Microsoft’s upcoming custom chip will be made by Intel

Intel’s relatively new foundry division — formerly known as Intel Foundry Services until earlier today — just landed a notable order from a big name. According to Bloomberg and The Wall Street JournalMicrosoft CEO Satya Nadella announced that his company will use Intel’s latest 18A (1.8nm) manufacturing process for upcoming in-house chip designs. But given Intel’s process roadmap, that means we probably won’t see Microsoft’s new chip until 2025.

While neither company disclosed the nature of said silicon, Microsoft unveiled its custom-made Azure Maia AI Accelerator and Azure Cobalt 100 CPU server chips last November, with an expected deployment some time “early” this year to bolster its own AI services. The Cobalt 100 is based on Arm architecture, and coincidentally Intel has been optimizing its 18A process for Arm designs since April last year (it even became an Arm investor later on), so there’s a good chance this collaboration will lead to the next generation of Cobalt CPUs.

In addition to the usual performance improvements with reduced node size, the Intel 18A also offers “the industry’s first back-end power solution,” which according to IEEE Spectrum, separates the power connection layer from the data connection layer on top and moves the former under the silicon substrate – as the name implies. This obviously allows for improved voltage regulation and lower resistance, which in turn allows for faster logic and lower power consumption, especially when applied to 3D stacking.

In Intel’s Q4 earnings call, CEO Pat Gelsinger confirmed that “18A is expected to reach production readiness in the second half of 24.” Given that Intel’s own 18A-based processors — “Clearwater Forest” for servers and “Panther Lake” for clients — won’t arrive until 2025, chances are that will be a similar timeframe for the next chip in Microsoft.

At Intel’s event earlier today, the executive shared an expanded roadmap for Intel Foundry technology, which includes a new 14A (1.4nm) node enabled by ASML’s “High-NA EUV”. (high digital aperture extreme ultraviolet) lithography system. According to AnandTechthis 14A jump could help Intel catch up after its late adoption of EUV for its Intel 4 (7nm) node, although risky production won’t happen until late 2026.

Intel Foundry is the brainchild of Gelsinger, who launched the department immediately after taking over as CEO in February 2021 as part of his ambitious plan to pit Intel against the likes of TSMC and Samsung in the chip manufacturing market. contract. Before Microsoft, Intel Foundry’s client list already includes MediaTek, Qualcomm and Amazon. The company still purposes to become the “second largest external foundry by 2030” in terms of production revenue, which it believes is achievable as early as this year.

This article contains affiliate links; if you click on such a link and make a purchase, we may earn a commission.