Slashdot Mirror
Sun Working to Eliminate Circuit Boards
lokedhs writes "Sun Microsystems is coming out with new chips without connectors. According to the article, this will have a lot of advantages: 'Performance, for instance, could greatly escalate because the speed of transferring data among chips and the number of channels for the transfers would increase. Energy consumption could also decline. Just as important, overall costs could fall, because defective chips could be removed like Scrabble tiles.' This technology will also lead to new CPU's without cache: 'The technique could also allow designers to remove the cache--the large pool of memory currently found on the processor--and put it on a separate chip. Caches were integrated onto processors to amplify bandwidth. Adding cache, however, bumps up manufacturing costs, as it greatly increases the number of transistors. With the bandwidth constraint gone, caches could once again be made independent without it having an impact on performance.'"
so basically they want to stack the chips? umm, heat?
I am the Alpha and the Omega-3
With so many chips so close together, they are certainly going to have heat problems.
Interesting technology, thought.
This idea has merit, but the explanation is oversimplified. Moving the cache 2 centimeters from on top of the processor to a separate chip may be feasible, but it will increase latency (increasing the number of clockcycles a cache fetch takes and re-affirming a need for an on-chip cache to cache the "cache"). Other applications of this technology (like the fact that any part with issues can be easily replaced) seem more relevant.
Pedant mode on:
They want chips without connectors to communicate with each other? Is that even possible?
This post brought to you by the Slashdot Association of Pedants.
What they need, instead is VioletTooth (wireless chip-to-chip communcations). That way, they won't have to worry about alignment problems and such!
Did you even read the article before posting it here. The article talked about eliminating the pin that is used to house the chip. Due to the size of the pins, it limits the number of I/O paths a chip can expose to the motherboard. Instead they can implement transmitter/receivers using capacitive inductence to increase the I/O paths a chip can expose. Thereby increasing the bandwith a chip can utilize.
I figured Sun would have laid off their entire R&D department by now :)
Love,
Zaq
This wireless chips integrated for a purpose thing reminds me of Replicators
Either way, I'm thrilled and spooked.
Um...who repairs motherboards anymore? At around $100 a pop, most people just get a new one.
If there's a high-end application for this technology, great, but getting rid of high-end hardware is one of the biggest reasons people are also getting rid of Sun...
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.
Just as important, overall costs could fall, because defective chips could be removed like Scrabble tiles.
With my luck I'll get a dead Pentium Z or Q that I just can't get rid of.
The angel in the oatmeal.
Wow, this announcement reminds me of an awesome book I just read: http://www.kuro5hin.org/prime-intellect/mopiidx.ht ml
So all the communications inside the computer will be implemented with radio waves? It will be even easier for them to eavesdrop all the extremely secret activities I use my computers for. Naturally I cannot tell you what they are, but they are related to a certain Internet service that - under a legitimate- and innocent-looking cover, of course - is designed to harm certain people by making their servers catch fire as well as making their bandwidth bill increase unbelievable amounts.
Sounds like the isolinear chips in Star Trek. It is amazing that an idea that was once science fiction is now coming to reality.
Most likely, the capacitive coupling of signals is only targeting chip to chip data signals, not the supply of power to the chips.
PJRC: Electronic Projects, 8051 Microcontroller Tools
And in other news, scientists are developing a computer with no electronic parts at all!
"Who are in control, they are not in control of anything - they don't even control themselves!" - Glen Beck
Eliminating connectors also removes a problem that pops up in microchips - without connecting wires, you don't have most of the parasitic capacitances that crop up on chips. A capacitor is anything with two metal contacts, so wires (especially parallel wires) cause very small capacitances, usually in the order of several picofarads. The problem is that on microchips, the capacitor values that are being used are generally in that range too, so parasitics can be very problematic.
Heat transfer is also a problem, but I'm not convinced that it would be so different from current heat considerations. The heat should still spread through the chip, regardless of connectors, and dissapate on the chip's surface. Whether they get overly ambitious with stacking these chips is another question.
