Forgot your password?
typodupeerror
Sun Microsystems Wireless Networking Hardware Technology

Sun Working to Obsolete Motherboards 228

Posted by CowboyNeal
from the look-ma-no-solder dept.
perl_camel_jockey writes "Sun is developing a new technology that promises to increase computing power by eliminating the need for physical, soldered chip-to-chip connections on the motherboard. Called 'proximity communications', it portends the ability for chips to talk to one another wirelessly just by being next to each other. Potential applications in computer design abound. Apparently this is part of Sun's Hero program, recipient of a $50 million grant from DARPA's High Productivity Computing Systems program to rejuvenate supercomputing in the US and regain the lead lost to Japan, in particular to NEC's Earth Simulator, ranked as the most powerful supercomputer in the world."
This discussion has been archived. No new comments can be posted.

Sun Working to Obsolete Motherboards

Comments Filter:
  • by Anonymous Coward on Friday August 06, 2004 @03:40AM (#9897330)

    Slashdot Microsystems is working hard on a system to eliminate duplicate postings [slashdot.org]. They hope to have the system working by early 2008.

  • DUPE (Score:5, Insightful)

    by Drakon (414580) * on Friday August 06, 2004 @03:41AM (#9897333) Journal
    ...
    You'd think that since they ask subscribers to email them if there's any problem with the story, they'd pay attention when we email them...
    *sigh*
    • Oh, wait. This isn't that replace motherboard with tiles of CPUs.

      It's some equally uninteresting pipedream...
      Yay.
    • Re:DUPE (Score:5, Interesting)

      by johnhennessy (94737) on Friday August 06, 2004 @04:23AM (#9897458)

      How about allowing subscribers to moderate stories before they hit the main site.

      People who are really busy could browse at +5 "Don't do anything else until you read this !!!" while people with loads of time (or in college) could browse at normal levels.

      Oh, and as a plus, you would eliminate dupes as well.
      • Kuro5hin already has a system where stuff gets seen by a whole lot of people before it even appears on the front page, let alone one of the section pages.

        Since /. moves a lot faster (note - maybe not, last time I submitted a successful story, I waited 3 days for it, but that must be the omlet), the subscribers could be allowed to tag a discussion with a dupe tag, or perhaps the -1 - +5 system. X number of -1 or dupe taggings would delay the story.
      • Re:DUPE (Score:3, Interesting)

        by Karma Farmer (595141)
        People who are really busy could browse at +5 "Don't do anything else until you read this !!!"

        It would be awesome if Slashdot moderation worked like that at all. But, it doesn't. Moderators don't decide "this post is worth a 3, while that one is worth a 5, and that one is worth a -1." Moderators are only given three choices for a post: +1, 0, and -1. Slashdot uses an insanely boneheaded algorithm to map those three moderation choices to seven different thresholds: -1, 0, +1, +2, +3, +4, and +5.

        As

        • It would be awesome if Slashdot moderation worked like that at all. But, it doesn't. Moderators don't decide "this post is worth a 3, while that one is worth a 5, and that one is worth a -1." Moderators are only given three choices for a post: +1, 0, and -1. Slashdot uses an insanely boneheaded algorithm to map those three moderation choices to seven different thresholds: -1, 0, +1, +2, +3, +4, and +5.


          There may be easier (and less contraversial) ways of implementing a moderation scheme.

          One could reward
          • You've just described an expanded karma scheme. Given all the stupidity that the love of karma causes already, I can't imagine that making it more important would solve anything at all.
        • by abb3w (696381)

          Slashdot is designed so that browsing at +5 means "show me the posts that have been seen by the largest number of readers with mod points."

          So, here's still yet another [textfiles.com] modest proposal [art-bin.com] for "repairing" the Slashdot moderation scheme.

          When mod points get assigned, users are given six at a time (instead of the current five). Mode Points then may be used in one of two ways.

          1. A "moderation" link, much like "meta-moderation". You're given 10 posts to look at; mod as many of them as you see fit, up to the num
      • The problem is, there would be more moderators per story than with comments. Things will turn out to be either -1 or 5, with few in between. Maybe they should be modded in increments of 0.1 or so?
      • How about allowing subscribers to moderate stories before they hit the main site.

