back to article Intel eggheads put bits in a spin to try to revive Moore's law

With silicon near its development headroom, Intel has been putting its boffins to work on replacements, and one potential technology revealed in a Nature paper uses room-temperature quantum materials. Chipzilla claimed its magneto-electric spin-orbit (MESO) technology's important characteristics are low voltage (as much as …

  1. Korev Silver badge
    Joke

    Moore's law being revived? Thank goodness for that, I was worried I'd have to learn to programme efficiently...

    1. Anonymous Coward
      Anonymous Coward

      Or even program efficiently

      1. Aladdin Sane Silver badge

        I prefer to write code.

        1. Michael Wojcik Silver badge

          I prefer to write efficient code, though it's also important to write code fairly efficiently.

    2. NoneSuch
      Facepalm

      "operates via spin–orbit transduction", which is "the coupling of an electron's angular momentum with its linear momentum".

      Not that old chestnut. :P

  2. Anonymous Coward
    Anonymous Coward

    Amazing stuff

    Humans are capable of quite amazing things....

    And then we have brexit and trump to balance it out.

    1. Steve Crook

      Re: Amazing stuff

      They're both amazing too.... Just tell me you not looked at either one and not been amazed...

      1. jmch Silver badge
        Facepalm

        Re: Amazing stuff

        Damn you, AC... and everyone else who followed the bait!!

        How the hell did the comments section about a genius technical innovation end up being full of effin' Brexit again? There's plenty of such threads on articles that actually ARE Brexit-related.

        To get back on track - the article mentions

        "magnetic spin is non-volatile"

        "current is determined by the direction of an electron's spin. That provided an efficient way to read out the state of the multiferroic"

        "an electric field alters or flips the dipole electric field throughout the material, which alters or flips the electron spins that generate the magnetic field"

        etc

        All of these seem to be applicable to memory / storage, but not to logic circuits. How can the state of one of these atoms be used to direct current flow one way or another in a logic gate?

        1. Michael Wojcik Silver badge

          Re: Amazing stuff

          How can the state of one of these atoms be used to direct current flow one way or another in a logic gate?

          Not my area by any means, but AIUI:

          - We're not talking about "the state of one of these atoms". We're talking about the state of a lattice of several atoms, per the illustration in the article. That's an important distinction, I think; there's a junction of different materials in play.

          - MESO is one type of spintronics. The Wikipedia article on spintronics discusses one way of building a transistor. The base scatters current to an extent dependent on its spin, so the spin can turn it "on" or "off" for appropriate current levels. It appears there's at least one patent on this sort of "magnetic tunnel transistor", which probably has more information if you want to dig.

    2. Ian Michael Gumby Silver badge
      Coat

      Re: Amazing stuff

      Dude! You had to go political.

      I'll skip your dig at Trump. He's a big boy and I'm sure if your rated, he'd tweet about you. :-P

      But lets take a look at Brexit.

      It was explained to me by a Brit who lives in both the US and London that people got fed up with some non-UK citizen sitting in Brussels (I think???) telling everyone what they can and cannot do.

      Hmmm. Over 240 years ago, a group of colonists got fed up paying taxes and being told what they can and cannot do by some guy sitting on the other side of the pond.

      On the positive... you got a chance to vote yourself out of a bad deal.

      All I can say is " Karma Bee-itches! "

      Mine is the jacket with the asbestos liner with enough space for a class 4 (plate) body armor.

      1. Anonymous Coward
        Anonymous Coward

        Re: Amazing stuff

        Hey, you can't compare Brexit with the American war of Independence - one was about a bunch of powerful, wealthy patriarchs who were determined to avoid paying their taxes, and the other......er... hang on a minute....

      2. BinkyTheMagicPaperclip Silver badge

        Re: Amazing stuff

        That's the dumb high level overview.

        Each EU member has input in and voting rights on EU law. The UK has used its power of veto many times.

        Fundamentally there are 28 (27 soon..) EU countries which means politics via consensus and compromise. It's a matter of opinion whether consensus or majority rule is the better method of governing.

        There are some exceptions, but by and large the people that argue about 'laws from Brussels' :

        Can't actually name any laws that are unfair

        (except possibly for untrue stories such as passport colour being enforced)

        Don't realise that the UK happily voted yes to this law

        In some cases where it's a law they wanted passed, don't realise it's the UK which vetoed this law (i.e. steel import tarrifs)

        For the vanishingly small number of cases where they can actually name a law the EU passed that the UK only grudgingly voted yes to, they have no concept that being part of a large trading block does involve - tradeoffs.

