Tesla has recently sparked a major stir in the autonomous driving chip market, with rumors of a potential partnership between Tesla and AMD. These whispers led to speculation that Tesla might be shifting its reliance from NVIDIA, a long-time partner in computing chips for self-driving systems. However, during GTC China, NVIDIA CEO Jensen Huang made it clear that even if Tesla used other chips, he would still buy a Tesla car. This statement was shared with a car electronics editor, highlighting the ongoing competition in the automotive tech space.
The brain of any self-driving system is the computing chip, and it has once again taken center stage. Despite claims by Groffont (a company associated with AMD) that the “fake news†was unintentional, the excitement around the topic shows how critical these chips are to the success of autonomous driving technology.
Several industry giants, including NVIDIA, Intel, and Qualcomm, as well as long-standing semiconductor suppliers and emerging startups, are all racing to develop their own autonomous driving chips. The goal is to gain a competitive edge and control the future of self-driving technology.
**1. Is Tesla’s “searching for new love†real?**
Last week, GlobalFoundries (GF), a foundry previously part of AMD, accidentally revealed that Tesla, AMD, and GF were working together on an autonomous driving project. This news caused a big reaction in the autopilot community, with AMD’s stock rising and NVIDIA’s falling slightly.
According to reports, Tesla plans to develop its own autopilot chip based on AMD’s IP, with GF acting as the foundry. However, GF quickly denied direct collaboration with Tesla, emphasizing only the "indirect relationship." Both Tesla and AMD have remained silent, neither confirming nor denying the claim.
Tesla is unlikely to confirm such a partnership for several reasons:
- Tesla currently uses NVIDIA’s Drive PX 2 in its models.
- The Model S and Model X are still parked in NVIDIA’s garage.
- NVIDIA’s DGX-1 deep learning system was donated to OpenAI, funded by Musk.
Despite this, Musk is known for his bold moves. His previous split with Mobileye serves as a cautionary tale, and now, while NVIDIA is surprised by the shift, it’s unclear whether Tesla will fully move away from them. However, Tesla’s goals seem to stretch far beyond just one company.
There are three main reasons why Tesla is interested in developing its own autonomous driving chip:
- Greater control over core hardware.
- Creating a unique advantage through hardware acceleration.
- Reducing costs at scale, as NVIDIA’s Drive PX 2 is power-hungry and expensive.
This makes self-developed chips a strategic choice for Tesla. In fact, Tesla already has a chip team of over 50 people, led by Jim Keller, who joined last year. Keller is a legendary figure in the industry, having worked at AMD, Apple, and more recently, contributing to the Zen architecture that powered the Ryzen processors.
**NVIDIA's Response**
At GTC China, Jensen Huang was asked if he would still buy a Tesla if the company developed its own autopilot chip. He didn’t immediately deny the possibility, leaving room for interpretation. The involvement of Groffont suggests that Tesla’s chip development may be further along than previously thought.
However, Tesla is unlikely to challenge NVIDIA’s GPGPU dominance in deep learning. Instead, the chip is expected to be a dedicated solution for autonomous driving.
**2. The Battle Among the Giants**
The Tesla-NVIDIA relationship is just one piece of the larger puzzle in the autonomous driving chip war. Companies like Intel, NVIDIA, and Qualcomm are leading the charge, each taking different technical approaches—GPU, FPGA, and ASIC—to meet the demands of self-driving systems.
**Intel: A $32 Billion Investment**
Intel has been struggling in the deep learning space due to the rise of GPUs. To counter this, it acquired Altera, a leading FPGA company, for $16.7 billion in 2016. FPGAs offer flexibility and efficiency, making them ideal for real-time applications in autonomous vehicles.
Intel also launched its own autonomous driving platform, Intel Go, and integrated Altera’s FPGA chips into Audi’s A8. Additionally, Intel acquired Mobileye for $15.3 billion, gaining access to the ADAS market and solidifying its position in the self-driving ecosystem.
**NVIDIA: Performance and Ecosystem**
NVIDIA has dominated the AI computing space, especially in deep learning. Its GPU technology offers unmatched performance, and its TensorRT 3 platform supports various deep learning frameworks. While NVIDIA has had some presence in the automotive sector via Tegra, its role in actual autonomous driving is still evolving.
**Conclusion**
The race for the next-generation autonomous driving chip is heating up, with Tesla, NVIDIA, Intel, and others all vying for dominance. As the industry continues to evolve, the battle for control over the brain of the self-driving car is just beginning.
Optical fiber connector is a detachable (movable) connection device between optical fiber and optical fiber. It connects the two end faces of optical fiber precisely, so that the optical energy output by the transmitting optical fiber can be coupled to the receiving optical fiber to the maximum extent, and the impact on the system caused by its involvement in the optical link can be minimized. This is the basic requirement of optical fiber connector. To a certain extent, optical fiber connectors affect the reliability and performance of optical transmission systems.
Fiber Fast Connector,Fiber Quick Connector,Fast Connect Fiber Connectors,Fiber Optic Quick Connector
Ningbo Fengwei Communication Technology Co., Ltd , https://www.fengweicommunication.com