Slashdot Mirror


Another Step In Quantum Computing: A Functional Interconnect

New submitter Gennerik writes: According to a recent article in the MIT Technology Review, a team of international physicists have been able to create a quantum computing interconnect. The interconnect, which is used to connect separate silicon photonic chips, has the important feature of preserving entanglement. This marks a vital step in creating quantum computers that don't have to work in isolation. According to the article, the trick that The trick that [University of Bristol Researcher Mark Thomson] and pals have perfected is to convert the path-entanglement into a different kind of entanglement, in this case involving polarization. They do this by allowing the path-entangled photons to interfere with newly created photons in a way that causes them to become polarized. This also entangles the newly created photons, which pass into the optical fiber and travel to the second silicon photonic chip.

43 comments

  1. The summary by Maritz · · Score: 5, Funny

    Summary would have read better if the submitter had read over it. ;)

    --
    I do not want your cheap brainburning drugs. They are useless for work. And I am a working man today.
    1. Re:The summary by jeffb+(2.718) · · Score: 4, Funny

      It's in a superposition of edited and non-edited states. If anyone had bothered to read it before posting it through to the main page, it would have collapsed.

    2. Re:The summary by MadCow42 · · Score: 4, Funny

      it's two entangled stories, because quantum physics.

      --
      I used to have a sig, but I set it free and it never came back.
    3. Re:The summary by Anonymous Coward · · Score: 2, Funny

      The article is what it appears to be. Slashdot does not play dice.

    4. Re:The summary by Anonymous Coward · · Score: 0

      you can avoid that by using graphene architecture fused by fentonanolasers
      or in engineering layman terms, reverse the disk array polarity
      windows may have or may have not performed and illegal operation and will have to restart...or not

    5. Re:The summary by bughunter · · Score: 1

      His copypasta got entangled in the cookpot.

      --
      I can see the fnords!
    6. Re:The summary by Big+Hairy+Ian · · Score: 1

      So if I alter the spin of one story it will alter the spin of the other story faster than the speed of light. Jeeze don't let Trump know about it!

      --

      Build a Man a Fire, and He'll Be Warm for a Day. Set a Man on Fire, and He'll Be Warm for the Rest of His Life.

  2. did you mean recursion by Anonymous Coward · · Score: 0

    https://www.google.com/search?q=recursion

    someone got lost somewhere while writing teh summary

  3. Does it run Windows yet? by Anonymous Coward · · Score: 0

    This whole thing is pointless if we cannot run our favorite applications on it.

    1. Re: Does it run Windows yet? by Anonymous Coward · · Score: 0

      Just think: Before writing a paper, it is already in memory in a superposition of perfect and garbage; Before playing a game, it is already in memory in a superposition of highest score ever and lost at the first challenge.

    2. Re:Does it run Windows yet? by Dunbal · · Score: 1

      Everyone can run all of their favorite applications at the same time. That's the beauty of it. Of course it just looks like a blank monitor but I assure you it's doing something incredibly quantum.

      --
      Seven puppies were harmed during the making of this post.
  4. just a friendly reminder by nimbius · · Score: 5, Funny

    Hey, its me, your boss (please, call me a thought leader for our synergy group.) Ive recently been made aware of our tubes, and im concerned to find out we arent keeping abreast of the latest technology in quantum...functional....photon quantum...inter...

    ...Listen. Im told we need to get our functional quantum polarity to be a photon. I read this on a technical article and --as I understand this -- none of our current infrastructure supports paths. Ive looked at my desktop several times today and cant find the icon for quantum entangler danglers or the polar photons. Ive put a ticket in to entangle the photons I ordered (they should arrive sometime friday) so we can start computing quantums.

    PS, if quantum photon tangler re-danglers are activated, let me know which light on the server indicates this and how i can check to make sure the tangler is launching fresh photons.