Does anybody know of any good explanations of how or why capacitive coupling works?
Also, how close do these chips have to be to each other? Could there be enough of a gap to allow something to cool them down?
What is that property of photons where they some how join then split them and what ever happens to the one will happen to the other, even if it is on the other side of the universe? I don't see why people aren't using this for a mode of communication or to replace wires. But anyway, all this r&d is just going to negate any savings that the "cheaper" chips would have, and there would defiantly still be a way to secure it, otherwise every time someone bumped into your desk you would blue screen.
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.
AMD has put their memory controller on the CPU. Wouldn't that be a step towards reducing the cache needed with mismatching Bus Frequency and Memory Speed?
Cache isn't what I aim to remove from the equation. My aim is to remove latency through the integration of both the North and South bridge onto the CPU. Then lets remove all the bus frequency and caches in that fashion.
This is a pretty nifty idea really, near as I can decipher it. The idea of a supercomputer is a lot of chips, all grinding away at a complex problem. Two approaches are currently used. This concept would seemingly let you do both, and mix or match on demand to fit the situation. It would also I guess let you use normal (well, wireless enabled) ram for the remote cache.
Ah!!! So you must work for Gator!
-nB
whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
This was posted back in September of last year :
m l?tid=102&tid=137&tid=187
http://slashdot.org/articles/03/09/22/1055244.sht
jdb2
It uses capacitance links to communicate instead of a direct wire. "Radio waves" aren't used. Two wires near (depends on currents, frequencies, etc. and in this case we're talking about microns) each other can have effects on each other as if they were linked when operating at high frequencies. Remember that a capacitor looks like a wire at high frequencies and essentially two wires next to each other are connected by a capacitor.
OK, I did not read the announcement, but it sure has the ring of incredulity. It sounds like Sun is desparate for some uplifting press. This faces so many fundamental obstacles it's not even funny. When the printed circuit board came out, the prediction was that it would eliminate connectors. We may stack a few chips, printed circuit boards aren't going away nor are connectors.
I would worry more about devices that purposefully introduce noise in the same bandwidth your cpu uses to do I/O. Some one could make your computer crash/freeze/hang on command. Or at the very least tamper with it's performance.
Faraday, may your cage will protect us!
This article didn't explain a whole lot to me, it was too high level. I was wondering if anyone could explain how using radio waves form one chip to another does not require extra hardware (for transferring and receiving) and if this technology would be more susceptible to interference?
With this connectivity, there would seem to be a need to standardize CPU chip speeds. Otherwise, a multi-CPU system w/disparate chip speeds would need a sophisticated register design to allow the faster chips to 'idle' while the slow one occupies a needed memory address.
If designed, however, this could allow admins to assign quickie chips to the OLTP (or DSS batch loads @ night) systems, and the slower CPUs to the less intensive tasks (like sys admin).
Wow. It will be even easier to bug a computer, just drop a survailiance device in it, or near it (preferably with a small flashing led on it, to the Mission Impossible soundtrack).
When asked why, the answer is almost always: "It's 2014".
Here
Are any of the EE's out there scared about their jobs becoming obsolete? Everything is going to be single chip solutions, not to mention moving away from electronic design into light-wave design ... I just hope I can hold on long enough and retire before any enormous changeover occurs ..
Seems like all that capacitive coupling would cause heat and e/m interference problems. Why couldn't you use LED lasers and sensors built onto the chips to optically couple adjacent chips through a simple optical connection? Each side of a square package could have a laser transmitter and a receiver so it could communicate with up to four adjacent chips. Dust in the sensors would be a problem. So would misaligned components. But, that would do the same thing, no?
Just wondering 'cause, you know, I got nothin' better to do with my morning than armchair-crituique the designs of ACTUAL EEs who know what they are doing.
"Lawyers are for sucks."
