Hacker News Reader

Comment by whatshisface | Hacker News Reader

Quantum computing is not yet territory for engineers. So far only the most rudimentary devices are being built, although IBM is wasting no time in their attempts to confuse as many people as possible with their PR. Quantum computer research is a reasonable field to try to get in to, but of course it's inside of academia (so you'd have to go back to school for that career change.)

Among researchers there remains some concern that no practical quantum computer will ever be built. The concerns are centered around practical limits in error correction (analogous to circuit noise). So, while now is a great time to be writing PR pieces about it, and an OK time to be researching it, it's a pretty bad time to be trying to engineer with it.

Here's Scott Aaronson, the most famous quantum information theory person there is: https://www.scottaaronson.com/blog/

View on HN

Everything I know about quantum computing I learned from Scott's comic: https://www.smbc-comics.com/comic/the-talk-3

I am not even sure how one can have a consumer level quantum computer (e.g. desktop, laptop, mobile device) at all. The physics behind a quantum computer doesn't seem to allow such consumer devices ever possible for a true QC. Our surrounding is full of interference. Of course, never say never right? I am just skeptical I guess.

They aren't meant to be consumer devices, and in fact can perform worse than a classical computer for most everyday tasks that are sequential, like loading a webpage. What they are good it is simulating outcomes, since a qubit can be in more than just 0 | 1 states.

That's why quantum computers would be good at breaking crypto, running risk simulations for business, AI/ML, etc and not sequential tasks like most consumer devices are meant for.

What about gaming?

I don't think anybody would have forseen that the primary use of massively parallel processors would be playing video games. I wouldn't be surprised if quantum turns out to work, that graphics programmers find a use for them.

This is one of the rare instances where I would say a cloud service makes sense. You're bulky, expensive, highly specialized quantum computer sits in a data center and consumers connect to it as needed.

I listened to this [0] podcast a few days ago and that is one of the things said - it's not practical to try and put a quantum chip in every desktop/PC, but a Quantum computing service in Azure is the endgame. Right now only a simulation is offered, but the idea is that when a real Quantum service is offered, the same code that runs in the simulator can run on the Quantum service.

[0] https://www.hanselminutes.com/618/quantum-computing-detangle...

if it's not in the territory for engineers, who's designing and building the quantum computers, and coming up with solutions to these noise problems?

I think at this stage, it's scientists designing and building them. Some scientists have a lot of overlapping skills with engineering, but it's not really the same thing.

Reply on news.ycombinator.com