    --
    Good people go to bed earlier.
    1. Re:just a friendly reminder by Inzkeeper · · Score: 1

      PS, if quantum photon tangler re-danglers are activated, let me know which light on the server indicates this and how i can check to make sure the tangler is launching fresh photons.

      Well, we have designed it so that the photons will automatically light up. I am particularly proud of that design.
      Unfortunately, we have not exposed them outside the box so you will not be able to see them from the outside.
      However, we are working on software that will represent their state in a meaningful fashion.

    2. Re:just a friendly reminder by Anonymous Coward · · Score: 0

      Yessir. I assume you also want the superpositioned quantum re-danglers. I'll order them with the company credit card. You'll notice the vendor on the expense report is Bugatti.

  5. Big news also in boson sampling by JoshuaZ · · Score: 3, Interesting

    In related news on quantum computing 6-photon boson sampling has also been performed (incidentally also by researchers at Bristol with some overap between the two groups). See http://www.scottaaronson.com/blog/?p=2435 for details and discussion. Boson sampling is an important idea which involves estimating the probability distribution of non-intersecting photons. Crucially, boson sampling may be substantially easier to construct since they don't require nearly as much in the way of complicated machinery and error correction as full-power quantum computers, but there are also strong reasons to believe that boson sampling cannot be done efficiently on a conventional computer. That paper is http://arxiv.org/abs/1505.01182 (which also has some other very cool results - they've made essentially reconfigurable chips for this rather than having to make new ones for any specific photon sampling procedure). The original paper which proposed boson sampling is http://www.scottaaronson.com/papers/optics.pdf.

    1. Re:Big news also in boson sampling by Anonymous Coward · · Score: 0

      Boson Sampling? Boring. Now, Boosom Sampling, THAT I can get into.

  6. Entanglement by Anonymous Coward · · Score: 0

    All this talk of entanglement sounds like something out of Calvin and Hobbes.

  7. And pals by Anonymous Coward · · Score: 0

    ] and pals have perfected is to convert the path-entanglement into a different kind of entanglement, in this case involving polarization. They do this by allowing the path-entangled photons to interfere with newly created photons in a way that causes them to become polarized. This also entangles the newly created photons, which pass into the optical fiber and travel to the second silicon photonic chip.

  8. ansible by Noah+Haders · · Score: 0

    now that we have entanglement working, can we skip the quantum computing and go straight to the ansible?

    1. Re:ansible by JoshuaZ · · Score: 1

      Unfortunately, it doesn't work that way. There's no known way to get faster than light communication using entanglement and if we're correct in our understanding of physics then it is entirely impossible. One can take two particles that are entangled but if one changes the state of one of them, it doesn't alter the other's state, it simply breaks the entanglement.

    2. Re:ansible by Noah+Haders · · Score: 1

      One can take two particles that are entangled but if one changes the state of one of them, it doesn't alter the other's state, it simply breaks the entanglement.

      this doesn't make sense. I thought the definition of entanglement was that two particles were linked and a change to one caused a change in the other, regardless of the distance between the particles, and this change happened at FTL.

    3. Re:ansible by JoshuaZ · · Score: 3, Informative

      No. That's not how entanglement works. A better way of thinking about entanglement is imagining two fair coins that can be any distance apart and the first time you flip them, you are guaranteed that they'll either both be heads or both be tails. This isn't a perfect description, but this is close enough. If one wants to be mathematically rigorous then we'd say that two particles are entangled if we cannot describe their combined state simply as the tensor product of the state of each one https://en.wikipedia.org/wiki/Quantum_entanglement#Quantum_mechanical_framework. If you want to read a good introduction to a lot of these issues, I recommend Scott Aaronson's "Quantum Computing Since Democritus" which is essentially aimed as an introduction to quantum computing for non-experts with a some math background (essentially assumes is ok with basic linear algebra and basic calculus). Scott is an absolutely fantastic writer.

    4. Re:ansible by Noah+Haders · · Score: 1

      we're saying the same thing. what I'm hearing is that each of us have a coin, and when we flip them they land on the same value. but the trick is, you say it only happens the first time you flip them, but I bet it happens every time you flip them, no? but then how will we know to flip them at the same time if we are a good distance apart (light years)?

