Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
Intel

Intel Unveils 'Breakthrough' 49 Qubit Quantum Computer (extremetech.com) 204

Long-time Slashdot reader cold fjord writes: Extremetech reports, "At CES 2018 this week, Intel's CEO Brian Krzanich declared the company's new 49-qubit quantum computer represented a step towards "quantum supremacy." A 49 qubit system is a major advance for Intel, which just demonstrated a 17-qubit system two months ago. Intel's working with the Netherlands-based Qutech on this project, and expanding the number of qubits is key to creating quantum computers that can deliver real-world results... "Qubits are tremendously fragile," Intel wrote in October. "Any noise or unintended observation of them can cause data loss. This fragility requires them to operate at about 20 millikelvin -- 250 times colder than deep space." This is also why we won't be seeing quantum computers in anyone's house at any point."
Krzanich also thanked the industry for "coming together" to address the Meltdown and Spectre vulnerabilities. "The collaboration among so many companies to address this industry-wide issue across several different processor architectures has been truly remarkable."
This discussion has been archived. No new comments can be posted.

Intel Unveils 'Breakthrough' 49 Qubit Quantum Computer

Comments Filter:
  • by 110010001000 ( 697113 ) on Saturday January 13, 2018 @03:57PM (#55922759) Homepage Journal
    The tipping point for quantum computers is around 30 qubits. This opens up research on breaking traditional encryption, AI, autonomous driving and blockchain. Very exciting.
    • by ceoyoyo ( 59147 )

      Possibly 30 fully error corrected and connected quibits would be interesting. On the other hand, IIRC there are already simulators that can do that many reasonably.

      It's unlikely that these quibits are fully connected or error corrected.

    • The tipping point for quantum computers is around 30 qubits

      [Citation needed]

      Some chap from IBM told me you need lots of bits for Shor's algorithm - at least the length of the key and quite possible several times more for quantum error correction.

      Looking at Slashdot's Let's Encrypt certificate it has an RSA key size of 2048 bits. So you need 2048*n

      E.g. if you did need to have 2x as many bits for error correction n is 2. Of course it might be you don't need error correction and some clever algorithm might let you crack the 2048 bits in two runs of 1024 bits e

    • The tipping point for quantum computers is around 30 qubits.

      It's going to take about 1,048 qubits to break encryption efficiently enough to destroy things like Bitcoin. Not a lot considering quantum computer qubit densities are increasing at about the same rate as Moore's law - Bitcoin has about 5-8 years before it becomes worthless.

    • Re:Huge breakthrough (Score:4, Informative)

      by aleksander suur ( 4765615 ) on Saturday January 13, 2018 @08:06PM (#55923791)
      Nope, you can easily simulate 30 qubits on home computer, 49 qubits, not so much. If a quantum computer has enough qubits that no classical computer could ever simulate it, then it's useful because it can do some things classical computer will never be able to do. That limit is thought to be around 50 qubits. The whole fura over quantum computers is that it can bring down computational complexity for some problems, not reduce it to zero and break all the encryption in the world, but bring it down nevertheless and it's a huge thing for some problems making the difference between computable and non-computable. Taking advantage of a quantum computer and doing something useful with it is obviously huge pain in the rear, but with this new machine by Intel we just might start hearing a few interesting successes soon.
      • The whole fura over quantum computers is ...

        Fura are millet dough balls eaten in Nigeria.

        Did you mean furor?

    • by Opyros ( 1153335 )
      So, have you now become a "quantum computing nutter"? Or are you just being sarcastic?
    • by Z80a ( 971949 )

      Don't worry, it only works when nobody is looking at it.

  • by llamalad ( 12917 ) on Saturday January 13, 2018 @04:00PM (#55922775)

    It's funny (sad, really) that Intel is so diligent about trying to bamboozle everyone into thinking that Spectre (which effects many manufacturers' processors) and Meltdown (which is intel-specific) are the same.

    Such bullshit.

    • by 110010001000 ( 697113 ) on Saturday January 13, 2018 @04:02PM (#55922787) Homepage Journal
      What is even more sad is that the Intel CEO dumped all of his stock after learning about the flawed processors, and it looks like he is getting away with it.
      • by Megol ( 3135005 )

        Citation needed. While the timing is suspect selling stock is a normal thing and needn't be related.

