Slashdot Mirror


One Step Closer To Speedier, Bootless Computers

CWmike writes "Physicists at the University of California at Riverside have made a breakthrough in developing a 'spin computer,' which would combine logic with nonvolatile memory, bypassing the need for computers to boot up. The advance could also lead to super-fast chips. The new transistor technology, which one lead scientist believes could become a reality in about five years, would reduce power consumption to the point where eventually computers, mobile phones and other electronic devices could remain on all the time. The breakthrough came when scientists at UC Riverside successfully injected a spinning electron into a resistor material called graphene, which is essentially a very thin layer of graphite. The graphene in this case is one-atom thick. The process is known as 'tunneling spin injection.' A lead scientist for the project said the clock speeds of chips made using tunneling spin injection would be 'thousands of times' faster than today's processors. He describes the tech as a totally new concept that 'will essentially give memory some brains.'"

19 of 249 comments (clear)

  1. spin computer by Anonymous Coward · · Score: 2, Insightful

    am i the only one who read the title and thought that PR firms and politicians could be in serious trouble?

  2. Bad summary again... by Facegarden · · Score: 4, Insightful

    So, this is becoming a trend. Bad summary. It's not an outright lie, just misleading. From reading the article, one might get the sense that we might see this in products in 5 years. However, the article actually states that the guy said:
    "I'm one of those researchers that really cringes at the thought of saying this [new technology] can be useful. I think for us, maybe within five years we can get one device working."

    So, the guy is realistic, and not a douche. "We can maybe get one working in 5 years" is not the same as seeing it in devices in 5 years (which, again, wasn't explicitly stated in the summary, but i feel like thats what people would think).

    In reality, we might get something in products in 10 years.
    -Taylor

    --
    Worldwide Military budgets: $2100 billion. Worldwide Space Exploration budgets: $38 billion. Really, world? Really?
    1. Re:Bad summary again... by Zouden · · Score: 4, Insightful

      It's also terribly pedestrian to say that this could lead to "speedier, bootless computers", like as if this technology will be implemented in the next Intel chip and suddenly Windows will load instantly and we'll all get high framerates in Crysis. Really, this technology is similar to quantum computing - eventually it'll find its way into extremely specialised applications, and by the time (or if) it does make it into our homes, computers will be very different things, almost unrecognisable.

      Also, "mobile phones and other electronic devices could remain on all the time." Guess what? My mobile phone already remains on all the time, because I recharge it every few days. If the reporter is talking about devices remaining on without charging, what does he think is going to power the antenna and the display? The scientists haven't invented a free energy device.

      --
      "A week in the lab saves an hour in the library"
  3. Re:I wonder though by 0123456 · · Score: 4, Insightful

    But the one about 'You can keep them powered on', it's like a game changer from out of left field. Maybe booting will become irrelevant by then?

    Not if they're running Windows. Doesn't it still have to reboot whenever you update the freaking PDF viewer?

  4. Cool stuff but... by PmanAce · · Score: 2, Insightful

    Will this new technology finally bring us to our beloved flying cars?

    --
    Tired of my customary (Score:1)
    1. Re:Cool stuff but... by pushing-robot · · Score: 2, Insightful

      Yes. Right after it gives us the Matrix.

      --
      How can I believe you when you tell me what I don't want to hear?
  5. Can't we get that already with memristors? by jcr · · Score: 2, Insightful

    For that matter, wouldn't any non-volatile, high speed memory device do the job?

    -jcr

    --
    The only title of honor that a tyrant can grant is "Enemy of the State."
  6. Re:we dont need more processing power tho by theheadlessrabbit · · Score: 4, Insightful

    even today's mainstream cpus are far more powerful than what our everyday tasks involve. even the fps-hungry gaming crowd has been reaching perceptive limits in regard to human eye, and the frame rate has become a sport, a statistical value.

    unless society takes on seti, parallel computing etc as hobbies, we wont need more processing power in our daily lives.

    Just wait till the next version of windows hits the shelves...

    I'm fairly certain that computing power is like hard drive space or time 'till the deadline , we will always find ways to fill it, no matter how much we think we have in the beginning.

    --
    -I only code in BASIC.-
  7. Re:Wishful thinking... by Relic+of+the+Future · · Score: 5, Insightful

    1000x in 5 years IS wishful thinking, but it isn't THAT drastically off from Moore's law, which predicts a 1000x increase every 10 to 15 years. And it's never happened overnight, but in steps every few months. Many of the "1000x-predicted" technologies that /. covered 10 years ago probably have been part of the 1000x-actual increase of the last 10 years.

    --
    Those who fail to understand communication protocols, are doomed to repeat them over port 80.
  8. Re:Graphene Revolution by ozmanjusri · · Score: 2, Insightful

    Windows update?

    --
    "I've got more toys than Teruhisa Kitahara."
  9. Re:Graphene Revolution by sempir · · Score: 2, Insightful

    'will essentially give memory some brains.'" Now if they can develop this for human consumption think what it would do for people with Alz........Aaahhhhhh......whassitcalled? ....

    --
    A closed mouth gathers no foot.
  10. Is the Software Ready for This? by mdm42 · · Score: 2, Insightful

    So, all very nice, we'll be able to have always-on computers that don't pig out on energy, BUT...

    How much of the software we use can handle running for long periods of time without crashing? Not many, in my experience.

    What with memory leaks, bounds overflows and who knows what else, some of which may be an oversight in your own code, but more likely is a bug inside some library you're using, or a compiler bug, or linker bug, or...

    As anybody who has tried it and knows, writing software that runs for weeks and months on end without restarting is really quite hard. And it's no bloody use if the hardware can stay up for months on end if the software can't.