    5. Re:ansible by Anonymous Coward · · Score: 0

      Protocol?

    6. Re:ansible by JoshuaZ · · Score: 1

      No. You get one such flip. It won't actually matter if you flip them at the same time or not. As long as you keep your coin carefully in a little box (where keeping it carefully in a little box is essentially a metaphor for keeping it in a little box that doesn't let any stray photons in), you can do your flip whenever you want. But let's say we had a billions coins so we could each get a billion flips. We still can't use that to send information faster than the speed of light because we have no way to control how the coins flip. All we'll know is that for my nth coin, it flipped the same way as your nth coin, and that that was true for every n. But if I tried to manipulate how my coin will flip then the entanglement goes away.

    7. Re:ansible by Anonymous Coward · · Score: 0

      you say it only happens the first time you flip them, but I bet it happens every time you flip them, no?

      Then you would lose the bet, as measurement of that type forces that entangled state into a particular state that is no longer entangled. And there is no way to know if the other end has been measured or not, as fundamentally, the measurement of one half of an entangled state gives you the same distribution of probabilities as measuring a non-entangled state (i.e. you will see heads 50% of the time and tails 50% of the time either way). It is only the correlations between the measurements at both ends that is special. And you can't force one end to something in particular, as that forces the system into a non-entangled state and the other end goes back to being random by itself.

      This is also one of those situations where the math is really straightforward and simple, and you can show a university freshman how it works properly quicker than trying to write some analogy that ultimately fails when pushed too far (or many quantum entanglement analogies that fail almost instantly...).

    8. Re:ansible by Agent0013 · · Score: 1

      I don't think it matters that we both flip them at the same time, they have already been flipped but the result is in both states at once. It is more like looking at the result of the flip. Once I look at the result then your coin result is also set and when you look you will see the same result even if it is an hour or day later. To do another case we would need new coins where the result is unknown and they are entangled.

      --

      -- ssoorrrryy,, dduupplleexx sswwiittcchh oonn.. -Quote found on actual fortune cookie.
    9. Re:ansible by Anonymous Coward · · Score: 0

      Actually it is like two coins that are linked together and are currently in the air twirling around (super position).

      If you sample one of the coins by catching it in your hand and slapping it on the wrist you see the that it is tails. If someone else picks the other coin and puts it on his wrist it will show tails as well. The result of the coin is purely random you cannot use it to send a message.

      You don't know when it comes out of super-position, just looking at it will cause the coin to land on your wrist. So you never know if you are the first or the second person to look at one of the coins. So you cannot use it for signalling.

      Because you can neither use it for signalling or for sending messages you cannot have faster than light communication with an entangled pair.

      The entangled pair can be part of a calculation, such as in a quantum annealer. In a quantum annealer many q-bits are linked together with different resistance/amplifications. The system as it comes out of super-position into a resting-position resolves (it is somewhat random) into the lowest possible power state. You can read the q-bits in resting-position and you know what the answer is.

      Because of entanglement it is like each of the q-bits occupy the same space, so there is no spatial-communication delay between each q-bit. But this may just be a mathematical concept.

    10. Re:ansible by ClickOnThis · · Score: 1

      No. That's not how entanglement works. A better way of thinking about entanglement is imagining two fair coins that can be any distance apart and the first time you flip them, you are guaranteed that they'll either both be heads or both be tails. This isn't a perfect description, but this is close enough

      Perhaps even better (per the analogy to particle-spins) is to imagine that one coin is guaranteed to be in the opposite state of the other, i.e., if one is heads, the other is tails.

      Another important point is that you cannot control the outcome of the observation: you can't make your coin produce a head or tail, you can only flip it and see what happens. If your coin shows say, a head, then you know immediately that the other is a tail, no matter how far away it is -- but you can't control the outcome, so you can't use the observation to send a faster-than-light signal.