        The fundamental rule of all politics is : you don't get everything you want, and you have to choose the least worst choice.

      3. Jamie Jones Silver badge

        Re: Amazing stuff

        "It was explained to me by a Brit who lives in both the US and London that people got fed up with some non-UK citizen sitting in Brussels (I think???) telling everyone what they can and cannot do."

        I'm not going to downvote you (it was unfair that anyone did!), but I will say:

        Yes, that explanation is EXACTLY how the brexitters perceive the situation.

        It is, of course, absolute bollocks.

        For years the goverment have blamed the EU for their cockups.... And for years, some sheep have believed it.

        There is a European parliament. We have British members of that parliament, that are voted for by British people (thought I bet most of the brexitters can't even name their MEP (member of the european parliament))

        The UK has a strong veto in the EU. We had a lot of influence. There are 751 MEP's, representing the 28 European countries... Guess how many of them represent Britain?

        73.

        Almost 10% of the EU 28 is controlled by britain, so if any country should be pissed at "eu control" it's certainly not us.

        Again, you can bet your last dollar the brexitters don't know this, either.

        The brexitters were scared into being scared about immigrants stealing our benefits and our jobs (sound familar?) even though the EU has fuck all to do with "all those mooslims", and even though the UK is able to expell EU citizens who aren't contributing to the economy after 3 months (the UK chose not to do so)

        It is very much like with trump. Scare the people into voting for the very people who will screw them more - tax breaks for the rich etc.

        Jacob-rees-mogg, who's father wrote a book on how to profit from a collapsing britain [ Blood in the streets - that book, by the way, is staight out of the neocon handbook - Paul Ryan etc. would agree with it 100% ] .. has an investment company that recently moved to Ireland "to better serve a post brexit economy". His plans for brexit are:

        From: Who is behind the push for a post-Brexit free trade deal with the US?

        The priority areas for removing “anti-competitive” EU regulations articulated in Plan A+ are unpicking data protection rules (GDPR) introduced by the EU to ensure privacy, and allowing the free flow of data across borders, which would let big tech companies use our data – or abuse it. EU regulations that require exporters using chemicals to present safety data to the European commission before obtaining authorisation for imports are a burden on business, including the plastics sector, the thinktanks say, and should go. Likewise pharmaceuticals companies find the requirements on transparency around clinical trials onerous and should be allowed longer patents – that is, more expensive drugs for the rest of us.

        Anyway, if you're interested, all this information is out there, and explained far better than I could, but basically, if you think of a brexitter as a gullible scared dimwitted trump voter, you won't be far wrong.

        EDIT:Blinky got in there before me, whist I was writing this!

        1. cornetman

          Re: Amazing stuff

          > There is a European parliament. We have British members of that parliament, that are voted for by British people (thought I bet most of the brexitters can't even name their MEP (member of the european parliament))

          The European parliament is a largely toothless body. The commission is the body that holds the power in the EU and it is most certainly not elected. It is basically a closed-shop star chamber of elites.

          The truth is that there are a lot of reasons to love the idea of an organisation like the EU, just not *this* one. The corruption is off the scale, it is extremely undemocratic (for the reasons mentioned above) and is now trying to incorporate countries that have radically different ideologies to the main block.

          Some kind of collapse was pretty much inevitable.

          1. Jamie Jones Silver badge

            Re: Amazing stuff

            Oh, I don't think the EU farts glittery rainbows. :-)

            There's lots wrong with it, but we have chance of influencing any change now!

            This does beg the question... Why didn't the leave campaign concentrate on any of the real issues? Why did they just focus on scaremongering and lies?

            1. cornetman

              Re: Amazing stuff

              > There's lots wrong with it, but we have chance of influencing any change now!

              I think that fundamentally the EU suffers from the same problems that all bureaucracies suffer from: their tendency to expand until they implode under their own weight.

              I think that's a peculiarly human problem and I have no answer to that. The people that have the power to change things have no incentive to do so while their snouts are in the trough.

    3. Anonymous Coward
      Anonymous Coward

      Re: Amazing stuff

      >Brexit and Trump

      I came to read the Reg to get away from those two subjects, this article is about some clever applied quantum physics. Any chance you could toddle off and discuss it in the Guardian comments section ?

