I get the impression that all these quantum computing "results" are actually just quantum experiments dictated by the hardware and they don't have enough degrees of freedom in the system to control it enough to be called a computer.
And of course simulating a complex quantum system classically is intractable, so the comparison to classical computers is deceptive.
rebanevapustus 109 days ago [-]
This is 100% it.
D-Wave is great at solving issues that directly map to what it is physically solving i.e approximating the state of lowest energy of some entangled (sparse) lattice with fixed topology.
The point is, what if certain problems can be mapped to it such that business value could come out of it?
So far this hasn't been a thing, but at least it can do something non-trivial. There's no other quantum computing device that is as close to attaining real-world usefulness than D-Wave.
adastra22 109 days ago [-]
It is exactly because simulating quantum phenomena on a classical computer is so difficult that the comparison makes sense.
feverzsj 109 days ago [-]
Like other quantum supremacy claims, it cannot be proved by others.
raverbashing 109 days ago [-]
"It works on my computer" is actually an ok claim when you have only one computer of that type
boothby 109 days ago [-]
Not just one, though. We've sold a few to customers, not to mention the computers in our Burnaby lab that you can easily access through our cloud service.
adastra22 109 days ago [-]
It must be tiring to wade through the comments on these topics.
boothby 109 days ago [-]
If I felt obliged to correct every single misconception I read here, it certainly would be tiring. Instead, I generally allow myself one comment and maybe a response. And that's not limited to work-related posts; it's how I budget social media.
adastra22 108 days ago [-]
Sage advice. If I may ask, what do you do at D-Wave? Do you ever think of switching to working on gate-based systems?
ackbar03 109 days ago [-]
When you check the results change
caturopath 109 days ago [-]
I'm mostly just waiting for Scott Aaronson to tell me any of these sorts of claims are true
searealist 109 days ago [-]
In other news: Blending a newspaper outperforms a "classical" computer simulating blending a newspaper.
cwillu 109 days ago [-]
dw;dr
Rendered at 05:59:00 GMT+0000 (Coordinated Universal Time) with Vercel.
And of course simulating a complex quantum system classically is intractable, so the comparison to classical computers is deceptive.
D-Wave is great at solving issues that directly map to what it is physically solving i.e approximating the state of lowest energy of some entangled (sparse) lattice with fixed topology.
The point is, what if certain problems can be mapped to it such that business value could come out of it?
So far this hasn't been a thing, but at least it can do something non-trivial. There's no other quantum computing device that is as close to attaining real-world usefulness than D-Wave.