Apple and Intel could become early adopters of TSMC’s 2nm node

0

Taiwan Semiconductor Manufacturing Corp. is set to start high-volume production of chips using its N2 process technology (2nm class) in late 2025 and deliver the first batch of these chips in early 2026. According to two media outlets and sources Within the financial community, the first customers to adopt N2 will be Apple and Intel.

Apple and Intel are the first to use TSMC’s N2

Apple has been TSMC’s biggest customer by revenue contribution for about a decade, so it’s no surprise that it’s also an alpha customer for N2. As for Intel, the company intends to use TSMC’s services to manufacture graphics processing units (GPUs) and various SoCs, two types of applications that benefit from cutting-edge nodes. So it’s no surprise that Intel is also an early adopter of N2, according to reports from DigiTimes and UDN. Moreover, given Intel’s volumes, it will quickly become one of the founder’s first customers.

Since the first batch of N2 chips are expected to ship in early 2026, it’s unclear which of Apple’s system-on-chips (SoCs) will use it then. Meanwhile, analysts at China Renaissance Securities believe Intel will use TSMC’s N2 for its Lunar Lake CPU graphics tile.

“We also see more clarity on TSMC’s N2 expansion timeline in Fab 20 (Hsinchu),” China Renaissance Securities analyst Sze Ho Ng wrote in a note for clients. “The tool move is expected to begin by the end of 2022, based on company plans, before late 2024E risk production with Intel (the Lunar Lake client PC graphics ’tiles’, while processor ’tiles’ are made using Intel’s 18A) with Apple being the primary customers for dedicated capacity support.”

Intel

(Image credit: Intel)

The N2 usage scenario for Lunar Lake is speculation at this point. However, Intel’s own slide that describes the graphics tiles for the Meteor Lake, Arrow Lake, and Lunar Lake CPUs makes it clear that the latter’s GPU will be manufactured externally using more advanced technology than N3.

AMD, Broadcom, Nvidia, and MediaTek have officially confirmed that they will use various TSMC N5 family nodes (N5, N5P, N4, N4P, N4X). MediaTek has already officially introduced its N5-based Dimensity 8000/8100 application processors and N4-based Dimensity 9000 SoC, while Nvidia will use a custom 4N manufacturing process for its Hopper and presumably Ada Lovelace GPUs. AMD’s Genoa and Raphael processors will also be manufactured using 5nm technology.

According to the DigiTimes report, all of these companies are currently in talks with TSMC about allocating N3 compatible capacity starting in late 2023 or sometime in 2024. In addition, these companies are also expected to start talks regarding N2 compatible allocations l next year, although they will certainly adopt N2 much later than Apple and Intel.

N2: coming in 2026

TSMC’s N2 will be the foundry’s first technology to adopt gate-all-around field-effect transistors (GAAFETs), years after Samsung’s 3GAE (2023) and more than a year after Intel’s 20A (2024) . So far, the world’s largest contract chipmaker hasn’t revealed what to expect from N2 in terms of improved power, performance, and area/transistor density over N3. . However, considering that it will be a completely new node, it is reasonable to expect tangible advantages over its predecessors. The new manufacturing process will continue to rely on proven extreme ultraviolet (EUV) lithography scanners with a 0.33 numerical aperture. In contrast, Intel’s 18A is configured to use ASML’s innovative Twinscan EXE EUV scanners with High-NA (0.55NA).

Share.

About Author

Comments are closed.