Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Researchers Design Microchip Ten Times More Efficient

ScuttleMonkey posted more than 6 years ago | from the proof-of-concept dept.

Power 113

WirePosted writes to mention that a new highly efficient microchip has been announced by researchers from MIT and Texas Instruments. The new chip touts up to 10 times more energy efficiency than current generation chips. "One key to the new chip design, Chandrakasan says, was to build a high-efficiency DC-to-DC converter--which reduces the voltage to the lower level--right on the same chip, reducing the number of separate components. The redesigned memory and logic, along with the DC-to-DC converter, are all integrated to realize a complete system-on-a-chip solution."

cancel ×

113 comments

Sorry! There are no comments related to the filter you selected.

Will we get these soon? (5, Insightful)

Naughty Bob (1004174) | more than 6 years ago | (#22776304)

The article doesn't say whether these chips are cheaper to make than the current technology. That will be the deciding factor regarding how soon these make their way into our portable devices.

Re:Will we get these soon? (0)

Anonymous Coward | more than 6 years ago | (#22776346)

But when can I put one in my TI-89?

Re:Will we get these soon? (3, Funny)

pilgrim23 (716938) | more than 6 years ago | (#22776362)

well whatever the outcome, the Moore the merrier. Thats the law.

Re:Will we get these soon? (0)

Anonymous Coward | more than 6 years ago | (#22776800)

I thought the law was:

Break the rule and face the bool!

Who run Processor Town? Moore run Processor Town!

Re:Will we get these soon? (3, Funny)

jd (1658) | more than 6 years ago | (#22778166)

There are twice as many jokes of that kind every 18 months.

Re:Will we get these soon? (1)

Iron Condor (964856) | more than 6 years ago | (#22778888)

There are twice as many jokes of that kind every 18 months.

Actually, according to Wikipedia the number has tripled in the last six months.

Re:Will we get these soon? (0)

Anonymous Coward | more than 6 years ago | (#22778490)

Agreed! How else can I get my p0rn rendered faster!

Re:Will we get these soon? (5, Insightful)

Shikaku (1129753) | more than 6 years ago | (#22776410)

The processors, chances are, are not very useful for things that require more processor horsepower. The article mentions use in things that could be powered by minuscule power input (i.e, body heat). I think it was invented for that purpose, that the power bill will be minuscule at worst, nonexistent at best for these things. Therefore, use in portable devices for now are not so much, except maybe a simple PDA that requires no batteries, only heat.

Re:Will we get these soon? (1)

nschubach (922175) | more than 6 years ago | (#22777696)

Some of us have a better coefficient for heat transfer. (Am I saying that right? Some of us "put off" more of our body heat than others.)

Like for instance. If I touch the girls here at work (...I'll likely lose my job *rim shot* [you know someone was thinking it]), they are usually "colder" than my body temp. Would this mean that the "warmer feeling" bodies would power these easier? Would those with "cold bodies" not have enough heat to power their PDA at all? Does it matter?

Re:Will we get these soon? (0)

Anonymous Coward | more than 6 years ago | (#22777894)

So what you're saying is that my wife's shoes might not be the ideal place for a computer? My pants on the other hand...

Re:Will we get these soon? (2, Funny)

Paradise Pete (33184) | more than 6 years ago | (#22778126)

My pants on the other hand...

If your pants are on the other hand, you're doing it wrong.

Re:Will we get these soon? (0)

Anonymous Coward | more than 6 years ago | (#22777942)

If I touch the girls here at work ... they are usually "colder" than my body temp.
How are alcoholics and pedophiles alike? They're always looking for an excuse to crack open a cold one...

Re:Will we get these soon? (0)

Anonymous Coward | more than 6 years ago | (#22778220)

damnit. posted too quick.

Re:Will we get these soon? (1)

John Meacham (1112) | more than 6 years ago | (#22779076)

Well, power doesn't come from heat, it comes from moving heat from a hot place to a cold place. so, if you are in 98 degree weather, your body can't power a thing. likewise, if you are in sub zero weather, you have a nice thermal gradient to exploit. I think the variance in environmental temperature far outweighs any contribution from abnormal body heat temperatures.

