SAN JOSE, July 19 — For decades, Moore’s law has been the quiet engine behind the devices communities across Asia rely on every day — from cheaper smartphones to more powerful laptops. Named after Gordon Moore, co‑founder of Fairchild Semiconductor and Intel, the “law” is actually an empirical observation: the number of transistors stuffed into an integrated circuit roughly doubles every two years with minimal increase in cost. It began with a magazine article.
In 1965, while at Fairchild, Moore wrote “Cramming more components onto integrated circuits” for the 35th anniversary issue of Electronics magazine. He projected that by 1975 it would be possible to pack as many as 65,000 components onto a single quarter‑square‑inch semiconductor.
At the time, component complexity was doubling every year, and Moore predicted that pace would hold for at least a decade. In 1975 he revised the forecast to doubling every two years — a compound annual growth rate of 41% — and that revised rate has held ever since, giving the observation its lasting name.
How it shaped the semiconductor industry
The story has deeper roots. In 1959, Douglas Engelbart presented a study on projected downscaling of integrated circuits at the International Solid‑State Circuits Conference, a session Moore attended. Then, in 1974, IBM’s Robert H.
Dennard formulated what became known as Dennard scaling: as MOS transistors shrink, their power density stays constant. Together, these insights gave chipmakers a roadmap for long‑term research and development planning.
The effects of Moore’s law have been anything but abstract. The steady doubling of transistors has driven down quality‑adjusted microprocessor prices, boosted memory capacity in both RAM and flash storage, sharpened sensors, and pushed up pixel counts in digital cameras. For families across Asia, this has meant access to increasingly capable electronics at lower real cost — a trend that has rippled through productivity, economic growth, and everyday social change.
The debate over its future
Industry experts now disagree on whether the law still holds. Since around 2010, microprocessor architects have reported a slowdown below the pace Moore predicted. In September 2022, Nvidia CEO Jensen Huang declared flatly that Moore’s law is dead.
Intel CEO Pat Gelsinger disagreed. No consensus has emerged; the future remains uncertain.
For communities in developing Asia that have leapfrogged into the digital economy on the back of ever‑cheaper chips, what comes next matters. If the pace stalls, would the sharp declines in device prices stall too? Would memory upgrades and sensor improvements slow?
The answers are still unwritten, but the semiconductor industry continues to use Moore’s law as a planning target — even as its end is debated.
