Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror
×
IBM

IBM Claims Quantum Computing Breakthrough (axios.com) 79

Axios reports: IBM has created a quantum processor able to process information so complex the work can't be done or simulated on a traditional computer, CEO Arvind Krishna told "Axios on HBO" ahead of a planned announcement.

Why it matters: Quantum computing could help address problems that are too challenging for even today's most powerful supercomputers, such as figuring out how to make better batteries or sequester carbon emissions.

Driving the news: IBM says its new Eagle processor can handle 127 qubits, a measure of quantum computing power. In topping 100 qubits, IBM says it has reached a milestone that allows quantum to surpass the power of a traditional computer. "It is impossible to simulate it on something else, which implies it's more powerful than anything else," Krishna told "Axios on HBO...."

Krishna says the quantum computing push is one part of his approach to return the company to growth.

"Some people think of it as science fiction," Krishna says of quantum computing in a preview clip from the Axios interview. "I think of it now as a feat of engineering."
This discussion has been archived. No new comments can be posted.

IBM Claims Quantum Computing Breakthrough

Comments Filter:
  • What the fuck (Score:5, Interesting)

    by backslashdot ( 95548 ) on Sunday November 14, 2021 @09:42PM (#61988735)

    This is the third Quantum Computing BS story today. IBM literally "simulates" a quantum circuit and then claims it is superior to classical. Well, no shit sherlock. As I said before this is the equivalent of claiming a pebble tossed in water is super to a classical computer because it accurately and instantly shows the interference and refraction patterns. There you go, call it "pebble supremacy." Or a camera and flash photography setup is superior to a classical computer in rendering photorealistic ray-tracing. Technically yes, but in reality bullshit.

    • Re:What the fuck (Score:5, Interesting)

      by backslashdot ( 95548 ) on Sunday November 14, 2021 @09:51PM (#61988755)

      There is a very easy way to prove actual quantum supremacy on a true quantum computer (ie, one without faux-qubits) .. factorize the product of two large randomly chosen prime numbers faster than GFNS. Someone call me when they do that.

      • by Anonymous Coward
        pretty sure your bones will have fossilized or turned to dust long before it comes back with a correct answer.
      • by sinij ( 911942 ) on Sunday November 14, 2021 @10:15PM (#61988813)

        Someone call me when they do that.

        Or even better, send you a singed message with your own key.

      • Re: (Score:3, Informative)

        by u19925 ( 613350 )

        The highest number factorized by quantum computer so far is "1,099,551,473,989 is equal to 1,048,589 multiplied by 1,048,601" from newscientist article. Let us see if this can break that record.

      • On the other hand, I wonder why the US and Chinese governments haven't placed an embargo on divulging further achievements in quantum computing.
        • by vbdasc ( 146051 )

          On the other hand, I wonder why the US and Chinese governments haven't placed an embargo on divulging further achievements in quantum computing.

          Maybe they already did, and we're only seeing the BS stuff...

        • There haven't been any meaningful achievements yet, why worry about more meaningless achievements?

      • But quantum supremacy must already be here, it's been announced at least a dozen times already just this year alone. Next year I expect at least two dozen announcements.

        And you can tell it's quantum because all of those announcements claimed that they were the first to achieve quantum supremacy, and all the future ones will also all be the first to achieve quantum supremacy. Something to do with quantum superposition of quantum supremacy announcements I believe.

      • by gweihir ( 88907 )

        There is a very easy way to prove actual quantum supremacy on a true quantum computer (ie, one without faux-qubits) .. factorize the product of two large randomly chosen prime numbers faster than GFNS. Someone call me when they do that.

        Not going to happen for bit-length that matter anytime soon and maybe never. QCs do not scale. Even the claimed 127 qbits (probably ~50 or so after error correction) took excessive effort and 50 years of research. If things continue to scale this badly, they will maybe be at 256 qbits (and 80 effective or so) in another 50 years. My programmable pocket calculator from 30 years back has more computing power.

      • There is a very easy way to prove actual quantum supremacy on a true quantum computer (ie, one without faux-qubits) .. factorize the product of two large randomly chosen prime numbers faster than GFNS. Someone call me when they do that.

        For a little perspective, executing Shor's algorithm to do the task you're requesting requires 2n + 3 qubits for every bit of the large semi-prime to be factored. For the 2048 bit modulus commonly used today in RSA-2048, the machine requires 4099 qubits to perform the factorization in "push the button, get the answer" time. And not just any 4099 qubits. All 4099 of them must be successfully generated fully coherently. That is, every single bit must be quantumly entangled. (Although now I'm reading that

    • by evanh ( 627108 )

      Huh, I'd always wondered how it worked, I think you've finally made me understand. Thanks, because the news items have all been entirely useless for all these years.

      I presume then the idea is it's like an analogue computer of old where they electrically simulated the desired maths with intertwined op-amps and passive tuning elements (resistors/capacitors/inductors), and the results could be viewed with an oscilloscope.