    (And, not having used Windows in about 14 years, I'm not talking about that piece of shite.)

    --
    New mod option wanted: -1 DrunkenRambling
    1. Re:Is the Software Ready for This? by fuzzyfuzzyfungus · · Score: 2, Insightful

      Depends on your context: for a user-facing computer, that is pretty much true. If X pukes itself, taking my graphical programs with it(or even if my browser pukes itself, taking my tabs with it) I might as well have rebooted the computer for all the inconvenience I've just been put to. There are even a few situations where(without good design) non-volatile memory could make things worse: today, if some peripheral gets confused and its internal processor stops talking, or starts talking nonsense to the outside world, you just cut the power. Unless it has been doing something unsafe, like a firmware update, it'll be fine when it comes back up. If it is nonvolatile, it'll be just as confused as when it went down. You'll need either markedly better firmware teams, or explicitly "state-flush" watchdogs or you'll be RMAing Wifi and graphics cards like there's no tomorrow.

      On the other hand, if I'm running a really low power system(not a "cry for me, I have to run off batteries" low power, more like "I have to run off a piezoelectric device catching vibrations" or "I have to run off a blood glucose fuel cell" low power), the ability to seamlessly drift between sleeping and waking, in milliseconds, would come in pretty handy.

      On a more consumer-relevant scale, laptops and phones that have a standby time limited only by the self-discharge rate of their battery chemistry would be nice, as would desktops for which "suspend" replaces "off" and "flush state" is a special error correction function only, rather than the default "off" state.

      On the server side, rapid supply scaling would be much easier if you could just keep the NIC alive and active(to skip DHCP, which takes a second or two); but able to wake the entire rest of the box, running the hypervisor or server of choice and ready to start handling requests in under a second. The NIC code, and whatever code(hypervisor or kernel) that manages the VMs or server instances would have to be rock solid; but you could kill and restart individual VMs and server instances as needed, and those would presumably be where most of the complexity lives.

  11. Speedy booting? So back to the 80s then by Viol8 · · Score: 2, Insightful

    I don't want to sound like your usual get-off-my-lawn but in the in the days of home computers you could switch it on and it would be ready literally in under a second. Yes I know the "OS" was probably only 16K in size or less but it was in ROM and the computer didn't bother with pointless self checking (you'll soon know if some hardware on your PC isn't working).

    Even early DOS machines could boot in mere seconds. So really all this very complicated technology is doing is bringing us back to where we were 20 or 30 years ago.

    Plus ca change.

  12. Re:Graphene Revolution by demonlapin · · Score: 3, Insightful

    When you start advocating the death, en masse, of your political opponents, you've moved outside the realm of civilized society. Stop. Even if it's "just a joke".

  13. Re:Wishful thinking... by mrnobo1024 · · Score: 2, Insightful

    Again, let's just look at the history. Computers are about 1000x faster than they were in 1980. What does software have to show for it? It's often more of a pain to use (I hate it when software tries to be "smart". Don't second-guess me, just give me an easy way to express what I want to do), and it's buggier than ever.

    Seriously speaking, 1000x faster starts getting near the level of human brains in raw power, so it should be able to run a real artificial intelligence on it.

    Even if this were true, we would have no clue as to how to write one. I have yet to see anyone satisfactorily define "intelligence", let alone propose a plausible algorithm for it. As far as AI is concerned, don't hold your breath.

  14. Re:Wishful thinking... by hairyfeet · · Score: 2, Insightful

    Uhhh...who boots anymore? Everyone has them set to sleep now, which wakes them in seconds. The only time I see a boot anymore is when a customer brings their desktop in, hell even the XP machines that cross my desk are set to sleep when you push the button. Unless your laptop is nearly dead why would you bother? And even then the new Win 7 hybrid sleep is still faster than a boot and has all your apps going, so why do it?

    Don't get me wrong, setting here with a quad and 8Gb of RAM I sure as hell enjoy fast, but to me it makes about as much sense as all those new distros that make a point to brag about their boot speed. Unless the thing crashed, why would you boot?

    And as for "1000x faster than the day before"? I think that won't happen until app programmers decide that "throw moar at it" isn't a valid strategy. While I've run stripped down OSes on monster machines, as well as built rigs for gamers that had truly insane specs, the OS was fast as greased lightning but the apps just got bigger and slower. Just look at how apps like Reader want to use quickstarters to cover up how bloated they've become. I think the only way you'd get 1000x speed is by starting a new OS from scratch and doing like Minuet and writing it for low level code to minimize the bloat. Because it don't seem to matter, FOSS or proprietary, all the apps are getting fatter.

    --
    ACs don't waste your time replying, your posts are never seen by me.
  15. Re:Wishful thinking... by divisionbyzero · · Score: 2, Insightful

    Is it wrong that as fast as things as changing these days, part of me still hopes for one of these '1000x faster in 5 years' technologies to live up to its full promise?

    I know it's coming; if not this tech than surely another one... I guess one hopes to live in interesting times, and I still dream for the day I wake up and there's a computer for sale that shatters Moore's Law. A computer 1000x faster than what was available the day before.

    Faster, please.

    (and thank you)

    If you create stuff, you should know that everything takes longer than you think it will; and, therefore, nothing happens as fast as you expect it to happen.

  16. Re:Wishful thinking... by Amouth · · Score: 2, Insightful

    Read up on Single mode fiber - in multi mode yes its the same problem as with Electrical waves on wires. But in Single mode fiber it's more like a serialized pipe.

    http://en.wikipedia.org/wiki/Single-mode_optical_fiber

    --
    '...if only "Jumping to a Conclusion" was an event in the Olympics.'