Slashdot Mirror
The Not-So-Cool Future
markmcb writes "Researchers at Purdue University and several other universities are looking to start work on a major problem standing in the way of future chip design: heat. The team is proposing a new center to consolidate efforts in finding solutions for the problem that is expected to become a reality within the next 15 years as future chips are expected to produce around 10 times as much heat as today's chips. The new center would work to develop circuits that consume less electricity and couple them with micro cooling devices."
I remember reading Timeline and they were talking about the limit of chips. No point in investing since they'll just get so small, that they'll burn themselves up.
Think about the people up in northern Canada, who need that precious heat! Unless this is some evil conspiracy to kill them off?
My UID is prime... is yours?
What this boils down to is "researches are looking at ways to make cooler chips." Well, duh, haven't they always?
Le français vous intéresse?
I thought the future of processors was going ot be photonic processors. I'm not sure if these will be producing any heat or not.
Yes Francis, the world has gone crazy.
As long as there's "work". There will always be losses (usually in the form of heat, but...)
I think that the solution to the heat problem will not come with better and more powerful cooling solutions, but rather radically changing how chips are designed and manufactured. The article doesn't contradict this, but I just want to emphasize that. Having some liquid nitrogen cooling unit is not the optimal, or even a good solution.
Canadian Cynic, canadian politics is less boring than you
they should look for ways to mass produce cheap diamonds.
Diamonds are about five times better at heat conducting as copper and could thus be used for passive cooling.
See pictures of tits
("ten times as much heat as today's processors")
I don't think that 1kW processors will be practical. Nobody is going to want to pay to run that, and nobody will want a heater running in their room all the time either.
I'd say that they should be looking to limit it to not much more than current figures (100W) - maybe 200W if we are generous. After that it gets silly.
For every expert, there is an equal and opposite expert. - Arthur C. Clarke
"Meanwhile, the cloud of electrons would be alternatively attracted to and repelled by adjacent electrodes. Alternating the voltages on the electrodes creates a cooling breeze because the moving cloud stirs the air."
Amazing, Purdue is developing the same technology used in such high tech devices as the Ionic Breeze air purifier.
Not that I claim to have a solution to the problem with overheating processors. But the power consumption of computers are starting to bother me.
I used to want the fastest computer around. But a few things have changed I guess.
First of all computers are starting to be fast enough for most needs.
Secondly, the way I use computers has changed with always on Internet. I never turn my computer off because I want to be able to quickly look something up on the web.
I also have a server that is running 24/7. Most of the time it is idling, but even when it is working I don't need it to be a speed demon.
So it is starting to be really important for me that a computer doesn't use a lot of power. I don't know if it affects my electric bill in a noticeable way, but it feels wrong.
The Internet is full. Go Away!!!
as other slashdot articles have proposed, future PCs (probably) won't be much more powerfull than today, but rather, like back in the mainframe days, dependend on some supercomputer selling it's processing power.
obviously such a mainframe can use massive parallel processing techniques were cooling is less of an issue.
See pictures of tits
We need to start working on the next generation of gerbil powered chips asap!!
Do not try to read the dupe, thats impossible. Instead, only try to realize the truth
What truth?
There is no dupe
The alliance proposed in the article, to me, seems similar to the AIM Alliance of the early 90s. Several companies united in a common goal. I've heard the AIM Alliance failed because competitors united in a common goal remain competitors, and as such tend not to fully disclose "trade secrets," even to further the common goal. If this proposed alliance takes off, I fear it will suffer the same fate as the AIM Alliance.
Raj Against the Machine! http://social-butterfly.appspot.com/
Not a joke.
The future is multi-core / multi-CPU boards where scaling comes from adding more pieces, not making them individually faster.
Yes, chips will always get faster and hopefully cooler, but it's no longer the key to performance.
Sig for sale or rent. One previous user. Inquire within.
I realized heat had gotten way out of control years ago when I got a spankin new Duron 800 and hooked it up without a heat sink. . . so much for that CPU.
Compare that to my trusty 400Mhz K62. The fan died the other day and all it did was reboot.
Now let's not even begin to discuss the P4. The heat problem is not years off, it's today and it is very serious already. In fact we're a few years into it already.
When I think of future problems that will happen to hardware, Hardware DRM comes to mind.
