After the IFS Direct 2024 event Pat Gelsinger, Intels CEO mentioned in a meeting, with the press and analysts that Chipzilla is incorporating TSMC’s advanced process node technology for their CPUs named Arrow Lake and Lunar Lake.
During the discussion, Pat highlighted Intel’s collaboration with TSMC for CPU designs emphasizing that future CPUs like Arrow Lake and Lunar Lake will shift from 5nm to 3nm nodes.
More About Intel and TSMC building Next-Gen CPUs
The upcoming Intel Arrow Lake CPUs are set for a release this year and will utilize Intel’s in-house 20A process node. Early reports have suggested that TSMCs 3nm (N3) process node will be used for the GPU tile. While maintaining the iGPU architecture from Meteor Lake (Alchemist “Xe LPG”) there are optimizations tailored for the mobility lineup with Xe LPG+ architecture. The transition from N5 to N3 is expected to deliver enhancements in efficiency and performance.
On a note Intel’s Lunar Lake CPUs are expected to embrace the P Core (Lion Cove) and new E Core (Skymont) core architecture likely produced on the 20A node, for the CPU tile.
The upcoming Lunar Lake GPU is set to receive an improvement compared to the Meteor Lake and Arrow Lake CPUs. By using Alchemist, Lunar Lake will adopt the next-generation graphics architecture known as Battlemage “Xe2 LPG.” Sources indicate that Intel plans to manufacture the Lunar Lake GPU tiles using TSMCs N3B process node.
Furthermore, reports indicate that Intel’s future client CPU, codenamed Nova Lake will also leverage TSMC’s advanced process nodes. Nova Lake CPUs are expected to utilize a 2nm node, though it’s unclear if a specific tile or the entire chip will be fabricated on the N2 node. Despite committing to a 4-node strategy over 5 years, Intel has demonstrated a strong reliance on TSMC to meet its supply demands for client CPUs.
During the IFS Direct 2024 event, the company rebranded its Foundry Services to Intel Foundry and introduced a new 14A node alongside various sub-variants of existing nodes. Intel’s discrete GPU family currently utilizes TSMC’s N6 process node, and this is expected to continue with the upcoming Battlemage discrete lineup, set for launch later this year.
