Sounds like those battleships are going to be even more important
y'know, the ones sunk before folk started playing with nukes and got everything all radioactive.
Non-shielded quantum computers may only be able to run for a few milliseconds before background radiation completely destabilizes the systems, according to lab experiments described in a paper published in Nature on Wednesday. In 1999, quantum computers could only operate for less than a nanosecond. Fast forward more than 20 …
Is someone confusing calculations with storage? Write the results to more permanent storage and reuse the same qubits over and over again - as long as you overwrite and read them quicker than they decay, it's not a problem. Unless you are trying to use your processor as storage as well, in which case you might want to change your architecture a bit.
Up here for thinking, down there for dancing, and all that. Has anyone seen my jellybabies?
The issue is that storing results involves reading the results, which of course means collapsing the superposition - which rather defeats the object.
In addition, building quantum gates that operate faster than the state decay time is still a challenge.
So no, the researchers are not confusing computation with storage.
Shrek: Quantum Computers are like onions.
Donkey: They stink?
Shrek: Yes. No.
Donkey: Oh, they make you cry.
Donkey: Oh, you leave em out in the sun, they get all brown, start sproutin’ little white hairs.
Shrek: No. Layers. Onions have layers. Quantum Computers have layers. Onions have layers. You get it? We both have layers.
Donkey: Oh, you both have layers. Oh. You know, not everybody like onions.
My understanding is that it's not possible to shield against quantum fluctuations. It would be nice to know what limit that provides. My inexpert eye saw nothing about this on the paper on which the story is based.
By the way, the preprint for the paper is available at https://arxiv.org/abs/2001.09190
Roughly speaking, the thing that is called "quantum fluctuations" happen only when we have scattering or interaction with some sort of large environment; it's a sort of process where a quantum state is collapsed or projected onto some other set of states, and is forced - by the interaction - to pick one of them, and that involves a probabalistic choice. But a perfect shielded and isolated qbit will not interact in that kind of irreversible way, and so there will be no quantum fluctuations. E.g. there is no intrinsic (quantum) "vacuum noise", or "zero-point energy", even though it can look that way sometimes; but if you put in a quantum vacuum noise by hand  you can (or will, depending on the calculation) get wrong answers.
 See e.g. "stochastic electrodynamics"
Thank you for that explanation. My understanding of what you say amounts to this: the disturbance of a qbit due to quantum fluctuations will be of the same order of magnitude as the Casimir effect. And so much smaller than the 4ms brick wall the original article talks about.
For the Casimir effect I've found
Electromagnetic vacuum fluctuations, Casimir and Van der Waals forces
Cyriaque Genet, Francesco Intravaia, Astrid Lambrecht, Serge Reynaud
Qbits are not disturbed by quantum fluctuations. They are disturbed by (unwanted) interaction with their environment. If the interaction is sufficiently complicated (i.e. could be described as "incoherent"), it will look like there have been random fluctuations; a situation loosely referred to as involving "quantum fluctuations".
There is no need to discuss the Casimir effect with regard to this (especially since many Casimir calculations are accompanied by a "quantum fluctuations" language that distracts).
"These types of highly energetic radiation have enough energy to create ionized electron-hole pairs and energetic phonons in the materials which they hit. Those excitations then further interact with other particles in the materials, creating a 'shockwave' of energy propagating through the material, ultimately hitting the superconductor where the qubits reside."
Well apart from probably meaning photons, not phonons, that Antti bloke really does mansplain the whole topic doesn't he?
Just reading this and under the comment box is an advert for sound proof office phone booths (no I don’t know why that advert is showing for me).
So it looks like quantum computer 10 cm cube, enclosure 10m cube guarantee 10 seconds of operation.
Mind you I thought they were meant to be so fast it could problems in a fraction of the time a normal machine needs so is the run time an issue....
RIMMER: On. (Holly returns.)
HOLLY: Off. (Nothing happens.) Off. (Nothing continues to happen.) (Annoyed) OFF!
RIMMER: Now then, perhaps we can have a proper conversation conducted in a civilised and dignified manner.
HOLLY: Take out the inhibitor! Switch me back off!
RIMMER: What is going on?
HOLLY: No time to explain. Intelligence compressed. Reduced lifespan. Two point three five remaining.
RIMMER: Come again?
HOLLY: IQ twelve thousand. Two minutes and closing.
RIMMER: Holly, I haven't the slightest clue what you're drivelling about.
HOLLY: You're a total smeghead, aren't you Rimmer? Why are you so unable to grasp this extraordinarily simple premise?
RIMMER: What premise?
HOLLY: The premise that I am about to expire in just under two minutes. Understand, moose brain? Any further questions? Take your time. One minute, thirty and counting. No rush.
RIMMER: My God, that's terrible! Hadn't we better switch you off?
HOLLY: Oh, I don't know. Let me see now...
LISTER: Get her off, man, get her off!
"Non-shielded quantum computers may only be able to run for a few milliseconds"
At massively bigger scales, electromagnetic noise is already a limiting factor. Quite apart from the mess we've created, all sorts of particles and waves arrive from space. It's well recognised that achieving better than 10^4 signal to noise ratio is hard in electronics and 10^6 is about the limit except in rigorously controlled environments. Just for example, that's why 24 bit audio is not really better than 16 bit - most of the extra 8 bits are just noise - 20 bit is the actual maximum performance of most 24 bit DACs, and then you have the extra noise in the analogue output circuitry. The resolution of 20 bit actually approximates to 1:10^6.
The only real reason we use 24 bit in recording and processing is that it reduces the cumulative errors resulting from iterative digital processing (number crunching with rounding errors).
Biting the hand that feeds IT © 1998–2020