Re:Will we get these soon? (1)

Clanked (1156473) | more than 6 years ago | (#22777702)

A pacemaker that never requires a battery change sounds good to me.

Pacemakers? (1)

camperdave (969942) | more than 6 years ago | (#22778710)

A pacemaker that never requires a battery change sounds good to me.

Haven't they got those things running off ATP [wikipedia.org] yet?

Re:Will we get these soon? (2, Interesting)

Serveus (1257850) | more than 6 years ago | (#22777866)

The thing about this is body heat is not a source of power. To have a source of power (for a thermocouple to work) you need a temperature differential, so you would need a part of the thermocouple sticking out of your body. Another thing is no matter how low the current drawn from batteries is they still need changing batteries only have a limited shelf life (with no current drawn.) Lower power means smaller batteries but I dont think thermocouples are a viable source of power for implants, perhaps they can get power from motion.

Re:Will we get these soon? (2, Funny)

dwywit (1109409) | more than 6 years ago | (#22778222)

I believe that one's testicles hang away from the body because they need to be a degree or two cooler than core body temperature - would that suffice? Could be a problem for the girls, though.

Re:Will we get these soon? (1)

steveo777 (183629) | more than 6 years ago | (#22778382)

Could be a problem for the girls, though.

Feet vs armpits. Should be a vast difference.

Re:Will we get these soon? (1)

bluefoxlucid (723572) | more than 6 years ago | (#22777898)

Processors are not engines, there is no horsepower. Electrons don't drive faster with more heat, they're running less efficiently.

Re:Will we get these soon? (1)

droopycom (470921) | more than 6 years ago | (#22778426)

Electrons dont run either.

Do you have a better analogy for relative measure of processing power or are you just being pedantic?

 

Re:Will we get these soon? (1)

bluefoxlucid (723572) | more than 6 years ago | (#22778794)

The requirement for "A LOT OF ELECTRICITY" to drive a processor is a serious fallacy. 180W Intel 1.2GHZ processors produce blazing amounts of waste heat; more modern, higher performance processors use a third of the power. A smaller circuit (RISC processor for example) can perform simple tasks faster with less power too (indeed some RISC processors can take 1.5 times the insns to do a logical task and blow through it in 1/2 the clock time anyway)

Re:Will we get these soon? (1)

jbengt (874751) | more than 6 years ago | (#22779618)

Power is power, electrical or horse.
Intel core 2 duo => approx 35 watts = 0.47 HP,+/-

Re:Will we get these soon? (2, Informative)

jbengt (874751) | more than 6 years ago | (#22779628)

gaahh!!
(preview is my friend, preview is my friend, preview is my friend . . . )
Power is power, electrical or horse.
Intel core 2 duo => approx 35 watts = 0.047 HP,+/-

Re:Will we get these soon? (3, Informative)

Gewalt (1200451) | more than 6 years ago | (#22776610)

Not really.... Design is usually more expensive than production. Production cost is pretty much insignificant. (Except memory)

Re:Will we get these soon? (1)

avandesande (143899) | more than 6 years ago | (#22776704)

Any examples where combining functions into a single (electronics) module hasn't paid off?

Re:Will we get these soon? (5, Funny)

misleb (129952) | more than 6 years ago | (#22777146)

Combining computer and girlfriend: FAIL

Re:Will we get these soon? (1)

nschubach (922175) | more than 6 years ago | (#22777744)

He said a "single" module, thus no counterpart and thus no "girlfriend."

Re:Will we get these soon? (3, Funny)

VGPowerlord (621254) | more than 6 years ago | (#22778372)

but if it did work, I'd make a note here: HUGE SUCCESS.

Re:Will we get these soon? (3, Interesting)

crgrace (220738) | more than 6 years ago | (#22777338)

Putting a power amplifier on the same chip as a radio transmitter has not been successful. A lot of money has been wasted going down that road so far...

