In this blog post, we will examine whether the pace of computer technology can continue to accelerate in proportion to the increase in the number of transistors, and explore its limitations and future possibilities.
Humans have developed countless technologies to lead convenient and comfortable lives. These technologies have significantly transformed human society through the Industrial Revolution and the Information Revolution, resulting in a substantial improvement in people’s quality of life. Among these, the development of computers has undoubtedly had the greatest impact on people’s lives. It is now very difficult to find any area of our society where computer technology is not used. Computers are utilized in a wide range of fields, from simple office tasks to finance, transportation, broadcasting, meteorology, and research, and as a result, the speed of task processing in each field has increased dramatically.
The greatest advantage of computers is their processing speed. For example, there is a huge difference in speed between writing a document by hand and creating one on a computer. Another advantage is that computers’ processing speeds can be made even faster. As the size of the various components used in computers has gradually decreased, it has become possible to fit more hardware into a single computer, which has ultimately led to improvements in the computer’s processing speed. So, what principles underpin the computer’s hardware to enable such fast processing speeds? Let’s examine these principles alongside the history of computer development.
The device we now call a “computer” was developed about 70 years ago, in 1950. However, computers at that time used vacuum tubes to perform calculations, and these are referred to as first-generation computers. Due to the nature of vacuum tubes, first-generation computers were very large, generated a lot of heat, and were prone to frequent breakdowns. These early computers were also very expensive to maintain. Subsequently, through two groundbreaking advancements, computers became much more compact and evolved into second-generation computers with faster processing speeds. Among the core technologies of second-generation computers, transistors and integrated circuits had the greatest impact on speed improvements. The integrated circuit was a technology invented by American electrical engineer Jack Kilby in 1958 while working at Texas Instruments. As integrated circuits came to be used in all computers, it became possible to achieve simpler structures and faster processing speeds. Kilby was awarded the 2000 Nobel Prize in Physics for his invention of the integrated circuit.
Following the advent of second-generation computers, computer development progressed hand-in-hand with the advancement of integrated circuits. As the density of integrated circuits increased, computer performance improved, and consequently, the range of tasks computers could perform expanded. The development of integrated circuits led to the miniaturization of CPUs and the emergence of minicomputers suitable for home use. Furthermore, companies such as Intel, Apple, and IBM, which conducted research to enhance CPU performance, developed high-density and ultra-high-density integrated circuits by further increasing the density of integrated circuits. During this process, CPU performance improved exponentially, a phenomenon described by Moore’s Law. Moore’s Law states that the number of transistors in a CPU doubles every two years, and development continues to follow this principle to this day.
A review of the history of computer development leads to the conclusion that a computer’s information processing speed depends on the development of efficient technologies and the continuous improvement of those technologies. Since the second generation of computers, the most efficient component in computers has been the transistor, and increasing the number of transistors has improved processing speed. So why is CPU performance determined by the number of transistors? It is because transistors are the core components of computer operations. The transistors used in integrated circuits such as CPUs are of the MOSFET (Metal Oxide Semiconductor Field-Effect Transistor) type, manufactured by depositing a layer of SiO₂ and a metal layer onto a substrate. Then, p-type or n-type semiconductors are placed on either side. Depending on the type of semiconductor used, they are classified as p-MOSFETs or n-MOSFETs. When a voltage is applied to a transistor, current flows or is blocked, and this can be used to control signals. In actual integrated circuits, c-MOSFETs—which combine two MOSFETs—are used to control electrical signals and form gates. Since these gates are used for computer operations, the more transistors there are, the faster the processing speed naturally becomes.
However, there are limits to reducing the size of transistors to increase their number. This is because the particles that make up matter are atoms, and atoms themselves have a certain size. In fact, starting in the mid-2000s, there was a period when further miniaturization was impossible because transistors failed to function properly when their size was reduced. The solution at that time was to manufacture transistors that were thin and tall, thereby increasing the area through which current flows. This allowed transistor sizes to begin shrinking again, and today they measure approximately 22 nm. Since a single CPU is composed of a dense array of extremely small transistors measured in nanometers, a vast number of transistors can be packed into a CPU, naturally increasing the computer’s processing speed.
Computer technology has advanced at an astonishing pace over the past 70 years. Since the transistor was first developed, steady research and development have led to an exponential increase in computer speed, and research to further reduce their size is still ongoing. Until better technologies are developed and commercialized, computers utilizing transistors will continue to evolve, and our daily lives will become even faster-paced.
