Connect with us

Hi, what are you looking for?


Autonomous Vehicles

Tesla Has “Raised The Bar For Self-Driving Computers,” Admits Nvidia

For any skeptics doubting the revolutionary capabilities of Tesla’s new Full Self Driving (FSD) compute hardware, why not tune in to what autonomy hardware rival Nvidia have to say about it? Nvidia has said that Tesla’s new hardware has just “raised the bar for self-driving computers.”

For any skeptics doubting the revolutionary capabilities of Tesla’s new Full Self Driving (FSD) compute hardware, why not tune in to what autonomy hardware rival Nvidia have to say about it? Nvidia has said that Tesla’s new hardware has just “raised the bar for self-driving computers.”

The easiest metric to grasp for non-technical folks is that the frame processing performance has increased from 110 frames per second in the version 2.5 computer to now 2300 frames per second in the FSD computer. There are also a lot of smart optimizations — including important energy efficiency optimizations — and other features of the chip design that are gone into in great detail in the full Tesla Autonomy Day presentation video.

Nvidia supplied Tesla with the hardware 2.0 computer that started getting installed in all Teslas produced from October 2016. This was based on the Drive PX 2 platform. It’s unclear what hardware the version 2.5 computer (from August 2017 onwards) was based on, but the vendor was likely still Nvidia, possibly with some version of its more recent PX Xavier or PX Pegasus hardware. Whilst no slouches, Nvidia has had to design all of these chips to work with a variety of sensors and autonomy software implementations, to appeal to a broad customer base of different automakers. Tesla, on the other hand, has already committed to a sensor hardware design for its autonomy system, and a particular software approach. Tesla can therefore purpose design and optimize its computer hardware to suit the particular needs of the overall system design.

The other main benefit of making their own chip is the ability to lower the cost for Tesla (by some 20%). The FSD computer was designed by Pete Bannon and his team — Pete is renowned for previously having designed Apple’s A5 through A9 CPUs, amongst other achievements. The FSD computer is made for Tesla at a Samsung fabrication facility in Texas.

Tech folks who are familiar with chip design can no doubt evaluate for themselves the merits of Tesla’s approach from the information contained in the presentation. If there are any lingering doubts for non-tech folks, it’s reassuring to know that Nvidia itself has acknowledged that the new Tesla computer has “raised the bar for self-driving computers.” At the same time, Nvidia noted that the company disagrees with Pete Bannon’s assessment that the Tesla computer is 7× as powerful as its own current PX Xavier computer. However, the key point may simply be that — for the needs of Tesla’s specific approach to vehicle autonomy — the FSD computer may be 7x more capable. (Editor’s note: Or perhaps it’s just that Tesla was comparing to the previous Nvidia system, not the current one?)

Over the past few weeks, the FSD computer has replaced the previous version 2.5 computer hardware in all Tesla vehicles being produced. Note that all Teslas that have Autopilot hardware version 2.0 (October 2016) or later can also be upgraded to the FSD computer. Elon Musk and the team believe that, with this computer and the existing sensor hardware now in place, the only remaining piece of the puzzle is the evolution of the software to enable Full Self Driving — and regulatory approval. The full presentation below also contains a lot of information about this, or check out Zach’s summary overview. Once the software is ready, Musk says, “The fleet wakes up with an over-the-air update.”

If you haven’t yet had a chance to watch the video, I strongly recommend it:

Appreciate CleanTechnica’s originality and cleantech news coverage? Consider becoming a CleanTechnica Member, Supporter, Technician, or Ambassador — or a patron on Patreon.

Don't want to miss a cleantech story? Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

Have a tip for CleanTechnica, want to advertise, or want to suggest a guest for our CleanTech Talk podcast? Contact us here.

Max is an anthropologist, social theorist and international political economist, trying to ask questions and encourage critical thinking about social and environmental justice, sustainability and the human condition. He has lived and worked in Europe and Asia, and is currently based in Barcelona. Find Max's book on social theory, follow Max on twitter @Dr_Maximilian and at, or contact him via LinkedIn.


You May Also Like


Tesla’s stock has remained a polarizing topic, especially as the company’s stock price dropped immensely throughout the last year. Bears and bulls are trying...


Following yesterday’s earnings call, shareholders of Tesla stock are pleased to see a “comeback” after a tough few months. Shares surged over 10% in...


In this article, I will explain how many households don’t have access to the US EV tax credit and some possible solutions. 81 Million...


On Tesla’s Q4 2022 and full-year 2022 conference call for shareholders (still occurring as I’m typing this), we received updates on Tesla Cybertruck production...

Copyright © 2023 CleanTechnica. The content produced by this site is for entertainment purposes only. Opinions and comments published on this site may not be sanctioned by and do not necessarily represent the views of CleanTechnica, its owners, sponsors, affiliates, or subsidiaries.