Re:Will we get these soon? (0)

Anonymous Coward | more than 6 years ago | (#22777066)

There shouldn't be any difference in cost. It's a fixed cost assuming the same die area.
He's just running lower voltages hence reducing power consumption like the square of the
voltage. P = C V^2 f (where C is capacitance, V is voltage, and f is frequency).

Re:Will we get these soon? (1)

SeaFox (739806) | more than 6 years ago | (#22777078)

The article doesn't say whether these chips are cheaper to make than the current technology.

And the cost to implement will also include patent licensing for chip manufacturers, not just the production costs as it may for current designs.

Actual Product Available NOW! (1)

cc22dd (1174901) | more than 6 years ago | (#22779324)

Note: I posted this earlier as a reply, but no one seems to have noticed it. Hence this repost. I'm a long-time-reader-very-rare-poster, so sorry if this is not the right way to go about it.

Although the study quoted by the OP got a lot of media attention because of MIT involvement, what is more interesting is this actual product that has been released last month: "One AAA battery! The boss must be kidding..." [embedded-computing.com]

This company (Silicon Labs) has managed to put a DC-DC converter in a microcontroller and have managed to do this on an actual product that you can buy now (not just a research project!). They claim to be able to run for years (even >15 years) on typical low-power applications such as data loggers that wake up for a short while take a measurement and go to sleep. This is also the first microchip that can run on one battery... if you think that adding an external DC-DC converter would do the same trick, you have to remember that the external DC-DC converter needs to be ON even during sleep mode so the micro can wake up again, which burns quite a bit of power. They claim to have eliminated this by putting the DC-DC converter on chip.

More articles on this micro: http://www.eetimes.com/news/semi/showArticle.jhtml?articleID=206801775 [eetimes.com] http://www.electronicsweekly.com/Articles/2008/02/26/43201/silicon-labs-microcontroller-features-integrated-dc-dc-for-portable-uses.htm [electronicsweekly.com]

Re:Will we get these soon? (1)

bgat (123664) | more than 6 years ago | (#22779822)

Yes and no. If it cuts the cost of the battery by 50%, then the chip could be considerably more expensive without affecting the overall product cost. Batteries and their related circuitry are expensive!

Dup? (4, Informative)

Bob-taro (996889) | more than 6 years ago | (#22776342)

I thought this sounded familiar. [slashdot.org]

Any chance of commercial success? (4, Interesting)

damn_registrars (1103043) | more than 6 years ago | (#22776348)

The first thing that came to mind when I saw this article was the Transmeta [wikipedia.org] Crusoe [wikipedia.org] processor. Which unfortunately never achieved much of any significant market penetration. Indeed, it seems that you really have to have something more than just an incredibly efficient chip in order to compete against the Intel - AMD behemoth.

Personally, I would love to see a chip that requires very low power make it into the mainstream market. I think it would great to have something like that for the miniITX form factor or something of that nature that hobbyists could tinker with and find fun applications for. The Transmeta, unfortunately, never realized that as far as I ever saw.

Re:Any chance of commercial success? (5, Interesting)

ejtttje (673126) | more than 6 years ago | (#22776444)

I don't think they're demonstrating a particular CPU, but a technology or design strategy that can be built into *any* chip. So Intel or AMD could pick up this research with their own chips. (subject to patents and licensing of course)

Also, from the article: "So far the new chip is at the proof of concept stage. Commercial applications could become available "in five years, maybe even sooner, in a number of exciting areas," Chandrakasan says

Re:Any chance of commercial success? (3, Informative)

Henneshoe (987210) | more than 6 years ago | (#22776606)

Sense the research was sponsored by TI, I am sure this technology will find its way into all sorts of embedded devices. Think everything from 32-bit uCs to Opamps. If it really does increase power efficency 10 fold, it wouldn't supprise me to see AMD and/or Intel license the technology from them for high speed uPs.

Yeah, but that was transmeta's fault (1)

WindBourne (631190) | more than 6 years ago | (#22776512)

The reason is that they needed to build up an industry to accept them. There were already other chips fabs that had a name. So what did transmeta do? Nothing. They should have spent a few bucks and looked for new ideas that used their chip. They only needed a few interesting ideas to make it. How much money? What a million / idea? Even had it cost them 3 million, then it would have been NOTHING in the long (or even short) run.

