At the GTC Europe 2016 conference on Wednesday, Nvidia officially announced a brand new system-on-chip (SoC) it is marketing as an “AI supercomputer.” The new chip, dubbed Xavier after the leader of the X-Men, marks Nvidia’s departure from the mobile SoC space in favor of the automotive industry, where the next generation of self-driving cars will require more than just horsepower.
Xavier is essentially the next generation of Nvidia’s Tegra chip, reports Anandtech, but with a few key improvements. Namely, it is capable of running on a fraction of the power and performing around 20 trillion operations per second. Nvidia’s aim with Xavier is to roll their existing Drive PX 2 system into one super powerful SoC.
What does that mean for you? Well, not much at this stage in the game. Xavier is still a long way out, it will not even be sampled until over a year from now. But then again, autonomous cars are also a little ways out for most consumers.
The announcement was about the long game. There had been questions about Nvidia’s SoC business ever since its public roadmap ended — with its previous SoC, Parker — so the announcement of Xavier was about more than showing off a new processor.
Nvidia’s expansion into the automotive space marks an official end to their mobile pursuits and Xavier illustrates that it is in it for the long haul. This latest SoC has some serious computational power, including two processors for 8K video input, which the next generation of autonomous cars are going to need in order read the road and process a variety of complex driving conditions.
Whether or not you will be able to use your car to run games remains to be seen, however.
