Moore’s Law may describe one of the most important drivers of tech over the last few decades – but for some time there has been a view that it is set to come to an end. A new technology advance may give Moore’s Law a new boost.
Moore’s Law was defined by Gordon Moore, the co-founder of Intel in 1964 when he said that the number of transistors on integrated circuits will double every two years for ten years. The saying was adapted, and popular mythology stepped in, to come up with our current understanding of what the law means: computers double in speed every 18 months.
Let’s just say that since that day in 1964 when Gordon Moore made his famous pronouncement, the average family saloon car had seen its top speed increase at a rate commensurate with Moore’s Law, then right now its top speed would be roughly the speed of light.
Moore’s Law matters because when computers double in speed every 18 months, we get used to replacing our machines regularly – and these days we update our smart phones every two years or so.
If computers stopped doubling in speed, then the need to upgrade every so often may go away, and the business model of computer hardware companies, and indeed smart phone companies, and even software companies will be decimated.
And for some time, there has been a view that Moore’s Law is indeed coming to an end. The problems relates to how small you can go. The more transistors you can pack into any given area, the faster the computer. But there is a view that we have reached the limit.
But new hope comes in the form of work carried out by the Department of Energy’s, Lawrence Berkeley National Laboratory. Researchers have managed to create a transistor with a one-nanometre gate. To put that in context, a nanometre is one billionth of a metre.
They achieved this by applying carbon nanotubes and molybdenum disulfide, an inorganic compound with the formula MoS2, which is a little like graphite and used as an engine lubricant.
Sujay Desai, a graduate student on the programme explained: “The semiconductor industry has long assumed that any gate below five nanometers wouldn’t work, so anything below that was not even considered. This research shows that sub-5-nanometer gates should not be discounted. Industry has been squeezing every last bit of capability out of silicon. By changing the material from silicon to MoS2, we can make a transistor with a gate that is just one nanometre in length, and operate it like a switch.”
Professor Javey, the lead principal investigator said: “We made the smallest transistor reported to date. The gate length is considered a defining dimension of the transistor. We demonstrated a 1-nanometer-gate transistor, showing that with the choice of proper materials, there is a lot more room to shrink our electronics.”
It’s work in progress, but these advances could give Moore’s Law a new shot in the arm – albeit, we may see a move away from silicon.
If the technology interests you, electrons flowing through silicon are lighter than those flowing through MoS2, which is normally considered an advantage. But at very small sizes – less than five nanometres – the electrons in silicon become much harder to control. “This means we can’t turn off the transistors. The electrons are out of control,” explained Sujay Desai. With MoS2 this problem is largely solved.
Of course, such advances, exciting though they are, do not provide the only hope that Moore’s Law can continue.
Ray Kurzweil, who these days is a big wig at Google’s parent company, Alphabet, looked at the idea that as Moore’s Law dies, we will see a new Moore’s Law in its place. In a Ted talk, he said that computing devices have been consistently multiplying in power for more than 100 years. He begins with mechanical calculating devices used in the 1890 US Census, moves on to Turing’s relay-based Robinson machine that cracked the Nazi enigma code, then to the CBS vacuum tube computer that predicted the election of Eisenhower, to the transistor-based machines used in the first space launches, to the integrated-circuit-based personal computer that we use today. He said: “Every time one paradigm ran out of steam another one came from left field to continue the growth. When we see the end of the paradigm it creates pressure to create the next paradigm.” He says that when we see the end of Moore’s Law we won’t see the end of computers becoming more advanced.
One option, is that we may move towards quantum computers – which in turn, could lead to a rapid acceleration in the speed with which computers increase in speed.