        People who are really busy could browse at +5 "Don't do anything else until you read this !!!" while people with loads of time (or in college) could browse at normal levels.

        Oh, and as a plus, you would eliminate dupes as well.

        How about allowing subscribers to ignore stories which they don't want to read -- even when they hit the main site?

        People who are really busy would not be forced to read dups and to pa

    • Re:DUPE (Score:3, Insightful)

      by Anonymous Coward
      I know the reason there are so many dupes.

      Go to the search page and try and find something, it totally and absolutely sucks.

      Google searches slashdot better than slashdot searches itself. Try it.

      How are you supposed to stop dupes when you can't even search your own database?

      More to the point, you subscribers aren't paying for much. The only thing they do is give articles the nod, and they still can't get it right. Let alone make their pages compliant, stop the 503 and 500 errors, make the search work o
    • Re:DUPE (Score:5, Funny)

      by cachorro (576097) on Friday August 06, 2004 @05:28AM (#9897608)
      I wonder how much more Sun would need to pay to get this article posted a third time.
    • Nah, that would imply they cared that their site is becoming a bigger and bigger joke.

      But lets face it, they don't.
    • No! They just take your $$$ and buy beer and hookers!
    • Besides paying for the privilege to see dupe stories in advance, are there any other advantages to a Slashdot subscription? Because frankly I can think of better things to do with my money.
  • by Anonymous Coward
    Now I can get the CPU into a separate cooling unit from the rest of the system so that I can pay less for more cooling where I need it.
  • by Anonymous Coward on Friday August 06, 2004 @03:44AM (#9897348)
    I think this has happened before...
  • by WegianWarrior (649800) on Friday August 06, 2004 @03:44AM (#9897353) Journal

    Cost of all the chips and stuff that makes up an über-1337 computer - 1000$

    Cost of fancy cabinet w/ window to artisticly put all your wireless chips in - 200$

    The look on your face as your motehr fires up her old vacuum cleaner, blanketing the area with RF-noise - priceless.

  • by sien (35268) on Friday August 06, 2004 @03:45AM (#9897355) Homepage
    To obsolete grammar!
  • Pride (Score:5, Insightful)

    by Hypharse (633766) on Friday August 06, 2004 @03:52AM (#9897370)
    Nothing like damaged pride to get the government to push along technology. Without Japan's competition this would probably not have happened. All we need is for Russia to cure a type of cancer using stem cells and congress would pass a law funding it the next day.

    About obsolete motherboards I have my doubts though. The Von Neumann (may be spelled wrong) model has been around for a long time because it has proven to work and it also supplies a lot of companies with revenue. If you only have a single chip, then a lot of companies are going to lose money and they won't like that. In fact there may be a silver lining in this that it will push motherboard manufacturers even further for fear of being wiped out by this type of technology.

    • Von Neumann [vt.edu]

      Nice to meet someone who has heard of 'Von Neumann'. I have a hard time to immediately connect physical implementation to the architecture [wikipedia.org] though and I doubt that you followed up there --> Dieudonné [1981] is a little more generous with words but appears to confuse the concept of the stored program concept with the wiring of computers: "Dissatisfied with the computing machines available immediately after the war, he was led to examine from its foundations the optimal method that suc
    • Re:Pride (Score:2, Informative)

      by ahfoo (223186)
      I doubt that part about stem cells.
      You may be surprised how close this already is to reality. Mainland China has already commercialized a form of gene therapy that has had powerful results on a broad range of cancers. It has a brand name even, it's called Gendicine. Google it if you doubt it could be real and you could have not heard of it.
      In fact, Bush's science advisor has publicly stated that he thinks mortality is essential for human dignity. This is an extension of the belief that human moral
    • Re:Pride (Score:3, Interesting)

      Maybe I'm being stupid, but I don't see how this work obseletes the Von Neumann architecture.

      Sure, there's no wires connecting the chips together, but the basic ingredients of a Von Neumann machine are still there, i.e. memory and processor.

