this post was submitted on 16 Apr 2024
326 points (98.8% liked)
Technology
59739 readers
3736 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
For those not in the know: The big issue with quantum computers is decoherence. This is (simply put) noise produced in the system, which interferes or overwrites the calculation / signal we want to get out of the computer. A large part of this is thermal energy, all that energy bouncing around destroys any chance of reading out the signal. So the solution would be to cool the machine within a fraction of a degree of absolute zero, which is hard but not impossible. Then there's also EM radiation coming from all around us (wifi and cellphones, but also things like radio), this is relatively easy to shield against. A bit of a pain, but still something that can be done. But then there's cosmic rays, there's a real chance a cosmic ray hits with enough energy to disrupt the calculation within milliseconds. Milliseconds isn't enough to do a useful calculation, so that's a problem. Shielding against this is also pretty hard, since cosmic rays can have a lot of energy.
Then there's the issue of measurement itself, measuring automatically means putting in energy to the system. This means it's very hard (or maybe even impossible) to read out the results, without destroying them. Even if you get the damn thing stable enough to do a useful calculation.
The more qubits in a system, the more powerful it becomes, and you need quite a number of them to do anything useful with the machine. But the more qubits the bigger the decoherence issue.
This is why some people (me included) don't believe the current form of quantum computers we are researching can actually work in the real world. We need some kind of big breakthrough on this to create an actually useful quantum computer system. With all the cooling and shielding requirements we certainly won't be using them at home any time soon.
But of course as with anything these days the marketing department and media runs with everything they can, spouting out nonsense about quantum computers becoming mainstream any day now and all the amazing things they can do. This can make it hard to figure out what the actual level of development is right now. Plus anybody working on this is putting in billions of dollars and sure as heck won't share anything with anybody. So maybe someone has already made a breakthrough, but I doubt it.
And then there's some people (me included) who bet a whole beer on quantum computers being inherently impossible. Not the "get them to calculate" part, but the "shave a factor off the asymptotics of computers using ordinary physics" part. The argument is simple: It could very well be that the more data you try to squeeze into a qbit, the fuzzier the result is going to get, so if you put ten million numbers each into two qbits and somehow make the qbits add them, you'll get ten million results that are ten million times fuzzier than if you'd put in a single number. To the best of my knowledge I've not yet lost that bet, it has not been demonstrated that researchers won't run against a wall, there, essentially that the universe has a limited computation capacity per volume of space (or however you measure things at that scale).
Other fun thing to annoy people with: Claim that deciding between P = NP and P /= NP is undecidable.
I think in general the goal is not to stuff more information into fewer qubits, but to stabilize more qubits so you can hold more information. The problem is in the physics of stabilizing that many qubits for long enough to run a meaningful calculation.
Argh it's been a while. The question is whether an n-qbit system actually can contain arbitrary (k <= 2^n^) amounts of n-bit states for arbitrary values of n and k: Such a system might work up to a certain number, but then lose coherence once you try to exceed what the universe can actually compute. As far as I know we simply don't know because noone has yet built a system that actually pushes boundaries in earnest. The limiting factor is more n than k I think but then I'm not a quantum physicist.
It would still mean ludicrously miniaturised computing, in fact, minimised to a maximum extent, but it would not give the asymptotic speedup cryptologists are having nightmares about.