- Doug McKenzie
Well, this is offtopic, but there's no place to discuss the problems Slashdot has been experiencing, so why not here? If the admins won't provide an appropriate forum, we have no other choice.
Does anyone have any idea what's going on here? I can't be the only one who wishes for a front-page story explaining why Slashdot is so amazingly unreliable and broken lately--especially for subscribers who are paying $ for this service.
If you stack chips, then each chip can only communicate with chips right next to it (am I wrong?) So you'd be able to communicate directly with 2-6 chips, otherwise the chips would have to relay through each other, like a daisy chain. Hopefully none of them would break that communication, otherwise you could have a motherboard "chip" relay messages between them all.
Let's see, in 2 weeks the've anounced that they were looking a buying Novell and getting rid of circuit boards. I guess a positronic brain will be next.
"Eve of Destruction", it's not just for old hippies anymore...
This side up.
Couldn't they solve the alignment problem with a tongue-and-groove type arrangement to keep chips from moving relative to each other? I knew all my experience playing with Legos would come in handy some day... now I can snap together my own computer!
"Freedom means freedom for everybody" -- Dick Cheney
Some time back I posted this, of which the the present Slashdot Article reminded me. Some of it sound similar. And some is different, of course.
could this make the chips automagically scalable? could adding an extra 1GHz of processing power be like adding an extra stick of RAM?
this kind of flexibility would be the only point going down this path imo.
OT: why is slashdot's uptime now a matter of hours?
Have you ever tried to go to BestBuy and buy a "motherboard" for a SunFire 6800?
I think that they meant "directly" through a conductive substrate, like a thin sheet of the proper material... Radio transmitters would not be practical at all.
So, the chips would be mounted ON tinfoil themselves and restricted to it. No need to wrap it up.
The heck with their CPU technology! Can't they do something useful, like cut down on the fan noise coming from my Sunblade? :-P
Sun Corporation anounced today the release of new Pokesun collectible computer tiles!
Catch them, arrange them, build a super-computer! But you gotta collect them all!!
Maybe the'll work on eliminating the 504's from this farce of a site.
As an aside, I guess I'll have to stop Gaff Taping my CPU into the socket.
[% slash_sig_val.text %]
nope cheap and effective to prevent this from happening:
just give your computer an earthed tinfoil hat(more of a ski mask really)
seriously.
each [sheilded] wire as its own infinante bandwidth. With shared band width (no wires) you run into channel allocation problems which implies bandwidth size problems.
So while you maight not have 32 wires for a 32 bit bus, you'll have wire, and all the frequencies in between.
I don't know how they say wireless will be less power consumptive. Even if you bypass the need for a transmitter and reciver, inductance power decreases exponentially with range...
You mitigate that by putting a shielded wire back in.
The cable company did this years ago. It's called cable. What Sun is promissing is satalite. And we all know satalite is more expensive in hardware costs.
You mitigate that
Slashdot's rate-of-post filter: Preventing you from posting too many great ideas at once.
I would expect the elimination of connectors and circuit boards to follow.
Maybe they could come up with a man who doesn't need a brain, and who lives without a pulse.
Oh, wait, they did. Sorry, Darl.
Ok, they are using "capacitive coupling" and amplified voltages
purportedly to communicate...if the bandwidth is that high, and a bank of these things is supposed to work together, how are they going to prevent crosstalk?
I'm having a hard time picturing a bank of 10cpus/cache/IO chips
stacked together working without interfering with each other...
Of course I must just be pessimistic...
"There is no ecache problem...pay no attention to the man behind the curtain..."
"You can't fake memory bandwidth that isn't there."
vodka, straight up, thank you!
Maybe the next step is for all the components to come in lego bricks. The self-organizing bus won't care what order you put them together in, so whenever you add a component (50 GB flash drive, GPU, firewire ports, ect.) you just click it on somewhere, and it adds it into the system.
Can anyone tell me how to set my sig on Slashdot?
