In Italy they have a saying that ‘when two men fight, the third wins.’ Proverbs are convenient to have at hand but, pre-packaged as they are, they risk coming off banal and sweeping. This one nicely sums up the story we’re about to tell though. In the great races of technology and business, whoever comes first snatches the biggest slice of the pie and has a good chance of pushing the others out of the running. The race for first place in these industries is frenetic; it’s not just prestige at stake, but prizes on a colossal scale. Our two competitors in the realm of supercomputers are the world’s biggest players in IT: Google and IBM. And the third wheel? Well, let’s begin at the beginning.

In October 2019 Google announced it had attained so-called quantum supremacy, meaning it had finally created the first quantum computer that can do very quickly what traditional computers can’t even do in a reasonable time. Attempts at new computing devices based on quantum architecture have been made for decades now. The best example for understanding them is a maze with only one possible route between the entry at point A and the exit at point B. A traditional computing system like the smartphone you use every day, or Eni’s supercomputer HPC5, tries every fork in the path, taking one turn then the other if the first doesn’t work. It goes on doing that until it finds the right route. But a quantum machine, with its different modus operandi, tests every possible direction simultaneously, saving massively on time and energy. In other words, while a traditional computer has only two separate states, or bits, namely 0 and 1, a quantum computer’s states, known as qubits, can be both numbers at the same time, and almost infinitely. So Google’s machine, christened Sycamore, could beat them all with its ability to do calculations a normal computer can’t in a reasonable time.

Just a few days later IBM published an article disputing Google’s supposed supremacy. The company argued that supercomputers around the world today are more or less capable of keeping up with the calculating speeds announced by the giant in Mountain View. In fact, though, the quantum computers built so far have not been able to overtake traditional machines in their effective calculating capacities. Quantum computing has two intrinsic downsides. First of all, it’s by no means sure that their calculations will get results. In fact the very uncertainty of the process means it’s as likely to fail as to succeed. It goes round in circles, in other words. Secondly, the software needed to run a quantum machine has to be written using a suitable logic, but only another quantum machine can put it together. To borrow another proverb, it’s a chicken and egg situation. We don’t know if real quantum supremacy will ever exist. Every classic machine is perfectly able of doing any calculation. Where the variable lies is in the time it takes to do the operation. Google claims its machine can do in 200 seconds the same operation that would take a traditional supercomputer 10,000 years. IBM contests that, saying its supercomputer Summit could do the operation in under two days, which is a reasonable amount of time. It will be interesting to follow the technological developments, all of them to the benefit of the useful purposes supercomputers are put to. For now traditional computers are used to pick out fields of hydrocarbons, run complex systems, hone new drugs and regulate global financial transactions. Quantum machines are still in the experimental phase and can only do the specific operations they’re designed for.

Now the third wheel comes into play: China. Already leading on other points of technology, the country’s now marching quickly forward when it comes to supercomputers too. They now have at their command a teeming network of supercomputers without global equal in terms of integrated computing capacity. What they plan to do with them is anyone’s guess, apart from some certain examples: a vaccine against Covid-19, ever lighter and stronger batteries, and more sophisticated and better-performing telecommunication systems. We don’t have any news of Chinese quantum computers yet, but it’s not a closed book. It’s all to play for in this game, with at least three teams on a pitch the size of the world.