      • Here is the link to the paper on how their previous quantum supremacy claim worked (with 53 bits instead of 127): https://www.nature.com/article... [nature.com]

      • If you think it's obvious now then you definitely don't know what's going on. It's superposition and other weird magical quantum shit that pretty much defies understanding but appears to be the way the universe works.

        • by evanh ( 627108 )

          Quantum behaviour wasn't the question. The question answered is one of how it applies to computing. Something I'd never seen answered in a single sentence like that before. For that, I thank backslashdot's comment.

          • You can't separate "quantum behavior" from quantum computing. If you think you've got it straight, or that OP described it accurately, you're probably wrong.

            • by evanh ( 627108 )

              Analogue computers are nothing like a digital computer. Maybe that's where the disconnect in this conversation is. I guess there are so very few in the media that would know what one is that they'd never use it as the reference. That adds up.

    • by Anonymous Coward

      Yeah. Reality is the best model of reality. No other model is 100% accurate.

      Modded you insightful but funny too.

    • by ShanghaiBill ( 739463 ) on Sunday November 14, 2021 @10:13PM (#61988807)

      But the summary says it will help to sequester carbon. That must be true because IBM would never spew BS about something as important as sequestering carbon and the relevance of quantum computing to carbon sequestration is totally obvious to everyone.

    • Technically yes, but in reality bullshit.

      Found the operator of the Quantum Bullshit Detector Twitter account.

    • The big difference between the quantum supremacy experiment and your pudding experiment is that the quantum supremacy experiment solved an unambiguous, well-posed mathematical problem. While people sometimes describe the computational task as âoesimulating the physical Sycamore computerâ, thatâ(TM)s not right. The actual task was calculating the output of an abstract quantum logical circuit, of which the Sycamore computer was an approximate physical instantiation. The difference is subtle but
    • Please read this: http://www.gagliardoni.net/#qu... [gagliardoni.net]

      I am not defending IBM's claims, I haven't even read the press release yet. But please be aware that "quantum supremacy" has a very precise meaning that is not what 99.999% of the people (including Slashdot users) would expect.

      TL;DR: 1) "quantum supremacy" is a moving target 2) it is totally possible that IBM has achieved "quantum supremacy" as we currently define it 3) this has almost surely no practical implications yet 4) integer factorization is
      • by znrt ( 2424692 )

        "quantum supremacy" has a very precise meaning that is not what 99.999% of the people (including Slashdot users) would expect.

        i'm afraid that's not how languages are supposed to work, and that might be the reason that it sounds like bs.

        • by jbengt ( 874751 )

          i'm afraid that's not how languages are supposed to work . . .

          I'm afraid that is how languages do work, though.

        • by jythie ( 914043 )
          Well, it sorta is, and sorta isn't. It highlights the problem of cross domain language and reporting. Domain specific terms can have precise meaning, but through reporting it hits the popular meaning and people forget that the two domains are not using it the same way. I guess the key word here is languages, rather than language.. we are talking about what are essentially two dialects of english and people reading a story written in one and read in another.
    • by gTsiros ( 205624 )

      I believe the technical term is Quantoturfing?

  • I'm not a dummy, but I can't seem to learn/find anything about quantum computing at all on-line (I haven't looked in a while though); 'It can handle two states at once' is as far as I hear. I'll ask my fellow Slashdotians, is there a text/video you'd recommend in order to get the gist of how it works without taking many high-level courses?
    • Re: (Score:2, Funny)

      by Anonymous Coward
      Extraordinary Popular Delusions and the Madness of Crowds, Charles Mackay.
    • by backslashdot ( 95548 ) on Sunday November 14, 2021 @10:27PM (#61988835)

      I'll ask my fellow Slashdotians, is there a text/video you'd recommend in order to get the gist of how it works without taking many high-level courses?

      Not really. Take a linear algebra course first. There is no way to simplify it without making it wrong -- but I will try. Do you know what an interference pattern is? Well in a quantum computer, you create a circuit through which a special kind of interference pattern occurs. The magic is that you create parallel interference patterns (think of it as the same computation occurring with slightly different settings across multiple universes in parallel ultimately affecting the interference pattern in your universe.) The resulting pattern will tell you the answer to certain types of problems. Problems that can be mapped such that they affect the interference pattern. Specifically, ones that can utilize quantum Fourier transform or amplitude amplification.

      How Shor's algorithm works: https://quantum-computing.ibm.... [ibm.com]

      • by kackle ( 910159 )
        I both get it and not get it at the same time. Ha!

        Seriously, though, that was helpful; it gave me more understanding than anything else during the last few years. Coincidentally, I took linear algebra twice because of an interest in 3D graphics, back when. Maybe that will help me out when I check out your link. Thank you.
    • by Aighearach ( 97333 ) on Sunday November 14, 2021 @10:49PM (#61988889)

      https://youtu.be/OWJCfOvochA [youtu.be] Explaining quantum computing at 5 levels of difficult (by IBM)

    • by SirSlud ( 67381 ) on Sunday November 14, 2021 @11:02PM (#61988905) Homepage

      here's a great book that explains quantum computing, how it works, the theory and the math that only takes high school math (mostly trig and matrix math)

      https://mitpress.mit.edu/books... [mit.edu]

      it's an easy read, and you will come away understanding what it is and how it works

      • by kackle ( 910159 )
        I appreciate the recommendation. I need to be in the "Twilight Zone" so I have more reading time...
      • by kackle ( 910159 )
        Thanks, but I don't trust a left-handed guy explaining right-handed vector math. ...Maybe he's both left and right-handed at the same time.
    • TBH I've spoken to few folks in the industry and most of them work under NDA. No one is sharing knowledge they do not need too.
  • Or is this just a fearmongering tactic designed to drive down crypto prices?
    • That's one hell of boogeyman you're flinching at.