Sun is the company I hate to hate. They have some of the brightest people in-house and create some amazing tech and ALWAYS seem to crap the bed on the business side. What good is a beautiful baby boy when it ends up being still-born? Man, I wish IBM just officially turn them into an R&D department.
-_-
Now when the guy downstairs fires up the microwave your computer can go ape shit along with the wireless network.
My old sig was REALLY stoopid.
As for the technique of capacitive coupling, that is how signals used to pass through low voltage amps virtually since the triode tube. The technique has been used for isolation amplifiers for many years. The signal on one side of the voltage barrier is digitised in some way (perhaps just PCM) and transmitted across a voltage barrier using very small capacitors, to where it is decoded. In some cases, power for the input side is also transmitted by capacitive coupling across the barrier.
Because the transmitting and receiving side of the capacitors is so tiny and the electric field therefore so constrained, it is not going to be possible to read the signals with an external aerial.
I believe Philips, among others, earlier suggested using LEDs and photodiodes along the edges of packages, but appart from requiring power they could only be unidirectional. Capacitive coupling itself absorbs begligible power and can be fully bidirectional.
Panurge has posted for the last time. Thanks for the positive moderations.
The problem with capacitive connections is that you are, for all intents and purposes, using small radio links. This causes several issues to come to the fore:
- Your immunity to cross-talk goes down. Misalignment will exacerbate these problems.
- Capacitive receivers will also be able to pick up local RF fields. The computer will be much more vulnerable to external interference than it was before.
- The computer will also radiate much more than it did before, creating more RFI and leaking information that might be crucial (like crypto keys).
Making the chips the meat in a sandwich with metal sheets for the bread would help this a lot, because tightly coupled ground planes attenuate both radiation and reception. As long as you're putting a ground plane on top of the assembly it might as well do double duty as a cooling device, though I wonder what effect the heat-transfer compounds would have on transmission and crosstalk.Sustainability and energy independence essay
I might be misunderstanding the article, but I did read it. "Radio waves" aren't being used, for one thing. And for another, the link is so sensitive to alignment (and presumably range) that small amounts of vibration confuse things. I don't think someone is going to eavesdrop.
_Any_ alternating current will cause electromagnetic radiation. When current changes in a capacitor, it creates an electric field, which creates a magnetic field, which creates an electric field, and so forth. Although it may not use "radio waves" specifically, it is the unavoidable byproduct of AC.
Oh, and our computers (save acryllic cases and such) are already encased in faraday cages. We've been wrapping them in "tinfoil" for years and years, and for the exact same reasons.
and they just trained me on the wire bonder last week
http://research.sun.com/async/Publications/KPDisc
jdb2
Have a carrier frequency of 30 GHz and use waveguides.
You describe the current cache situation pretty well. But this SUN BS article claims that it won't be necessary. That you won't need multiple levels of cache, so no cache chip.
This article is just complete BS. They're marketing their stock now.
SUN is not going to suddenly change from a workstation company to a chip technology company.
I like the concept. It seems Sun is slowly getting on track again.
The difference is water cooling can move heat a significant distance. Peltiers cannot. Peltiers move some heat a short distance, and then throw in more heat of their own into that small area. In short, they suck. They have very few practical uses. That's why you don't see them around much.
early crays did this by layering chip on chip with metallic layer welding between the GaAs nuggets. that's why crays had to run in freon baths or liquid air. this can be expected to raise the price of sun servers built with that type of layering technology. substantially.
isn't the market moving in the opposite direction, towards demanding cheap commodity solutions?
just asking....
if this is supposed to be a new economy, how come they still want my old fashioned money?
Isnt chips without connectors like write-only memory?
This is a very cool idea and it's the kind of thing I expect from Sun. Once it's stated, the solution almost looks obvious. While there's lots of work needed to make the idea practical, I admire the way they took a big noise problem and used it to propagate signal. It's too bad they are run by someone who thinks that they are going to make their money by licensing software instead of selling chips and licenses to very cool and real inventions.