      As somebody has already said it's not going to be just one chip, however even single-chip computers still follow the model. Yes, the processor and memory reside on the same chip, but they're still logically separate units. Indeed most modern processor chips are Von N
  • by physicsphairy (720718) on Friday August 06, 2004 @03:52AM (#9897372) Homepage
    "it portends the ability for chips to talk to one another wirelessly just by being next to each other."

    Well, unless each chip comes equipped with its own miniaturized nuclear reactor, aren't they still all going to have to have leads running to the powersupply?

    I don't mean to be a heckler, but are you really "doing away" with the motherboard or just reducing it to a voltage bus with transmitters and receivers replacing some of the input and output pins?

    • A damn good point, the only way I can see them powering such devices wirelessly is if they use a coil and place it in a magnetic field. Of course such a strong magnetic field will play havoc with anything close by so the PC cases will need good magnetic shielding.

      So there you have it, this is a tech which we aren't going to see for 10-20 years. Sun desperately needs some good ideas now, they won't be around in 10-20 years if they don't sort out their near future.
    • by syukton (256348) on Friday August 06, 2004 @04:53AM (#9897529)
      The system could be inductively charged.

      You can transmit power as well as signals without wire. Really, all a singal is, is waveform-modified electromagnetic radiation. Radio transmission towers have their outputs measured in Watts, computers have their power consumption measured in, you guessed it, watts. Whether it's induction or using RF technology to energize the chips, it's entirely feasible *and* possible.

      I'm all for doing away with the motherboard and the wires all together anyhow. And jumpers too, I hate those little bastards.
      • I'm all for doing away with Phoenix and their We-Don't-Trust-You BS.
      • And who gives a crap about efficiency? This device needs 400w, i could run a wire to it and send 400w or I could build a nuclear reactor and blast it with 15 gigawatts!

        Oh crap I just developed cancer, guess that sending 400w over a copper wire wasn't such a stupid idea after all.
    • by Anonymous Coward
      This technology is only for interconnecting the different dies together. Currently you can buy flash & SRAM stack chips on a single package (for the cell phone market). This is done with stackchip technology using die to die connection.

      This technology is useless at the system level because of tight mechanical alignment issues. Think optic bench experiments - heavy table, stabilized and 3D alignment to line chips together to form a big system. Oh yes, power, misc signals etc.

      PCB will still be used fo
  • One thing is for sure. If they can get this to work and if heat production can be cut down, this would make computing equipment and electronics much smaller. The printed circuit board is one of the big things holding us back from much better electronics miniaturization.
  • Worried... (Score:4, Interesting)

    by rpbailey1642 (766298) <robert...b...pratt@@@gmail...com> on Friday August 06, 2004 @03:58AM (#9897388)
    I might just be horribly ignorant, so please inform me...security? Even though all it is transferring is system calls and such, I am certain that there are ways to back-engineer what the computer is doing at that time by reading the (very faint, I'm sure) wireless signals. Again, I do not know, so will someone please enlighten me? What exactly is going on, what are the security ramifications?
    • Just put your computer into a metallic case.
    • Re:Worried... (Score:5, Insightful)

      by ryouki (209039) on Friday August 06, 2004 @04:38AM (#9897498) Homepage
      I dont think there is much to be worried about. The system the article discribes seems to be using capasitors spaning 2 chips to interconnect components on the system. One plate on each chip. The gap between the chips woudld work like a dieletric on a normal cap.

      I don't think they are trying to make a small computer with bluetooth or wifi glueing components together. Cables have more bandwidth than wireless.

      The security riskes for these "wireless" connections would be no more than that of a normal capacitor of the same size.
    • Re:Worried... (Score:5, Informative)

      by Asic Eng (193332) on Friday August 06, 2004 @05:27AM (#9897604)
      You have many hundreds of transmitters next to each other in the space of just a few square milimeters. Their transmission power needs to be so low that they don't interfere with their neighbours. The mix of signals which radiate off from that arrangement should be close to impossible to decode.

      In any case, in conventional devices the pins will also work as (very low-power) transmitters, too. So once you've found a method of decoding this signal mix, you can probably get at the information on conventional chips, too.

