Image: Intel

As part of today’s Intel Accelerated event for reassuring investors that Intel is poised to return to its former glory and reclaim its product leadership crown in 2025, Gelsinger revealed a brand-new road map that confirms Intel has completely changed its node naming. Instead of the traditional 7 nm and 7+ naming schemes, those products are now called “Intel 4” and “Intel 3,” respectively, while future 5 nm and 5+ products have been renamed to “Intel 20A” and “Intel 18A,” respectively. The “A” in those names refers to angstrom, a new era of semiconductors that will leverage two breakthrough process technologies also revealed by Gelsinger today: “RibbonFET, Intel’s first new transistor architecture in more than a decade, and PowerVia, an industry-first for backside power delivery.” Intel estimates that Intel 20A will ramp in 2024, followed by Intel 18A in 2025.

Image: AnandTech
  • Intel 7 delivers an approximately 10% to 15% performance-per-watt increase versus Intel 10nm SuperFin, based on FinFET transistor optimizations. Intel 7 will be featured in products such as Alder Lake for client in 2021 and Sapphire Rapids for the data center, which is expected to be in production in the first quarter of 2022.
  • Intel 4 fully embraces EUV lithography to print incredibly small features using ultra-short wavelength light. With an approximately 20% performance-per-watt increase, along with area improvements, Intel 4 will be ready for production in the second half of 2022 for products shipping in 2023, including Meteor Lake for client and Granite Rapids for the data center.
  • Intel 3 leverages further FinFET optimizations and increased EUV to deliver an approximately 18% performance-per-watt increase over Intel 4, along with additional area improvements. Intel 3 will be ready to begin manufacturing products in the second half of 2023.
  • Intel 20A ushers in the angstrom era with two breakthrough technologies, RibbonFET and PowerVia. RibbonFET, Intel’s implementation of a gate-all-around transistor, will be the company’s first new transistor architecture since it pioneered FinFET in 2011. The technology delivers faster transistor switching speeds while achieving the same drive current as multiple fins in a smaller footprint. PowerVia is Intel’s unique industry-first implementation of backside power delivery, optimizing signal transmission by eliminating the need for power routing on the front side of the wafer. Intel 20A is expected to ramp in 2024. The company is also excited about the opportunity to partner with Qualcomm using its Intel 20A process technology.
  • 2025 and Beyond: Beyond Intel 20A, Intel 18A is already in development for early 2025 with refinements to RibbonFET that will deliver another major jump in transistor performance. Intel is also working to define, build and deploy next-generation High NA EUV, and expects to receive the first production tool in the industry. Intel is partnering closely with ASML to assure the success of this industry breakthrough beyond the current generation of EUV.
Image: Intel

The industry has long recognized that traditional nanometer-based process node naming stopped matching the actual gate-length metric in 1997. Today, Intel introduced a new naming structure for its process nodes, creating a clear and consistent framework to give customers a more accurate view of process nodes across the industry. This clarity is more important than ever with the launch of Intel Foundry Services. “The innovations unveiled today will not only enable Intel’s product roadmap; they will also be critical for our foundry customers,” Gelsinger said. “The interest in IFS has been strong and I’m thrilled that today we announced our first two major customers. IFS is off to the races!”

Source: Intel (via AnandTech)

Don’t Miss Out on More FPS Review Content!

Our weekly newsletter includes a recap of our reviews and a run down of the most popular tech news that we published.

Join the Conversation

16 Comments

  1. Well I know why Intel is doing this, it’s losing face in the numbering scheme, so if you are losing, just re-invent the whole system and make it work for you!

  2. ON a serious note though, the definition of what a “nm” is has been all over the map for a long time now, definitions of what a nm is measuring vary a lot between manufacturers. For example, TSMC’s 7nm being really what Intel’s 10nm is, and Samsung’s 8nm being closer to what TSMC’s 12nm or 10nm is, and what a half node is, etc…. it’s really all over the place. It’s more complex than that now with FinFET technologies and other things coming down the road like gate all around.

    We do need a strong definition, we need a line in the sand, so to speak, on this measurement and get everyone to adhere to it, that’s the hard part. I think when you look at it, density is probably the more important factor. Maybe that should be the benchmark.

