A way to manipulate electrons and transmit information quantum-mechanically

rochester.edu

17 points·DoreenMichele··7 comments

Direct link to paper, preprint: https://arxiv.org/abs/1904.05372 and published: https://www.nature.com/articles/s41586-019-1566-8

> “You and I are quantum systems. The particles in our body obey quantum physics. But, if you try to compute what happens with all of the atoms in our body, you cannot do it on a regular computer,” Nichol says. “A quantum computer could easily do this.”

That sounds too good to be true.

It is technically false. Quantum computers do not add computing capabilities, they are still a Turing machine.

However, in practice, they could solve some problems that would take too long for a electronic computer. That is, if we manage to build a universal quantum computer with a big enough working set, which for the moment we don't and we aren't near that point either.

Quantum circuit is used to model quantum computers these days, not Quantum Turing Machine. QTM is also a different beast than Turing Machine, see e.g. BQP or PostBQP.

Quantum computers are capable of computing things which aren't classically-computable. For a clear explanation I recommend "Simulating Physics with Computers" by Richard Feynman. It presents one example of such computation and step-by-step walkthrough. An excerpt: "That's why quantum mechanics can't seem to be imitable by a local classical computer".

It probably is. There's bound to be some law that says you can't simply simulate a system with a strictly smaller system.

Simulate to a certain extent I think yes, emulate, obviously no, but in the simulation extremely critical discoveries would be lost that only appear in a full emulation. Simple distinction maybe, but worth noting I think.

Reply on news.ycombinator.com