The Tegra X1 chip is a "mobile super chip" from Nvidia based on the company's Maxwell GPU architecture. Designed to outperform other mobile CPUs, the chip is being touted as "the world's first mobile super chip."
The Tegra X1 will be responsible in generating the required horsepower which is necessary for processing those images that vehicles can see and use them to analyze on-road situations. It can be comparable to a supercomputer in a car since it can understand the environment surrounding the vehicle, be aware of the situation at all times and decide on what appropriate action to take.
"The future car is going to have an enormous amount of computational ability," said Nvidia Chief Executive Jen-Hsun Huang. "We imagine the number of displays in your car will grow very rapidly."
Huang added that a platform where two of the X1 chips would be combined is soon to hit the market. This will allow the platform to gather data from up to 12 HD cameras that monitor traffic, blind spots and other safety conditions found in driver assistance systems.
Dubbed as the Drive CX car platform, Huang said that its pair of Tegra X1 chips will help in making navigation a more interactive experience. Moreover, the chip will also allow the 4K video system in cars. It would then enable car makers to use the Drive CX when they need to develop infotainment or in-car navigation features.
"It's really important to have that level of object recognition," said Patrick Moorhead, principal analyst at Moor Insights and Strategy. "That's were Nvidia is going to shine."
As a way to demonstrate the chip's power, Huang did a demo of the gaming engine Unreal 4 as it runs the game Elemental IV. While the result had been amazing, Huang noted that the Tegra X1, being the world's first teraflop-capable processor, will shine the most when utilized in cars.
"We believe future cars will be the most advanced computers in the world," said Huang.
One possible partner for the Tegra X1 chip is Tesla which has always been known for its continuous search for the latest and greatest in automobiles.
There's also a plan to use the chip in self-driving cars. Huang noted that since most of the sensors currently found in cars are replaced by cameras, auto-pilot cars would only work when they have the capacity to become "self-aware." By combining image processing with the so-called "deep neural net computer vision," future cars can have the ability to be aware of the situation as a result of a vision-based system of learning.
