Slashdot Mirror


Intel Experimental Processor Runs On Solar Power

An anonymous reader writes "For the IDF keynote, Intel showed an experimental processor that is solar powered (incandescent light shining on a solar panel). The whole computer itself still runs on regular power; only the processor itself is solar. From the article: 'The concept processor, code-named Claremont, can run light workloads on solar power by dropping energy consumption to under 10 milliwatts, said Justin Rattner, chief technology officer at Intel, during a keynote address at the Intel Developer Forum in San Francisco. That is low enough to keep a chip running on a solar cell the size of a stamp.'"

104 comments

  1. Claremont? by Jeremiah+Cornelius · · Score: 1

    I think that's French, for "Koh-e-Noor".

    --
    "Flyin' in just a sweet place,
    Never been known to fail..."
  2. Imagine... by tenco · · Score: 2, Funny

    ...a Beowulf cluster of those (on your roof).

    1. Re:Imagine... by Anonymous Coward · · Score: 0

      It becomes handy on the roof of my spaceship when an ancient, mysterious probe shaped like nightstick from the sky is passing by, emanating devastating energies, calling for whales and heading to Earth. An ad hock solar-sale..should keep me alive..and the Beowulf cluster.. working.

    2. Re:Imagine... by maxwell+demon · · Score: 1

      An ad hock solar-sale

      You want to sell the sun?

      --
      The Tao of math: The numbers you can count are not the real numbers.
    3. Re:Imagine... by Anonymous Coward · · Score: 0

      Why's this marked "Funny"? It's actually quite a good idea.

    4. Re:Imagine... by tenco · · Score: 1

      "Hi, this is Dave calling from UTC+1. Can you do a quick google search for me?" "Sorry, can't do. My proto-sentient house called it a day and went into hibernation a few minutes ago..."

      But: innermost shell of a Matrioshka Brain?

  3. 10mW chip running off 60W bulb by robot256 · · Score: 1, Troll

    10mW cpu / 60W incandescent bulb = 0.16% efficient. Go green technology!

    1. Re:10mW chip running off 60W bulb by blair1q · · Score: 3, Funny

      It's not for you.

      It's for people who actually see the sun sometimes.

    2. Re:10mW chip running off 60W bulb by catprog · · Score: 1, Insightful

      60W incandescent bulb =

      1.26watts light
      58.74 watts heat
      http://en.wikipedia.org/wiki/Incandescent_light_bulb#Efficiency_and_environmental_impact

      Look a how much of the light is shining off the panel and the efficiency of the light->cpu is even better

      And is the cpu using all the power available from the panel?

      --
      My Transformation Website
      Kindle Books http://www.catprog.org/rev
      Interactive CYOA http://www.catprog.org/st
    3. Re:10mW chip running off 60W bulb by TheInternetGuy · · Score: 0

      And on top of that, The photons generated from that 1.26W are spreading radially from the lamp, which means only a very small percent of the light generated by the lamp will hit the "stamp sized" solar PV cell. And also, I did not see 60W mentioned anywhere in the article.

      --
      If my comment didn't sound as good in your head as it did in mine, then I guess we all know who's to blame
    4. Re:10mW chip running off 60W bulb by Anonymous Coward · · Score: 1

      I believe those are called daywalkers.

    5. Re:10mW chip running off 60W bulb by Anonymous Coward · · Score: 0

      yes. I often deride superior engineering and total efficiency because I am completely ignorant about the world and far too important to concern myself with the veracity of my position. I also must assault my political and social enemies with the truth of my delusion. We are victorious! Nothing will defeat our advanced critical thinking abilities!!

    6. Re:10mW chip running off 60W bulb by jiteo · · Score: 1

      The point is that lighting is something that you *will have anyways*, so why not run a processor off of it. They're not proposing adding light bulbs to your house just to power your laptop. I honestly can't believe this got modded 3 insightful at the time of this writing.

    7. Re:10mW chip running off 60W bulb by Mashiki · · Score: 0

      Except you don't always have lights. And in plenty of places you don't even have basic candle light. This is a pretty show piece, nothing more.