        • You are right. It is normal for a CEO to SELL ALL OF HIS STOCK and options to the minimum level allowed right before a major business event. Perfectly normal. My, my, don't YOU look stupid?
      • It is normal for somebody to cash out eventually.

        It isn't improper to sell stock just because some computer people were going to be mad at the company. Insider trader only covers things that affect the stock price. Nothing about this situation suggests that big companies are going to stop buying intel products, and so there is no reason to think that the bad PR will affect the stock price.

        In fact, their recent stock price is the highest it has ever been other than a brief spike during the .com boom/bust.

        If

        • You are right. It is perfectly normal. Just disregard all the analysts who have said they have never seen a CEO do it before. What do THEY know?
          • Every fall, we get some pattern of rain and sun that causes 25% of locals to declare, "I've never seen it do this before!" Yes you have, dummy. You just didn't notice it, because it isn't a significant thing.

        • Insider trader only covers things that affect the stock price. Nothing about this situation suggests that big companies are going to stop buying intel products,

          If they just stop building all-intel networks and mix some AMD in, which is actually pretty likely going forward since they have more PCIE lanes and the same performance (after vulnerability mitigation) and cost less.

          and so there is no reason to think that the bad PR will affect the stock price.

          I think it probably will, it just hasn't sunk in yet.

          In fact, their recent stock price is the highest it has ever been other than a brief spike during the .com boom/bust.

          So far, Intel's PR is working. The investors believe their lies about performance impact. When it continues to emerge that they are lying about this, I think you will see some impact.

          We'll find out.

          • Sure, but why would they do that? There is nothing about this that suggests that diversification of CPUs is going to save them money somehow.

            This means, computers they bought in the past were 1% more expensive than they had originally written down. Supply diversification doesn't help that at all.

            I think it doesn't matter if their PR works or not, either way it is not a significant hit on their business. Probably why they didn't put a lot of effort into it. PCs are mostly commodity now, and datacenter custom

            • Sure, but why would they do that? There is nothing about this that suggests that diversification of CPUs is going to save them money somehow.

              Of course it will. AMD CPUs are cheaper. How will that not save them money?

              This means, computers they bought in the past were 1% more expensive than they had originally written down. Supply diversification doesn't help that at all.

              That's not how it works. People built whole clusters out of intel processors. Now that mitigation will cost them 10-50% depending on workload, they will need 10-50% more nodes. This failure on Intel's part has also created a whole lot of work for administrators because they botched the patch, it's failed on a lot of Win10 machines (can't apply the update, can't remove the update.) They botched the silicon and then they botched the pa

      • It was a normal, automatic, scheduled stock sell off. I think most CEOs are making too much for what they bring to the table, but there wasnâ(TM)t anything particularly insidious about this.

    • by Junta ( 36770 )

      Sadly I don't have a citation, but I am told at least one of the ARM vendors took a similar optimization and as a result is in the same boat as Intel with respect to meltdown.

      The story has been 'intel v. amd' but there are a lot of other players out there.

      The optimization on the face of it doesn't seem *that* obvious of a bad idea: do the access check only if the result would issue, and in all cases you flush the obvious access points. The fact that something that is still inaccessible is in cache and not

      • You mean the ARM processor that isn't out yet, that Intel co-designed with ARM? That one?
        • by Megol ( 3135005 )

          Are you making things up or just being _extremely_ badly informed?

          • No, I am talking about the ARM Cortex-A75 which isn't released yet and Intel had a hand in designing and is susceptible to Meltdown. My, my, don't YOU look stupid?
    • Exactly what I was going to say - Krzanich obviously got the memo from Intel's brass - make sure you mention Meltdown and Spectre in the same breath, and mention that it's an industry-wide problem.

  • by Anonymous Coward

    But not one has demonstrated true quantum solutions in compute time faster than classical computing - yet. I'm not saying it is impossible, just that this feels very much like cold fusion in the 90s. We have been one year from a quantum computer for fifteen years. Before anyone jumps up and down screaming DWave, we know these are not true quantum computers, more likely magnetically paired computers. There are entire papers written about them.

    Intel and IBM spark my interest because they tend to actually

    • Considering that 56 qubit computer has been simulated on a supercomputer its no wonder you don't hear faster than classical news yet. Any quantum computer with less than 56 qubits, that supercomputer can beat it.
  • Predictions (Score:5, Informative)

    by ChatHuant ( 801522 ) on Saturday January 13, 2018 @04:06PM (#55922815)

    " This is also why we won't be seeing quantum computers in anyone's house at any point."

