Close

Who needs a supercomputer when an Xbox will do?

Share on facebook
Share on twitter
Share on linkedin
Share on whatsapp
Who needs a supercomputer when an Xbox will do?

It’s always nice to get a bargain, and one researcher is currently rather pleased with himself for realising that an Xbox chip could be used to save the University of Warwick thousands of pounds on parallel processing.

Dr Simon Scarle, a researcher in the university’s WMG Digital Laboratory, wanted to model how electrical excitations in the heart moved around damaged cardiac cells in order to investigate cardiac arrhythmias.

The problem was that conducting these simulations using traditional CPU-based processing would have meant booking time on a dedicated parallel processing computer or spending thousands on a parallel network of PCs.

But Scarle had a better idea, having previously worked as a software engineer for Rare, part of Microsoft Games Studios. He reckoned that the Xbox 360’s graphical processing unit could do the job for a few hundred pounds.

And it worked. “Although major reworking of any previous code framework is required, the Xbox 360 is a very easy platform to develop for, and this cost can easily be outweighed by the benefits in gained computational power and speed, as well as the relative ease of visualization of the system,” he said.

Unfortunately, the results of the research itself were less happy, with his study demonstrating that it wasn’t in fact possible to predict the rise of certain dangerous arrhythmias.

Of course, his technique may lead to a few tricky situations. “I can imagine some cheeky people trying to get games consoles past their department heads on rather flimsy pretexts,” Scarle told TG Daily.

So what next? Nuclear physics on a Playstation? Well, yes, actually.

“There’s a few other people that have been looking at using the graphics chip on a games console or desktop PC,” said Scarle. “I know of one person who’s been using a Playstation 3 to model black holes.”

The results of his work have been published in Computational Biology and Chemistry.