      --
      If it weren't for deadlines, nothing would be late.
    11. Re:ansible by Noah+Haders · · Score: 1

      well what do you want me to say? at least I'm trying to solve a problem and make something new and exciting. you're just throwing up your hands and saying something won't work. Is that was science is all about? shutting down prominent avenues of research? what if somebody had shut Einstein down, or Copernicus, or Darwin? where would we be now?

    12. Re:ansible by ClickOnThis · · Score: 1

      Is that was science is all about? shutting down prominent avenues of research?

      Science is about observing nature and forming theories and laws that agree with those observations. Prominent avenues of research can explore speculative areas and can even challenge previous results. However, the stronger those previous results are, the harder it is to challenge them. I'd be absolutely thrilled if someone came up with a way to achieve faster-than-light signalling. But I'm not expecting it to happen anytime soon, because our current understanding of physics says that it's impossible.

      what if somebody had shut Einstein down, or Copernicus, or Darwin? where would we be now?

      Einstein, Copernicus and Darwin were perfectly capable of shutting themselves down (so to speak) if their prospective theories or laws did not have observational support.

      --
      If it weren't for deadlines, nothing would be late.
    13. Re:ansible by JoshuaZ · · Score: 1

      No. I'm attempting to explain that people have thought about this a lot, and we understand why it won't work. I also gave you references on what to read that will explain it in detail. I strongly recommend Scott's book I mentioned earlier. If you do think that your idea has any chance of working, the best thing to do is to try and actually go read a bit on the subject and see if you can make it work. But in general, it is worth keeping in mind that the vast majority of ideas *don't work* and understanding why they won't work is important. I'm curious, would you have a similar reaction if I were having a discussion with someone who had an idea about using a dynamo powered by a generator to generate an infinite amount of electric power?

    14. Re:ansible by towermac · · Score: 1

      Here, I'll help you out.

      He's right; it's never going to happen in a 4 dimensional space-time manifold. Or perhaps more accurately; attempting to use mass (photon, electron, what have you) bound within the confines of said manifold to somehow trick its own existence... you see how one gets out into the weeds there?

      You have to think bigger. If these dimensions don't allow something, then you'll have to go around them; under, over, bypass somehow.

      How to get outside this manifold we seem to be bound to? Well, it will be tricky. How would you even know, if you somehow pulled it off? In any case, it might start with a singularity. Punch right out of these 4 dimensions. It happens in nature at the center of supermassive black holes. We see hints of dimensional permeability in virtual particles that arise out of nowhere and instantly annihilate themselves. We theorize about it as stings.

      However, conservation of mass will always be obeyed, even outside of relativity and quantum mechanics. Call it a faith in science that still allows for change. The mass that you are able to put into a singularity goes somewhere, and it will not be bound by any speed of light or anything like that. Things like location and velocity are irrelevant here.

      Now I'm not saying you could drive into it. But what if you built this singularity machine, and turned it on, and then built another, and turned that on, and then grazed it with an electron beam aimed at a loudspeaker. You might be able to hear background noise the singularity is picking up from beyond our 4 dimensions. I bet the Milky Ways' center is pretty loud. There is still probably some distance factor, but that factor is galactic. And remember, you really can't aim this device. But you might be able to hear Morse code tapped into the first one, while listening to galactic background noise on the second one...

      Hey, you wanted to solve the problem. I'm just trying to give you a bit of an idea what you're in for. At least I didn't tell you that it can't be done. :)

    15. Re:ansible by ToasterMonkey · · Score: 1

      No. That's not how entanglement works. A better way of thinking about entanglement is imagining two fair coins that can be any distance apart and the first time you flip them, you are guaranteed that they'll either both be heads or both be tails. This isn't a perfect description, but this is close enough

      Perhaps even better (per the analogy to particle-spins) is to imagine that one coin is guaranteed to be in the opposite state of the other, i.e., if one is heads, the other is tails.

