Intel and Nvidia collaborate on the development of 'Intel x86 RTX SOCs' for PCs incorporating Nvidia graphics, along with custom Nvidia data center x86 processors. As part of a significant agreement, Nvidia invests $5 billion in Intel's stock.
In a groundbreaking development, Intel and Nvidia have announced a partnership to jointly develop multiple new generations of x86 products. The collaboration is set to revolutionize the tech industry, as Intel will fabricate custom x86 data center CPUs for Nvidia, which Nvidia will sell as its own products.
At the heart of this partnership is the Intel x86 RTX SOC chips, which will feature an x86 CPU chiplet tightly connected with an Nvidia RTX GPU chiplet via the NVLink interface. This connection allows for uniform memory access (UMA), enabling both the CPU and GPU to access the same pool of memory, offering improved performance and efficiency.
The use of the NVLink interface for CPU-to-GPU communication provides a significant advantage over PCIe, offering up to 14 times more bandwidth and lower latency. This enhanced communication will lead to a more seamless and efficient workflow, particularly in data-intensive tasks.
Nvidia's investment in Intel is also noteworthy. The tech giant has purchased $5 billion in Intel common stock at $23.28 per share, representing a roughly 5% ownership stake in Intel. Softbank, too, has shown faith in Intel's future, purchasing $2 billion worth of Intel stock at $23 per share.
The partnership is in the early stages, with product releases and specifications to be disclosed at a later, unspecified date. However, it is clear that this collaboration will find tighter integration using the NVLink interface, paving the way for innovative products in the consumer gaming PC market and beyond.
It's worth noting that Intel has a history of working with AMD for interoperability between their hardware and software, and the new partnership will continue this tradition. AMD, however, is unlikely to adopt NVLink Fusion, as the company is heavily invested in its own interconnect initiatives like Infinity Fabric and Ultra Accelerator Link.
Nvidia remains committed to other announced product roadmaps, including the Arm-based GB10 Grace Blackwell processors for workstations, Nvidia Grace CPUs for data centers, the next-gen Vera CPUs, and products on its internal roadmaps that haven't been publicly disclosed yet.
Both Nvidia and Intel will hold a webcast press conference at 10 am PT for more details on the partnership. The future of x86 products is undoubtedly an exciting one, and this partnership promises to bring about significant advancements in technology.
However, there is some uncertainty about the production locations for the Intel x86-RTX SoCs due to delays and cancellations in Intel's plans for new factories. As more details about the partnership and the production process emerge, we can expect to see a more clear picture of what this collaboration means for the tech industry.
Read also:
- Reporter of Silenced Torment or Individual Recording Suppressed Agony
- EPA Administrator Zeldin travels to Iowa, reveals fresh EPA DEF guidelines, attends State Fair, commemorates One Big Beautiful Bill
- Musk announces intention to sue Apple for overlooking X and Grok in the top app listings
- Cybertruck's Disappointing Setback, Musk's New Policy, Mega-Pack Triumphs, Model Y's Anticipated Upgrade Prior to Refresh (Week of January 25 for Tesla)
