when his laptop battery, which he'd been led to believe would last hours, conked out after only about 30 minutes.

"It just shut off on me," said the 30-year-old from San Jose, who is studying dentistry. "I almost started yelling. It's kind of frustrating, but what can you do?"

Consumer complaints about being misled on how long their laptop batteries will last aren't new, of course. But the issue has had the Web roiling lately, with much of the hullabaloo centered on those two longtime chip-making antagonists, Intel and Advanced Micro Devices.

Sunnyvale, Calif.-based AMD triggered the brouhaha in recent blogs and media interviews, contending that laptop ads often contain unrealistic battery-life claims because the data come from tests where laptops aren't used as actively as many people use them.

And last month, a class-action lawsuit filed in U.S. District Court in San Jose accused Intel of essentially rigging those tests to inflate the battery life of laptops powered by its chips.

"A reasonable consumer expects battery-life measurements to reflect the way consumers actually use laptop computers," the suit said. Because of the Santa Clara chip maker's alleged contrivance, it added, "Intel has wrongly increased its profits from the sale of laptops with Intel processors."

But Intel officials called the suit groundless and characterized criticism of the tests as wrongheaded. They claim the tests accurately reflect the way most people use laptops. And while Intel would consider other ways to test the batteries, company spokesman Bill Kircos disputed claims that the public is confused.

"Anyone who criticizes consumers' intelligence when shopping for laptops is underestimating the consumers," he said.

Testing service

The controversial tests are specified under a benchmark dubbed MobileMark, which was adopted several years ago by the nonprofit Business Applications Performance Corporation, or BAPCo, whose members include AMD, Intel, Dell, Hewlett-Packard and other tech companies.

The class-action suit filed June 26 by the Girard Gibbs law firm of San Francisco contends the tests were pushed on BAPCo by Intel. Intel denies that, adding that the same law firm unsuccessfully filed a class-action suit against Intel several years ago. A statement issued by BAPCo said MobileMark was "vigorously debated and cooperatively developed" by its members.

Whatever the case, critics say the tests leave much to be desired.

MobileMark assesses how long batteries work while a laptop is in three operating modes, which include playing a DVD movie, accessing a spreadsheet and doing other tasks. And according to data provided by AMD, batteries in laptops equipped with Intel's chips consistently last longer in the tests than laptops with AMD's chips.

Various factors determine how long a laptop battery lasts, but chips play a big role and Intel has focused on making its chips energy-efficient. But AMD, which specializes more than Intel on making graphic-oriented chips, argues that MobileMark generally doesn't reflect use of power-gobbling laptop features such as graphic-heavy video games.

As a result, even though laptop ads typically caution that battery life varies with use, AMD claims consumers often find their machines run out of juice long before they thought they would.

"The measurements in the best case are confusing; worst case they are misleading the consumer," said Patrick Moorhead, AMD's vice president for advanced marketing.

Rob Enderle, a technology analyst with the Enderle Group in San Jose, agrees.

"Everyone in the industry knows this benchmark is wildly optimistic and that the actual battery life you'll get is often less than half what MobileMark suggests," he concluded in a recent note on the controversy. "This is because MobileMark measures battery life much like you might measure gas mileage if you started the car, put it in neutral and coasted down a long hill."

Other sources

AMD says a better idea would be to give consumers battery-life data from MobileMark as well as from some other test where laptops are run harder. When AMD did such a comparison, it said the battery-life difference between its chips and Intel's virtually disappeared.

Officials at Hewlett-Packard, the world's biggest seller of personal computers, declined to comment on the dispute.

Sensible move

But Ketan Pandya, head of AMD-based products at Dell, said AMD's suggestion to augment MobileMark with another battery-life measure makes sense.

"It's something that is good for the industry and definitely good for the consumer," Pandya said.

Internet commentators remain divided on the subject, however.

"Shouldn't AMD be trying to make its chips use less power when they're not doing anything?" an article in laptop magazine wondered aloud last month.

But a recent article by Notebooks.com agreed with AMD, concluding that battery-life advertising claims based on low-power laptop configurations "leave those who order notebooks with graphic cards, high-speed hard drives and other extras completely in the dark."