Moore's Law: How Long Will The Trend Continue?

Moore's Law, an empirical relationship, states that the number of transistors on a microchip doubles about every two years with minimal cost increase. It is not an actual law but an observation made by Intel co-founder Gordon Moore in 1965. The law has guided the semiconductor industry in long-term planning and setting targets for research and development. However, there is no consensus on when Moore's Law will cease to apply. While some believe that physical limits will be reached in the 2020s, others argue that advancements in technology and materials will extend its applicability.

The physical limits of Moore's Law

Moore's Law, an empirical relationship, is based on the observation that the number of transistors in an integrated circuit (IC) doubles about every two years. However, it is not a law of physics, and there are physical limits to how long it can continue.

The law, created by Gordon Moore in 1965, was initially based on the idea that the complexity for minimum component costs would increase at a rate of roughly a factor of two per year. However, this was later revised to every two years, as the complexity of components increased.

The problem for chip designers is that Moore's Law depends on transistors shrinking, and eventually, the laws of physics will intervene. For example, electron tunnelling prevents the length of a gate—the part of a transistor that turns the flow of electrons on or off—from being smaller than 5 nm. Additionally, as the number of transistors on a chip increases, so does the heat produced, increasing the chance of malfunction.

Other physical limits to transistor scaling include source-to-drain leakage, limited gate metals, and limited options for channel material. These factors mean that other approaches are being investigated, which do not rely on physical scaling. For instance, spin-based logic and memory options are being developed in labs, and there is potential in using alternative materials for transistors.

While Moore's Law has been a driving force of technological and social change, productivity, and economic growth, it is unlikely to continue indefinitely. Gordon Moore himself predicted that it would reach a physical limit, probably by around 2025. However, this does not mean that technological progress will come to a halt. There are several reasons to be optimistic about the future of computing, even as Moore's Law slows or ends.

The future of computer chips

The end of Moore's Law?

Some industry experts argue that Moore's Law is already coming to an end or will do so in the near future. In 2013, Robert Colwell, the director of the microsystems group at the Defense Advanced Research Projects Agency, predicted that Moore's Law would end in 2020 or 2022 as chip manufacturers reached the physical limits of transistor size. Colwell's prediction was based on the belief that the exponential growth curve of Moore's Law was unsustainable and that economic factors, rather than technological ones, would ultimately bring it to an end.

In 2022, Nvidia CEO Jensen Huang declared Moore's Law dead, citing the increasing costs and technical challenges of keeping up with the law's predictions. However, Intel CEO Pat Gelsinger disagreed, stating that Moore's Law was still alive and well. This disagreement among industry leaders highlights the ongoing debate and uncertainty surrounding the future of Moore's Law.

The impact of Moore's Law on technology

There is no denying that Moore's Law has had a significant impact on the technology industry. The law has guided long-term planning and set targets for research and development, leading to advancements in digital electronics such as reduced microprocessor prices, increased memory capacity, improved sensors, and enhanced digital cameras. These changes have driven technological and social progress, improved productivity, and fuelled economic growth. As a result, Moore's Law has touched almost every facet of modern life, from smartphones and tablets to transportation and healthcare.

While Moore's Law may be approaching its natural end, it doesn't mean that technological progress will come to a halt. There are several avenues that researchers and engineers are exploring to continue advancing computer chip technology:

• Specialised chips: One way to overcome the slowing of advances in general-purpose chips is to develop more specialised processors, such as graphics processing units (GPUs) and custom processors for neural networks, computer vision, voice recognition, and the Internet of Things devices.
• Cloud computing: With cloud computing, the heavy lifting of computational tasks can be carried out in large data centres, utilising the power of many times the number of transistors in a regular single computer. This allows machines to become exponentially smarter without constantly upgrading their processors.
• New materials and configurations: Researchers are investigating the use of materials other than silicon for future chips, such as elements from the third and fifth columns of the periodic table that offer better conductivity. Additionally, new configurations, such as 3D patterns, are being explored to pack more transistors onto a circuit board.
• Quantum computing: Quantum computers are an experimental and expensive technology that deals with quantum bits, which can be 0, 1, or both 0 and 1 at the same time. This superposition of states could make quantum computers much faster and more efficient than classical computers.
• Other innovations: History has shown that predicting the future of technology is challenging, and there may be innovations in the coming decades that we cannot even imagine today.

While the future of computer chips beyond Moore's Law may be uncertain, it is clear that the quest for faster, smaller, and more efficient computing will continue to drive innovation and shape our world in ways we can only begin to imagine.

The impact of Moore's Law ending

Moore's Law is an empirical observation that the number of transistors in an integrated circuit (IC) doubles about every two years. It is not a law of physics but rather an empirical relationship. It is named after Gordon Moore, the co-founder of Fairchild Semiconductor and former CEO of Intel, who in 1965 noted that the number of components per integrated circuit had been doubling every year.

Moore's Law has been used in the semiconductor industry to guide long-term planning and set targets for research and development. Advancements in digital electronics, such as the reduction in quality-adjusted microprocessor prices, the increase in memory capacity, and the improvement of sensors, are strongly linked to Moore's Law. These changes in digital electronics have been a driving force of technological and social change, productivity, and economic growth.

However, there are physical limits to the ability to continually shrink the size of components on a chip. As chips approach the atomic level, engineers will face challenges in manufacturing and cost. Many people predict that Moore's Law will end in the early 2020s, when components on chips are only around 5 nanometers apart.

