There is yet another contender for most powerful quantum computer ever. IonQ has announced a new machine with 32 quantum bits, or qubits, that the firm says is more practically powerful than any of its competitors.
Unlike ordinary bits, which can store only a 0 or 1, qubits can store a combination of these states, allowing quantum computers to process far more information. IonQ’s machine doesn’t have more qubits than any other computer, but the firm says its quantum volume – a metric coined by IBM that accounts for the number of qubits, their error rate and a few other key properties – far surpasses that of any other. The firm’s device has a quantum volume of over 4 million, compared with the previous record of just 64.
This quantum volume means IonQ’s computer is closer to being able to solve practical problems. “I think it’s not flimflam, it’s not hype,” says Peter Knight at Imperial College London. “This will be regarded as a tremendous advance.”
Advertisement
IonQ has managed to create a device with such a high quantum volume by trying to make qubits that are as error-free as possible. Rather than using superconducting chips that are similar to ordinary computer circuits, as many other quantum computers do, IonQ uses single ions as qubits and manipulates them with lasers.
“The qubits themselves are perfect, although there are some errors that we add when we use them,” says Chris Monroe, IonQ’s chief scientist. The error rate of the device is less than 1 per cent, he says, making it more accurate than most of the other quantum computers on the market. The qubits are also remarkably well-connected, which means it is easier to run more complex algorithms.
“It doesn’t have more qubits than anybody else, but they’ve reduced their errors in such a way that they’re almost perfect, and that gives them this additional oomph, this additional power,” says Knight. “With a modest number of qubits, they’ve achieved a great deal more.”
But whether IonQ’s device is actually better than its competitors’ isn’t necessarily clear, says John Preskill at the California Institute of Technology. “I’d say the most obvious competitor is Google’s Sycamore device,” he says. In 2019, Google claimed that Sycamore used 53 qubits to solve a problem that couldn’t be solved or simulated in a reasonable amount of time by any classical computer, a milestone called quantum supremacy.
Monroe says that the IonQ device is “sort of on the edge” of achieving quantum supremacy. A classical supercomputer could still solve the problems that this quantum computer can, but it might take weeks to months to do so, he says. “But as long as the quantum computer is faster, it’s still useful.”
“If you want to measure power of a quantum computer in terms of how difficult it is to simulate with a classical computer, it doesn’t smell right to claim that a quantum computer with 32 qubits is more powerful than a 53 qubit device,” says Preskill. It is hard to directly compare the two devices because they are fundamentally different types of quantum computer, but the difficulty of simulating a quantum computer grows exponentially with an increase in the number of qubits, he says.
Monroe says that IonQ is already working on doubling the number of qubits, which will make this device a billion times harder to simulate with a classical computer. Since the company has focused on producing high-quality qubits, even increasing their number by a small amount should provide a significant boost to quantum volume.