GoatPigSheep, the 3 most important food groups
Where does the heat really come from in chips? What I mean is - wires in my house have current running through them, but they don't need cooling, so why does a chip on a much smaller scale?
Why are CPUs so different from a lot of electronics out there? Is it the component count, the tiny size, or the close proximity of the components that makes them hot?
And another few points: why are they in such large packages and why aren't the pins (metal that passes closer to the core than anything else) designed to aid cooling?
Working on the latest generation of computers, its no suprise that the cheaper/generic fans are very noisy trying to turn faster to compensate for the greater cooling requirements.
An efficient and inexpensive cooling solution would be more desireable, IMHO.
Has anyone else experienced the "jet engine" noise comeing from newer systems?
Guess if you make the chip with less need for cooling requirements, we'll solve the puzzle also, however, that may be the more expensive road to the solution of fan noise, no?
Especially for those of us with newer motherboards who want a completely silent system with as few fans as possible
First it was CPUs with cooling and big/slow/no fans and big heatsinks, then PSUs GPUs and now MOBOs. My current custom box (now 14 months old) was built to be silent and I had a hard time settling on a motherboard that was state of the art, stable, and still used a passive heatsink to cool the board chipset fan-free. I finally settled on an Asus P4P800.
I can definately believe heat becoming even more of an issue. For those of us who want power/performance and quiet at the same time, this will become even more of a challenge as time goes on. I for one hope not to rely on expensive and/or complicated cooling devices, like peltier units, water pumps and the like. I hope the focus is on efficient chips that only clock up/power up as they need to, like the pentuim M.
my 2 cents.
uR iGn0ranc3, Their Power
No guys please stop this research...I really prefer a cpu fan, a northbridge fan, a graphics card fan, and two case fans that sound like a freaking 747. Why don't they just design a case where one whole side is a freaking fan?
Whoa that's cool, now it means no more petrol is needed.
If i take out my CPU cooler it reaches about 100'C. Now lets see, 100 x 10 = 1000'C in only 15 years of chip industry. If we manage out to put this heat into work, lets say we can have 'PC + hairdryer' packages or 'PC + free home-heating' winter offers or even 'PC - burn-a-pizza' boxes. Think about it, its only good news.
Funny, -1
Fucking a fat girl is like riding a scooter... it's fun 'til someone sees you.
Comment removed based on user account deletion
I'd like to hear from some engineering types about why we can't use the excess heat from CPUs to do useful work. I know virtually all large-scale methods of generating electricity involve generating large amounts of heat through some process (nuclear reactions, burning coal or oil, etc), using it to create a hot gas, which turns a turbine, generating electricity.
I also have some vague handwaving idea that there are processes for generating electricity that have to do with harnessing temperature differentials, but I really don't know what I'm talking about.
Anyway, why can't we have little gas turbine generators (or some other method) in our machines that reclaim some of this lost energy, instead of wasting it? Seems like the aggregate energy amounts would be pretty large.
Take off all your....um...inefficient circuits! Its getting hot in here, take off all your inefficient circuits!
Using energy creates heat. If they use less energy there is less heat. I think they should ignore the direct problem and fix the indirect problem.
I thenk you need to get a bigger room. the 300W or so your computer uses would hardly be enough to heat a toilet...
See pictures of tits
Sig response: Actually, yes mine is - thanks for asking!
James Tiberius Kirk: "Spock, the women on your planet are logical. No other planet in the galaxy can make that claim."
w00t, no more heaters! now we just need a new way to cool my house...
"The microscopic cloud of ionized air then leads to an imbalance of charge in the micro-atmosphere, and lightning results. "
Using lightning to cool a CPU?
Doesn't EMF pose a problem here?
Guess you could shield, but thats counter productive isn't it?
>problem that is expected to become a reality within the next 15 years as future chips are expected to produce around 10 times as much heat as today's chips.
This is bullshit. I am never even considering buying a >>100W CPU for my desktop, certainly not 1000W.
I'd rather see a less fans in my machine, not more.
Looking into heat/area is more reasonable as area will decrease for a while still.
Or perhaps I'm grossly physics-impaired.
Thermal conductivity chart.
Please note plane parallel thermsl conductivity of graphite which greatly exceeds diamond.