      --
      Om, nomnomnom...
    8. Re:10mW chip running off 60W bulb by dbIII · · Score: 2

      Is a solar powered pocket calculator a "pretty show piece"? The first calculators I got to use plugged into wall sockets for power. A solar powered laptop is the obvious progression of that idea that was bound to happen as soon as the power requirements could be reduced enough.
      And as for your objection, nobody said the thing couldn't have a rechargeable battery. If the thing is going to be in darkness forever you use something else for that very contrived edge case.

    9. Re:10mW chip running off 60W bulb by justforgetme · · Score: 1

      Think at the bright side!
      this will bring nerds to the outsides! Except for those who will do the obvious and use 400 lightbulbs to drive their new light powered® computer and drive their electricity costs up 70000%.

      --
      -- no sig today
    10. Re:10mW chip running off 60W bulb by justforgetme · · Score: 1

      The first calculators I got to use plugged into wall sockets for power.

      Hah! the first calculator I got to use was full of people!

      joke aside you are right that is seems just logical for processor developers to find more and more energy efficient designs but I never expected the main manufacturers (Intel|AMD) to come up with it. Not with the never ending Flop fever© of the computer industry.

      --
      -- no sig today
    11. Re:10mW chip running off 60W bulb by maxwell+demon · · Score: 2

      Is a solar powered pocket calculator a "pretty show piece"?

      As long as the rest of the computer needs conventional power, a solar-powered CPU doesn't give you any advantage. What I think they actually wanted to show is "see, our processor needs that little energy that you can even put it on a solar cell!" Given that processors are usually hidden somewhere in a case, having a solar cell attached to it wouldn't usually do much good anyway. Of course if all components of the computer can be made to consume that little energy, the whole computer may be solar cell driven; however in that case you'd certainly not put separate cells for each component; instead you'd power the whole computer from a central solar cell unit.

      Also note that processors needing little energy are useful even without solar cells.

      --
      The Tao of math: The numbers you can count are not the real numbers.
    12. Re:10mW chip running off 60W bulb by TheRaven64 · · Score: 1

      The first calculator I used had a handle on the side that you turned. Addition and subtraction were fine, but multiplication could get really tiring... It did help teach me about logarithms though. Also about gearing when I took it apart.

      --
      I am TheRaven on Soylent News
    13. Re:10mW chip running off 60W bulb by Tomato42 · · Score: 1

      I doubt people would be content with 300x200 monochrome screen in their laptop.

      The biggest power drains in laptops are the screens and hard disks. The latter is fixed with SSDs, the former not so much. Though I think it would be actually possible to create a sub 50mW computer if you used e-Ink display, SSD storage and CPU from the article.

    14. Re:10mW chip running off 60W bulb by robot256 · · Score: 2

      Dear Internet Poster,

      I am sorry to hear that your sarcasm meter is broken. I would like to take this opportunity to explain that my comment was made purely as a joke, and that I am actually a strong supporter of alternative energy and efficiency. I understand the technical feats involved in this project, but I just thought it was hilarious that their "super energy efficient processor" was being powered by an incandescent light bulb. (Not that making fun of climate denialists isn't fun too, but it's not what I was thinking...)

      TC

      P.S., this post was written in a light-hearted manner as well. Don't take it personally.

    15. Re:10mW chip running off 60W bulb by Mashiki · · Score: 1

      Actually yes. Considering that the slide rule could still do more, and more efficiently. Until the rest of the components catch up, a CPU running in that state is nothing more than a pretty show piece.

      --
      Om, nomnomnom...
    16. Re:10mW chip running off 60W bulb by dbIII · · Score: 1

      I get it - you're one of the "don't wake me up until it's made in China and sold at Walmart" mob. Your point is noted but I consider it of very little worth because products do not appear in a single day by magic. Your comment tells us far more about yourself than anything to do with this low power consumption component.

    17. Re:10mW chip running off 60W bulb by ChrisMaple · · Score: 1

      Try again. 10 mW / 60 W is 0.016%.

      --
      Contribute to civilization: ari.aynrand.org/donate
  4. But can it... by Anonymous Coward · · Score: 0

    Nevermind. I'm not gonna bother asking

    1. Re:But can it... by Anonymous Coward · · Score: 0

      Thanks. We appreciate your sacrifice.

