Intel Confirms Ongoing GPU Product Development and NVIDIA Partnership Enhances Product Roadmap

Intel has assured the public that it will maintain its GPU products in the future, notwithstanding the newly announced collaboration with NVIDIA.

Intel’s Commitment to GPU Development Amid NVIDIA Collaboration

In a significant announcement today, NVIDIA and Intel revealed a partnership aimed at developing next-generation x86 chips that merge Intel’s CPU technology with NVIDIA’s leading GPUs. This collaboration will target various applications, including client PCs and enterprise-level AI and HPC solutions.

During the press conference featuring NVIDIA CEO Jensen Huang and Intel CEO Lip-Bu Tan, Intel’s intention to uphold its own GPU lineup was not explicitly detailed. Currently, Intel boasts two primary GPU families: the Gaudi and “Shores”series for enterprise AI, alongside the Arc series aimed at client and business applications. Despite initial hurdles, these GPUs have begun to establish a solid foothold in their respective markets.

Concerns arise regarding Intel’s potential reliance on NVIDIA’s GPUs for future chiplets in its products, which could diminish the need for Intel’s internal graphics division. However, Intel has responded to these worries, providing clarification to PCWorld and HotHardware:

“We’re not discussing specific roadmaps at this time, but the collaboration is complementary to Intel’s roadmap and Intel will continue to have GPU product offerings, ”

Intel Spokesperson

This statement reinforces Intel’s commitment to its GPU product lines, indicating that their roadmap remains intact despite the NVIDIA deal. Notably, Intel is preparing to launch its 3rd Generation Xe architecture, known as Xe3 or codename “Celestial, ”alongside Panther Lake CPUs. Although these upcoming chips are unlikely to integrate any NVIDIA technology, next year’s Nova Lake CPUs might be the first to capitalize on NVIDIA’s GPUs.

Earlier reports mentioned Intel’s Nova Lake-AX chip, which positions itself against high-end SoCs like the AMD Ryzen AI MAX series. The Nova Lake-AX is expected to deliver enhancements, including a larger integrated GPU, increased cache, and innovative packaging technologies. Speculation suggests that up to 48 Xe3 iGPU cores could be included, but with the NVIDIA partnership, Intel may reconsider and instead implement an RTX GPU from NVIDIA’s upcoming generations.

This strategy could mirror Intel’s previous Kaby Lake-G initiative, which combined multiple chiplets, including an AMD Radeon GPU based on RX Vega architecture. While Kaby Lake-G didn’t achieve significant success, it laid the groundwork for future SoCs such as AMD’s Strix Halo and Intel’s AX series, which effectively utilize advanced packaging solutions developed by TSMC. Intel’s proprietary packaging technology, Foveros, is expected to play a crucial role in the design of these Intel and NVIDIA collaborative chips.

While NVIDIA’s RTX GPUs will enhance certain high-performance models, Intel’s standard offerings are likely to remain rooted in its own Xe architecture lineup, including Xe3 and the forthcoming Xe4. Nevertheless, lingering questions about the future of Intel’s Arc discrete GPUs remain pertinent.

Intel Assures They Will Continue To Have GPU Product Offerings In The Future, NVIDIA Deal Complimentary To Product Roadmap

Intel remains committed to advancing its Arc discrete GPUs, as integrating NVIDIA technology would offer little strategic advantage and could cede ground in the discrete market, which currently sees NVIDIA holding a staggering 94% market share.

Jensen Huang highlighted plans for utilizing NVLINK and other advanced technologies in AI applications, although further details on their impact on Intel’s Jaguar Shores AI accelerator line remain scarce. As Intel embarks on this collaboration with NVIDIA, excitement builds for upcoming product announcements anticipated at CES and other significant events, signaling promising developments on the horizon.

Source & Images