In the fast-paced world of technology, few concepts have proven to be as influential as Moore’s Law. This observation was originally made by Gordon Moore in 1965, co-founder of Intel, which has guided the exponential growth and progress the digital world has made for over half a century.
Back in 1965, Gordon Moore first made his observation in an article published in Electronics magazine. In the article Moore observed that every year the number of transistors that could fit onto a microchip increased every year. Furthermore, he noted that the number of transistors had approximately doubled the last few years. From this observation he made his original prediction that this trend would continue. Later in 1975, he adjusted his prediction to doubling every two years. This prediction would later come to be known as Moore’s law and has since served as a guide for long-term planning in the semiconductor industry. As an example, for the growth the industry has undertaken, the guidance computer of Apollo 11 contained around 17.000 transistors, while the IPhone 14 pro had a transistor count of 16 billion(!).
Before continuing, it is important to understand how transistors work. As you might know our computer works in binary code, zero’s and one’s. Transistors are used to be either equal to such a zero or a one. Simply put, If a transistor is letting an electrical current through, it equals a one, otherwise it equals a zero. The more transistors are located on a chip, the more complex information it can process and the faster it is able to do it.
While Moore’s Law has been extremely accurate for the past decades, experts predict that Moore’s Law will eventually reach its end. Gordon Moore even stated himself in 2005 that his prediction cannot hold forever, stating that ‘The nature of exponentials is that you push them out and eventually disaster happens.’ The primary reason are the physical constraints that accompany the size of transistors as they approach atomic dimensions.
An effect which occurs at this size of scale, quantum tunneling, makes it immensely more complicated to maintain the same pace of miniaturization and performance improvements. Quantum tunneling occurs when electrons have a finite probability of crossing a barrier, even though classically they do not have enough energy to do so. In the case of transistors, this effect becomes significant since the size of the transistors will only be a few nanometers and thus the barrier’s thickness decreases. This allows electrons to go through the barrier even when they should not, leading to a current going through the transistor resulting in the transistor being equal to a one or zero when they should not.
Luckily, this does not mean that computers will simply stop getting better. Reaching the end of Moore’s Law simply means that we cannot continue making better computers using the same technique. As a result, researchers and engineers are actively looking for new technologies which can help sustain the growth of our computing power beyond the limits of the traditional transistors we use at the moment. Several potential replacements include things like quantum computing and nanoelectronics, but special materials such as graphene1 also have extraordinary electrical properties that could play a major role.
Moore’s Law has been a driving force behind the impressive advancements made in recent history. It has inspired the rapid miniaturization of transistors, leading to increasingly more powerful and efficient computers. However, as we approach the physical limits of our traditional transistors, it opens up new opportunities for innovation to overcome the limitations of traditional transistors.
Footnotes:
1: A revolutionary material discovered in 2004 which is extremely strong, conducts heat and electricity and is one atom in thickness.