    2. Re:But can it... by webmistressrachel · · Score: 0, Troll

      The troll is talking to itself again...

      --
      This tagline was transcoded to result in at least one smirk. If you experience failure to smirk, please consult your Gen
    3. Re:But can it... by webmistressrachel · · Score: 1

      Relax, or die, idiot. The overwhelming majority of human beings frown on your behaviour. You are fucking American bully scum and your treatment of the rest of the world (including me, today) will not carry on forever.

      Noone was pretending to be a mod. Noone was even sure (Even the editors can never to be sure!) whether they were posted by the same person, but it damn well looks like it, what with the first cue-ing the second so well!

      In fact, your voilent reaction seems to imply that you DID do it, and didn't like being called for it! Oh yes. and the fact you're lying about gender again also makes me wonder just how many ACs there - not that many, since you troll an awful lot of my posts with this bullshit that you think I look like a man. Fucking die, already.

      --
      This tagline was transcoded to result in at least one smirk. If you experience failure to smirk, please consult your Gen
  5. Information Void? by UnresolvedExternal · · Score: 1

    Urgh - a quick google unearths nothing more than copy-pastes of this article

    Anyone got something more interesting on the actual tech?

    1. Re:Information Void? by Ken_g6 · · Score: 3, Informative

      AnandTech has some more details.

      --
      (T>t && O(n)--) == sqrt(666)
    2. Re:Information Void? by Guy+Harris · · Score: 2

      Urgh - a quick google unearths nothing more than copy-pastes of this article

      Anyone got something more interesting on the actual tech?

      Well, a quick Google for "claremont site:intel.com" found this page, but I suspect somebody else has already discovered it and posted it here, or the site's running from that processor and somebody turned the light off, as it's responding rather slowly. From the Google summary, it's a "Near Threshold Voltage Processor".

    3. Re:Information Void? by UnresolvedExternal · · Score: 1

      What a de-lightful post :P

  6. Matrix!!! by wilhil · · Score: 0

    Lets block out the sun before it's too late! Rise of the machines!

  7. "Incandescent"?! by smoothnorman · · Score: 1

    (incandescent) so... we've finally found a reason not to use CFLs? ;)

    1. Re:"Incandescent"?! by Khyber · · Score: 1

      Damnit where are my mod points for this?!?!?!

      --
      Still waiting on Serviscope_minor to wake up to fucking reality and realize that Jessica Price isn't going to fuck him.
    2. Re:"Incandescent"?! by Anonymous Coward · · Score: 0

      The processor is actually running at 60 watts, and is 0.0167% efficient.

  8. Light workloads by Curate · · Score: 5, Funny
    The concept processor, code-named Claremont, can run light workloads on solar power...

    That makes sense. Now what if you want to run dark workloads?

    1. Re:Light workloads by Anonymous Coward · · Score: 0

      You have to remove the box cover and expose the CPU to direct sun light to get it working but once it starts working the CPU fan cools it quite a bit, which is why you can use it only for light work loads.

    2. Re:Light workloads by Anonymous Coward · · Score: 0

      no man it doesn't make sense u can be here to know why http://bit.ly/px6KsY

    3. Re:Light workloads by Sulphur · · Score: 1

      The concept processor, code-named Claremont, can run light workloads on solar power...

      That makes sense. Now what if you want to run dark workloads?

      The concept processor, code-named Claremont, can run light workloads on solar power...

      That makes sense. Now what if you want to run dark workloads?

      I find your fear of the dark workload disturbing.

      --

      I'm sorry, we can't print your check; the light bulb burned out.

    4. Re:Light workloads by Anonymous Coward · · Score: 0

      That makes sense. Now what if you want to run dark workloads?

      Then it runs on Dark Energy of course!

    5. Re:Light workloads by quenda · · Score: 1

      How light? Would it be powerful enough to run a pocket calculator?
      I bags the patent on that idea.

    6. Re:Light workloads by Anonymous Coward · · Score: 0

      Wait until sunset.

    7. Re:Light workloads by PiSkyHi · · Score: 1

      Yes, if it were the size of a postage stamp, I could attach the device to my wrist and carry around my solar powered arm based difference engine, its screensaver could display something like the time, the date and possibly current weather conditions.

