The so-called “Moores Law” is named after Gordon E. Moore, who in 1965 formulated an empirical observation about the number of transistors on a given area of silicon. This became known as “Moores law”. The law predicted that this number would double every two years and has been proved correct for more than 50 years. Transistor counts are one way to measure the progress of semiconductor technology. They are also used as an indicator of how much processing power can be squeezed into a given space, which is important for everything from personal computers to mobile phones. In general, the more transistors you can fit onto a given area of silicon, the more powerful and efficient your chips will be.
The history of Moores law is often told as a story of ever-increasing transistor counts. But this focus on transistor counts can be misleading. It gives the impression that chip designers are simply finding ways to pack ever more transistors onto their chips. In reality, they are doing much more than that. They are constantly finding ways to improve the performance of those transistors and to reduce their cost per unit of performance. As transistor sizes have shrunk, chip designers have been able to pack ever more of them onto a single piece of silicon. At the same time, they have continued to find ways to make those transistors faster and less power hungry.
All these improvements in transistor performance have had a profound impact on computing devices and applications. Over the past 50 years, we have seen desktop computers become smaller and more powerful; laptops become thinner, lighter and cheaper; and smartphones become ubiquitous worldwide. All these changes would not have been possible without continual improvements in semiconductor technology – specifically, in transistor performance. And all those improvements can be traced back to Moores law.