Intel isn’t giving up on ARC graphics for Nvidia chiplets
Intel will not stop making GPUs – Nvidia’s GPU options will be “complementary”
Intel and Nvidia have struck a deal that will have Intel create custom x86 CPUs for Nvidia, and Nvidia create GPU chiplets for Intel’s CPUs. One of the questions this deal raises is this. What does this Intel/Nvidia deal mean for Intel’s ARC GPUs? The answer is simple: nothing.
Intel has stated that Nvidia’s GPUs will be “complementary” to Intel’s own GPU products. “Intel will continue to have GPU product offerings.” In other words, Intel has no plans to abandon its ARC/Xe graphics architecture.
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 won’t stop making ARC GPUs because they can’t
Intel won’t give up on ARC for one simple reason: because it can’t. Even if Intel uses Nvidia GPU chiplets for high-end gaming products, it still needs its own GPU IP for lower-end products. Nvidia won’t want to create low-end GPU chiplets for Intel. The margins are too low for that. Intel also doesn’t want to be a slave to Nvidia.
So does Intel ARC have a future? Yes! That said, it remains to be seen how this deal will impact Intel’s plans for ARC. This deal diminishes the need for a “high-end” or “discrete” Intel ARC GPU option. That said, Intel still requires robust GPU IP for its integrated graphics solutions, which necessitates strong drivers and software support. If Intel abandons its planned ARC high-end products, its lower-end products will suffer. Furthermore, it will give Nvidia considerable leverage.
You can join the discussion on Intel’s plan to continue creating GPUs on the OC3D Forums.