      • Re:Worried... (Score:3, Informative)

        by songbo (614466)
        Actually, there are already methods available to figure out what the chip does currently. I have heard of one that tries to measure the power usage, at each fetch-execute cycle, and based on that, can figure out what kind of instructions have been run. The CPU already gives out EM radiation which can be detected. It may be just a matter of figuring out what kind of signal is radiated when each instruction is being executed to know what instructions are being run. So the same security concerns already exist
  • by Henk Poley (308046) on Friday August 06, 2004 @04:02AM (#9897394) Homepage
    Honey, I'll warm you something in the microwave!

    Noooo...

    [Beep] rebooting... grmbl...
  • by DupyMcCopy (803233) on Friday August 06, 2004 @04:02AM (#9897396)
    Dust & dirt. I would imagine that at such low voltage levels, induced current would require a damn near perfect level of alignment between the chip and the "socket". This is admitted in the article. What they don't admit is that it's going to be nearly impossible to get the damn thing in the socket without letting dust or dirt inbetween the chip and the socket. And a more interesting topic is their consistent mentioning of taking the cache of the chip. That's a nice dream and all, but where the hell are you going to put it then? Hardwired onto the motherboard? That's going to dramatically increase the cost of mobo's (so they are simply shifting who gets to eat the high sticker price on their products). And what if I buy a quad capable mobo, but only put 2 processors on it, I'm effectively wasting 2 sets of cache, rather than simply wasting 2 cpu sockets, and the sockets are a hell of a lot cheaper than the cache. I suppose you could fix this by going back to COAST (cache on a stick, yeah i know you remember that nasty stuff). But that brings in a whole new problem: These days, cache is only fast because it's so close to the cpu. If they move it off the die, it's just going to be put back on in 2 years because we can't access the cache fast enough ever since we moved it off the die. I'm no super computer engineer, but these guys better have an entire family of rabbits they plan on pulling out of their asses or this fucker's gonna flop.
    • What they don't admit is that it's going to be nearly impossible to get the damn thing in the socket without letting dust or dirt inbetween the chip and the socket.

      You just need to develop an airtight transport and allignment system. There will be some complexity involved.
    • And a more interesting topic is their consistent mentioning of taking the cache of the chip. That's a nice dream and all, but where the hell are you going to put it then? Hardwired onto the motherboard?

      You are failing to completely grasp the significance of this technology. The only reason why a processor even has cache right now is because of the slow bus that connects it to main memory. Connect the processor and memory directly and now your memory is your cache.
  • by murr (214674) on Friday August 06, 2004 @04:03AM (#9897399)
    Sun should find this project rather easy going - their motherboards ARE already pretty obsolete anyway.
    • I was just reading an article about their new amd workstations and it had a picture. I wish I could find it now but a quick search didn't turn it up. They have their cpu's on a seperate board, their chipset is also on another board. The two are connected to the main board. Makes it easier and cheaper to keep the workstation up to date. I wouldn't call that outdated.
    • Sun should find this project rather easy going - their motherboards ARE already pretty obsolete anyway.

      I know, funny...

      [/voice=Nelson Muntz]HA HA![/voice]...

      But in all seriousness, Sun is pretty much at the top of the game when it comes to motherboard manufacturing. Take a look at the specs on their new AMD Opteron workstations [sun.com].

      2 AMD Opteron 200 series processors. (up to 2.4 ghz.)

      4 PCI-X busses.

      Up to 16GB PC3200 memory with 12.8GB/sec. total memory bandwidth!

