October 24, 2023
This week, Reuters revealed a new threat to Intel's historical stronghold in the PC and workstation market. It's coming from the GPU maker NVIDIA. “Nvidia has quietly begun designing central processing units (CPUs) that would run Microsoft’s Windows operating system and use technology from Arm Holdings,” the news agency reports.
NVIDIA declined to comment on the report.
A New Threat to Intel
Reuters' report says “[NVIDIA's] new pursuit is part of Microsoft's effort to help chip companies build Arm-based processors for Windows PCs.” This indicates NVIDIA is now striking at the heart of Intel's core business—CPUs for personal computers and workstations.
Processor market analyst Jon Peddie, President of Jon Peddie Research, said, “It's not different than what Qualcomm is doing, or what Apple does with their PCs. NVIDIA has the talent to do it, and the brand to support it.”
Though originally a chipmaker for mobile devices, Qualcomm is now elbowing its way into the PC market with its Arm-based CPU, named Qualcomm Snapdragon X. Today, Qualcomm executives revealed the company's new Snapdragon Elite X chip “will be available in laptops starting next year and has been redesigned to better handle artificial intelligence tasks like summarizing emails, writing text and generating images,” reported Reuters.
Apple began using its own M1 CPUs for its Mac PCs in 2020. The move freed Apple from its dependence on Intel. Another chipmaker, AMD, began offering AMD Ryzen Threadripper CPUs for PCs and workstations in 2017. These represent cumulative challenges, slowly chipping away Intel's dominance.
Mercury Research's breakdown of the CPU market (with IoT and SoCs included), as cited by Venture Beat, shows Intel currently leads the sector with 68% in 2023 Q2. The second runner up AMD holds 31%.
With its outsized reach in AI, film, game, and engineering, GPU maker NVIDIA's reported CPU strategy adds another threat to Intel's lead.
Data Center Rivalry
NVIDIA has already developed Arm-based CPUs for data centers. Announcing the processor codenamed Grace at NVIDIA's annual GTC (GPU Technology Conference) 2021, NVIDIA CEO Jensen Huang described it as “a new kind of computer, the basic building block of a modern data center, the world’s first CPU designed for Terabyte-scale accelerated computing.” On the decision to choose Arm architecture, he said, “it's super energy-efficient, has an open licensing model, and is used broadly in the mobile and embedded segments.”
While Grace CPU challenges Intel's position in data center sector, it was not designed to be in personal computers and workstations. But Reuters' report indicates these dynamics are about to change.