NVIDIA’s Drive PX 2 is the onboard supercomputer that is installed in all-new Tesla vehicles since October 2016 and that powers the automaker’s second generation Autopilot with the ‘Tesla Vision’ image processing technology.
We had to take Tesla at their words that the high-tech piece of equipment was in the vehicle until now. An owner removed the panels of the new car and we got our first look at the system this week.
NVIDIA describes the Drive PX 2 as “the world’s first AI supercomputer for self-driving cars”. Its computing power is comparable to about 150 MacBook Pros and the company estimates that one can support a level 4 self-driving system while two would be necessary for a fully self-driving level 5 vehicle, but Tesla is aiming for its software to be efficient enough to run level 5 on one.
But if that doesn’t work out, the company made sure that it is relatively easily accessible so that it can be swapped.
It is mounted on the passenger side of the dashboard right above the outer panel of the glove box. By removing it, it can be accessed and technically swapped for a more powerful computer if the need ever arise.
A few owners removed the panel to reveal the Drive PX 2 and this week, they shared the picture on the Tesla Motors Club forum:
It is a little more elegant that in NVIDIA’s own test car, which uses the same computing platform, but again it is a test car while Tesla is installing it in a production vehicle:
But it’s still a big difference. Tesla’s current fully self-driving vehicle program is being tested using an hardware-ready platform. Granted, the company is still working on the software and maybe it has a long way to go, but when you look at Tesla’s self-driving video demonstrations that they released and compared them to other automakers, like Ford for example, it’s interesting to note with what they are powered.
Here are Ford and Audi’s computers in their self-driving test cars:
Those are so demanding that they need their own energy source. That’s a problem that is going to be addressed.
I also took some pictures of the bare Drive PX 2 without any casing or liquid cooling system at NIVDIA’s CES booth earlier this month (click on the pictures for higher resolutions):
Tesla built it self-driving platform so that it could use different computers, but at the end it came down to NVIDIA versus AMD and the former was the better choice at the end, according to Tesla CEO Elon Musk.
But if it was to change, Tesla could swap the unit as an upgrade.
Additionally, the company is reportedly working on its own custom SoC (System on Chip) especially for self-driving cars and looking at Samsung to manufacture it. The chip would likely be designed by famous chip architect Jim Keller who now leads Tesla’s Autopilot hardware team.