CPUs

Intel Plans to Launch 1.5nm CPUs in 2029, 2nm in 2027 & 3nm in 2025: Backporting to Older Nodes and 11th Gen Rocket Lake Lineup

Intel may be in a sorry state of affairs at the moment, but that doesn’t mean the game is over for Team Blue. Being 10x bigger than AMD does have certain benefits such as maintaining thin profit margins and even staying in a loss for a few years.

Turns out Intel hasn’t given up on Moore’s Law just yet. As per slides revealed by AnandTech, Intel plans to return to its two year Cadence by 2021. This means a new node every two years along with a revamped core architecture as well. Going by this strategy, the company expects:

  • 1.5nm node in 2029
  • 2nm in 2027
  • 3nm in 2025
  • 5nm in 2023
  • 7nm in 2021

Backporting Newer Architectures

Here’s something new. Those rumors about backporting we’ve been hearing, all true. Intel has somehow found a way to backport newer core architectures to a more mature form of the previous one.

This means all the benefits of newer architectures on the same ol’ mature port, without the drawbacks of a new one. This is actually pretty smart. If you look at the new 10nm node that Ice Lake leverages, yields are still a problem. Achieving high-frequencies at reasonable voltages is another challenge. This somewhat mutes the advantages of the new Sunny Cove core, especially in gaming where clock speeds have a strong impact on frame rates.

By backporting the new architecture to the most mature form of the previous node, all this can be avoided. High boost clocks without breaking the power or voltage barrier with all the bells and whistles of the new microarchitecture.

This has technically never been done and if actually works out, it’ll help Intel combat AMD’s Zen microarchitecture without curbing the prices.

11th Gen Intel Rocket Lake: 14nm+++ with new Willow Cove Core Architecture

Now that backporting has been more or less confirmed, it actually makes sense if Intel would backport the Willow Cove core (originally based on 10nm) to the 14nm+++ node. Honestly speaking, the company doesn’t have a lot of options. The Skylake core is more than five years old and there’s a certain limit to which you can push a microarchitecture.

The 10nm process yields are still not good enough for volume production and even if they were, the frequencies aren’t scaling anywhere as much as the 14nm node. Backporting the Willow Cove core to the 14nm process takes care of both the problems. This means that Intel’s 11th Gen lineup might actually put up a fight and at least keep them in the competition, if not win.

I for one, am looking forward to the Rocket Lake-S CPUs and this backporting technology. It’ll be interesting to see if it provides any tangible benefits in the real-world or not.

Tags

Areej

Computer Engineering dropout (3 years), writer, journalist, and amateur poet. I started Techquila while in college to address my hardware passion. Although largely successful, it suffered from many internal weaknesses. Left and now working on Hardware Times, a site purely dedicated to. Processor architectures and in-depth benchmarks. That's what we do here at Hardware Times!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Back to top button
Close