TSMC announces FinFlex tech for N3 node, nanosheet-based N2 in 2025

TSMC announces FinFlex technology: ultimate performance, power efficiency, density, and flexibility for the future of chips.

Comment IconFacebook IconX IconReddit Icon
Gaming Editor
Published
Updated
3 minutes & 30 seconds read time

Taiwan Semiconductor Manufacturing Co. (TSMC) has just unveiled its updated process technology roadmaps, some new information on the next-gen N3 process nodes, and more at its 2022 TSMC Technology Symposium.

TSMC announces FinFlex tech for N3 node, nanosheet-based N2 in 2025 05

TSMC unveiled its new FinFlex technology for N3, which allows chip designers -- like Apple, AMD, NVIDIA, Qualcomm, and others -- to choose the best options for the key functional blocks on the same die, using the same design tool set.

Godfrey Cheng, Head of Global Marketing, TSMC, explained on the official TSMC website that TSMC's new FinFlex technology: "extends the product performance, power efficiency and density envelope of the 3nm family of semiconductor technologies by allowing chip designers to choose the best option for each of the key functional blocks on the same die using the same design toolset. These options include a 3-2 FIN, 2-2 FIN and 2-1 FIN configuration with the following characteristics".

TSMC announces FinFlex tech for N3 node, nanosheet-based N2 in 2025 03
  • 3-2 FIN - Fastest clock frequencies and highest performance for the most demanding compute needs
  • 2-2 FIN - Efficient Performance, a good balance between performance, power efficiency and density
  • 2-1 FIN - Ultra Power Efficiency, lowest power consumption, lowest leakage and highest density
TSMC announces FinFlex tech for N3 node, nanosheet-based N2 in 2025 04

We're already seeing the start of this with hybrid CPUs, where there are now CPUs with high-performance CPU cores that are matched with power-efficient CPU cores + and then GPU cores and I/O blocks. But now with the introduction of TSMC's new FinFlex in N3 technology, product designers can "choose the best FIN configuration for each of these functional blocks, optimizing each block without affecting others, all on the same die".