      Another important point is that you cannot control the outcome of the observation: you can't make your coin produce a head or tail, you can only flip it and see what happens. If your coin shows say, a head, then you know immediately that the other is a tail, no matter how far away it is -- but you can't control the outcome, so you can't use the observation to send a faster-than-light signal.

      It's like mashing two TINY potatoes together and setting the remains aside. They are now entangled. When you observe one, you can infer the state of the other if neither had interacted with any potatoes in the meantime because if your potato is missing a hunk, it's probably stuck to the other one. We observe these tiny potatoes by throwing other little vegetables at them and observing the results, thus disentangling them in the process, because now you have broccoli guts all over.

      It's "spooky" because you can separate the potatoes by any distance, and when you measure one, you know something about the other one even if it was eleven lightyears away. In other words, the potatoes are in a superposition of states until you measure them, and once you do, the other potato picks the opposite state, because I think Copenhagen interpretation of QM is absolute bulls*t.

    16. Re:ansible by Anonymous Coward · · Score: 0

      Except the claim you put forward directly contradicts observation. People like Einstein have been shut down at times by stuff like that, because no matter what you're trying to do, the ultimate test is observation, and that is entirely what science is about. However, what people like Einstein do instead, is to continue to move forward, by being aware of what has already been done and solving actual problems with theories and unexplained observations.

      It often helps if you can at least read a Wikipedia summary, or really helps if you actually read up on a topic, before trying to use it to solve things anyway. It will save you a lot of time from throwing out ideas that may have been thought of or tested decades ago, and just hoping someone will reward you nonetheless.

  9. There is the theory of the Moebius by Anonymous Coward · · Score: 0

    A twist in the fabric of space where time becomes a loop where time becomes a loop where time becomes a loop where time becomes a loop where time becomes a loop where time becomes a loop where time becomes a loop where time becomes a loop from which there is no escape.

  10. Now you can send data before you know what it is. by dsmatthews9379 · · Score: 1

    If I can share entangled states via an interconnect I can send a member of a pair to another location before I cause the state of it's twin to be resolved. This allows me to build a pair of quantum buffers, which gives you zero lag links and caches etc.

  11. Entanglement is disprovable, this is bunkum by Anonymous Coward · · Score: 0

    How entanglement is claimed:
    The claim is that two photons are entangled, that when you detect a property of one, it *sets* the same property of the other.

    This is typically "proved" by experiments like Quantum Eraser. But of course it violates causality and breaks time and all sorts of things.
    https://en.wikipedia.org/wiki/Quantum_eraser_experiment

    This is how it actually works:
    The detector 1 and detector 2 are joined by a "Coincidence circuit", so when the polarizing filter is placed in front of D1, only photons with the correct polarization are let into the detector, and only the corresponding photon is counted as detected at D2.
    So in effect the photon is tagged by time, and the link between the two photons is the coincident detector.

    No entanglement, just a piece of man made electronics.

    And to disprove it, you can simply measure *all* photons, and *simulate* the coincidence detector later in a spreadsheet or computer program. Is the photons state now being set by the spreadsheet? No, its not.

    What this group have built is a link that they claim does not permit them to know the state of the photon as it travels, and thus doesn't count as 'detection' under the vague concept of detection. But a vacuum also fits that bill.

    1. Re: Entanglement is disprovable, this is bunkum by Anonymous Coward · · Score: 0

      There are a long list of experiments and setups using entanglement, including the simplest ones in an undergrad lab course, which do not use coincidence counting. If your beef is with the coincidence counter, you're not even close to disproving entanglement, and likely don't even understand the principles behind it. The result is a straw man.

    2. Re: Entanglement is disprovable, this is bunkum by Anonymous Coward · · Score: 0

      1. You haven't disclaimed anything I've said about the 'entanglement detector' in those experiments.
      2. Those experiments are the main claim to proof of entanglement.
      3. You haven't addressed the issue with the proton
      4. There is NO experimental proof of entanglement that doesn't have exactly the same issues.
      5. You have not offered one either.

