Pretty much quantum physics applied to movement and location.
But will it be able to run Dwarf Fortress?
Not without a complete rewrite. It's after all written in a language that only supports bits. To use qubits you'd need quite a different language. And then there's the question if the current algorythems can take advantage of the quantum superpositions, which they probably can't since they're made for linear processors.
That is very true- it is a problem that we see today going from 32-64 bit processors...
Assume that we write a qprogram that emulates a modern environment or what have you with some inefficiency- say, 10 qoperations per simulated operation. Then we run it on a machine that is capable of 1,000 qoperations (a value I've picked for no sound reason) for each 1 operation that a normal computer is capable of.
Even with our inefficency, we still have an emulation that is one hundred times as fast as our original process, without re-writes. Now, who wouldn't mind a 1,000 FPS fortress in a 8x8 area with ocean, magma, 200 cats, and an invasion?
Now, don't get your hopes up, though.
They've been able to make fusion in labs for decades now, but it's still at least decades away from even starting to work.
And frankly, I suspect that this "singularity" thing belongs in the class of science fiction next to treatises on Cavorite. If only for the fact that the only people able to invent computers capable of replacing computer designers would be computer designers. But more important than that- I think there are a lot of elements of true humanlike intelegence that we don't understand.
I think that's comparative to someone from 1890 asking if a thermo-nuclear power plant can power a satalite for more than 100 years.
I think it's a bit more like the first monkey with a wierd thumb asking... well, asking if dyson spheres are practical for planet simulation.
One thing I'd like to point out is that simulating a planet would require that it simulates any moons it might have. In order to simulate it's moons, one would have to simulate the local solar system- which would need to be part of the local cluster.
Basically, in order to accurately simulate a single particle, you need to accurately simulate the universe.