This is just the beginning for Nvidia at CES 2016

Jan 5, 2016 09:55 GMT  ·  By

Nvidia has just revealed at CES 2016 a new water-cooled supercomputer named Nvidia Drive PX 2 that is built specifically for cars.

It’s clear that the next item on the list of objects or machines that will become really smart in the near future are cars, and Nvidia wants a piece of that pie. CES 2016 (Consumer Electronics Show) in Las Vegas is the perfect opportunity to launch something like this, even if its makers boast with pretty incredible features for this piece of technology.

The presentation at CES 2016 was made by Nvidia’s CEO, Jen-Hsun Huang, who showed the new DRIVE PX 2 and revealed that Volvo would be the first ones to implement the new technology, who will use it in 100 cars, in 2017, in the city of Gothenburg, Sweden.

DRIVE PX 2 is powerful

Nvidia’s CEO explained just how powerful the new unit really is. First of all, it’s built on a 16nm process, unlike the Titan X which uses the 28nm process, making it about six times more powerful. DRIVE PX2 has 12 CPUs cores, capable of 8 teraflops of processing power and 24 teraflops of deep processing operations, and Jen-Hsun Huang said that it’s the equivalent of 150 MacBook Pro.

“DRIVE PX 2's deep learning capabilities enable it to quickly learn how to address the challenges of everyday driving, such as unexpected road debris, erratic drivers and construction zones. Deep learning also addresses numerous problem areas where traditional computer vision techniques are insufficient -- such as poor weather conditions like rain, snow and fog, and difficult lighting conditions like sunrise, sunset and extreme darkness,” Nvidia explained in the official announcement.

The new hardware doesn’t come alone. In fact, Nvidia made a second announcement for NVIDIA DriveWorks, which is a piece of software that will help car manufacturers better integrate the new technology.

Besides Volvo, other companies have expressed interest in the DRIVE PX 2, like Audi, BMW, Ford, Mercedes-Benz, ZMP, Preferred Networks and AdasWorks. It’s likely that more will want to at least test it. It remains to be seen just how successful Nvidia will be, but it’s in a market with little to no competition.