Intel had another disastrous quarter, but the chipmaker remains optimistic about its IDM 2.0 plans. During its Q4 2022 earnings call, Intel CEO Pat Gelsinger and CFO David Zinsner reiterated that the chipmaker is on track to mass produce its advanced process nodes through 2023 to 2025. Team Blue promises a volume ramp of five EUV nodes in the next four years, namely Intel 4, Intel 3, 20A, and 18A.
The 14th Gen Meteor Lake family leveraging the Intel 4 process is ready for mass production, and the volume ramp is planned for the year’s second half. While the 4nm node will power the next-gen Core lineup, Intel 3 will be used to fab the 6th Gen Xeon Granite Rapids-SP and Sierra Forest cloud processors later the same year.
Looking ahead to the future, the 20A (2nm) and 18A (1.8nm) nodes are making good progress with the engineering release already in the hands of lead customers. The test chips for both process nodes have been taped out, and the final production release is expected in the next few weeks.
With 4nm (i4) Meteor Lake and 3nm (i3) Granite Rapids set for a late 2024 launch, the next major product on Intel’s roadmap is Lunar Lake. Allegedly based on the Lion Cove (Royal Cove) core architecture designed by Jim Keller, it will be a mobile-centric platform focusing on power efficiency and all-day battery life.
The 15th Gen Lunar Lake family is on track for “production readiness” in 2024, with the first designs already taped out. It’ll keep the P-core count unchanged or perhaps reduce it from six to four and squeeze in more E-cores. Expect a wide range of ultra-thins and convertibles in the future.
Intel plans to achieve “process performance parity” in 2024 and “unquestioned leadership” by 2025 with the 18A or 1.8nm process node. On the client side, we have the first chiplet lineup in the form of Meteor Lake later this year, followed by Lunar Lake in 2024. The data center segment will see the launch of Emerald Rapids, a Sapphire Rapids refresh, towards the end of the year, and Granite Rapids in 2024.
Source: SeekingAlpha (Via: @Harukaze)