    "There is no reason anyone would want a computer in their home."

    Ken Olsen, founder of Digital Equipment Corporation, 1977

    • Plus my phone has more processing power than those DEC computers. And since we know that past performance always predicts future performance, we will all have quantum iPhones soon. Progress is inevitable. Just look at Moore's Law.
    • " This is also why we won't be seeing quantum computers in anyone's house at any point."

      "There is no reason anyone would want a computer in their home."

      Ken Olsen, founder of Digital Equipment Corporation, 1977

      Whelp, if someone did ever get a quantum computer in thier home, one thing I'm sure of is 640k of memory ought to be enough for anyone.

    • by SeaFox ( 739806 )

      " This is also why we won't be seeing quantum computers in anyone's house at any point."

      "There is no reason anyone would want a computer in their home."

      Ken Olsen, founder of Digital Equipment Corporation, 1977

      I think the problem here is the refrigeration equipment required for stable operation (you know, that thing you're leaving off to take the summary's quote out of context), not that no one would want a quantum computer in their home. The Apple II came out in 1977, and could be set up on a small table.

      • by tsa ( 15680 )

        Quantum biology is slowly becoming big. Proteins are the answer to the cooling problem.

      • I think the problem here is the refrigeration equipment required for stable operation

        The first vacuum tube computers had very similar problems: they needed industrial grade power supplies, required cooling, tubes would fail regularly (the ENIAC engineers considered it a success when they reduced the tube failure rate to one every couple of days). And, to be fair, what Ken Olsen probably meant was nobody would want a machine like his company's PDP or VAX computers, built with discrete transistors and/or relatively simple integrated circuits.

        Both you and him are however making the s

        • You are right, since vacuum tubes were supplanted by transistors, the current quantum bit technology will also be replaced by something better too. After all, technology only gets better and better.
        • by ceoyoyo ( 59147 )

          There were no fundamental physical barriers to scaling down classical circuits to make a modern microprocessor, although you did have to have some imagination to see the possibilities.

          There are very good reasons to think that it may not be possible to make quantum circuits that operate high temperatures.

    • by tomxor ( 2379126 )

      " This is also why we won't be seeing quantum computers in anyone's house at any point."

      "There is no reason anyone would want a computer in their home."

      Ken Olsen, founder of Digital Equipment Corporation, 1977

      People wont want it because they probably wont even know they are using it, it's not comparable to a normal computer in terms of it's visibility, it's more like an extra component for very specific applications. Not to downplay how amazing they are but a quantum computer doesn't magically mean it can do all traditional computing faster (it can't), it can do very specific esoteric problems much faster and has other interesting unique esoteric properties - but it's often hard to figure out how to do them. Qu

    • Since Olsen was referring to home automation, given the security problems, he may have been right.
  • by cold fjord ( 826450 ) on Saturday January 13, 2018 @04:10PM (#55922839)

    Intel’s New Chip Aims For Quantum Supremacy [fastcompany.com]

    . . . those 49 qbits can already do something that almost no traditional computer, even a supercomputer, can: solve a sorting problem that has up to 5.63 trillion possible outcomes. This is right around a milestone called “quantum supremacy”–the point at which a standard computer can’t even simulate a quantum machine.

    The promise of a quantum computer comes from one of those spooky aspects of quantum physics, whereby each qbit can hold two (or more) values simultaneously. Each time you add a qbit, you raise two to a higher exponent value. In this case, two raised to the 49th power yields 562,949,950,000,000 variations that can be compared at once. This technique is ideal for certain types of calculations, like speculative chemistry research that simulates the interactions of dozens of electrons, around multiple atoms, at once. It’s also useful for cracking digital encryption.

    • by tomxor ( 2379126 )

      Each time you add a qbit, you raise two to a higher exponent value. In this case, two raised to the 49th power yields 562,949,950,000,000 variations that can be compared at once. This technique is ideal for certain types of calculations

      Be careful not to loose sight of how much emphasis should be applied to the specificity in "specific calculations"... Almost all of the computations that happen on your regular computer 99% of the time will not be sped up by quantum computing.

