China's quantum computer 1 million times more powerful than competitors
The closest competitor is Google's Sycamore quantum computer, with 55 qubits
A team of Chinese researchers claims to have built two different types of quantum computers that outperform their nearest competitors in terms of speed, and can perform calculations that are impossible for non-quantum computers.
Quantum computers work in a fundamentally different way than classic computers. They use qubits (quantum bits) as the basic building blocks of computing. Unlike regular binary digits (bits) that are used to store a 0 or 1 on everyday computing machines, qubits can store a combination of both through superposition.
The research team, led by quantum physicist Pan Jianwei from the University of Science and Technology of China (USTC), said they have designed a 66-qubit programmable superconducting quantum computer that is about 10 million times faster than the world ' s fastest digital supercomputer, and a million times more powerful than Google ' s 55-qubit Sycamore quantum processor.
Researches have named the system 'Zuchongzhi 2.1' after the noted 5th century Chinese mathematician and astronomer.
Zuchongzhi 2.1 is an upgraded version of ZuChongzhi - a 62-qubit programmable superconducting quantum prototype that Chinese researchers unveiled in May.
In 2019, Google said its Sycamore processor was able to achieve 'quantum supremacy' for the first time, surpassing the performance of conventional devices. Sycamore performed a specific task in 200 seconds that would take the world's best supercomputer nearly 10,000 years to complete.
In addition to Zuchongzhi 2.1, the Chinese team also claims to have built a new light-based photonic quantum computer, named 'Jiuzhang 2.0', that can perform calculations 100 trillion times faster than the world ' s fastest existing supercomputer, and large-scale Gaussian boson sampling (GBS) 1 septillion times faster.
The researchers used Zuchongzhi 2.1 and Jiuzhang 2.0 in experiments to calculate the probability that a specific input configuration may lead to a particular output configuration.
Such probability calculations are not feasible for classic computers.
The researchers said that for a system with 1,043 possible outcomes, Jiuzhang 2.0 can sample the output 1,024 times faster than a 'standard' supercomputer.
They also noted that a sampling calculation using Zuchongzhi 2.1 is about 1,000 times more difficult to perform on a classical computer.
"We estimate that the sampling task finished by Zuchongzhi in about 1.2 h [hours] will take the most powerful supercomputer at least 8 yr [years]," the scientists wrote in the study paper.
"This indicates that our research has entered its second stage to start realising fault-tolerating quantum computing and near-term applications such as quantum machine learning and quantum chemistry," the study ' s co-author Zhu Xiaobo told Chinese state media.
The team ' s findings are detailed in two papers published in the academic journal Physical Review Letters.