Nvidia and Intel shocked the tech world last year with a partnership announcement, detailing a collaboration on new x86 SOCs with integrated RTX GPU chiplets. Now, new rumours suggest that Nvidia's next-next gen Fenyman AI GPUs could see this friendship go much deeper than previously thought.
Technical details of the Intel and Nvidia partnership
Industry sources speaking to Digitimes have indicated that the I/O die on upcoming hardware will "partially utilise" Intel's advanced manufacturing. While the core GPU die itself will still be manufactured by TSMC, Intel is expected to play a significant role in the production chain.
According to reports via Notebookcheck, the technical breakdown of this potential collaboration includes:
- The use of either Intel's 18A or 14A process for the I/O die.
- Intel providing roughly 25% of the advanced packaging for the units.
- Continued reliance on TSMC for the primary GPU die production.
What we know about Nvidia's next-next gen Fenyman AI GPUs
The Fenyman GPU architecture was nominally revealed at GTC last year, though specific details have remained scarce since the announcement. Currently, there is a posited 2028 release date for these units.
However, given the current state of the global tech supply chain, all long-term release dates are up for debate. This is particularly true as the Rubin architecture, which precedes Fenyman, is only just beginning to see its time in the sun. It would not be a massive surprise if Nvidia's future hardware ended up using Intel Foundry tech somewhere in the chain, especially as the two companies appear to be close allies.
Intel's foundry ambitions and financial turnaround
Intel's most recent financials have looked particularly strong, marking a significant turnaround after several years of industry strife. CEO Lip-Bu Tan has highlighted the critical importance of securing foundry customers and partnerships. A primary goal for the company is getting more high-profile tech built on its 18A and 14A processes.
With Nvidia currently cruising from strength to financial strength, they are an ideal partner for Intel to hitch their wagon to. While an I/O die may not be the most vital part of an AI GPU, seeing Intel's technology integrated into Nvidia's next-next gen Fenyman AI GPUs would represent a massive step forward for "team blue" if these rumours prove true.