Friends don't help friends install M$ junk.
I wonder if this could lead to new treatments of spinal injuries? Say you could place a chip intervening between a severed spinal cord. Instead of having to physically attach all those millions of nerve endings, you could have the chip do it by proximity, and carry the signals on past the gap.
BTM
That was the turning point of my life--I went from negative zero to positive zero.
I see it can reduce pin space problems but fail to see why it should reduce heat A TX/RX pair, and an amplifier, required in place of every pin? A transfer across a soldered connection does not in itself create any heat.
Obviously it's something like fiber optics using cheap ever-lasting LEDs to transmit data. However it could also use RFID. Much lower power and higher bandwidth capacity.
JA-MSP
Ironically, without Americans there would be no September 11th. Ironically, without Muslims there would be no September 11th. This is a meaningless statement. Without Jews there would have been no mass genocide in WWII, also. The situation in the Middle East is the result of years of complex political developments that you clearly don't understand much of - Zionism was largely a reaction to European anti-Semitism, and large scale settlement in the Palestinian Protectorate began as anti-Semitism grew in Europe, leading up to the formation of Israel after World War II. The Islamic population continues to be unhappy over a tiny slice of land that they don't want to share, and the conflict is fueled by extremists on both sides. Well, actually, many Palestinians these days would be content to share (i.e. two-state solution or something comparable) apparently, but the extremists and the incompetent government on both sides won't seem to let that happen.
Hmm .. independent chips, that communicate using capacitive inductance?
Isolinear chips anyone?
"Don't worry about the problems you have in mathematics, I assure you mine are much greater." - Einstein c.1919
This is a Dupe .
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.
Multi-chip modules have been around for a long time. The Pentium Pro, for example, was a multi-chip module. There's a multi-chip module Linux computer in a single package from ETRAX. Multi-chip modules are expensive and hard to manufacture, and they're generally used only when you need to combine chips that couldn't be manufactured on the same substrate, like a fast CPU and flash memory. They usually cost more than the chips packaged individually. That's why this isn't a mainstream technology.
This new approach might revive the multi-chip module market. Might. This has to become a cheap process before it will be used outside the supercomputer world. A whole generation of automated assembly machinery has to be developed to assemble and align chips in multi-chip modules before this is more than a demo technology. But this looks more promising than the way multi-chip modules are currently made. If it becomes cheaper to put two chips in one package than to put two chips in two packages, this is a significant development. Otherwise, not.
its negative effect on the market, read this one again and be happy.
:)
OSS is bringing down the overall value of computing, which is a good thing for all of us. The increased competition means the big players must begin to really innovate of die slow. The stuff we use everyday should be cheap. Intel did its job on the hardware side of things, OSS is working hard on the Software side.
This is the Sun I am used to seeing. I have said before, their value is in their people --nice to see them putting it to use.
Blogging because I can...
Didn't you just assume it was terrorism? Geez man, get with the times.
Sun should be working on making their very expensive UltraSparc chips as fast as ultra cheep Intel and AMD processors.
Dean G.
"I have a great mind to believe in Christianity for the mere pleasure of fancying I may be damned."-- some guy named George
It's only a special service given to subscribers. All of us freeloaders have to put up with the normal, always-on service.
Sound to me like serialize data and some sort of packetize communications between the chips (TCP?)
Gee, does each chip then has it own IP address? that sure will use up all the IPv4 address space.
I don't know if this is related...but the links between stories after you choose an article are gone, forcing me to have to go back to the homepage to get to the next article. I had a couple ideas. Did they not do some hardware upgrades not long ago? If so, maybe they are forcing stress testing on them to see whats wrong? Or maybe they sold out and are going for front page hits circa 1999. I always noticed in the subsections you could not traverse articles and always had to go back to the main page, maybe thats why. Anyway, /. /.'ing itself seems like cybersuicide.
Pretty much the same way one aligns a glass fibre in it's termination point...
