“If we are going to have this technology in five to 10 years, we will finally be able to do things that are consuming a lot of processing power today,” explains Patrick Moorhead, analyst at Moor Insights and Strategy. “This could really open up our concept of what a computer is … The phone would be smart enough not to just be a computer, but to be my computer.”
Intel researchers in the company’s laboratory in Barcelona have created a prototype of a 48-core chip for smartphones and tablets.
Enric Herrero, a research scientist at Intel’s labs in Barcelona, Spain, said that so far they have built a prototype of a 48-core chip designed to run different applications on different cores on a smartphone or tablet.
Today, some small mobile devices use multi-core chips. However, those multi-core CPUs can be dual or quad core working with few GPUs. Having a 48-core chip in a small mobile device would open up a whole new world of possibilities.
At this point, however, the researchers are working to see exactly how to best utilize that many cores on a single device.
“Normally a processor with a core does the jobs one after the other,” explains Herrero. “With multiple cores, you can divide the work between them.”
This expert explains that with many cores, someone could, for example, encrypt an email while working on other power-hungry applications at the same time. It could be done today, but operations could slow down as they would have to share resources.
Tanausu Ramírez, another Intel researcher working on the 48-core chip, also pointed out that if someone were, for example, watching high-definition video, a 48-core chip would be able to use different cores to decode video in different frames. at the same time, giving the user a more fluid video experience. Ramírez also explains that many cores can work in parallel in different projects and consume less energy.
Justin Rattner, Intel’s CTO, thinks a 48-core chip for small mobile devices could hit the market “long before” the 10 years the researchers predict.
“I think the desire to go to the more natural interfaces to make the interaction much more human-like is really what is going to drive these computing needs,” he explains. “Having a large number of cores to generate very high performance levels is the most efficient way to use power to achieve those performance levels.”