  3. big_D Silver badge
    Coat

    So this explains...

    Why they are now 4 years behind on their tick-tock improvements to their processors and haven't released anything "new" since Skylake, just tock improvements to the existing architecture.

    Mine's the one in quantum flux.

  4. Pascal Monett Silver badge

    Okay, we're going to be using spin now - then what ?

    Is this the end of miniaturization ?

    It seems this tech straight out of science-fiction is edging towards reality, and I applaud that is it can be made to work. I fear that Moore's Law will effectively reach its EOL date at that point.

    I'm sure we'll be improving hardware organization for decades to come, but once you're down to electron spin, I doubt that you go any further.

    1. Jamie Jones Silver badge
      Coat

      Re: Okay, we're going to be using spin now - then what ?

      Don't be fooled by the article.. It's just Intels spin on things.

      Ok, ok, I'm going!

    2. This post has been deleted by its author

    3. Little Mouse

      Re: Okay, we're going to be using spin now - then what ?

      "but once you're down to electron spin..."

      New Scientist did an interesting article on the "ultimate" computer around 25 years ago, assuming that Science & Ingenuity would always find a way around the limits of technology for speed / storage / etc, until you reached the physical properties of the atomic/subatomic particles that the thing was actually made of.

      Fortunately we've still got quite a way to go yet.

      1. Voyna i Mor Silver badge
        Boffin

        Re: Okay, we're going to be using spin now - then what ?

        "Fortunately we've still got quite a way to go yet."

        Base 6 arithmetic using quarks is going to be a PITA (or rather more politely a pain in the {up down charm strange top bottom}).

        1. Rol Silver badge

          Re: Okay, we're going to be using spin now - then what ?

          Binary switching became the de facto means of processing very early in computing, as attempts to switch at anything higher was thwarted by the technology of the day, and not long after, binary began to prove itself more accessible than analogue, to the point that pseudo real-time digital processing was almost indistinguishable from analogue and so even that branch of processing fell by the way.

          Well, technology has since moved along at a pace, and it isn't inconceivable that a quarternary switching processor (4 states) could be made, or denary, or even analogue.

          Clearly all the money is in binary, but as the sheep start worrying that the grass can't be grown any quicker, I hope some of them start eyeing up the scrub land over the fence, that just needs a bit of fertiliser and some patience, to produce a crop that generates yields greater per acre by factors of 4 to, well, infinity.

    4. 2+2=5 Silver badge
      Boffin

      Re: Okay, we're going to be using spin now - then what ?

      > I'm sure we'll be improving hardware organization for decades to come, but once you're down to electron spin, I doubt that you go any further.

      Encode data as a vibration on a near infinitely long super string. Just like delay memory but halfway across the universe. :-)

  5. m0rt Silver badge

    @pascal

    "I fear that Moore's Law will effectively reach its EOL date at that point"

    Conversely, I look forward to Moore's law being reached as then it will force a shift in other directions for processing. Who knows what will come out of it? I think it is exciting.

  6. Voyna i Mor Silver badge

    How many generations?

    We had magnetic core, spinning rust, bubble memory. Is this 4th generation magnetic storage or have I missed anything important?

    1. elDog Silver badge

      Re: How many generations?

      Off the top of my pointy head, I think we could throw in wire memory (bits thrown on at one end are read at the other), paper tape/cards, RAMDAM (or something like that).

      I also think we should not expect this last one to be the end of the road.

    2. Persona

      Re: How many generations?

      You have missed the Laddic which is/was a magnetic device for performing logic. Was it important? Well I've seen (actually stood inside) a computer built with Laddic that was controlling a nuclear reactor, so arguably yes.

  7. defiler Silver badge

    The problem with replacing silicon

    ...is that it's been so finely developed over so many years. Starting with something new is going to be a money pit for as long as it takes to develop it to compete with silicon. It may be significantly better in the long run, and it may be that the principles behind it are far superior, but it takes a lot of inertia to pass a worse idea that's been really well developed.

  8. steelpillow Silver badge
    Trollface

    ?

    Nobody is talking switching speed yet. I wonder why not.

    Nor is anybody talking about noise immunity - lower voltages mean greater susceptibility.

    1. Michael Wojcik Silver badge

      Re: ?