      I'm so glad I bought IBM shares, instead of crypto! I don't really understand the physical mechanism, but crypto seems to be really bad for mental health.

  • They only metion one thing, hacking. But hacking is illegal! Can it run hello world? Can it run anything?

    • You need thousands of qberts to hack anything.

      What this is, is a guy talking about the 100 qbert machine, and then next to those statements also talking about his perceived future of quantum computing. And Axios, and of course slashdot, got confused and though he was saying they built a thing that does a thing. But no. There is no hello world, there is no software problem.

  • hmmm (Score:5, Funny)

    by gravewax ( 4772409 ) on Sunday November 14, 2021 @10:05PM (#61988787)
    One person with no background in Quantum computing interviewing someone else with no background in quantum computing who is making bold statements that I am sure his engineers are cringing at.
  • by sconeu ( 64226 ) on Sunday November 14, 2021 @10:34PM (#61988859) Homepage Journal

    Imagine a Beowulf Cluster of these!!!

    But does it run Linux?

  • by sjames ( 1099 ) on Sunday November 14, 2021 @10:38PM (#61988863) Homepage Journal

    A cat poop is also way too complex to model at the quantum level on a conventional computer. That doesn't say anything about the computational capability of cat poop.

  • by methano ( 519830 ) on Sunday November 14, 2021 @10:43PM (#61988873)
    Is it me, or does this quantum computing stuff starting to sound like a bunch of hooey? What are they even going to calculate with it? It never seems to have any real world application. Why do I have this iMac on my desktop, when a quantum computer is billions of times better? Can they make a little one for me that's only a 1000 time better than my iMac? And how the hell is it gonna solve the battery problem. That's gonna take people making things and testing them? 127 qubits, my ass!
  • I expect the "short term" problems that quantum computing are things that take a "a lot" of computing power and people have a lot of interest in: DNA sequencing, drug interaction modeling, simultaneous repathing problems/traffic modeling, materials testing/simulations, explosion modeling.

    The press just talks about cryptography because everyone has at least heard of it.

    One of the bigger problems not many people have been talking about is the quantum-to-computer interface and operating system. These quan
  • by Anonymous Coward

    If we were to believe all that Watson could do 15 years ago, we already would have better batteries, less emisions and World Peace at least 20 years ago.
    I don't think of this as science fiction but more like bloated sales talk.

  • ...that one of the Many Worlds is the one where IBM actually has made the breakthrough.

    Job done. Pay me my consultant's fee, IBM.
  • As a layman, anytime I read articles about quantum computers and their calculations, it seems that it can only handle and interpet very specific instructions for very specific tasks, it can't do what general processors do.

    It reminds me of Zoolander and being able to only turn right. Their quantum CPU can turn left, and other computers can't, so clearly it means it's more powerful.

    I have a high end desktop PC, and no other standard desktop PC can make coffee. However, there is a small computer in my coffee m

    • Any article that can easily and clearly describe 'what can quantum computing do for you' should win an award or something, as one has yet to be written.
      • by Tyr07 ( 8900565 )

        Yeah I always just figured it was beyond my understanding that I didn't see any current value / practicality in it. It all felt mostly like marketing fluff. Maybe someday we truly harness 'quantum computing' whatever that will look like or mean when we do. Part of the issue I have is any verification is destructive. I guess if they made two identical devices that gave the same result, and they verified one they'd know the other was true too.

  • Please read this: http://www.gagliardoni.net/#qu... [gagliardoni.net]

    I am not defending IBM's claims, I haven't even read the press release yet. But please be aware that "quantum supremacy" has a very precise meaning that is not what 99.999% of the people (including Slashdot users) would expect.

    TL;DR: 1) "quantum supremacy" is a moving target 2) it is totally possible that IBM has achieved "quantum supremacy" as we currently define it 3) this has almost surely no practical implications yet 4) integer factorization is
  • Nothing else.

  • Omnission (Score:4, Insightful)

    by thunderclees ( 4507405 ) on Monday November 15, 2021 @11:26AM (#61990319)
    Conspicuous by its absence is the mention that a real driver behind this is providing meaningful answers in rel time to queries from those mile long data storage that the alphabet mafia is keeping on everyone.
  • It's never aliens, or a real quantum breakthrough. Unless of course it's powered by "amazing new battery that researchers just developed but you can't actually buy it yet" or "fusion breakthrough". Then it's totally aliens.

"Pok pok pok, P'kok!" -- Superchicken

Working...