Speaking at a press conference earlier that evening, the general director of NVIDIA, Jensen Huang and head of Intel Lip-Bu Tan, asked questions from the media about the newly announced cooperation between them.
While both general directors refused too far in terms of questions about what process the created systems would be created (or whether Nvidia would oblige to operate the Intel foundry in the near future), they both had a lot to say about the design and scope of the market that is trying to capture.
“In relation to the size of the market,” said Huang, “PCs market and PC market are large, and we will build revolutionary products, the first such products. Nothing of this kind has been previously built on the X86 market.”
It is said that these products are both NVIDIA-CUSTOM X86 processes integrated with AI infrastructure platforms to operate data centers and system system (SOC) with integrated NVIDIA RTX GPU chipletes for the consumer market.
“I think they will be revolutionary products,” continued Huang. “I know that everyone is working on it, we are very excited, architects working on it are very excited. And we can’t wait to tell you more over time.”
Despite the reluctance to provide the exact details of the project, Huang gave some tips on how two architecture can cooperate for consumer chips based on SOC:
“Intel has the ability to pack Multi Technology Foveros and it really allows it. The reason is that, as we all know, NVIDIA GPU technology is based on the TSMC foundry. And this is one of the extraordinary things you can do, combining NVIDIA chiplets with Intel CPUS in the ability to pack multi-technology and multiple packaging technology.
“So this is a really great way to mix and match technology, and this is one of the reasons why we will be able to introduce innovations so quickly and build these extremely intricate systems and deliver them as a package of multi -spike systems.”
The modern systems will probably be a uncomplicated fusion between GPU technology based in TSMC based in NVIDIA and mainly TSMC technology in Intel.
So it seems that the modern systems will probably be a uncomplicated merger between GPU technology based in NVIDIA TSMC and mainly TSMC Intel technology. Once, when asked about the perspective of Nvidia, he uses Intel as a foundry partner for advanced AI, Intel CEO, Lip-Bu Tan, said:
“As for the Intel foundry, we are still making progress. And then in terms of yield, 18a, 14a, we clearly want to qualify. Jensen and I will check it, but in general we will continue to conduct our success on the side of the process, and then we will win the client’s trust and trust, and then one step.”
Nevertheless, thrilling things, right? I do not think that any of us had NVIDIA/Intel cooperation in our books this morning, but it seems that they both work together longer than anyone expected.
“Two technological teams discuss solutions and architect, they probably take place up to a year,” said Jensen. “We have kept the partnership really quietly.”
No jokes!
The best PC Build 2025 computer