      Holy schnikes Batman! Sun's new workst

  • i like when i see a headline and go "i wonder how many 'dupe!' posts there will be?" don't eliminate dupes... what would /. readers do with all their spare time?
  • by doktorstop (725614) on Friday August 06, 2004 @04:21AM (#9897456) Homepage Journal
    Let me visualise this... you need the chips but you don't need them to be on a motherboard... how would a computer look? A plastic bag full of chips? you dig in, take the processor out and throw in a new one to upgrade? groovy
  • by Julian Morrison (5575) on Friday August 06, 2004 @04:38AM (#9897497)
    It's about increasing chip-to-chip bandwidth by using capacitative coupling instead of (comparatively huge) physical wires. This means the chips would have to be more closely connected, probably slotting together like lego bricks.
  • by bani (467531)
    i guess /. really, really, really wants to make sure everyone reads this story.

    i'm still trying to figure out why though.
  • Is that it represents the first step towards real ubiquitous computing. With this technology, there's really no need to have the computer be bound to the actual physical box. You can put the chips in just about anything as long as they can communicate, basically. So you could literally be surrounded by a computer when stepping into a room with chips embedded in the walls. Now that's pervasive computing, we could have real "smart" buildings!
  • I still fail to see how this is not going to mix up the signals from various different chips without modulating the ones and zeros, and thus, how it is going to be faster than direct electrical connections at light speed. But they've probably spent some time thinking about it so i'll believe it for now. What I think is spectacular about the idea are the possibilities to cluster. <imagination>Now if they make them transparent and solar powered and spherical, building a supercomputer may get as simple a
  • C=64 Reminder (Score:3, Interesting)

    by incog8723 (579923) on Friday August 06, 2004 @05:30AM (#9897611)
    This reminds me of an issue with a C=64 I used to have that I had removed the aluminum foil casing from (inside the case). The 6502 processor wirelessly queried the SID chip for reassurance and understanding.

    Sincerely,
    Ass-embly Language
  • Chip Issues (Score:3, Interesting)

    by drakyri (727902) on Friday August 06, 2004 @06:14AM (#9897684)
    One of the issues we run into where I work (we make oscillators) is that chips behave differently over their specified temperature operating range. Commercial parts are rated from 0 to 70 Celsius (Industrial : -40/+85, Military : -55/+125). Commercial range is pretty decent for most applications; the average user wouldn't expect their computer innards to be heating up past 150 Farenheit. Anyhow, the simpler chips we make operate differently at different temperatures (the part I'm working on now is rated +- 100 parts per million).

    A more complicated way of making these is to make what's called an Oven-Controlled model - you basically create a little oven that responds to the temperature of the chip, keeping it at a certain optimal temperature. These parts are much more stable and accurate; they vary in parts per billion. Dust is a big concern during manufacture; they're pretty sensitive, but once they're sealed, they're more or less set to go.

    On a completely seperate note, I have to wonder what kind of issues Sun will be having with crosstalk on their new mobo's.
  • I don't know if slashdot works like this, but when a new story is submitted, previous postings (say, of 1-2 days) can be searched if they match. If they do, then moderators receive the new story with a remark that it may be a dupe. Then the moderator searches the last 2 days' stories, and if it finds a dupe, the new story is rejected.
  • Sun Microsystems: The air between stuff is the computer.

  • I'm sure they have this worked out but I wonder how much outside RF interference is needed before it becomes an issue. A lightning strike wipes my TV from 5 miles away. I wonder how easy it will be for the black hats to come up a jammer. A focused narrow beam aimed at a Home Depot from the parking lot (hehe). Oh what fun!
  • Prime Intellect (Score:3, Interesting)

    by skinfitz (564041) on Friday August 06, 2004 @07:54AM (#9897854) Journal
    This idea reminds me of the free online novel Prime Intellect [kuro5hin.org] which I can strongly recommend.
  • I learned about this a while back on that one site......you know the one. Slashdot. CowboyNeal should read it sometime. They have some great material there, maybe they could us it for stories on this site.

  • Though I was concidering chip-to-chip connections being optical, the benifit there being you can take your signals some length with a bit of fibre-optic.

    But you can't get rid of a motherboard, youre always going to need power, optical power is probably a long way off, and proximity power is only really used in electric toothbrush rechargers.
  • I personally welcome our new replicant overlords, and hope that I can prove useful.
  • by Baldrson (78598) on Friday August 06, 2004 @09:06AM (#9898206) Homepage Journal
    DARPA's history of supercomputing initiatives isn't quite as bad as NASA running the Shuttle program but its up there.

