Intel is rethinking how it releases its semiconductor innovations — and brands — CEO Pat Gelsinger announced today on the company’s Intel Accelerated webcast. The announcement outlines the next half-decade of Intel’s processor roadmap, new chip and packaging technologies, and a promise of an “annual cadence of innovation,” with the ultimate goal of helping Intel recapture its leadership in the processor space by 2025.
Future Intel products (which will already start in the upcoming 12th-generation Alder Lake chips later this year) will no longer use the nanometer-based node nomenclature that both they and the rest of the chip industry have been using for years. Instead, Intel is introducing a new naming scheme that it says will provide “a more accurate picture of process nodes across the industry” and how Intel’s products fit into that landscape.
How that works in practice is that those new third-generation 10nm chips will be called “Intel 7”, rather than being given a 10nm-based name (like last year’s 10nm SuperFin chips).
At first glance, it looks a lot like a cheap marketing tactic designed to make Intel’s upcoming 10nm chips look more competitive next to AMD products, which are already on TSMC’s 7nm node, or Apple’s 5nm M1 chips. And while that’s technically true, it’s not as unfair a comparison as it necessarily seems. In modern semiconductors, node names don’t really refer to the size of a transistor on a chip: thanks to advances such as 3D packaging technologies and the physical reality of semiconductor design, that hasn’t been the case since 1997 (as noted by ExtremeTech).
And from a technical perspective, Intel’s 10nm chips are largely comparable to “7nm” branded hardware from competitors such as TSMC or Samsung, which uses similar manufacturing technologies and offers comparable transistor density. That’s something that translates to commercial hardware as well: We’ve already seen that Intel’s current 10nm chips are still competitive with, say, AMD’s cutting-edge 7nm Ryzen chips.
That’s just to say that Intel’s rebranding here isn’t entirely unfair to watch, even if it makes it harder to parse when those bigger “node” change advances happen with the new nomenclature.
Here’s a look at Intel’s new roadmap and what it all really means.
• Intel 7 is the new name for what would have been Intel’s third-generation 10nm technology and the successor to Intel’s 10nm SuperFin (aka Intel’s second-generation 10nm chips, mainly found in 11th-generation Tiger Lake chips). Intel says the new Intel 7 hardware will offer about 10 to 15 percent improvements in performance per watt compared to the previous generation — or, as is always the case, improved power efficiency and battery life if hardware manufacturers prefer to keep performance the same.
The first Intel 7-based products will arrive as early as this year, with the already previewed Alder Lake chips in late 2021 for consumer products and the upcoming Sapphire Rapids chips in 2022 for data centers.
• Intel 4 is the architecture formally known as Intel’s 7nm process, which Intel infamously had to delay until 2023 last summer due to manufacturing issues. Originally slated for 2021, it’s the next big technology leap for Intel, using EUV (Extreme Ultraviolet) technology — something already used by Samsung’s and TSMC’s 5nm node products, for comparison. It will still use the same broad FinFET transistor architecture that Intel has been using since 2011. Thanks to all those improvements, Intel 4 is expected to have a transistor density of about 200-250 million transistors per mm², compared to about 171.30 million transistors per mm² on TSMC’s current 5nm node.
Intel says Intel 4 will provide a performance-per-watt jump of about 20 percent, while reducing overall area. Production is slated for the second half of 2022, with the first Intel 4 products slated for 2023 (Meteor Lake for consumer products and Granite Rapids for data center).
• Intel 3, Ready for production in the second half of 2023, is the new name for what would have been a second-generation 7nm product under Intel’s previous naming scheme. Like Intel 4, it is still a FinFET product, although Intel says it will offer additional optimizations and use of EUV for about 18 percent higher performance per watt compared to Intel 4. No release date or product names for Intel 3 chips already announced, but presumably they won’t be available until 2024.
- Intel 20A is the name for the next generation of Intel technologies that, under the old scheme, would have been the architecture that followed the previously branded 7nm node. It’s also the most substantial announcement Intel has made today, technologically speaking, one that will see Intel debut its first new transistor architecture since FinFET in 2011, dubbed “RibbonFET”. The new architecture will be Intel’s first gate all-round transistor, a fundamentally new transistor technology for the company that promises greater transistor density and smaller size. In addition, 20A will see the introduction of “PowerVia,” a new technology that allows wafers to be fed from the back of the chip, rather than having to round up power to the front.
The “20A” in the title is intended to evoke the “Ångstrom era” of semiconductor design – an ngstrom is a unit of measurement smaller than nanometers. (20Å = 2nm, although, like the other renamed Intel names above, Intel 20A does not refer to any specific measurement on the products themselves.)
Intel’s 20A isn’t expected to rise until 2024, and like Intel 3, it doesn’t have a formally announced release date or products yet.
- Intel 18A is the furthest future from Intel’s roadmap and will feature the second generation of Intel’s RibbotFET technology for “another giant leap in transistor performance.” Intel says Intel 18A is in development for “early 2025,” and that it expects this generation of technology to restore its leadership in semiconductors.
In addition to all the process roadmap news, Intel also announced two major updates to its Foveros chip stack packaging technologies (the second generation of which will debut in Intel 4’s Meteor Lake in 2023). Foveros chip stack combines several hardware elements in a single die, such as Intel’s Lakefield chips, which merge five CPU cores, an integrated GPU and DRAM into a compact stack to save internal space compared to a traditional design.
Foveros Omni allows for more variety in stacked chips by making it easier to mix and match tiles regardless of their specific size – for example, by allowing a base tile that is smaller than the top tile in a stack. And Foveros Direct provides direct copper-to-copper bonding between components, reducing drag and reducing impact. Both new Foveros technologies are slated for production in 2023.
Intel’s new names may help the company recontextualize its current and future products more accurately against the competition, but the fact remains that Intel is behind. Even if we accept that the Intel 7 is on par with 7nm products from other foundries, those foundries are already past their 7nm chips and on 5nm hardware. Which means the companies that rely on those outside foundries — like Apple, AMD, Nvidia, Qualcomm, and pretty much every other big tech company — can still get chips that are more advanced than Intel’s best work. Apple’s superlative M1 Macs, for example, already use TSMC’s 5nm chips – handily outperforming Intel’s comparable products. AMD is rumored to work on 5nm Zen 4 processors as early as 2022, which could provide similar competition to Intel from its already advancing competitor.
Even with the ambitious annual cadence for its roadmap, Intel is playing from behind; it doesn’t expect to fully catch up with the rest of the industry until Intel 20A in 2024. And it doesn’t expect to regain leadership in the semiconductor industry until 2025 with Intel 18A. And everything That assumes Intel will no longer experience delays or production issues like the ones that held up both the 10nm and 7nm processes (which put the company in its current situation in the first place).
However, after years of setbacks, it is clear that the revitalized Intel is not going down without a fight. But the coming years will show whether the efforts are sufficient.