--dave
davecb@spamcop.net
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?
With the abolishment of mainboard bottlenecks, who needs cache anymore? It'll be just as fast to pull data from main memory every time you need it.
However, it's probably not a place to discuss it unless you have something to contribute to resolving it.
The next 503 error will be ready soon, but subscribers can beat the rush and see it early!
Uh... I have nothing to contribute, but I can't resist the opportunity to post here anyway. :-)
Yeah, I got some 503 errors trying to access slashdot about 20 min ago. It was especially weird for me because I haven't been to slashdot or tried to access it in any way for over a month. And the first time I do, what do I get? A slashdotted slashdot.
Furry cows moo and decompress.
The question is, did Sun pay Roger Williams for the idea?
The Web is like Usenet, but
the elephants are untrained.
Reading between the lines, I think that this could be a ploy by Sun to solve some of their biggest problems-the high prices and late deliveries caused by producing chips with huge cache sizes-which would be a boon for Sun in the marketplace. No longer would Sun be losing the performance wars due to the delays inherent in producing chips with eight-megabytes of cache, instead a generic Ultrasparc chip could just have an appropriate chip stored externally. This would allow Sun to serve a greater variety of customers at a lower cost because it could sell the exact same chip in servers with 512k, 2MB, 4MB, and 10MB of cache as opposed to having the different chip lines it does now.
There is a research paper here that gives a lot more information than the article linked (ironically enough, I happened to be reading it yesterday). They address many of the issues people have brought up (alignment, dust, etc.), and the paper really isn't a hard read.
They actually have a bunch of interesting papers in the parent directory here, mostly covering stuff about asynchronous/clockless computing.
My server
I've managed to 'Eliminate Circuit Boards' without even trying in the past. I need a better soldering iron... and to control my temper
the idea of using fiber optic connections between chips? far less latency over long distances, no heat or capacitive coupling problems, no noise, incredible bandwidth, only one connection instead of 900+ little metal dots.
Impressive, green one.
If these chips can do what they say, it will be very very nice.
http://github.com/gbook/nidb
No. A wave is a combination of an electric and a magnetic field. This capacitive coupling relies exclusively on the electric field. Yes, Maxwell's equations state that a fluctuating electric field induces a magnetic field, but the field is small.
Theoretically, a very small amount of radiation could escape in a radial direction from the gap between the capacitor plates. But this is true of any type of capacitor.
The chip is not using "radio waves" to communicate. It is using an electric field.
By the way, it is very simple to spy on signals in wires, because the current flowing in a wire produces a magnetic field. This can be detected without tapping the wire.
In other words, I can already spy on your computer if I have sophisticated enough equipment.
I admit it's a tough field (hehe) to really get a grasp on, but seriously... Don't comment with false authority on things you barely understand.
So will they fix that ugly ass file chooser that we are stuck with in Linux?
That has to be one of my biggest gripes. It really sucks. This is one area, and I admit it's a small one, that Windows does a better job. I just hate the damn thing. That component has to go.
I hate the look. I hate the non relative pathing. I hate the fact that once I start typing I've lost info. It's a total piece of crap.
Wouldn't such a design be jammable?
I don't know about you, but I like the idea that my machines will keep chugging away barring a massive EMP.
I would hate to turn on my blender, and have my server go insane...
Lodragan Draoidh
The more you explain it, the more I don't understand it. - Mark Twain
Sun service engineer: I'm trying to fix this CPU, but all I have is sky pieces, anyone have a piece with a little bit of a boat?
It would be good for hobbyists in that the shift to tighter pins and ball grid arrays is making it extremely expensive to fabricate single circuit boards. Imagine buying a bunch of ASICs from digikey, stacking them together with duct tape, and instantly having a custom circuit board. The sides of the chips would only need transmit, recieve, and clock plates.
In manufacturing, the trend is still to integrate more and more on a single die. The cache will still be on the CPU but in addition, so will the system memory, graphics chip, and power supply. One day the 120V AC power cord will plug directly into the CPU.
