Tesla is apparently looking to switch to AMD “Navi” chip for its onboard media computer (MCU), according to a new leak.
Originally, Tesla used Nvidia Tegra chips in its Media Control Unit (MCU), which is different from its Autopilot or Full Self-Driving (FSD) computers that use Tesla’s own chip.
As Tesla started working to transition away from using Nvidia’s chips for its Autopilot computer, the automaker also switched from Nvidia to Intel for its onboard MCU media computer.
Now, we’ve learned that Tesla is again looking to make a change in its MCU, and it is looking to move to AMD.
Patrick Schur shared a document showing that Tesla was looking to move to AMD’s new Navi 23 chip before removing the document as it could potentially reveal the source:
This is AMD’s latest chip powering its new GPU cards, and they claim more power and efficiency:
“Groundbreaking AMD RDNA™ 2 gaming architecture delivers up to 2X higher performance and up to 54 percent higher performance-per-watt compared to AMD RDNA™-based graphics cards“
Higher performance-per-watt is something that Tesla is always looking at closely since the automaker is looking to improve efficiency through every sub-system inside its vehicles in order to improve overall efficiency and, therefore, range.
It’s unclear when Tesla would be moving to the new chip for its onboard computer.
Electrek’s Take
I wouldn’t put this in the “completely confirmed” column just yet, but it looks like a solid source.
Tesla is always looking to improve computing power inside its vehicles, and that’s a trend that I’ve been calling ever since Tesla hired Jim Keller back in 2016.
Cars are increasingly becoming computers on wheels, and it is going to be critical for automakers to always keep pushing for higher performance in order to compete.
Tesla seems to have understood that early by both working on its chips and looking to buy the latest technology from chip makers, like AMD.
FTC: We use income earning auto affiliate links. More.
Subscribe to Electrek on YouTube for exclusive videos and subscribe to the podcast.
