In December 1964, Gordon Moore, then a director at Fairchild Semiconductors, made a seemingly casual observation that would become the driving force behind the semiconductor industry for over half a century. Speaking to a professional society, Moore predicted that the number of transistors on integrated circuits would double every year. This projection, later dubbed “Moore’s Law,” wasn’t based on strict scientific principle but rather on observed economic and industry trends – a key detail often overlooked.
The Rise of the Microchip
At the time, computers were room-sized behemoths, and integrated circuits (microchips) were still in their infancy. The silicon transistor had only been invented a decade earlier, and the first rudimentary integrated circuits were just beginning to emerge. Moore noticed a clear pattern: transistor counts were doubling rapidly, from 16 in 1961 to 120 by 1964.
This exponential growth wasn’t accidental. Moore’s prediction wasn’t just about what could happen; it was a self-fulfilling prophecy. By setting this expectation, he spurred the industry to innovate relentlessly to meet it. In 1965, he further amplified the idea in an editorial for Electronics magazine, boldly predicting 65,000 components on a single chip – a staggering number for the era. Today, chips contain trillions of transistors.
From Prediction to Industry Standard
Moore co-founded Intel in 1968, and his “law” transformed from a prediction into a core business objective. The pace was later revised to doubling every two years in 1975, a more sustainable rate that nonetheless fueled decades of innovation. This relentless push for miniaturization is the foundation of nearly all modern electronics, from smartphones to supercomputers.
The Limits of Growth
For years, experts predicted Moore’s Law would hit its limits, yet it persisted through ingenious engineering solutions. As Moore himself noted in 2016, “There always seems to be an impenetrable barrier down the road, but as we get closer to it, people come up with solutions.” However, physical reality eventually intervened.
The “law” began to falter around 2016 when Intel struggled to shrink transistor sizes from 14 to 10 nanometers within the predicted timeframe. This breakdown wasn’t a failure of innovation but a consequence of fundamental physics. As transistors shrink, quantum effects like quantum tunneling become unavoidable, causing electrons to leak between transistors and disrupt functionality.
Moore’s Law, in its original form, is no longer viable. While the industry continues to push boundaries, the exponential growth of the past is unsustainable. The era of doubling transistor density every two years is over, as the laws of physics impose real limitations on how small we can go.
