Skip to Content
Uncategorized

Watching the Clock

December 1, 2002

You can’t always believe everything you read on a computer box. Even clock speed-that gigahertz-expressed metric most computer buyers associate with a machine’s power-may not mean what it used to. A spat between computer titans Sun Microsystems and IBM over just how misleading the measure is has led to a fledgling industry effort to find a more accurate way to gauge a computer’s abilities.

The problem is a result of new trends in chip design. Clock speed refers to the frequency with which a chip’s central processor does basic operations. Specialized new chips have multiple processors that can end up competing for access to memory, creating bottlenecks that may worsen with future multiprocessor chips, says David Yen, Sun’s executive vice president of processor and network products. But while they may perform at a slower clock speed than today’s chips, the multiprocessor chips Sun plans to build will run many software programs faster because the additional processors let the chips handle more incoming and outgoing data. Clock speed, Yen asserts, is often “used in a misleading way and overhyped. We’re looking for a more appropriate yardstick.”

Clock speed, however, is still a legitimate measure of multiprocessor chips that have wide processor-memory connections and therefore don’t suffer bottlenecks, says Ravi Arimilli, the chief technology officer behind the development of a new IBM multiprocessor chip. Arimilli dismisses Yen’s call for a new benchmark, saying business customers already look at a range of metrics-including how fast their own software applications will run-before they buy a new computer.

But consumers don’t necessarily have the same computer savvy. And although today’s debate centers on high-end chips used in business servers, it may not be long before PCs have multiple processors. Yen says Sun is talking with chipmaker Advanced Micro Devices about a new way for all users to evaluate multiprocessor chips-a measure “simple enough that people still intuitively get it.” That could be a challenge, but it’s clear that the gigahertz has had its day in the sun.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.