In Memory Of Gordon Moore: Reflections On The Eventual End Of Moore’s Law – Analysis

By He Jun

On March 24, Intel and the Gordon and Betty Moore Foundation announced on the official website of Intel that Gordon Moore, the co-founder of the company, passed away peacefully at his home in Hawaii at the age of 94.

Moore was born in San Francisco, California, on January 3, 1929, and received his Ph.D. in chemistry in 1954. In 1968, he co-founded Intel with his colleague Robert Noyce and served as its executive vice president until he became the president in 1975. In April 1979, Moore became chairman and chief executive officer, holding that position until April 1987, when he became chairman. He was named chairman emeritus in 1997 and stepped down in 2006.

Moore was a pioneer in the semiconductor industry and is known for his famous “Moore’s Law”. In 1965, he wrote in an article that due to technological progress, the number of transistors on a semiconductor chip doubled every 18 to 24 months since the invention of integrated circuits a few years earlier. In other words, the performance of processors approximately doubles every two years, while the price is reduced by half. This prediction, based on observations, became known as Moore’s Law. It is in reality not a mathematical or physical law, but rather a prediction of the trend in the development of the integrated circuit industry, summarizing the speed of information technology progress as an empirical rule.

The importance of Moore’s Law for the global semiconductor industry lies in its role as a guiding principle for the industry’s development and competition. It prompted Intel and other chip manufacturers to actively invest in research and development resources to ensure the continued progress of the industry.

In his 1965 paper, Moore also predicted the future, saying, “Integrated circuits will lead to such wonders as home computers – or at least terminals connected to a central computer – automatic controls for automobiles, and personal portable communications equipment”. This prediction was 20 years earlier than the personal computer revolution and more than 40 years earlier than Apple’s iPhone. In an interview in 2005, Moore said, “It sure is nice to be at the right place at the right time. I was very fortunate to get into the semiconductor industry in its infancy. And I had an opportunity to grow from the time where we couldn’t make a single silicon transistor to the time where we put 1.7 billion of them on one chip! It’s been a phenomenal ride…”. He stated in 2008, “All I was trying to do was get that message across, that by putting more and more stuff on a chip we were going to make all electronics cheaper”.

During the more than 50 years of the application of Moore’s Law, Moore’s extraordinary insights and vision have become a reality. Over half a century, chips have become more efficient and cheaper at an exponential rate, driving most of the world’s technological progress. Computers have gone from mysterious and unreliable behemoths to tools that most people cannot do without. Information technology has moved from the lab into countless ordinary households, and the Internet has connected the world where multimedia devices have enriched everyone’s lives. Based on integrated circuits, Internet giants such as Apple, Facebook, and Google have also emerged. In China, after entering the 21st century, its information industry has developed rapidly, giving birth to technology giants such as Huawei, Tencent, and Alibaba, whose development is highly dependent on the development of the semiconductor industry.

Intel Chairman Frank Yeary considers Moore to be a brilliant scientist and a leading American entrepreneur. “It is impossible to imagine the world we live in today, with computing so essential to our lives, without the contributions of Gordon Moore,” he said. Pat Gelsinger, Intel CEO, said, “Gordon Moore defined the technology industry through his insight and vision. He was instrumental in revealing the power of transistors, and inspired technologists and entrepreneurs across the decades. Such views are indeed accurate. As a trailblazer in the semiconductor industry, Moore’s contributions have greatly influenced the world of technology.

Following the passing of Moore, the question of when Moore’s Law will come to a close has resurfaced. It is logical that Moore’s Law, which describes the development trend of the integrated circuit industry, will eventually reach its limit. The crucial factor is determining when that limit will be reached. Moore himself predicted in 2015 that it would cease to be effective within a decade, specifically by 2025. Presently, the industry’s view on the limit of Moore’s Law is slightly more optimistic than Moore’s, with some estimating that it will reach the end by roughly 2030.

Researchers at ANBOUND have analyzed the development characteristics of the integrated circuit industry and industry opinions, concluding that there are six reasons behind the eventual obsolescence of Moore’s Law: (1) physical limitations, as the chip size will eventually encounter a physical limit; (2) engineering limitations, as chip manufacturing technology and processes will reach their limits; (3) heat dissipation, since as more transistors are integrated into a chip, more heat will be generated, presenting challenges for cooling; (4) the quantum tunneling effect, which occurs when the chip process reaches 1nm and causes a phenomenon known as “current leakage” within the industry; (5) low power consumption requirements, as terminal devices require increasingly efficient energy use; and (6) economic factors, since as chips become smaller, their production costs increase and the marginal benefit of producing and using advanced chips decreases, making it difficult for the market to sustain.

When Moore’s Law reaches its limit, how will the silicon-based semiconductor industry develop in the future? Perhaps there will be new choices in semiconductor materials, or breakthroughs will emerge in other technological paths such as quantum computers, photon computers, and biological computers. Even for silicon-based chips, innovation can be considered in the field of chip applications, shifting the focus from continuously upgrading chip processes to how to better use existing chips.

The American entrepreneur Steve Blank, sometimes known as the Godfather of Silicon Valley, once said the limit of Moore’s Law provides a direction for thinking. We should not focus on “how well we can develop chips,” but rather on “what we can do with such good chips”. What we are facing now is “excessive performance”, and we need to give more thought to how to make full use of chips as the direction of future innovation. Although Moore’s Law will become ineffective, there is still significant room for innovation in new computer architectures, chip packaging technologies, and low-power memory. In addition, hardware innovation has not stopped. In fields such as virtual reality, display technology, sensors, and wearable devices, the technology has yet to reach the point where it can match current chips. Therefore, thinking about what can be done with chips may be more meaningful than trying to increase chip performance.

Final analysis conclusion:

The passing of Gordon Moore and the end of Moore’s Law mark an important milestone in the development of the global semiconductor industry. It signifies that the industry has lost a great pioneer and that a glorious era has come to an end. However, this does not mean that the development of the semiconductor industry has come to a halt, but rather that it is brewing new innovations. The exact time frame for these innovations to materialize is uncertain, however, they are bound to occur sooner or later, given the inevitability of progress in the semiconductor industry.

He Jun is a researcher at ANBOUND