Them being Sun, how could you tell?
lets see...
CPU plus cache times triple points equals 492 points.
oh yeah? raid array, cpu, 802.11g, gigabit NIC, dual ATI X800s times triple points equals OWNED BITOCH!!!
___________________ Check out really bad kung fu videos @ http://shittykungfu.com
gee, I thought Sun was about eliminating Linux. guess they've lowered the bar.
I think this is a really cool idea. You bond you chips to some sort of substrate that that provides power and some rudimentary i/o. Then most of the chip-to-chip data flows through capacitive coupling of their edges.
If you use square chips then each chip can talk to only 4 neighbors. However if the chips are hexagons then there are 6 neighbors which ups the grid bandwidth.
I must say I'm skeptical about going to serial links for things like memory access. If you go from a 256 bit parallel bus running at 800 MHz to a serial link, your motherboard traces are going to have to carry a signal at something like 200 GHz to get the same bandwidth. Your circuit board is going to need to be a millimeter-wave waveguide, and what are you going to make the transducers shoveling that data over the motherboard out of? You can generate 100 GHz-THz carriers using Gunn diodes, but that's not a signal.
You'd need optical links, and not very long ones. It probably wouldn't reduce the cost of the motherboard, anyway.
Read here for what the problem is and a work-around..
Sometime last year there was an article pointing to The Metamorphosis of Prime Intellect. I couldn't find the Slashdot article in the archives, but here is the story: http://www.kuro5hin.org/prime-intellect/
To sum up the point of my reference, they made full processors with only 3 pins for power, and the chips used the "Correlation Effect" to communicate with each other. I immediately thought of this story when I read this article.
A good store nonetheless. Just goes to show that sometimes Science Fiction isn't too far stretched from Science, given enough time.
Jeremy
I just couldn't resist :-)
This technology pertains more to chip manufacture than motherboard manufacture. The alignment difficulties alone will prevent this from being seen in the field. According to the research paper, the scientists first aligned the chips using a 10x stereo microscope, then used a Vernier measurement system to align them to within a few microns. There's no way that process will be seen outside of a lab or manufacturing plant.
What this will let chip makers do is to manufacture the cpu and cache on separate silicon wafers, then stack them together and package them as a unit. The researchers claim a speed of 21.6 Gigabits/second using a 4x4 matrix of transmitters. Perhaps we'll see processors being sold with X Gig of memory on-board, with X being the amount of memory that can be manufactured in the same space as the CPU. Perhaps additional processors could be stacked together as well. Imagine putting 4 CPU's and 4 Gig of memory into a spot on a motherboard that takes 1 CPU today. You will still need a printed circuit board to connect to the circuitry that handles the external devices, ports, slots, etc.
There was once an article about a guy who had filed a patent on a novel shape of brick. One problem was that any garden patio or driveway could deform due to flooding and heavy objects. Various solutions including placing layers of clay, cement and sand below ordinary shapes bricks. His solution was to have bricks in the shape of truncated tetrahedrons; take a tetrahedron and clip off two adjacent edges. This will leave two rectangle aligned at 90 degrees to each other. Placed side by side, the bricks would lock each other into shape, and thereby prevent any deformation.
Vintage computer adverts: http://www.vintageadbrowser.com/computers-and-software-ads
If the chips are to be as pinless as a Scrabble tile, how will they be powered?
Sun could get rid of the computers altogether and be really efficient!
My new
Your opinion might be different, but to me, it's a close call.
Bus speeds haven't kept pace with processor speeds, to be sure. However, saying they haven't increased is completely untrue. Whether or not you can say the the improvements in bus speeds have failed to increase "steadily" is a matter of what scale one is using to measure the improvements. From 8 bits at under 5 megahertz to 64 bits at 1.6 gigahertz is quite an improvement.