  3. Density is probably as close as we’re going to get. Won’t be perfect, but we’re not getting perfect since there’s no way to compare the same IC ‘blueprint’ across nodes.

  4. It was only a matter of time before Intel joined the “Marketing naming” for process nodes trend. All of their competitors have been doing it for ages.

  5. [QUOTE=”Dan_D, post: 38406, member: 6″]
    This isn’t confusing at all.
    [/QUOTE]
    [QUOTE=”Burticus, post: 38424, member: 297″]
    Clarification via obfuscation.
    [/QUOTE]

    And you expected a descriptive name based on their great history of descriptive CPU names in the core era? :p

  6. Bad marketing – tricking people into buying your crap

    Good marketing – making people feel good about buying your crap

  7. [QUOTE=”Zarathustra, post: 38434, member: 203″]
    And you expected a descriptive name based on their great history of descriptive CPU names in the core era? :p
    [/QUOTE]

    Intel hasn’t known how to name shit since the Pentium IV. One could even argue that their inability goes all the way back to the Pentium’s initial successor.

  8. [QUOTE=”Dan_D, post: 38448, member: 6″]
    Intel hasn’t known how to name **** since the Pentium IV. One could even argue that their inability goes all the way back to the Pentium’s initial successor.
    [/QUOTE]

    I always assumed they intentionally obfuscated CPU names in the core era to make it difficult for consumers to know what they were getting, at least without googling detailed specs.

    A good CPU product name would have something that could easily decode the following directly in the product name:
    – CPU Generation
    – Core count
    – Clock speed (maybe even max turbo)
    – Whether or not it has hyperthreading

    In the beginning you sort of could, As I recall in Nehalem and Sandy Bridge Core i7 meant 4C-8T, Core i5 meant 4C-4T, Core i3 meant 2C/4T and Celeron was 2C/2T.

    But then they went ahead and muddled that like crazy as time went on.

  9. [QUOTE=”Zarathustra, post: 38457, member: 203″]
    I always assumed they intentionally obfuscated CPU names in the core era to make it difficult for consumers to know what they were getting, at least without googling detailed specs.

    A good CPU product name would have something that could easily decode the following directly in the product name:
    – CPU Generation
    – Core count
    – Clock speed (maybe even max turbo)
    – Whether or not it has hyperthreading

    In the beginning you sort of could, As I recall in Nehalem and Sandy Bridge Core i7 meant 4C-8T, Core i5 meant 4C-4T, Core i3 meant 2C/4T and Celeron was 2C/2T.

    But then they went ahead and muddled that like crazy as time went on.
    [/QUOTE]

    Pentium was chosen as a name since the courts told Intel they couldn’t copyright numbers. Pentium and Pentium Pro made sense. Pentium II onward didn’t. Core 2, Core 2 Quad was OK, but Core i3/i5/i7 never did and as you rightfully pointed out, it’s far more confusing now.

  10. [QUOTE=”Zarathustra, post: 38457, member: 203″]
    In the beginning you sort of could, As I recall in Nehalem and Sandy Bridge Core i7 meant 4C-8T, Core i5 meant 4C-4T, Core i3 meant 2C/4T and Celeron was 2C/2T.
    [/QUOTE]

    This is all good and well, the issues come when you have to deviate from that, and for some reason don’t want to use certain numbers.

    AMD is also pretty bad at naming though, they are all over the place especially with their CPU’s and APU’s having same series nr but different gen’s

  11. [QUOTE=”Denpepe, post: 38472, member: 284″]
    This is all good and well, the issues come when you have to deviate from that, and for some reason don’t want to use certain numbers.

    AMD is also pretty bad at naming though, they are all over the place especially with their CPU’s and APU’s having same series nr but different gen’s
    [/QUOTE]

    The solution is this:

    Never ever under any circumstance give marketing what they want. Those nitwits will always muck things up. Set a predictable engineering and stick to it no matter what even if you don’t like how it looks or sounds, no exception.

    Agreed. AMD has sucked at this lately as well. And yes, that mismatch in generation and product numbers has really annoyed me. It seems misleading.

Leave a comment