  9. MSP430 Low Power Demo using Grapes as a Power Sour by Anonymous Coward · · Score: 0

    ce:

    http://www.youtube.com/watch?v=ZxGZIiyyxrM

  10. Now those case windows have a purpose by Joshua+Fan · · Score: 1

    As does in-case lighting... wait a minute...

  11. What does it matter? by Intropy · · Score: 1

    Why would a processor "care" where it gets its electricity? Anyone who owns a computer and has rooftop solar panels or buys electricity from a provider using solar generation presumably has probably run a processor on solar power. Heck, so has anyone who's used a two dollar solar powered pocket calculator. A CPU and its electrical power source are just not coupled concepts.

    1. Re:What does it matter? by timeOday · · Score: 3, Insightful

      If a device's appetite for power falls below what can be gathered passively from the environment, then it doesn't need a battery or power cord, and can run practically forever, which is a big impact.

    2. Re:What does it matter? by Anonymous Coward · · Score: 0

      Indeed... the low power-draw is cool. The fact that they provided it from a small solar panel is neat, but was mainly done for publicity. The wow-factor doesn't reduce the achievement or technological innovation, and comments about this being useless in the dark are clearly missing the point.

    3. Re:What does it matter? by tmosley · · Score: 1

      Yup. I run my laptop off of a solar panel I have rigged up in my front yard. I really need to get another one and a larger battery, though, as I have to switch it back to grid power a couple of hours after sundown.

    4. Re:What does it matter? by AmberBlackCat · · Score: 1

      But does this thing perform any better than the processor in a solar powered calculator?

    5. Re:What does it matter? by Intropy · · Score: 1

      Yes, but that is a story about improvements in the computation power to electrical power consumption ratio. The solar power bit is immaterial.

    6. Re:What does it matter? by Anonymous Coward · · Score: 0

      Because the processor, in the wild, must FIGHT for its survival if it is to be the best of the best to get selected to go in to computers.
      Don't you read computer science wild monthly?

    7. Re:What does it matter? by Anonymous Coward · · Score: 0

      If a device's appetite for power falls below what can be gathered passively from the environment, then it doesn't need a battery or power cord, and can run practically forever, which is a big impact.

      It can run practically forever without a battery or power cord doing what? If the input and output and storage and memory devices it is interacting with still require external power, what is the point of it?

  12. ive got one of those.. by Anonymous Coward · · Score: 0

    from casio...its called a calculator

  13. Impressive... by fuzzyfuzzyfungus · · Score: 1

    The "solar" bit is pretty much a gimmick; but I suspect that the underlying technology will be something that Intel finds very handy indeed, for chips of a variety of power levels.

    Apparently, Intel has been working on bringing down the Vcore as sharply as their process capabilities allow. Lower core voltage, substantially lower power consumption, all else being equal(as people overvolting their CPUs tend to find out quickly...) It remains to be seen if Intel will be able to do this cheaply enough to actually push their power use down across the board, or if this will end up being a cherry-picked-and-blessed-for-10x-the-price ultramobile and very high density compute thing only; but being able to shove Vcore down is a nice piece of process research.

    1. Re:Impressive... by Anonymous Coward · · Score: 0

      This is basically a slimmed down P4 with all the impuritys removed, to overclock use a magnifying glass.

    2. Re:Impressive... by Electricity+Likes+Me · · Score: 1

      This isn't really something Intel can do optionally though. With the rate power consumption of CPU cores has been increasing, they need to do something about it pretty much now since it's not going to be possible to air-heatsink that much heat if it keeps going up.

    3. Re:Impressive... by petermgreen · · Score: 2

      With the rate power consumption of CPU cores has been increasing

      You are a few years behind the curve. Since the pentium D CPU TDP (roughly power consumption under the highest normal load) has stayed pretty much flat while core counts and performance of individual cores have gone up (despite the drop in clockspeeds)

      The power consumption per core has been going down in recent times. The pentium D 965 extreme has two cores, a clockspeed of 3.73 GHz and a TDP of 130W. The i7-990x has six cores.a nominal clockspeed of 3.47 GHz (plus turbo boost) and a TDP of 130W.

      Comparing the two processors (made about 5 years apart) the later processor has the same TDP, three times as many cores and slightly lower clockspeed (but each core has far better performance at a given clockspeed)

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    4. Re:Impressive... by Electricity+Likes+Me · · Score: 1

      The point still stands though - super-low power consumption has to be an active area of research for them since we've been at about the maximum manageable power consumption by a CPU for a while now. Increases in transistor counts need to accompany decreases in power consumption per transistor.

  14. yes, yes it does! by Gravis+Zero · · Score: 1

    The chip is an experimental Pentium CPU and ran on a PC with the Linux operating system.

    --
    Anons need not reply. Questions end with a question mark.
  15. Great by Dunbal · · Score: 1

    Now botnets can really run 24/7 when people never sleep or shut down their machines to save electricity!

    --
    Seven puppies were harmed during the making of this post.
    1. Re:Great by Baloroth · · Score: 1

      A solar-powered processor running 24/7? I didn't know polar bears and penguins owned computers.

      --
      "None can love freedom heartily, but good men; the rest love not freedom, but license." --John Milton
  16. Re:MSP430 Low Power Demo using Grapes as a Power S by mug+funky · · Score: 1

    those grapes can be a little sour when they're unripe.

  17. I proposed this at lunch at IBM Research ~ 1999 by Paul+Fernhout · · Score: 1, Interesting

    I suggested IBM could make chips that get power from a solar cell integrated with them, and that communicate with each other via either light or radio (so, no need for a backplane or wire harnesses, and potentially the light could even be directable to build ad-hoc networks across an open central space if the chips were on the inside of a sphere). No one took it very seriously. In college, around 1984, I suggested a desktop computer that was the desktop and was a monolithic several centimeter thick optical computer (the reaction was mostly just bemusement). A couple years earlier I'd suggested in a physics term paper how optical computer links between chips would probably be needed to do AI (the professor could not understand what that had to do with physics).

    Guess it's just hard being ahead of your time. :-) The wages of reading too much sci-fi I guess -- where I first read about optical computing and communications ideas. Of course Hal 9000 had an optical computing core, and IBM helped with some ideas there, so maybe ideas can come full circle?

    Glad to see people are finally making them, even if not IBM. Although maybe not so glad, as they could soon become "smart dust" and there goes the rest of our privacy (see Vernor Vinge who uses the smart dust theme in at least two stories).

    --
    A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.
    1. Re:I proposed this at lunch at IBM Research ~ 1999 by Anonymous Coward · · Score: 0

      Nothing to see here, move along. Your future will be ready in 20-50 years, same as last time, and before that, and before that.

      Maybe if you spent less time reading SF, and more reading TFS, you'd realize that this has no connection to any of your ideas about optical communication and/or computation. It's a low-voltage processor running on electrical power, computing with electronic gates, and communicating to the rest of the electrically powered systam with electric signalling. Not a shred of photonics to be seen, unless, as a publicity stunt, you use an external PV cell to provide the electrical power to the processor.

    2. Re:I proposed this at lunch at IBM Research ~ 1999 by Anonymous Coward · · Score: 0

      I suggested...

      I suggested...

      I suggested...

      Guess it's just hard being ahead of your time. :-)

      No, I think you'll find what's "hard" is actually making things that other people talk or dream about :-)

    3. Re:I proposed this at lunch at IBM Research ~ 1999 by wagnerrp · · Score: 1

      That's fantastic and all, but you're completely missing the point. The fact that it was powered by a PV cell is irrelevant. What they are showing off is a Pentium core and DDR3 memory idling at 10mW. They are showing off a processor that has a hundred fold or more difference between idle and load. You no longer shut off your PC, you no longer put it in standby, you just let it go into low power mode indefinitely. The PV cell was just a demonstration of how low power it was.

    4. Re:I proposed this at lunch at IBM Research ~ 1999 by Anonymous Coward · · Score: 0

      To be blunt: no one took your ideas seriously because they were stupid ideas that showed you had no idea what you were talking about, and they were all too nice to laugh in your face. Sorry. You're now patting yourself on the back for something completely unrelated, so it seems someone really needed to do the laugh-in-your-face part before you keep going on about it again.

    5. Re:I proposed this at lunch at IBM Research ~ 1999 by drinkypoo · · Score: 1

      That's fantastic and all, but you're completely missing the point. The fact that it was powered by a PV cell is irrelevant. What they are showing off is a Pentium core and DDR3 memory idling at 10mW.

      Try reading TFA with your English parser active. The memory experienced a sevenfold power reduction and the CPU "can run light workloads" at "less than" 10mW (which in my book has me counting nines.) DRAM still has to be refreshed and SRAM still sucks power. (Where's my MRAM?) Also, a classic weakness in Intel solutions has been the power consumption of the chipset. When the Athlon 64 came out and eliminated the north bridge the desktop offering was lower-power than Intel's powerful mobile chip of the time in a complete solution because the TDP of CPU+Chipset was lower.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    6. Re:I proposed this at lunch at IBM Research ~ 1999 by Asic+Eng · · Score: 1

      Well, I don't see much value in stating great goals without any concrete ideas of how to achieve them, unless you are really the first person to actually think of the goal itself. Saying it would be great to have a processor consuming only a few mW is not very helpful. Yes of course it would be great, people have been working on reducing power consumption for a long time.

      Actually having an idea how to reach that goal - now that's something interesting to hear about.

      Also you seem to have misunderstood the article, this device neither integrates a solar cell, nor does it have anything to do with optical computing.

    7. Re:I proposed this at lunch at IBM Research ~ 1999 by Paul+Fernhout · · Score: 1

      "No, I think you'll find what's "hard" is actually making things that other people talk or dream about :-)"

      True, but that is also hard to do when most of the social resources to do that are diverted into "me, too" redundant competition, paperwork related to that like patents, or, alternatively, military arms races.

      I would have been happy to work on the details of all sorts of neat socially-useful devices. (I've ended up do mostly software because it was cheaper to do that as a small independent compared to stuff like robotic hardware or things requiring lab chemistry.)

      Essentially, there is little support for fundamental or basic civilian research anymore, including at places like IBM Research. The few slots to do that are intensely fought for these days, meaning the best social infighters tend to get them (not saying some of them are not good scientists, too).

      So, what is really hard is actually making things with essentially no budget while also having to do something else so you can get ration units to pay your bills and also trying to be a nice person. :-)

      Still, the scale of our society is so big that fundamental stuff happens anyway here and there. But it is such a lie compared to the picture I was painted in the 1970s when I was in school. Part of the reason:
      http://www.its.caltech.edu/~dg/crunch_art.html
      http://disciplinedminds.com/

      That's one reason much better fundamental science will flourish with something like a "basic income".

      Here is a 12 minute YouTube video I just made that talks about a balance between five interwoven economies that shifts with cultural change and technological change:
            http://www.youtube.com/watch?v=4vK-M_e0JoY
      A PDF file of the presentation is here:
            http://www.pdfernhout.net/media/FiveInterwovenEconomies.pdf

      Also indirectly related to life in big organziations:
      "Smile or Die"
      http://www.youtube.com/watch?v=u5um8QWWRvo
      "Have Fun at Work"
      http://infohost.nmt.edu/~shipman/org/hfaw.html
      "It is dangerous, and often fruitless, to try and solve problems without considering the underlying social system. ... On a purely practical level, this book is an excellent survival manual for results-oriented engineers who have developed attitude problems about the structural barriers to success in their work environments. Livingston discusses how to evaluate your social structure's potential for success, ways to get working projects out the door in spite of these barriers, and how to tell when you're wasting your time even working there."

      --
      A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.
    8. Re:I proposed this at lunch at IBM Research ~ 1999 by Paul+Fernhout · · Score: 1

      "Your future will be ready in 20-50 years, same as last time, and before that, and before that."

      True:
      http://en.wikiquote.org/wiki/William_Gibson
      "The future is already here -- it's just not very evenly distributed."

      "Maybe if you spent less time reading SF, and more reading TFS, you'd realize that this has no connection to any of your ideas about optical communication and/or computation. "

      Maybe. But, sadly, I don't read much sci-fi these days (no time). About the only sci-fi thing I read in the past few months, and coincidentally related to Intel:
      http://techresearch.intel.com/newsdetail.aspx?Id=30
      "The Last day of Work" by Douglas Rushkoff"

      While the comments are right that for whatever reason I was not paying attention that the solar cell was not on the chip, none-the-less, what about low-power operation is not related to the idea of chips that can run on incidental light? Such projects may take multiple innovations to make happen, as is common with fundamental research. The point is that with such chips, you can build a massive system cheaply because you just drop the chips on a surface without much infrastructure (putting them into a sphere might require glue though).

      In any case, documenting previous discussions can potentially serve to invalidate later patents (and I'm past the limits of my confidentiality non-disclosure agreement on that time).

      --
      A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.
    9. Re:I proposed this at lunch at IBM Research ~ 1999 by Paul+Fernhout · · Score: 1

      True overall, including on my misperception on the TFA, but see my other comments in this thread.

      Also, on: "Well, I don't see much value in stating great goals without any concrete ideas of how to achieve them".

      That kind of ignores that notion of "fundamental research" which you would think a big research organization ideally would do more of. Related:
          http://en.wikipedia.org/wiki/Pure_research

      Also, it ignores the notion of research as a social endeavor.

      Why should people, especially in a research setting, be expected to have the answer when they raise the question?

      --
      A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.
    10. Re:I proposed this at lunch at IBM Research ~ 1999 by Paul+Fernhout · · Score: 1

      AC wrote: "To be blunt: no one took your ideas seriously because they were stupid ideas that showed you had no idea what you were talking about, and they were all too nice to laugh in your face. Sorry. You're now patting yourself on the back for something completely unrelated, so it seems someone really needed to do the laugh-in-your-face part before you keep going on about it again."

      Thank you, AC. :-) It's good to stay humble, true.

      For reference:
      http://web.archive.org/web/20010405020550/http://www.cascadepolicy.org/dee_hock.htm

      --
      A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.
  18. RCA's CDP-1802 did that 30+ years ago by stox · · Score: 1

    Glad to see Intel finally catching up.

    --
    "To those who are overly cautious, everything is impossible. "
  19. Sea-Change by Cogent91 · · Score: 1

    WOW. A stamp sized solar array accompanying a processor provides an indefinite power source?? If so, combine that with pico-electic power from a watch and you might be able to get a wristwatch sized processor to have suffecient power to do simple things like communicate with wi-fi and render simple apps. I think this potential is going to be a sea-change; I can hardly wait.

  20. Retro Processor by Anonymous Coward · · Score: 0

    RCA, 1966 patent

  21. Why solar? by Anonymous Coward · · Score: 0

    SCADA uses aside (and even these would be dubious, how would you power the rest?), why solar? Why not make use of the fact that you have a CPU that sips 10mW and just feed it conventional power? I don't know about you guys, but the idea of leaving a computer item in direct sunlight (why shine a 60W bulb to transfer 10mW energy?) kind of seems like it goes against logic.
     
    Apparently any time you can stick "green" or "solar" into a summary, it's auto-frontpage....

    1. Re:Why solar? by Electricity+Likes+Me · · Score: 1

      What is the major mental deficit that leads to a bunch of people hearing that Intel in a demonstration indoors in a conference hall used a 60W light bulb to power there demo low powered CPU, concluding that obviously Intel are saying that under normal circumstances you would use a 60W light bulb to power said processor.

    2. Re:Why solar? by Lumpy · · Score: 1

      Well... if you can get a PC to run off of 10mw I have a power source that would be cool, and it's not Solar...

      4 surplus pacemakers from the 70's and 80's those use a nuclear battery How would you like a 20 year battery for your laptop?

      --
      Do not look at laser with remaining good eye.
  22. 10mW, you can get that from... by BillX · · Score: 1

    While the whole solar-powered thing is kind of a gimmick, the fact that they have an (x86-compatible?), 'real' CPU (not microcontrollers) operating on 10mW is pretty impressive. This is a level where powering from ambient motion and a user's body heat is also feasible.

    --
    Caveat Emptor is not a business model.
  23. Would the last person to leave... by thegarbz · · Score: 2

    Would the last person to leave, please DON'T turn off the lights.

    1. Re:Would the last person to leave... by Anonymous Coward · · Score: 0

      What about those motion detector, turn the lights off set-ups??? (Am I the only one annoyed by those?)

  24. Yes, well done Intel, but ok..... by Anonymous Coward · · Score: 0

    The CPU was running on solar power, and hung when it got a bit dark.

    The rest of the lashup was connected to a conventional PSU. You know, things like the graphics card, the logic glue on the motherboard and so on consume LOTS of power. A CPU that has a low power consumption is pretty. Call me back when an entire general purpose computer can run off solar power.

    Wait - what about the Raspberry Pi? 1 watt power consumption, runs Linux. Connect one of those solar recharging units. What more do you need?

    *grin*

    1. Re:Yes, well done Intel, but ok..... by Lumpy · · Score: 1

      Ring Ring...

      Pick any computer you like from Dell or other sources and I can make it run on solar with no effort. just add enough solar panels and inverter to meet it's watt need.

      Running on solar is the 2011 equivalent of 1999's innovation called "on the internet"

      Look I'm a genius! I'm doing accounting..... ON THE INTERNET!!!!

      --
      Do not look at laser with remaining good eye.
  25. Go all the way! by spectrokid · · Score: 1

    Booh! Picture shows processor connected to solar panel with 2 wires. (+ and - I presume) Why not go all the way? Solar panels ARE made of silicon. Interweave the transistors into the panel itsels, generating power where needed. And get rid of the 2 ugly wires! Imagine having your CoreI whatever on the roof! (Or more realistically, the processor on the outside of your netbook.)

    --

    10 ?"Hello World" life was simple then

    1. Re:Go all the way! by klingens · · Score: 1

      CPUs are made of monocrystalline Si, solar panels are made by polychristalline Si. Dotation (sp?) is most certainly different as well.

    2. Re:Go all the way! by vnsnes · · Score: 1

      Seems like research supports this: ieeexplore.ieee.org/iel5/16/5896011/05773485.pdf?arnumber=5773485

  26. Solar Cars contest, Solar CPU contest? by Anonymous Coward · · Score: 0

    There is a Solar Car contest, let them do a Solar CPU contest. If an ARM CPU is allowed to take part then you better get a Samsung NC215S solar netbook.

  27. What's that movie by Anonymous Coward · · Score: 0

    Ow yeah the matrix. Now we just have to destroy the sky....ow wait.

  28. Ignore the solar part, read the mV part. by Anonymous Coward · · Score: 0

    Running processors from solar panels has been around ever since, well, the solar powered calculator. With modern semiconductor processes and lower voltage, having a processor capable of decoding x86 from a postage stamp sized solar panel isn't too much of a stretch.

    The near-threshold-voltage thing is the awesome part... according to this link, the thing can *run* at 10mV!

    http://www.anandtech.com/show/4820/rattner-shows-off-near-threshold-voltage-intel-architecture-cpu

  29. Oh good god...... by Lumpy · · Score: 1

    Why is everyone doing this "IT RUNS ON SOLAR!!!! crap?

    The Pentium III runs on solar! Give me $600.00 and I'll go to harbor freight and buy 3 of their solar panels kits and run an entire computer from 1999 on solar! Look it's the first solar Desktop computer!

    This is nothing special. I have a VIA C5 processor in a motherboard that draws less than 15 watts when going full boat from 5 years ago that will run on solar. That makes VIA far better at this because they are 5 years ahead of intel!

    everything in my house will run on solar!

    --
    Do not look at laser with remaining good eye.
    1. Re:Oh good god...... by luther349 · · Score: 1

      yea i know whats a ssd single core atom eeepc draw 35 watts. so a cheap ass 50 watt could charge and run it.

  30. That's what this is all about by Quila · · Score: 1

    The solar aspect is just a visual demonstration to get across how low-power the chip is.

    They might have powered it with a potato if they could get it low-power enough.

  31. Nice ... by Anonymous Coward · · Score: 0

    now we can finally shelf our solar pocket calculators from the 1970ies!

  32. Why Not? by Anonymous Coward · · Score: 0

    30 years ago they ran calculators the size of postcards with solar panels.