    If you want fast computers developed in the US, buy them from the US market and try to mould your tax incentives so that they simulate, as closely as possible, a net asset tax as described in "A Net Asset Tax Based On The Net Present Value Calculation [geocities.com]".

    The reason Cray Computer Corporation's gallium arsenide fab went out of business wasn't for lack of funding -- it was for a lack domestic market for the end product, supercomputers, in the wake of the end of the cold war. One could also chalk it up to Cray's fixation on supercomputers since the output of the GaAs fab line could have been altered to serve high speed telecom markets, but if DARPA wants fast supercomputers, there was help available from private capital sources.

    Its never a good idea for government to compete with private capital sources in high technology.

  • Does anybody have a fear or feel paranoid about all of the wireless devices that we currently use, the radio waves that they produce and their affect on the human body?
  • One of the chief difficulties in developing the technology comes from the environment where computer chips live. Heat and vibration in this environment can cause chips to get out of the precise alignment needed for proximity communication. Sun is currently tinkering with different techniques and different packages to prevent, or correct, these effects.

    Solution [lego.com]
  • Beowulf clusters of chips just sitting in a great big heap!... So in the year 3200 we'll have Self-aware garbage dumps. Another interesting point would be eaves-dropping on processors ala Tempest radiation, but now the processor will be doing the transmission.
  • That will be great when we can just buy chips and literally drop them into a box with i/o ports. But since that isn't going to happen - what's the real advantage to the consumer of having no motherboard?

    Won't we just be fastening our add-on cards, etc to some sort of board? And won't these chips need to be fastened to a 'board' so that they can communicate with each other properly? ie. be the proper distances away from each other, not be rolled around like bingo dice?

    In the least, we will still require
  • "By contrast, proximity communication relies on capacitive coupling--the ability of two electrically charged devices close to each other to interact."

    So... eventually I could build a powerful machine by throwing a bunch of chips into my pockets.... mmm I can see it now:

    Hey there, is that a beowulf cluster in your pants, or are you just happy to see me?

  • As I wrote the last this was posted: [slashdot.org]
    • "Sun is not "coming out with new chips without connectors". Sun has demonstrated a new kind of interconnect in a lab. They might use it in a DoD funded supercomputer project. Maybe.
    • You're not going to "stack chips like Scrabble tiles". The unpackaged chips have to be aligned within a few microns and held in position. That's going to be done in an IC packaging facility. The result will be a multi-chip module, a single package containing several chips."

    • If this is really one of the "first" (I know it isn't *the first* attempt, but another attempt) attempts to take CPU and supporting chipsets "3D"? That is, manufacture the IC chips (the bits of silicon), then "stack" them aligning the connectors (or whatever) on the edges to allow for more processing power in the same (or nearly same) space? It may not be this, exactly, but perhaps the technology developed will help lead to this?
      • by Animats (122034)
        This isn't a "3D" stacking technology. The chips that communicate have to be mounted face to face. See the illustration in Sun's technical paper. [sun.com]

        For an example of true 3D chip stacking, see Infineon's SOLID technology [lightreading.com]. Infineon announced that in 2002. Intel and Sharp have also played around with similar approaches.

        The Infineon approach is interesting because it puts a layer of copper between the chips. Getting heat out of the middle of the stack is a major problem with all stacking schemes. Infine

        • I see what you're saying - looking at the Sun tech paper shows that the "connections" are at the edges of the chip, and face-to-face is needed (that is, the connections don't go through the silicon substrate). Infineon's tech is true stacking - so I wonder if the two techs couldn't be combined in some manner? That is, allowing the capacitive "connecting" surfaces (I know they are truely connected, just a capacitive junction, like is used in audio and rf) to be placed on top and bottom surfaces of the chip s
  • In other news, General Electric was found working on a cordless extension cord. However, they hit a snag when one of the lab techs were killed while attempting to walk through the arc.
  • I can throw my new GPU right into the box, close the case and not have to worry about anything else!
  • does this mean my 2.4ghz phone will crash my computer instead of just my network connection?
  • Now i gorra wear lead underwear to use the computer.

This is a good time to punt work.

Working...