The end of Moore's Law does not mean the end of technological progress. Here are some potential impacts and outcomes:

• Slower rate of improvement: The speed of improvements in technology will likely slow down, but it will not come to a complete stop.
• Better algorithms and software: There will be a higher priority on squeezing more performance out of existing chips, leading to the creation of better algorithms and more elegant software.
• More specialized chips: Chip designers will focus on creating more specialized processors, such as graphics processing units (GPUs), custom specialized processors for neural networks, computer vision for self-driving cars, voice recognition, and Internet of Things devices.
• Cloud computing: With cloud computing, a lot of the heavy lifting for big computational problems can be carried out in large data centers, making machines exponentially smarter without having to change their processors every 18 months or so.
• New materials and configurations: Researchers are investigating future chips that could be made of materials other than silicon. For example, Intel is experimenting with 3D transistor configurations. Other materials, such as those based on elements from the third and fifth columns of the periodic table, could be used due to their better conductivity.
• Quantum computing: Quantum computers are a different type of computer that deals with quantum bits, which can be 0, 1, or both 0 and 1 at the same time. This superposition could make quantum computers much faster and more efficient than current mainstream computers. However, there are challenges to making quantum computers a reality, such as the need to keep them incredibly cold.
• Unforeseen innovations: Just as few people predicted smartphones, Google, or Amazon in the 1980s and 1990s, we cannot know exactly what innovations will arise in the future of computing. Computers in a few decades may look entirely different from those we use today.

While the end of Moore's Law may bring challenges, there are also opportunities for creativity and innovation. The semiconductor industry will need to adapt and find new ways to drive progress and improve technology.

Predictions for the end of Moore's Law

Moore's Law, which states that the number of transistors on a microchip doubles about every two years, is expected to reach its physical limits in the 2020s. This is because the components on a chip cannot be made smaller than the size of an atom, and there is only 1.5nm of space left to print on, depending on the element.

In 2005, Gordon Moore, the law's namesake, predicted that the law would eventually end, saying:

> It can't continue forever. The nature of exponentials is that you push them out and eventually disaster happens... We're pushing up against some fairly fundamental limits so one of these days we're going to have to stop making things smaller.

Some experts believe that Moore's Law has already ended, citing the increasing challenges and costs of making transistors smaller. For instance, the cost of a fabrication plant (fab) is rising at around 13% a year and is expected to reach $16 billion by 2022. The number of companies with plans to make the next generation of chips has shrunk to three, down from eight in 2010 and 25 in 2002.

However, some companies, such as Intel, remain optimistic about extending Moore's Law. In 2024, Intel began receiving parts for a machine that can create technology that "pushes Moore's law forward". This machine, a High NA Extreme Ultraviolet Lithography system, can print transistors as small as 2nm.

While there is no consensus on when Moore's Law will end, its impact on technological and social change, productivity, and economic growth has been significant. The end of Moore's Law will likely lead to a shift in focus to areas such as new chip architectures, quantum computing, and AI and machine learning to drive technological progress.

Alternatives to Moore's Law

While Moore's Law has been a guiding principle for the semiconductor industry, it is not a law of physics and has its limitations. As the pace of semiconductor advancement slows, industry experts predict that Moore's Law will cease to apply beyond 2025. However, this does not signal the end of technological progress. Here are some alternatives and emerging technologies that will shape the future beyond Moore's Law:

• 3D Computing: Intel introduced its 3D chip technology, stacking chips vertically instead of placing them side by side. This approach allows for a significant increase in transistor density and improves the integration of chip functions. As a result, 3D chips can offer much higher speed and power in a compact space and are expected to be up to 1,000 times faster than current chips.
• DNA Computing: DNA computing leverages the incredible information storage capacity of DNA. A thumbnail-sized DNA computer could, in theory, be more powerful than today's supercomputers. While still in its early stages, DNA computing has the potential to revolutionize computing.
• Quantum Computing: Quantum computers use quantum bits or qubits, which offer exponentially greater computational power than traditional bits. A small quantum computer could, in theory, exceed the power of a conventional computer the size of the Milky Way galaxy. Quantum computing could be the key to achieving ambitious goals like controlling the weather or colonizing Mars.
• Specialized Chips: As Moore's Law slows, chipmakers are focusing on developing specialized processors for specific applications such as graphics processing units (GPUs) and custom processors for neural networks, computer vision, voice recognition, and the Internet of Things (IoT) devices.
• Cloud Computing: With cloud computing, computationally intensive tasks can be offloaded to massive data centers, leveraging the power of many more transistors than a single computer. This approach allows devices to become exponentially smarter without frequently upgrading their processors.
• New Materials and Configurations: Researchers are exploring alternatives to silicon, such as elements from the third and fifth columns of the periodic table, which offer better conductivity. Additionally, new configurations like 3D patterns for transistors are being investigated to pack more transistors onto circuit boards.
• Yule's Law of Complementarity: This law states that when two attributes or products are complements, the demand for one complement is inversely related to the price of the other. For example, reducing the price of printers increases the demand for ink cartridges. This law can guide strategic pricing and product development.
• Hoff's Law of Scalability: This law suggests that the potential for scalability is inversely proportional to customization and directly proportional to standardization. In other words, highly customized products are less scalable, while standardized products are more scalable. This law has been applied successfully in the automobile and computer industries.
• Evans's Law of Modularity: This law advocates for modularization to simplify complex and incompatible technology structures. By creating modular components with common interfaces and standards, companies can make their products more flexible and protect their customers' investments. This approach has been successfully applied in software development, leading to more efficient and responsive methodologies like agile development.
• The Law of Digitiplication: This law states that digitizing a resource or process increases its potential value in a multiplicative manner. For example, digitizing a retail store allows many customers to access its products and services simultaneously, and additional features like search functionality further enhance the customer experience. Digitization creates new opportunities for value creation.

Frequently asked questions