If you don't want to peruse the linked material...thermal conductivity(W/m - K):copper = 401, diamond = 895, Graphite = 1950.
Begging forgiveness in advance for possibly over-generalising:-
The alarming heat output of modern processors is to a significant degree caused by the fact that they have a multi-GHz clock which must run at all times even when the chip is idle. Asynchronous chips have no clock - which means when they are idle, they do not generate anywhere near as much heat. And as a general rule, RISC processors are more efficient than CISC processors running at the same clock speed, so I predict these factors in combination will push the industry towards asynchronous RISC designs in preference to trying to run current-model processors at insane temperatures. Googling for asynchronous processors will provide a good variety of info on this interesting subject.
It's called cogeneration. The big problem is that source of cold. You want a large temperature differential so you need something really cold or are willing to let your cpu get really hot, hot enough to incandesce. The latter would let you replace your leds with light from the cpu. You could then replace your incandescent light bulbs with your surplus leds.
Are we getting dumber of something? Or more likely is this just academic masturbation?
;-)
The solution has been around for a long time. I feel like I should keep it a secret and patent it (again) for this particular purpose. But you know what? I don't give a sh*t, cause the whole patent system sticks to high heaven too. And I'd just get abused by some asshole money man in the end anyway. Heh, can you tell I'm a bright but very poor man living in a sea of captialist pig-sharks. No, I'm not cynical
Anyway the soultion is.... *drum roll please*...
THE STERLING ENGINE!
Idiots.
What would it take to replace x86 with another chip like Crusoe or MiPS and make it better for desktop PCs?
-jpeg
hello!!! work on that.
I am the Alpha and the Omega-3
I think your 470C number is too high.
If your chip current runs at ~45C, and your room temperature is ~25C, then it's running at 20C above RT. Ten times that means that in the future it will run at 200C above ambient, which is 225C (which is still alot!). (And actually, the temperature would be even a bit lower, since the higher the temperature gradient, the faster heat dissipation occurs, which reduces the gradient more efficiently... so 10X heat production generally means less than 10X final temperature excess.)
Of course, the whole point is that 225C is way too much. The 225C assumes that the heat production is 10X higher, but the cooling solution has not changed. The point of all this research is to find cooling solutions that can keep up with these heat dissipation requirements, so that your processor of the future will still run at ~50C even though it's generating tons of heat.
"Mechanical engineers at Purdue have filed patents for ... "
"The patents arose from a research project funded in part by the National Science Foundation."
The idea of getting the NSF funding, (in part), the research that will later lead to mechanical engineers getting the patent would be a great way to make money at the expense of others.
Should not the patent rights be shared among those who funded the project?
This would let you get all the benefits of existing tried-and-tested cooling methods, but would eliminate the bugbears of the chip's casing being an insulator and the possibility of condensation screwing everything up.
A variant on this would be to have the chip stand upright, so that you could have a cooling system on both sides. The pins would need to be on the sides of the chip, then, not on the base.
A second option would be to look at where the heat is coming from. A lot of heat is going to be produced through resistance and the bulk of chips still use aluminum (which has a relatively high resistance) for the interconnects. Copper interconnects would run cooler, and (if anyone can figure out how to do it) silver would be best of all.
A third option is to look at the layout of the chips. I'm not sure exactly how memory chips are organized, but it would seem that the more interleaving you have, the lower the concentration of heat at any given point, so the cooler the chip will run. Similarly for processors, it would seem that the more spaced out a set of identical processing elements are, the better.
A fourth option is to double the width of the inputs to the chips (eg: you'd be looking at 128-bit procrssors) and to allow instructions to work on vectors or matrices. The idea here is that some of the problem is in the overheads of fetching and farming out the work. If you reduce the overheads, by transferring work in bulk, you should reduce the heat generated.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
Put three 100W light bulbs in a large cardboard box, and stand by with a fire extinguisher.
tasks(723) drafts(105) languages(484) examples(29106)
Oh for the love of god, where do you people learn this?
Human brains, being as powerful processors as they are don't run as hot... Therefore as a pc chip doesnt _need_ to either surely?
because I want the G8 to go into Powerbook first when its release. I tired of this whole G5 fiasco
You don't have to be smart to use a Mac, you just have to be smart enough to buy one
Erase software that sucks many many resources ... .... ... ... ... ... ... ...
