3nm

3nm is a semiconductor process node designation used to describe a generation of manufacturing technology intended to produce smaller, more power-efficient integrated circuits. The term is part of industry naming conventions and does not map directly to a single, fixed physical dimension on a chip. In practice, 3nm tooling enables higher transistor density and improved performance or reduced power compared with 5nm processes, using advanced lithography and transistor architectures.

Industry context and developers: The leading semiconductor foundries—Taiwan Semiconductor Manufacturing Company (TSMC) and Samsung Electronics—have developed 3nm-class processes. TSMC markets a 3nm process known as N3, with an enhanced variant N3E. Samsung has announced 3nm offerings for select products. 3nm processes are targeted at high-end applications, including mobile application processors and graphics devices, where performance-per-watt and transistor density offer tangible benefits.

Technology and design: The 3nm family employs advanced lithography, including EUV for critical layers, and uses 3D transistor structures to improve density and electrical performance. Depending on the vendor and variant, transistor architectures may include nanosheet or gate-all-around designs, replacing or augmenting earlier FinFET approaches in some programs. The overarching goals are higher performance and/or lower power while maintaining manufacturability and yield.

Impact and challenges: Adoption has expanded in premium mobile devices, AI accelerators, and high-performance computing. Realizing 3nm benefits requires substantial investment in equipment, supply chains, and design optimization. Yield, variability, and cost challenges accompany the transition from 5nm to 3nm as fabs scale production capabilities.

Notes: The label "3nm" reflects a node designation rather than a precise physical measurement; actual device geometries vary by vendor and product. The 3nm transition signifies a broader industry shift toward more advanced logic and memory processes as manufacturers pursue greater efficiency and performance.