In chip manufacturing, the smaller the better. Intel, Samsung, and TSMC are all in an unending race to find ways to reduce the size of their chip nodes while increasing the density of semiconductors, resulting in faster and more energy-efficient processors. than. And right now, TSMC’s customers are lining up to be the first to get their hands on the Taiwanese company’s latest breakthrough.
Accordingly, Taiwan Semiconductor Manufacturing (TSMC), the world’s largest chip manufacturer and top supplier to Apple, has upgraded its production line and promises to bring dramatic improvements to the process. company when starting mass production of 2nm (2N) technology chips at the end of 2025. If you don’t know, the latest iPhone and MacBook models currently use 5nm processors, while the new series uses them. 3nm chips are expected to launch later this year. It’s also not surprising that Apple will be one of the first customers to use the new technology – but there’s still another name: Intel.
You might wonder why Intel, which also has its own chip factories, looks to a rival for supply? According to reports by DigiTimes and UDN, Intel will use TSMC’s 2nm node in its GPUs and other SoCs. The new Intel-designed Arc series of GPUs are intended to use 3nm and 2nm process nodes from TSMC to gain an edge in the battle to regain market share from big names like Nvidia and AMD.
On the Apple side, the Cupertino giant has been TSMC’s biggest customer by revenue over the past decade, with TSMC-manufactured processors now in every phone, tablet, and model. Laptops use M1 chip. The latest chips are using 5nm processors, so it is likely that Apple will launch a bunch of new products before the N2 chip is ready in early 2026. TSMC’s 3nm chip will be available later this year, and so it will likely appear in the iPhone 14 when the handset is introduced this fall.
As for Intel, they believe that it is possible to return to the track after being left too far behind by competitors. CEO Pat Gelsinger emphasized last year that Intel will return to the top by 2025. He also plans products in the following years, in which products using 2nm (20A) nodes are expected to launch in 2024. Next, 18A (codename Lunar Lake), will use TSMC’s 2N for GPU and may launch earlier than planned.
2nm chips could also be available soon thanks to IBM; the company unveiled a 2nm chip with nanosheet technology last year – this is the technology of compressing 50 billion semiconductors into a chip the size of a fingernail. IBM says its 2nm process will be delivered to partner factories – including Samsung and Intel – by the end of 2024, giving Intel a rare lead over TSMC.
Apple and Intel aren’t the only companies looking to TSMC for cutting-edge 5nm N5 chips. According to Tom’s Hardware, MediaTek has announced two 5G smartphone chips – Dimensity 8100 and Dimensity 8000 – using TSMC technology, and Nvidia plans to use the 4N process for dedicated GPU Hopper in data centers, as well as line of graphics cards. Consumer graphics Ada Lovelace (RTX 40-series).
TSMC has not revealed the performance and power saving index of the N2 chip; however, the upcoming 3nm process is said to improve performance by 15% and reduce power consumption by 30%; so we could totally see similar upgrades with next-gen technology.