Build a toy quantum computer at home

Post by radioactivist » Wed, Jul 22, 2020, 3:03 PM UTC

shakow wrote: post #39

radioactivist wrote: post #11

While this is a nice demonstration of the polarization of light, this is not a demonstration of quantum mechanics, or quantum computing (though it does have pedagogical value, if qualified properly). Polarizers essentially just project the electric field of the wave onto some axis, zeroing out the perpendicular component. Keeping in mind that light intensity is the square of the electric field strength, all of this c…

> Polarizers essentially just project the electric field of the wave onto some axis Where goes the energy of the orthogonal component of the field? Absorbed by the polarizer, reflected, ... ?

Yes, those are both possibilities for building a polarizer [1].

For example, a simple polarizer could be a grid of thin metal wires whose spacing is smaller than the wave-length of the incoming light. For the component of the E-field parallel to the wires currents can be induced freely along their length, and so the grid behaves much like a solid metal plate and reflects that part of the wave. For the component of the E-field perpendicular to the wires, significant currents can't be generated (since the wires are thin) and that part of the wave passes through.

[1] https://en.wikipedia.org/wiki/Polarizer

[2] https://en.wikipedia.org/wiki/Polarizer#Wire-grid_polarizers

Post by jlokier » Wed, Jul 22, 2020, 3:25 PM UTC

thdrdt wrote: post #33

Is it possible / are there virtual or emulated quantum computers? I read about the Kyndi model but could not find any implementation.

Yes, it is possible to emulate a quantum computer, and this is done quite often for research and to try out quantum programs.

The problem is it takes O(2^n) classical computer resources to emulate a general purpose n-qubit quantum computer. In other words, exponential time or size.

(We can simulate some larger quantum chemistry systems on a classical computer, but those aren't general purpose. The simulations are quite restricted in what they can measure, and there's still a significant practical size limit.)

So we can only emulate very small general purpose quantum computers or other quantum systems. For larger quantum computers, in principle those can be emulated too, except you would need an impossibly fast and large classical computer to do it. So we can't do so in practice.

This is actually the motivation for building real quantum computers of significant capacity.

If a real quantum computer can be built with a large number of high quality, fully coherent qubits, it will be able to do calculations that can't be emulated on any classical computer we can actually build and run, just because of the O(2^n) practical limit.

Right now, there are no quantum computers like that. There are some dubious marketing claims around, and there are also some genuine, but smaller, devices.

Because we can't even simulate a large quantum computer, we don't know for certain whether such a device can even be built in the physical world. The abstracted maths of quantum mechanics, which has proven to be extremely accurate and correct for everything it's been used on, says it can (subject to practical engineering details), but the physical world may have a subtle limitation which we can't detect in smaller systems, that only happens with larger quantum computers and prevents it from being possible. The maths itself might even have a subtle reason (such as stability or entropy) why the system cannot work, but no such reason is known at the moment. We can't "run" the maths to find out its behaviour on a large system, for the same reason we can't simulate a large quantum computer without a large quantum computer in the first place. We can only reason about it in the abstract.

Re: Build a toy quantum computer at home

Re: Build a toy quantum computer at home