Quantum computing promises to conquer challenges out of reach of the conventional technology that powers smartwatches and supercomputers.
But quantum computing fans also hope to bring new tools to AI, logistics, and finance.
With a slowing Moore’s Law constraining conventional computing, that could mean important new progress.
IBM actually announced two new Osprey systems.
The first is fully tested, said Jay Gambetta, vice president of IBM’s quantum computing work.
All the qubits are good," Gambetta said.
“It’s another confirmation of the road map as we build larger and larger devices.”
The second incorporates improvements made to Osprey’s predecessor, Eagle, that extends the lifespan of calculations.
That’s long enough for hundreds of calculation steps.
IBM’s supercomputers, like those of Google and Rigetti, are called superconducting quantum computers.
They rely on very cold temperatures just a fraction of a degree above absolute zero and colder than space.
And they’re housed in increasingly large and expensive cylindrical refrigerators.
Now it’s got a new method, though, ribbons of communication links that are more compact.
“We solved a bottleneck in scale,” Gambetta said.
More scaling is on the horizon.
For 2023, IBM plans to build 1,121-qubit Condor.
After that comes 1,386-qubit Flamingo in 2024 and 4,158-qubit Kookaburra in 2025.
Correction at 8:32 a.m. PT: The story misstated the temperature at which IBM’s computer operates.
It runs at a fraction of a degree above absolute zero.
