Engineers have long lived by a simple, seemingly obvious rule when designing new computers: The machines have to deliver correct answers.
If asked to compute two plus two, a computer should answer four. But what if computers didn’t always have to answer correctly?
Nearly a decade age, a Houston computer scientist posed this heretical question. Today, it’s led to a movement dubbed “probabilistic computing,” which he believes will revolutionize the future of computing.
On Sunday, Krishna Palem, speaking at a computer science meeting in San Francisco, will announce results of the first real-world test of his probabilistic computer chip: The chip, which thrives on random errors, ran seven times faster than today’s best technology while using just 1/30th the electricity.
Just think: One need never again worry about draining an iPhone battery in a day or even a week.
“The results were far greater than we expected,” said Palem, a Rice University professor who envisions his chips migrating to mobile devices in less than a decade.
And hopefully some of the companies that will arise to design, manufacture, and use those chips will be located right here. Regardless, this is an exciting development.