Perhaps the improvements haven't been steady enough for News.com, but steadiness isn't the point. The overall improvement is. There's more improvement desired, to be sure. Which is better: watching bus speeds improve "steadily", or to watch a breakthrough occur so that bus speeds once again match processor speeds? That's one breakthrough not likely to happen -- at least not any time soon. However, such a breakthrough would be much better than improvement at a pointless steady rate.
Now...it seems like someone wants to use the property to couple signals. Coooool!!!! Let's work with nature instead of against her! However, this will be one hell of a feat of Physics to pull off on a large Multi-chip Module: something that Physicists and PHd EEs can really sink thier teeth in to. Good to see some real neet analytical science comming back to basic circuit design.
Looks like Sun has returned to their well-deserved place as *the* den for executive-level crack-smoking.
Does anyone else read this and have visions of the HAL 9000? Here's another picture in case the first one gets Slashdotted.
Send/track messages to 100K people: www.xPressAlert.com
Dude, you've got to plug your computer in...
These aren't the sigs you're looking for.
The tin-foil hat is a much discussed, and from what I can gather, much needed item in the counter-geek apparrel. Unfortunately, I can't seem to find any tin foil in the country in which I live, only aluminium foil.
... To avoid the government/illuminati mind control waves, does the foil have to be tin? Or will aluminium suffice? And if aluminium is not suitable, what can be done? It's impossible to find tin foil here, and tin cans are not very malleable, so shaping them into a hat presents new difficulties, besides which the tin can is mostly steel anyway.
So my question to the geek community is (and it's a technical question)
Your answers are greatly appreciated.
Im an Electrical Engineer for a large government contractor and right now im currently working on Support&Test systems for one of our programs. My main job right now is to come up with ways to easily test surface mounted chips such as BGA's (Ball Gate Array). The old generation of IC's which had pins that ran through the PC board lent themselves well to testing because we could just lay the PCB on a "bed of nails" which would probe the pins on the backside of the PCB. With surface mounted chips this isnt so, there are no pins to probe and so JTAG created IEEE 1149.1 which is boundary scan testing, basically build the test circuitry into the IC itself. This proved very usefull, but it quickly became apparent that it had a big shortcomming, it only supported digital I/O and you would be hard pressed to find circuits now a days that arnt mixed signal (ie Analog + Digital). So an update to 1149.1 was made to allow analog probing called 1149.4. Only problem is, nobody has adopted it yet! Why you ask? Well because nobody has made software to easilly support it. And why is that? Well because nobody has shown much interest DUE TO THE LACK OF SOFTWARE! So as you can see we are currently stuck in the chicken and the egg scenario.
When you are dealing with Mil-spec testing such I, your worst nightmare is a non-repeatable failure. That is something went wrong but you don't know what and you cant make it happen again. When you toss up something into space, it has to work the first time EVERY time, no second chances in my line of business. I could only imagine the nightmares the poor sap would have who has to test one of these things. Surface mount is bad enough, but now they are introducing a myraid of external sources of error to the soup when the signals relay on inductance and capacitance which probably isnt much higher than the parasitic levels. I can only help they keep us test guys in the back of thier mind during R&D
A single wire is currently used to transfer say 400Mbps. So you use 128 of them to get 6.4 gB per second of bandwidth to your main memory. (that's beyond the cache!)
If you'd use wireless, and you'd use the 2.4Ghz "band" there is on the order of 2.4Gbits per second of bandwidth. If you stretch it a bit, you'd be able to get about 2.4 Gbytes per second. Use more bandwidth, and you'd be able to get enough for ONE cpu and ONE memory, but no SMP. And you still need your cache.
Now while the first CPU is communicating with it's main memory, you want the second CPU to do the same to a separate memory. Well, they are going to interfere if you don't channel the data trhough those copper guides, commonly called wires....
Note that when you put conductors on a PCB, you get to send more signals just a couple of cm further. But with wireless, you would "spoil" a whole sphere for other uses. Nah. This won't go far.