Erase enterprises's royalties of hardware architectures
Erase the incompatibilities of many hardwares and drivers
Erase the copyrightes' owners in the hardware
Clean the hardware of dirties
Clean the software of dirties
Build a common hardware for everyones
Build a common software for everyones
It's the way of the future.
Now we know the real cause of global warming.
There is a major break in Moore's law per core, but they are getting around that with multiple cores. If you have twice as many active devices and you run them at 1 over square-root of 2 times the frequency the power per core will be halved and you get the same computing power. Less frequency reduction gives more computing power. This does help the heating problem by spreading the heat source over a larger area.
Another way to increase the cooling effectiveness is do draw heat out both the top and bottom of the chip. Perhaps the pins could carry heat out. Perhaps we need a package with a hole in the array of pins to give more cooling on the bottom.
The problem of putting in more conections to the CPU is that it takes up so much space. It may be necessary to increase clock speeds for transfers on and off chip and use multiplexing. The ultimate would be serial links on laser beams. Then, we could use fewer pins and gain more access to the chip for cooling.
The move to 64 bits is overkill for most desktop users, but is wonderful for servers. I believe we could best use a small number of 64 bit water cooled beasts with hundreds of thin clients running at 10 watts or so. Now, there is a power saving. 200W for server + N X 10 W for clients is much better than N X 100 W as long as N > 3 For N=100 clients, the cost is 1200W instead of 10000W. It will take the chip techs a long time to do that, but the network boys and girls can make it happen today.
A problem is an opportunity http://mrpogson.com
Most CPUs can run with a range of clockspeeds. The over-clockers go up. You can go down at less than peak loads. I work a lot with terminal servers. The load from each client is very spiked. There are burst of a second or two as a chunk of data is moved or processed. Some systems can be set to adjust clockspeed according to load just like a screensaver can blank the screen. The absolute best way to save energy is to network so that one powerful machine can serve a bunch of clients. A $1000 machine can handle 35 thin clients that need only 10 watts or so to run an LCD monitor with a low power CPU without fans built in. That's about as much power as a wall clock per client.
A problem is an opportunity http://mrpogson.com
Put the powerful server in the attic or basement and network to a fanless thin client. You can also try longer video/mouse/keyboard cables if you cannot go fanless.
A problem is an opportunity http://mrpogson.com
"So.. in 15 years, my PC will be 470C? (1,166F)" and "I think ill need to find a new hobby in 15 years."
Might I recommend becoming a McDonalds coffee server?
Higher temperatures increase noise levels. Room temperature is about .025 electron volt. 600C is about .075 ev I think silicon carbide or gallium arsenide are suitable semiconductors but they would need to be fabricated at higher temperatures, making precision much more difficult. On the mobo, you would have to insulate or cool all the heat sensitive components in the neighborhood
A problem is an opportunity http://mrpogson.com
Finally, someone has decided it's a good idea to slow the heat-death of the universe! Come on, electronics have to be one of the biggest heat producers on earth!
billions and billions of $. the software that runs on x86 won't run on MiPS or Sparc or vice versa. the SW manufaturers would have to recompile the code to match new chips. all the legacy of sw we now have will be gone.
estimate 100 million PC's in the US.
each having ~1500 bucks of SW
that is $ 150 billion right there.
and lets say HW also costs ~1500 per PC
that is another 150 billion.
* POOF *
add the cost of developing new compilers
and time
add shipping and handling
and may be a few things that i missed.
... recycling the bit-bucket?
Finding better ways to suck excess energy of a chip is very well and good, but it might be better to reduce the energy produced by the chip in the first place.
If every time a 1 is set to a zero, why not feed that into a bank of capacitors rather than the current solution (which I believe is to sink it to ground, thus producing heat)?
"Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
Why are you bringing this up again for? I wrote my answer to it in a previous SlashDot article about a YEAR AGO. Chips need to be designed with a series of integrated tunnelings thru the entire chip. This would INCREASE SURFACE EXPOSURE into the chip core, magnifying the cooling effect. I guess I should write Intel and AMD instead of SlashDot... but I thought Intel & AMD was ON SLASHDOT to pick up new ideas! So, here it is. Design your chips larger, leaving cooling tunnels interspaced throughout the cpu chips. If nothing else, think of hallways in a BORG SHIP.