      Nobody is talking switching speed yet. I wonder why not.

      Because research is a process, and not instantaneous?

      That's OK. Someone has to ask the stupid questions, or we'd be left wondering why no one had.

  9. John Savard Silver badge

    Other Good Consequences

    Even if Moore's Law coming to an end doesn't lead to ingenious new architectural ideas, it will have the result that computers won't become obsolete every few years. This would make it easier to justify spending the money on getting a good one.

    Or one possible result might be computers that could be upgraded incrementally by adding more CPUs. So you start with one 8-core CPU, then you put in another one, then you get two and add them, and you've got 32 cores to do your bidding.

    1. dbastianello

      Re: Other Good Consequences

      So you mean like what AMD has been doing since they introduced the AM# socket?

      I've been arguing this point for almost two decades now, at least it is green compared to throwing away 80% of the hardware in a system every time one wants an upgrade.

      You don't need to have spintronics for this, you can have it now... although being that AMD promised AM4 socket life up till 2020 (4 years commitment) I would say it is better to wait for the new socket to get best advantage of this model of computing. Plus you can say that you are an environmental tech as well, easy twofor.

    2. Kernel

      Re: Other Good Consequences

      "Even if Moore's Law coming to an end doesn't lead to ingenious new architectural ideas, it will have the result that computers won't become obsolete every few years. "

      Aah no - the hardware vendors will just need to have a talk with their software vending mates and all will be good (in their world, at least) again.

  10. John Savard Silver badge

    Oops

    Followed the link in the article:

    "MESO uses a multiferroic material that's both magnetic (like a common permanent magnet, the atoms are aligned) and ferromagnetic,"

    isn't quite correct. The reason that a change in the electric field can change the magnetic dipole moment is because the material is both magnetic... and ferroelectric.

    1. Brewster's Angle Grinder Silver badge

      Re: Oops

      The chunk of the article you quote is a TYPO; the final ferromagnetic should read "ferroelectric". I haven't got easy access to email at the moment or I would send it in. Somebody tell 'em. With that in place it makes sense.

    2. Michael Wojcik Silver badge

      Re: Oops

      Yup. The Berkeley article has this right. Presumably just a typo.

  11. IGnatius T Foobar !

    Why does anyone need computers?

    Just use the cloud!

  12. hellwig Silver badge
    Paris Hilton

    Hmm... Uh huh, oh of course, yep, Obviously....

    So I understand that things spin and you change or read the way they spin, I'm not sure how this is a replacement for CMOS logic and not just a non-volatile storage mechanism?

    Do these things act as Diodes? Wikipedia tells me Diode logic is a thing. It also tells me you have to have pull-up or pull-down resistors, which I don't think would scale very well.

    Man, I do not recall this from my EE courses.

    I'm so ignorant.

    1. Michael Wojcik Silver badge

      Re: Hmm... Uh huh, oh of course, yep, Obviously....

      I'm not sure how this is a replacement for CMOS logic and not just a non-volatile storage mechanism?

      Do these things act as Diodes?

      Yes, and you can build transistors from them too. See my post upthread for some links, or search for "spintronic transistor". Short version: You can build junctions where the spin-state of one material affects current transfer across the junction. At least that's as I understand it - this is Not My Area.

  13. BryanFRitt

    "between 10 and 30 times lower than CMOS", math???

    "and consequently lower power (between 10 and 30 times lower than CMOS)."

    10 times lower than a number, say X is X-10*X = -9*X. Notice the sign, the before has opposite sign of the new. Now are we now gaining power by using this, or was it gaining power before and now it's losing power, or was it zero power change the whole time? or perhaps someone(me?/you?) doesn't understand how this math/power usage works? This doesn't make since to me. What is it really?

  14. BryanFRitt

    "between 10 and 30 times lower than CMOS", math???

    "and consequently lower power (between 10 and 30 times lower than CMOS)."

    10 times lower than a number, say X is X-10*X = -9*X.

    Notice the sign, the before has opposite sign of the new. Now are we now gaining power by using this, or was it gaining power before and now it's losing power, or was it zero power change the whole time? or perhaps someone(me?/you?) doesn't understand how this math/power usage works? This doesn't make since to me. What is it really?

    Does the same apply to

    "low voltage (as much as five times below today's CMOS-based chips)"

    5*below a number, say X, is X-5*X =-4*X

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2019