Google is claiming Quantum Supremacy thanks to its Sycamore processor. This is the first time that a quantum computer has been able to outperform a conventional computer.
What has Google achieved?
People have been talking about quantum computing for three decades and a few experiments have been conducted to test the concept. So far though, it’s remained a concept rather than a reality. More science fiction than science fact.
Until now. Google has used a quantum computer based on its Sycamore processor to perform a task in 200 seconds that it claims would have taken 10,000 years for a conventional computer would have taken 10,000 years to perform. It has published the results of its experiment in the Nature journal.
IBM disputes the claim
IBM claims that its Summit supercomputer that Google is basing its 10,000 year claim on could perform the task in 2.5 days rather than 10,000 years. IBM says that Google’s calculations were based on processor performance alone and didn’t take other factors into account such as efficient memory management.
Having said that, even if IBM is correct and its Summit supercomputer could have performed the task in just 2.5 days, Google’s Sycamore processor would still have been 1,000 times faster. This is a major milestone in quantum computing.
What is Google’s Sycamore processor?
The Sycamore processor is a quantum processor. Quantum processors use qubits rather than the bits that conventional processors use. Unlike bits which can have a value of either one or zero, qubits can be both one and zero at the same time.
This means that quantum processors are capable of performing multiple calculations simultaneous which in turn means that theoretically, they should be much quicker than conventional processors. And now Google has proved that they are.
Impressively, although Google’s Sycamore processor contains 54 qubits one of them was not working and so it was switched off. The task was completed using the remaining 53 qubits. That’s the equivalent of winning a race in a car that’s only firing on three of its four cylinders.
The quantum effect on conventional processors
Intel has been shrinking the size of its processors for years. In 1971 when Intel’s original processor – the 4004 – was launched, the gap between the transistors on that processor was 10,000nm. Current Intel processors are based on a 10nm gap and the company is expecting to reduce that gap to 5nm over the next five years.
Smaller transistors are faster and more energy efficient. But there is a limit to how small those transistors can be. Once they get to a certain size, quantum tunneling kicks in. This is where the gaps are so small that the electrons can go where they want rather than where the processor wants them to go.
So quantum computing is not just taking the crown because it performs better. The quantum effect is also making conventional processor development much more difficult.