      • Yes, quantum computer is not a replacement for classical computers, but it should be able to do few things classical computers will never be able to do, provided good enough quantum computer and good enough algorithms.
  • Intel : '20 millikelvin ... why we won't be seeing quantum computers in anyone's house at any point'

    How do they know that no part of my house is below 20 mK? It's kinda cold in here right now.

    More seriously, how do they know that no other qbit technology will ever remove their low-temperature restriction?

    • You know where it is really cold? Mars. I am thinking these will be great for when Musk sets up his outpost on Mars.
      • Mars: The "Fragility" of quantum computers mixed with the high radiation enviornment of Mars, combined with a slow hardware upgrade rate (long travel time).
  • "operate at about 20 millikelvin -- 250 times colder than deep space" - I assume he's referring to areas of outer space that are a near-complete vacuum, and are far away from anything. Guess what - no temp there. Only 'things' can have a temperature, non-things can't. It would be like saying deep space was hairy, or had a nice singing voice.
    • Re: (Score:2, Informative)

      by Anonymous Coward

      Actually deep space does have temperature (2.725 K) mainly because of the cosmic microwave background. This seems to imply that 20 mK is 135 times colder than deep space.

      https://en.wikipedia.org/wiki/Cosmic_microwave_background

    • I give you the cosmic microwave background [wikipedia.org].
    • Deep space is not completely empty. Interstellar space averages about one million particles per m^3, and those particles to indeed have a temperature. In some places that temperature is high, but presumably they're talking about the lowest it can get, which is the temperature of the cosmic background radiation.

      • The cosmic microwave background isn't the same everywhere.

        Space has some large voids in it where time passes faster than our reference frame here, which stretches out the cmb wavelength to feel cooler there.
        Near a singularity, if you fought gravity and stood still near it, you would experience a very hot background due again to time passing much much slower there. If you look into space, the farther those places are away, the hotter the CMB is there due to expansion of space.
  • Quantum computing (Score:4, Informative)

    by burtosis ( 1124179 ) on Saturday January 13, 2018 @05:04PM (#55923061)
    This article reminded me it's about time I gave my kids the talk [smbc-comics.com].
    • Nice. In case you buy into some of the crappy analogies, a little common sense might help too. Suppose you worked out a problem with all possible inputs at once in parallel. The output would have all the possible results in parallel, so how would you pick the one you need?

      It's been a while since I studied Shor's algorithm in detail, but the gist of it would be that it's a lot more elaborate than elementary factorization by trial and error using primes. Simply running sha256 on a quantum computer won't ma

  • "This is also why we won't be seeing quantum computers in anyone's house at any point"

    About a hundred years ago, a computer cost several millions of dollars and took up a whole factory floor. "This is why we won't be seeing computers/telephones/radios/television/movies in anyone's house at any point" (variations of which were uttered by IBM, DEC, Western Union, Warner, AT&T, Popular Mechanics, ...)

    It may not be in my lifetime, but I fully expect quantum computing to come to the home within the next 100

    • Yes, because digital computers did this it is only natural that everything will. Eventually humans will cost 2 cents and be only a few millimeters wide.
    • by gtall ( 79522 )

      Yep, comparing apples and oranges is surely a way to success.

  • Any news on fixing Meltdown and Spectre?

    • Meltdown itself reflects a form of quantum computing within Intel CPUs. It performs multiple fetches at once and tunnels through permission barriers using spooky action at a distance. It used to be rather limited -- imagine what they can do when the entire CPU works the same way!
      • Ah, so the "encryption will be worthless when quantum computers arrive" doesn't come from them being able to factor every number easily and with trivial time consumption but from them being inherently so insecure that you can't keep any private key private?

  • by CaptainDork ( 3678879 ) on Saturday January 13, 2018 @06:31PM (#55923479)

    “And this is how you shall make it: The length of the CPU shall be 300 qubits, its width 50 qubits, and its height 30 qubits.” (Genesis 6:15)

  • since Intel built it.

    Take any claims about how amazing they are with a grain of salt these days until all the bugs and problems have been identified.

    **cough MELTDOWN**
    **cough SPECTRE**

    Well the patch to fix the vulnerabilities will only have a 30% impact on processor performance :|

  • I'm certain this will be one of the first tasks of the first quantum computers once the teams figure out how to ask the question.

How come financial advisors never seem to be as wealthy as they claim they'll make you?

Working...