I’m sure you all remember the famous Intel cycle called Tic-Tac. This was to build two successive microarchitectures, a first more powerful and brute, and then refine it by making it more efficient and a little more powerful.
Well, for the first time in more than 6 years, the company has once again named it in its Investor Day presentation after forgetting it for more than five years (curious that they do it just when AMD has shown muscle with its Ryzen 6000).
When exposing its next architectures after the current Intel 7 (Enhanced 10nm SuperFin), the company mentioned its successors, starting with Intel 4 (based on 7nm EUV), which offers electrical properties and transistor densities in the 5nm league from TSMC.
Intel 4 will debut with the Meteor Lake mobile architecture, scheduled for the first half of 2023, and mass production of wafers will begin in the second half of 2022.
The Intel 3 node is planned for a year later, in late 2023, and the server processor that succeeds Sapphire Rapids will be developed for this manufacturing process. Next, Intel, along with other companies, will enter the tricky sub-2nm class.
As for the 20A node (20-angstrom), this is being designed for a specific category of Intel processors planned for the first half of 2024. That same year, the company will debut the Intel 18A node (18 angles).
Intel announces a chip to mine cryptocurrencies 1,000 times faster than graphics cards
Intel has introduced this process under the name Arrow Lake, with product launches expected in 2024 and wafer production in 2023.
Note that Intel has spent the last 5 years going through a lot of problems when it comes to lowering lithography in its processors, getting stuck at 12nm for more than 3 years, while AMD went down without stopping in each generation.
For this reason, let’s take the information from Intel with a grain of salt and wait for these microchips to become a reality. Until then, it’s time to make the most of its Alder Lake generation, one of the best it has produced in the last decade.