    3. Re: Entanglement is disprovable, this is bunkum by Anonymous Coward · · Score: 0

      No the same AC, but nonetheless:

      1. You haven't disclaimed anything I've said about the 'entanglement detector' in those experiments.

      Whether you use a spreadsheet or coincidence detector doesn't matter, the former is the setup in some lab course based experiments that use less equipment, or trying to more emphasize what is going on. The coincidence detector in such setups just simplifies the results, giving you a simple count, but you can just as easily record everything and look for coincidences vs. non-coincidences afterwards, and it doesn't make any difference. The point is the statistics of how many times it does coincide vs. doesn't, something that can't work out with local hidden variables (depending on what a particular setup is exactly looking at, for some setup, you start every experiment with two photons, and the coincidence detector just stops you from counting background ones unrelated to the pair you sent out).

      2. Those experiments are the main claim to proof of entanglement.4. There is NO experimental proof of entanglement that doesn't have exactly the same issues.5. You have not offered one either.

      You say "experiments" but keep naming one, and that was not the original, nor the only claimed evidence, which goes back to experiments in the 70s. And there are many variations built on top of the eraser experiment, like delayed choice eraser experiment. There are plenty of others not based on that, including more interesting ones using the GHZ state of photons and ones not using particles like Josephsonn junction states.

      3. You haven't addressed the issue with the proton

      There is no issue with the proton. That the proton can behave non-locally is not because the quarks separate and go different ways. You don't see fractional charge at one location because the parts of the proton split up, you can't trap one of the quarks by itself. A system of particles can behave nonlocally in the same way that a point particles can, and it is quite distinct from the parts just spreading out. The scale over which a proton wavefunction can be spread out is much larger than the scale of the quarks within and how far they can separate. This is why there are still searches and theories being proposed for subparts to electrons and photons (no observational evidence yet from scattering experiments though), and it doesn't change or contradict how quantum mechanics work. Heck, quantum mechanics can now do non-local experiments with whole atoms, and it shows no evidence of involving the parts of atoms separating.

  12. You can even time travel by Anonymous Coward · · Score: 0

    See Delayed Quantum Choice Eraser experiment. Which QM believers claim affects things in the past, i.e. time travel. However there is a coincidence detectors in the experiment, which means you are simply filtering at t=0 and only looking at experimental results at t=-1 that fit your entanglement criteria at t=0!, really such a simple flaw, and yet they turn a blind eye to it.

    It's the flock of starlings thing again. If your detector can only see a flock and not a single starling, then all of these artifacts appear in your data, including time travel and probabilistic position as per QM model. We *already* have proof of the flock in the proton.

    Consider the proton, a QM particle that fits the QM model, and yet we already know from deep inelastic scattering that it is actually made of smaller particles.

    So the proton was never *AT* a position when it was detected, it was the effect of these smaller particles on the detector that was *AT* a position. So we already (since the 1970s when the proton was discovered to be made of smaller particle hypothesized to be Quarks) have proof that QM is simply the effect of a group or cloud of smaller particles on the detector. It does not set the position of the proton, because the proton didn't exist as a particle, it was just an effect of a group of quarks!

    So the proof is all there, no entanglement, QM simply describes the effect of a detector, detecting smaller cloud or groups of particles.

    1. Re:You can even time travel by Anonymous Coward · · Score: 0

      You seem to copy paste this to nearly every quantum mechanics article, yet only focus on the particulars of one particular experiment, and not addressing responses that point out the "starling" approach to a photon doesn't work. If experiments split the cloud of starlings up to point A and point B, you can't explain why those starlings suddenly disappear at point B when you make a measurement at point A... and they can never be detected again at point B. If the particle was really spread out, you could continue to detect it at other points. Maybe if you make things convoluted enough, you can get something that does make your cloud of starlings match experiments, but then you will just have converged on wavefunctions already used in quantum mechanics, already describing how particles spread out like clouds.