What Is Moore's Law?

Moore's Law is the empirical observation that the number of transistors on an integrated circuit doubles approximately every two years, accompanied by a proportional reduction in cost per transistor. Though often described as a law, it is not a physical constraint but a projection that shaped decades of semiconductor industry planning. The observation has served as both a benchmark and a target: chip manufacturers organized their research roadmaps around it, and the competitive pressure to meet it became self-reinforcing, driving the sustained miniaturization of transistors from thousands per chip in the 1960s to tens of billions per chip in contemporary processors.

Gordon Moore, then Director of Research and Development at Fairchild Semiconductor, published the original formulation in Electronics Magazine on April 19, 1965, in a paper titled "Cramming more components onto integrated circuits." As documented by the Computer History Museum, Moore drew a trend line through five years of data from 1959 to 1964 and projected that component density would reach 65,000 per chip by 1975, following a doubling roughly every twelve months. In 1975, after the projection had been validated, he revised the doubling period to approximately two years, the rate most commonly associated with the law today.

Origins and Industry Adoption

The physicist Carver Mead coined the phrase "Moore's Law" in the late 1970s, giving the observation the name by which it became a guiding principle of the semiconductor industry. The International Technology Roadmap for Semiconductors, a coordination document used by chipmakers worldwide, explicitly referenced the doubling cadence as a planning horizon for process node development. Intel, TSMC, Samsung, and other manufacturers oriented their capital investment cycles around delivering new process nodes on roughly two-year intervals. Research published in PLOS One on Intel chip density from 1971 to 2020 finds that transistor density followed a biphasic sigmoidal pattern, with rapid growth phases alternating with periods of near-stagnation, rather than a smooth exponential curve, suggesting the law describes an industry target as much as an underlying physical regularity.

Scaling Mechanisms and Process Technology

The primary mechanism sustaining Moore's Law was photolithographic scaling: the wavelength of light used to pattern transistor features shrank from ultraviolet to deep ultraviolet to extreme ultraviolet (EUV) as nodes progressed from micrometers to nanometers. Each node reduction required new materials, new photoresist chemistries, and new transistor architectures. The shift from planar transistors to FinFET three-dimensional gate structures at the 22-nanometer node in 2011 was a pivotal architectural change driven by the need to control electron leakage as gate oxides thinned to only a few atomic layers. Gate-all-around nanosheet transistors represent the next structural iteration, surrounding the channel on all four sides to maintain electrostatic control at 2-nanometer and smaller dimensions.

Physical Limits and Post-Moore Computing

As transistor dimensions approach atomic scales, classical geometric scaling faces hard physical boundaries. Quantum tunneling allows electrons to traverse gate dielectrics even when the gate is switched off, raising static power dissipation and limiting how thin the insulating layer can be. Heat density in dense chips has grown to levels that constrain sustained operating frequencies. The industry response has shifted from transistor count per unit area to system-level performance per watt, achieved through chiplet integration, three-dimensional stacking, and domain-specific accelerators designed for artificial intelligence, signal processing, or cryptographic workloads. The Intel Moore's Law press kit notes that advanced packaging now allows multiple chiplets fabricated at different process nodes to be combined into a single package, effectively extending the density trajectory through integration rather than pure lithographic scaling.

Applications

Moore's Law has shaped a range of fields, including:

• Consumer electronics design, enabling increasingly capable smartphones and laptops
• Data center infrastructure, where cost-per-computation determines cloud service economics
• Medical imaging and genomics, where processing density enables real-time analysis
• Telecommunications, where baseband processors handle increasingly complex modulation schemes
• Artificial intelligence hardware, where GPU and NPU designs depend on sustained transistor density growth