Microsoft in recent years made a big push to bring attention to themselves in the quantum space. With Google laying claim to quantum supremacy in October last year, we think the topic’s worth another State of Affairs update.
If you look here: Scorecard for Qubit Count Microsoft is actually near the bottom in the quantum computing field right now. Their devices are just simulated, while Google and IBM are on top with actual quantum computing devices. Intel and Rigetti are honorable mentions.
IBM offers a real quantum computer available for testing via Cloud>. Everyone else just offers quantum simulators to the public.
The company D-Wave, popularized in media, has many more qubits than Google or IBM, but D-Wave processors are not general purpose quantum circuits. They are annealing processors. You need 100x as many qubits in annealers to match the performance of the more general quantum gate processors. This is why Google and IBM remain in the lead, even though D-Wave’s systems have 2048 qubits (soon over 5000) as opposed to Google and IBM’s 53.
Google’s Hartmut Neven predits doubly-exponential growth in qubits every couple years. Our next quantum computers will likely go from 53 qubits, to over 200. However, it’s worth noting that Google lost one of its most renowned experimental engineers, John Martinis (haha, Martinis) who claimed to have a plan for hundreds or even a thousand qubits on a single quantum processor.
If you’re curious about the design of Google’s Sycamore processor, look here: https://www.youtube.com/watch?v=IWQvt0RBclw&list=PLQY2H8rRoyvwcpm6Nf-fL4sIYQUXtq3HR&index=7
The tl;dr: is that Google’s Quantum computer did in several minutes what would take anywhere from days (if you ask IBM) to 10,000 years (if you ask Google) for the world’s most powerful supercomputer to perform.
Unfortunately, programming in the quantum space is still difficult, as it often requires intimate knowledge of the hardware design, and is largely isolated to specific kinds of problems. It’s worth keeping an eye on, but currently quantum supremacy means very specific types of problems can be done by quantum computers that can’t be done by classical ones.
It does NOT (yet) mean that quantum computers are “generally” better than classical computers, and it remains an open problem to see if and when they will be.