Taiwan Semiconductor Manufacturing Co. (TSMC) on Thursday commenced its 2022 TSMC Technology Symposium, where the company released its chip development timeline as well as its future plans. The timeline reveals that the Taiwan-based chipmaker will introduce its leading-edge 3nm (N3) chips in the second half of this year and will also unveil the 2nm technology sometime in 2025. The new technologies will be used to make advanced CPUs, GPUs, and SoCs.
The 3nm nodes will come in a total of five 3 nm-class nodes, dubbed N3E, N3P, N3S, and N3X. These N3 variants are claimed to offer improved process windows, higher performance, increased transistor densities, and augmented voltages for ultra-high-performance applications. All these technologies will feature the TSMC’s proprietary FINFLEX architectural innovation that offers great flexibility to chip designers and allows them to precisely optimize performance, power consumption, and costs of Chips.
As AnandTech rightly points out, N3 is made for developed for brands like Apple which could take advantage of the increase in the performance, power, and area (PPA) delivered by leading-edge nodes.
If we talk about the 2nm technology, It will offer a remarkable improvement over the N3, with a 10-15% speed increase at the same power, or a 25-30% power reduction at the same speed, unleashing new standards of Efficient Performance.
N2 will come with the GAAFETs (gate-all-around field-effect transistors) to deliver a full-node improvement in performance and power efficiency. The N2 technology platform will feature a high-performance variant in addition to the mobile compute baseline version and comprehensive chipset integration solutions.
The first N3 chips are expected to enter into production in the coming months and will arrive on the market in Q1 2023. N2 is scheduled to begin production in 2025.
Related:
