Forgot your password?
typodupeerror
Technology Hardware Science

A Plasmonic Revolution for Computer Chips? 188

Posted by timothy
from the oooh-lookit-move-around-all-squishy-like dept.
Roland Piquepaille writes "Today, we're using basically two ways to move data in our computers: transistors carry small amounts of data and are extremely small, while fiber optic cables can carry huge amounts of data, but are much bigger in size. Now, imagine a single technology combining the advantages of photonics and electronics. This Stanford University report says a new technology can do it: plasmonics. (For more about plasmons, read this Wikipedia article.) Theoretically, it is possible to design plasmonic components with the same materials used today by chipmakers, but with frequencies 100,000 times greater than the ones of current microprocessors. There is still a challenge to solve before getting plasmonic chips. Today, plasmons can only travel a few millimeters before dying, while today's chips are typically about a centimeter across. Read this overview for more details and references about plasmonics, and to discover why it's one possible future for chips' circuitry."
This discussion has been archived. No new comments can be posted.

A Plasmonic Revolution for Computer Chips?

Comments Filter:
  • Alright (Score:5, Interesting)

    by OverlordQ (264228) on Monday April 04, 2005 @06:01PM (#12138371) Journal
    Lets keep it simple, put all of the Roland Piquepaille [thedarkcitadel.com] conspiracy posts here. :)

    Editors: GIVE HIM HIS OWN DAMN SECTION SO CAN HIDE HIS POSTS
    • Re:Alright (Score:1, Offtopic)

      by Doc Ruby (173196)
      Why would we hide his posts? They're interesting, and cost me the same to read as any other (non-reg) Slashdot linked stories: nothing. That said, I would like a feature that lets us block stories by submitter on our own Slashdot pages.
      • Why would we hide his posts?

        Because, the articles come from Roland as if he was just another Slashdot user, which then gives more credibility to the submission [1].

        In reality, it's quite clear that there is some sort of business connection between Roland and Slashdot. There is no other reason to accept 100% of Roland's submissions.

        It's a desceptive relationship.

        [1] Except that the credibility is undermined by the constant complaining of the Slashdot readers.
    • Re:Alright (Score:1, Insightful)

      by Anonymous Coward
      It's not that we need a "Roland Piquepaille" section; it's that we need the ability to filter stories by submitter.
    • If you're wondering why all this fuss about the Roland Piquepaille problem, check out the rpiquepa's Recently Accepted Submissions [slashdot.org].

      6 articles were submitted in the last month, NONE were rejected. If there were any Rejected articles, they would be displayed under a "Recent Submissions" section.

      What are the chances that the Slashdot editors accept 100% of Roland's submissions, when they reject the majority of submissions from other people.

      When was the last time YOU had a story accepted by the Slashdot crew
      • by Neil Blender (555885) <neilblender@gmail.com> on Monday April 04, 2005 @06:22PM (#12138610)
        6 articles were submitted in the last month, NONE were rejected. If there were any Rejected articles, they would be displayed under a "Recent Submissions" section.

        Not true. You only see your own rejected submissions. Other people can only see your accepted submissions.
      • What are the chances that the Slashdot editors accept 100% of Roland's submissions, when they reject the majority of submissions from other people.

        Who cares?

        The real question is whether or not the articles he submits are worth the time of Slashdotters to read and reply to them. I'm not entirely sure what the conspiracy theory is--that he has some arrangement with the editors to accept the stories is one thing I've heard--but I frankly don't care WHO submits a story or WHERE they link it to as long as

    • Re:Alright (Score:1, Offtopic)

      by CSMastermind (847625)
      In the mean time we can spread the word about what he's doing, not visit his site (I blocked it at our school) and say, "Thank you for the information, it's neat now I'm going to google the topic and find out about it".
      • Re:Alright (Score:2, Insightful)

        by Anonymous Coward
        Blocking many people's access to a web-site based on your personal opinion is censorship and is, in my opinion, a bad idea. You can educate people about what he is doing and your opinion about it, but removing other people's freedom of access to information is unethical.
    • Re:Alright (Score:5, Funny)

      by Stalyn (662) on Monday April 04, 2005 @06:21PM (#12138594) Homepage Journal
      i'll take Roland Piquepaille over Jon Katz any day.
    • "Editors: GIVE HIM HIS OWN DAMN SECTION SO CAN HIDE HIS POSTS"

      That'll never happen. Every time a Roland story pops on, a bunch of twerps come in and bitch about it. For every bitch-post, there's an ad served. For every ad served, bling bling in Slashdot's pocket.

      If you guys would take off your tard hats for a moment, you'd discover that the best way to get rid of the guy is to simply stay out of his threads.
  • The future is now. (Score:5, Interesting)

    by Sheetrock (152993) on Monday April 04, 2005 @06:01PM (#12138372) Homepage Journal
    Not only for plasmonics, but for mutable instruction sets. There has been a tendency in computing innovation to withdraw to yesterday's discoveries. Tried-and-true approaches offer the twin comforts of backwards compatibility and tested reliability, attractive propositions to the modern CEO or venture capitalist savvy enough to recognize the additional benefit of recognizing further gains on already completed research. Unfortunately, and in my opinion, this follow-the-leader approach has lead to stagnation in CPU development. I'll explain using a simplified analogy for the benefit of the less technically-inclined.

    Let us think of a computer processing unit as a juggler, and bytes as mangoes. Older CPUs would juggle one mango at a time, and frequently require modifications to the stage to boot. Around the 1980s, they could juggle two mangoes. Then four around 1990, and today as many as eight at a time! Now you would be expected to be quite impressed with each leap, notwithstanding the fact that you really wanted a juggler that could handle melons, grapefruit, or watermelon slices instead of (or in addition to) mangoes. In addition, the fact that you are juggling in a zoo where a primate is free to grab your fruit and substitute twigs (or worse!) mid-juggle owing to something called "stack smashing" in computer terminology is not supposed to discourage you.

    There is a movement towards something called mutable paragraphs, where as in English "words" (groups of bytes) can be of different lengths depending on need. This may mean the ability to exactly fill out a data page for better efficiency, or to allow the CPU to work with communication protocols in their element (if a common network packet is 68 bytes long, a word should be ½NP or 34 bytes in the I/O buffer.) It also means that you use no more CPU space than you absolutely need to for a computational step, decreasing wear and tear on your components.

    I guess what I'm getting at is that science fiction has nothing on practical interative design for real world technological improvement. Sure, we might get to the same place we read about 50 years ago, but not all in one step.

    • by Anonymous Coward on Monday April 04, 2005 @06:06PM (#12138426)
      +5 You Made All That Shit Up Didn't You?
      • by Anonymous Coward
        No, he's a cut-n-paste troll. I feel kind of bad about blowing the whistle on him, as he's a beautiful example of what all trolls should aspire to, but, honestly, aren't people going to realize that this talk of stack smashing and juggling mangos sounds familiar?
    • by Anonymous Coward
      Copied from his earlier post here [slashdot.org]
    • Let us think of a computer processing unit as a juggler, and bytes as mangoes. Older CPUs would juggle one mango at a time, and frequently require modifications to the stage to boot. Around the 1980s, they could juggle two mangoes. Then four around 1990, and today as many as eight at a time! Now you would be expected to be quite impressed with each leap, notwithstanding the fact that you really wanted a juggler that could handle melons, grapefruit, or watermelon slices instead of (or in addition to) mangoe
    • wouldn't plasmonics make van eck phreaking not feasible??
    • ... practical interative design ...

      Did you mean "interactive" or "iterative"? It's just one letter difference either way!

    • by bperkins (12056) *
      Try googling "Let us think of a computer processing unit as a juggler, and bytes as mangoes."
  • by Neil Blender (555885) <neilblender@gmail.com> on Monday April 04, 2005 @06:02PM (#12138376)
    But they might need to rephase the modulators and run in through some sort of tachion inverter feild.
    • That's good, but if we route a phased graviton pulse through the GNDN tubes, and output through the main deflector dish, we can probably get an order of magnitude increase. We'll need to divert warp power to the shields to protect against reflected Bertol Rays, though.

      Either that, or we change the gravitational constant of the universe. Take your pick.
    • That is where the flux capacitor comes in handy.
    • don't forget to modukate the shield harmonics- that's always the solution!
  • Heat (Score:3, Insightful)

    by TheKidWho (705796) on Monday April 04, 2005 @06:02PM (#12138387)
    How much heat are these things going to generate though. Because you know E = hf, if you have 100,000 times the frequency, your going to need to throw in 100,000 times the energy!! Of course that is simplification of what is really happening with these kinds of chips and it is much more complicated then just 100,000 times the energy needed. But it seems like these things might make the Pentium IV seems like a fridge!
    • Re:Heat (Score:3, Insightful)

      by markana (152984)
      This is why the plasma conduits in the Federation control panels keep blowing up in their faces...
      • No... the reason why "the plasma conduits in the Federation control panels keep blowing up in their faces..." is 'cos it looks cool on TV.
    • 6.63*10^-34 * 10^5... :) not much i would guess :)
    • Re:Heat (Score:3, Funny)

      by ralphclark (11346)
      Piffle! Just use a smaller value of Planck's constant! Easy. :o)
      • Re:Heat (Score:3, Insightful)

        by owlstead (636356)
        Fortune cookie for you, Mr Dallas

        The primary purpose of the DATA statement is to give names to constants;
        instead of referring to pi as 3.141592653589793 at every appearance, the
        variable PI can be given that value with a DATA statement and used instead
        of the longer form of the constant. This also simplifies modifying the
        program, should the value of pi change.
        -- FORTRAN manual for Xerox Computers
        • ...that many early ForTran compilers didn't check too closely to see if what they were assigning to was an LValue, and since a float was six bytes and a pointer to a float was 2 bytes, a compiler would typically store a copy of a constant somewhere and refer to is using the pointer, just as if it was a variable. The end result is that if you did a typoe and executed something that said:

          PI = R * R
          2 = PI

          then from that point onwards, both 2 and PI would assume the value of R squared, so if R started out bein

    • Re:Heat (Score:5, Informative)

      by barawn (25691) on Monday April 04, 2005 @06:29PM (#12138648) Homepage
      How much heat are these things going to generate though. Because you know E = hf, if you have 100,000 times the frequency, your going to need to throw in 100,000 times the energy!! Of course that is simplification of what is really happening with these kinds of chips and it is much more complicated then just 100,000 times the energy needed. But it seems like these things might make the Pentium IV seems like a fridge!

      Power does usually scale with the frequency, but it also scales with the signal strength (number of carriers: intensity in a photonic case, ~voltage in an electronic case). If you can up the frequency by a factor of two and cut the voltage (for instance) by a factor of two, it's the same power usage.

      Of course, using E = hf is completely wrong here - that's the energy of a photon, and in a completely photonic chip, wouldn't matter in the tiniest bit - because the photons are emitted at one point, and absorbed at another, so there's no net energy loss.

      Most of the places where the frequency dependence comes in are energy losses - like the resistance of a wire. With light, there's very little energy loss (in a fiber, for instance), so the chip will run very, very cool.
      • Power usage is generally linear with frequency, but proportional to the square of the voltage. Your example would use roughly half the power.
      • While I will agree there are other factors that affect a CPU, this formula does provide a bottom reference value to demonstrate that increased CPU frequency... regarless of effeciency... will always consume more energy.

        There are other fundimental constants in information theory that demonstrate a quantum effect for a single bit of data being manipulated... regardless of the effeciency of the device that is being used. There are fundimental information theory limits to how little energy can be consumed to
        • There are fundimental information theory limits to how little energy can be consumed to flip that bit, and the formula of E=hf is a good place to start and try to figure out just how much energy must be used to change a one to a zero and back. The emmission and absorbtion of photons will increase entropy, and will eventually lead to a loss via emmision into the IR band. This generates heat.

          Well, yes, but that energy is miniscule, even at optical frequencies. Even if you assume that it was changing a megab
          • If we are going to get CPU speeds to Terahertz or Petahertz frequencies, I will have to agree that the physical restraints of the speed of light are going to be a very major factor with CPU designs. Also, I don't know of any physical device manufactuer who is even remotely claiming even a 1% efficiency for storage and manipulation of a bit. (That would be a huge marketing ploy if it ever were achieved.) Physical devices, even optical systems, are far less efficient than that. There is no way that they a
            • Also, I don't know of any physical device manufactuer who is even remotely claiming even a 1% efficiency for storage and manipulation of a bit. (That would be a huge marketing ploy if it ever were achieved.) Physical devices, even optical systems, are far less efficient than that. There is no way that they are > 100,000 times more energy efficient than conductor/semi-conductor systems if simply because it would have been done already if the savings were that substantial.

              Optical systems are orders of ma
              • You realize that you are considering the speed of light to be a "soft" issue (as opposed to the "hard" limits from energy concerns). What makes you think that the speed of light is any less "hard" than the other concerns?

                I think that FTL communication is a virtual impossibility, and some very fundimental physics reasoning must be developed to make it happen.

                I only comment about FTL communication because the U.S. Patent & Trademark Office granted a patent for FTL communication. The examiner must ha

                • Someone (perhaps with a perpetual motion machine) must think this is possible to accomplish.

                  And if that person has a perpetual motion machine and FTL travel, then he can get around both the speed of light issues and the heat issues!
      • You ahve to charge a line with capacitance C above a certain threshold voltage V to open the transistor gate F times a second, giving you dissipated power of F*CV^2/2 just for that one line -- note V _squared_.

        Of course if you switch to superconductor logic you would not have to chagre the whole line (and there is no voltage except the instance Josephson jucntion switches), but this is a topic for another discussion.

        Other than that, yes, you are correct that you will need roughly 100,000 times more energy
      • Of course, using E = hf is completely wrong here - that's the energy of a photon, and in a completely photonic chip, wouldn't matter in the tiniest bit - because the photons are emitted at one point, and absorbed at another, so there's no net energy loss.

        Since that's the most intelligent remark I've seen for this story, I'm going to jack this thread. Plasmons still seem like magic to me, but I get to use the technology for something useful while electrical engineers figure out how to make plasmon comput

  • by MLopat (848735) on Monday April 04, 2005 @06:04PM (#12138408) Homepage
    Another use for this technology maybe rendering objects invisible to the observer. Using plasmons to stop light from scattering back to the observer's eyes. Unfortunately it can only be used to hide very small objects since the wave lengths of the light need to be near the size of the object that reflecting them.

    If anyone wants anymore info on this check out this link [slashdot.org].
  • Plasmonics does not sound like a perfectly cromulent word. Are you sure that the authors haven't embiggened the word a bit a bit? I mean, it's not unpossible...
  • by Valiss (463641) on Monday April 04, 2005 @06:08PM (#12138452) Homepage
    Today, plasmons can only travel a few millimeters before dying, while today's chips are typically about a centimeter across.

    Well, in that case, it sounds similar to my research. See, if you jump, you can fly. Now currently, I can only fly a foot or two. Of course, most people want to fly longer distances, but it's a start.
    • Aha! Another Yogic Flyer [mum.edu]!
    • Re:Sounds like.... (Score:3, Insightful)

      by RichardX (457979)
      Sounds a bit like my research into anti-gravity by using parachute jumps.

      Y'see, I started with a full size parachute, and each time I make a jump I use a slightly smaller chute than last time. Eventually I won't need a parachute at all!
    • The trick to flying is to jump and miss the ground. Perhaps this will improve your results.
    • Ahhhhh I see your problem... you're trying to fly by jumpin UP. This will never work....

      You need to start by throwing yourself at the ground from a high point, so gravity is giving you a big helping hand... and then the most important step... practise MISSING the ground... Therein lies the secret.
  • by Rightcoast (807751)
    I know next to nothing about this field but wonder if it would be possible to bridge the gap in distance using using metal or ceramic nano-wires embedded on-chip for the plasmons to travel across?
    • Or couldn't they build the equivalent of buffers to repeat the signal at defined intervals? Heck even if it slows down the signal propagation by 3 or 4, the increased speed in general would probably be worth it...
  • Wiki Free (Score:1, Interesting)

    by onyxruby (118189)
    Is it really too much to ask for a wiki free link for reference? Articles written by the tyranny of the persistant dont tend to have much to do with reality. Really, I'm sure some school, corp, journal or industry site probably has something about this. How about a link to a google search, or something with some shred of credibility?

    Wikipedia, because the tyranny of the persistant must be right.

    • Re:Wiki Free (Score:5, Insightful)

      by kebes (861706) on Monday April 04, 2005 @06:54PM (#12138875) Journal
      As someone who posts using links to wikipedia occasionally, I must say that I object. When I'm looking for a source to explain what I'm talking about, I simply reference the best URL I can find. Very often, that is wikipedia. It doesn't really matter who hosts the data if I know it to be valid.

      As someone who has done some research on surface plasmons, I find the wikipedia article on Plasmon to be accurate and useful, so I think it is a good reference. Not all wikipedia articles are so good, but then again I don't reference the bad ones.

      On the other hand, you are pointing out that we shouldn't accept wikipedia articles just because wikipedia is cool and lots of people edited the article so it must be right. Yes, that's valid. However, as with *all* sources of information, whether it is a wiki or slashdot or an encyclopedia or the local news, the end-consumer MUST use his judgement to decide if the information is valid or BS. It is an illusion to think that traditional sources of information are error free. In all cases, the reader must simply use judgement and double-check if things seem wrong.
      • Re:Wiki Free (Score:2, Insightful)

        by LiquidCoooled (634315)
        I don't think he is complaining about wikipedia in itself.
        More the fact that a person with an agenda could replace clean methodical bias free information with drivel changing the view for everyone in the process.
        The original source may not have the time or inclination to maintain his articles, so the biased view remains.

        It may be required in the long run to have a karma/points system - much like slashdot, where particular versions of articles can be rated and those written by established experts gain great
        • I don't think he is complaining about wikipedia in itself.
          More the fact that a person with an agenda could replace clean methodical bias free information with drivel changing the view for everyone in the process.
          The original source may not have the time or inclination to maintain his articles, so the biased view remains.


          This, by the way, is one of the things the Xanadu hypertext system was intended to prevent.

          A link-end was not just to a particular page, but a particular section of text (as small as a si
  • Plasmonics? (Score:3, Funny)

    by tekrat (242117) on Monday April 04, 2005 @06:11PM (#12138487) Homepage Journal
    Wasn't that a Punk Rock Band from the 80's with Wendy O'Williams?
  • by Sebastopol (189276) on Monday April 04, 2005 @06:14PM (#12138508) Homepage
    ... is making semiconductors???

    Oh, plasMONICS... my bad...

    (I know, I know: she's deceased)
  • by loqi (754476) on Monday April 04, 2005 @06:19PM (#12138561)
    "He worked at Plasmonics Institute, just an engineer in a sealed white suit... he did a good job in the computer race, but his bosses didn't like him so they shot him into space!"
  • What the...? (Score:5, Interesting)

    by barawn (25691) on Monday April 04, 2005 @06:19PM (#12138571) Homepage
    Claiming "100,000 times the frequency" is a little misleading. You're not talking about the processor running at terahertz speeds - simply put, you can't make things small enough to do that. Plasmonic signals, photonic signals, electronic signals - they all travel on the order of light speed. There really wouldn't be much point raising the clock frequency beyond the characteristic length of the processing unit (Pentium 4 designers understand this now - they had to put "drive" stages into the pipeline just to allow signals to propagate, and that deep pipeline lead to a very low IPC).

    This would be useful for things like memory and processor interconnects, because you could shove gigantic amounts of data. Hence the reason that the article stresses their use as high-traffic freeways. I'm not sure I see the point in an all-plasmonic chip (unless they've got power advantages) because of size concerns.
    • ...overclocking his fingernail computer by swapping the green CMD (Cuticle Mount Device) LED in it for a blue one. Geeks could have nail-size competitions and "hand clusters". The mind boggles.

      With a name like that, plasmonics doesn't have to have any point as long as it also looks cool. (-:
    • I get the moving lots of data around at massive speeds concept. However, how about some basic logic gate or even a transistor like functionality using plasmons? Wouldn't THAT be what would bring faster chips closer?
      I mean, having a pipe that can transport huge amounts of data at the speed of light is great, but wouldn't the feasibility of a chip depend on the capability of routing all that data based on basic logical conditions?

      Kinda like no use having a missile that can carry a million tons and fly at Mac
      • Or am I missing something here?

        The transistor speed that matters is the switching frequency (that is, how fast it can switch from a 1 to a 0, and vice versa). That's what the "GHz" means, right? OK.

        The point here is that it would be pointless to have a CPU made of transistors that switched at 100 THz (optical) frequencies. It wouldn't be faster at all.

        Why? Because that chip still has to be made of atoms, and atoms have a scale. That scale is on the order of nanometers. Let's just say 1 nanometer, to mak
  • by dfn5 (524972) on Monday April 04, 2005 @06:33PM (#12138680) Journal
    Today, we're using basically two ways to move data in our computers: transistors carry small amounts of data and are extremely small

    I don't know about your computer, but my computer uses wires to move data and transistors to process said data. I don't see how one can compare transistors to fiber optic cables.

  • by pyrotas (862419) on Monday April 04, 2005 @06:41PM (#12138744)
    Plasmons can be easily created in metallic nanotubes. Furthermore, it is possible to create them into an entangled state. This _in principle_ might be exploited on the quantum computation scale.
  • does anyone else think the term 'plasmonics' sounds like something you'd see in the movie 'Barbarella?'
  • "Today, plasmons can only travel a few millimeters before dying, while today's chips are typically about a centimeter across."

    Maybe its just me, but as electronics become smaller and smaller, wouldn't this be more and more possible?
  • by joey_knisch (804995) on Monday April 04, 2005 @06:59PM (#12138923)
    Hmm... Let's see...

    7 years: Develop efficient plasmonic tech.
    5 years: Create manufacturing process
    +3 years: Design cpu
    ________________________________

    15 years: Just in time for Duke Nukem Forever
  • You don't need a thesis, dissertation, white paper, or science magazine to tell you that the closer you put things together the faster their particles can travel.

  • Nah. Forget plasmonics, fiber optics, quantum computing, and all this other junk... Let's make computers that are 100% mechanical, but built out of MEMS. Tiny gears, levers, pistons, pendulums, and other mechanisms will perform computations not previously possible with electronic computers. Better yet, let's return to the days of relay logic, but using nanotechnology to built microscopic relays. Good idea.
  • Well.. (Score:5, Informative)

    by anethema (99553) on Monday April 04, 2005 @07:54PM (#12139322) Homepage
    I havent read the article (this is slashdot after all) but the summary is terrible. (unless its the articles fault)

    From the summary:

    transistors carry small amounts of data and are extremely small, while fiber optic cables can carry huge amounts of data, but are much bigger in size.


    Transistors are just switches in the digital world. Just like anything that would be modulating the optical carrier.

    Fiber optic cables arent switches at all, or even active. You cant even compare them with transistors at all. Compare transistors maybe with an optical switch (which are ususally transistor actuated) or compare fiber optic cable with wires, but not transistors with FO cables.

    • Roland's blog is also terrible . . . It is the shameless last link in his topic.

      Plasmons are generated when, under the right conditions, light strikes a metal. The electric field of the light jiggles the electrons in the metal to the light's frequency, setting off density waves of electrons. The process is analogous to how the vibrations of the larynx jiggle molecules in the air into density waves experienced as sound.

      This analogy makes no sense because this is not analogous to how the larnyx jiggles air

    • Fiber optic cables arent switches at all, or even active.

      Actually, some of them are.

      One really useful example is doping the fiber with small amounts of an atom that lases in the frequency band of the light being carried. Then you wrap a bit of the fiber around a lamp giving off a suitable higer pump frequency of light. Result: A repeater amplifier. Feed it a little power and it boosts your signal.

      There are several other hacks. (At least one of them is a logic gate.)
  • by StateOfTheUnion (762194) on Monday April 04, 2005 @08:06PM (#12139434) Homepage
    Today, we're using basically (Basically is my cop-out word so that anyone that actually knows this technology can't call me out for any errors) two ways to move data in our computers: transistors carry small amounts of data and are extremely small, while fiber optic cables can carry huge amounts of data, but are much bigger in size(Actually I used the word basically because I used a terrible analogy. Transistors are used for gates and switches. Optical fiber carries information. Perhaps I should have said electrical conductors versus optical fibers, but that wouldn't sound as impressive even though its a much better analogy. But I think that most /. readers are really stupid and won't see through my gobbldey-gook. I am actually French, so you can blame it on the fact that English is not my first language.). Now, imagine a single technology combining the advantages of photonics and electronics. This Stanford University report says a new technology can do it: plasmonics.(This technology is not really new. In fact its not a technology at all. Its actually a natural phonomena, but /. readers are too stupid to know the difference. Also, its not really new either. Its been known for at least a decade . . . at least thats what this link [qub.ac.uk] form 1996 implies) (For more about plasmons, read this Wikipedia article.) Theoretically, it is possible to design plasmonic components with the same materials used today by chipmakers, but with frequencies 100,000 times greater than the ones of current microprocessors. There is still a challenge to solve before getting plasmonic chips. Today, plasmons can only travel a few millimeters before dying, while today's chips are typically about a centimeter across. Actually the articles that are linked to in the topic say that heat, connectors and other issue have to be worked out first, but /. readers can't handle more than one concept at a time, so I'm going to dumb this down for them) Read this overview for more details and references about plasmonics, and to discover why it's one possible future for chips' circuitry. (shameless plug for my blog where I'm soliciting for "premium blogads" in the upper right side of the blog. But /. readers won't notice that I have a conflict of interest and I'm trying to launch a career as a blogger/ tehcnology writer)

    Come on Roland, give us a break . . . you obviously don't understand what your writing about. Your analogies make no sense, your summarize is full of gross holes and you're trying to "sex-up" plasmons by calling a natural phonomena a technology and saying that it's something "new" when it is not. And seems you're doing this to attract hits to your blog so that you can sell ad space.

    I know this post is harsh, but I have to say that it appears that you are attempting to exploit the /. community for your own personal financial gain. We /. readers aren't as gullible or stupid as you seem to think . . .

  • I'm curious what part of a CPU they were saying was a couple cm accross. The die itself, where all the transistors are is probably less than a square CM in most cases. Given the fact that half or more of that is cache memory and the fact that most computation takes place within an area probably less than a few mm in length, it seems all they would need to do is repeat the signal every stage or so. But maybe this is why it's so difficult.

Lisp Users: Due to the holiday next Monday, there will be no garbage collection.

Working...