back to article Boffins one step closer to solving nanoscale computer challenge

Scientists at the University of California, Santa Barbara have developed a design of a working nanoscale computing device that could complete the Feynman Grand Challenge. Richard Feynman’s famous talk “Plenty of Room at the Bottom” is considered a precursor to the boom in nanotechnology. Although there has been much progress …

  1. Anonymous Coward
    Anonymous Coward

    Not a commerical application

    But there is a big obstacle standing in the team’s way. Memristors are still being heavily researched and are not commercially available yet.

    How is "and are not commercially available yet" relevant - this is all about research, surely ?

    1. Pascal Monett Silver badge

      It's just as relevant as stating that hyperspace travel is possible but hyperdrives that would allow it are not commercially available yet.

      In other words, we're one step closer to the solution, but no one can tell how many steps remain for us to actually get there.

      Although I must admit that I feel confident that memristors will be available before hyperdrives.

  2. frank ly

    How high can you go?

    "The key to solving the Feynman Grand Challenge is that the two-dimensional circuit can be stacked to create a three-dimensional device within 50-by-50 nanometres."

    Maybe they need to respecify it as a 50nm cube.

    1. Adam 1

      Re: How high can you go?

      And while they correct the challenge to reflect volume rather than area, they could correct the measurements to be a more meaningful nanograpefruits.

    2. Steve Foster
      Headmaster

      Re: How high can you go?

      If you follow the link to the Grand Challenge, you'll be able to read that specifying it as a 50nm cube is exactly what they have done.

      1. defiler

        Re: How high can you go?

        "specifying it as a 50nm cube is exactly what they have done."

        Aww. I was reckoning on 50m x 50nm x about-a-yard.

  3. Brian 3

    No, surely if there is to be a new spec, it must have rounded corners, no?

  4. John Smith 19 Gold badge
    Unhappy

    To be clear that's a volum of 500 atoms on a side.

    So actual nanotechnology would not have a major problem with doing this.

    But basically semiconductor technology is having trouble with it.

    1. John H Woods Silver badge

      500 atoms on a side.

      ^hydrogen^... big atoms can be 5x bigger.

      1. lys
        Boffin

        Re: 500 atoms on a side.

        The atomic radius is less than 0.2nm for almost all atoms (and it does not necessarily increase with the atomic number).

    2. Yesnomaybe

      Re: To be clear that's a volum of 500 atoms on a side.

      "... actual nanotechnology would not have a major problem with doing this"

      I am thinking some kind of Babbage adding machine built from a couple of interlocking molecules? Heat vibration could provide the power.

  5. Mark 85

    How are they coming on the second part...?

    I followed the link and there's pieces to the prize. The one in this article and then there's a nano-scale robotic arm. I'm wondering if the computing device needs to come first.

    1. John H Woods Silver badge

      Re: How are they coming on the second part...?

      Unless you need the arm to assemble the device...

  6. John Smith 19 Gold badge
    Meh

    Rotaxane?

    As this years Nobel in Chemistry was won by 3 boffins, one of whom had assembled molecules (or groups of molecules) to perform computation functions.

    BTW most atom sizes are still measured in Angstroms, IE 0.1nm.

    According to http://antoine.frostburg.edu/chem/senese/101/periodic/faq/what-atom-is-largest.shtml this Cesium is the biggest atom at up to 0.273nm atomic radius, so 0.546nm in diameter.

    So a cube 50nm on a side is a cube 91.5 Cesium atoms on a side. That's 753571 whole Cesium atoms in volume.

    Seems like there should be enough space if you're doing proper molecular nanotechnology.

    1. Tom 7

      Re: Rotaxane?

      You might get away with something at near no kelvins at all but much warmer and its looking dodgy.

      Its a while since I did silicon design but we added 1 atom of doping in 10**14 to get things to work and having less than 10**6 to start with makes me think that even the one on a corner is still fucking with the one on the opposite corner.

      1. John Smith 19 Gold badge
        Unhappy

        "You might get away with something at near no kelvins at all but much warmer and its looking dodgy."

        For crystalline semiconductors perhaps.

        You are aware you were assembled by "nanomachines" running in every cell in your body, right?

        Semiconductors require the perfect crystal structure as well to function. Electrochemical systems are more forgiving.

  7. NanoMeter

    Richard Feynman liked to talk about jiggling things

    No - NOT boobs, but quantum particles.

  8. JeffyPoooh
    Pint

    "[nanoscale] Memristors... ...are not commercially available yet."

    I keep ordering nanoscale Memristors off eBay, but I can never find them in the shipping envelope.

  9. This post has been deleted by its author

    1. Anonymous Coward
      Anonymous Coward

      Re: Sorry ladies! Not this time!

      Surely you know the rules - the feminine suffix for professions is in decline more or less internationally.

      (Incidentally, how about Feynman's earlier challenge - to build an electric motor that fitted into a 1/64 inch cube. He didn't think anybody would do it until an engineer turned up one day - and, said Feynman, he said to himself "oh-oh" as the first thing the guy brought out was a binocular microscope.)

  10. DropBear

    I'm not entirely convinced that a challenge requiring a 8+8 bit adder is adequately fulfilled by a component that... well... accumulates the effects of current passing through it. Addition it may technically be, but still...

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