Re:Any chance of commercial success? (1)

J.R. Random (801334) | more than 6 years ago | (#22776574)

Well, if you read the fine article you will see that the applications they talk about are things like medical implants, where you'd like to avoid surgery every few years to replace the batteries. The article makes no claims that these chips will appear any time soon in your desktop computer. Since they save power in the usual way (by reducing voltage) they're probably slower than stock chips. This doesn't matter in a lot of imbedded applications but it won't attract the gamer crowd.

Re:Any chance of commercial success? (3, Informative)

damn_registrars (1103043) | more than 6 years ago | (#22776802)

Since they save power in the usual way (by reducing voltage) they're probably slower than stock chips.

Yes, I do recall that the Transmeta chips were a fair amount slower than the Intel / AMD chips that were out at the same time, though in some regards one could say they made up for it with far better battery life in laptops.

This doesn't matter in a lot of imbedded applications but it won't attract the gamer crowd.

I can't speak for everyone, but I wasn't planning to run duke nukem forever on a low-power system... But I can think of plenty of typical household applications that would be well suited to a cpu that consumes less power.

Re:Any chance of commercial success? (1)

Naughty Bob (1004174) | more than 6 years ago | (#22776892)

Additionally, when you make the voltage smaller, you can increase the complexity of the silicon, packing more functions per unit volume.

Re:Any chance of commercial success? (0)

Anonymous Coward | more than 6 years ago | (#22777012)

I wasn't planning to run duke nukem forever on a low-power system
By the time Duke Nukem Forever comes out, you'll be running it in a VR chamber with your home PC powered by a Mr. Fusion generator...

Re:Any chance of commercial success? (0)

Anonymous Coward | more than 6 years ago | (#22779150)

Awesome! So, like, in 2015?

Re:Any chance of commercial success? (4, Insightful)

PPH (736903) | more than 6 years ago | (#22776654)

That depends on youe definition of 'the mainstream market'. This technology may never appear in desktop/laptop PCs, but become popular in handheld devices where power consumption is a major issue. There is a limited amount of power saving economically feasible in PCs as long as the displays and other peripherals continue to be major power hogs.

Another interesting market might be in server farms. But I wouldn't count on this driving the market. CPU architectures specific to servers haven't sold well, so this isn't an economically viable niche.

Microcontrollers are a large enough market segment to justify the R&D. I forget where I read this, but if you take the total percentage of the uP and uCs installed in PCs and round it to the nearest whole percent, that number is zero.

Re:Any chance of commercial success? (5, Informative)

Phat_Tony (661117) | more than 6 years ago | (#22776664)

Transmetta had radically better power consumption for a while and might have some day come to dominate the portables market, had they retained an advantage like the one they had at their debut. Transemetta's problem was underestimating how rapidly Intel could improve the power efficiency of their chips. In response to Transmetta, Intel suddenly got serious about power consumption and got competitive so fast it left Transmetta with little to differentiate their chips from the competition.

Like anything, the commercial viability of this doesn't just depend on how much better it is than what's already out there, but on how long it'll take their competitors to catch up.

Transmetta didn't do so well, but the real winner of Transmetta's actions was the consumer. Transmetta drove Intel and AMD to improve efficiency much more rapidly than they had been. Let's hope this new technology makes it into production and does the same.

Re:Any chance of commercial success? (1, Insightful)

Anonymous Coward | more than 6 years ago | (#22776738)

Ugh. Transmeta. Two t's, not three. Sony and Intel are licensees of Transmeta's technology (the latter being the terms of a legal settlement, but no doubt Intel makes use of it). They were always a flop as a chipmaker, but seem to be doing all right as a smaller fabless concern.

Re:Any chance of commercial success? (1)

WATist (902972) | more than 6 years ago | (#22777444)

It always seemed to me that transmeta was a victim of the dot com boom and the 15 minutes of fame it gave them. That publicity raised expectations and warned competitors before they were ready.

Re:Any chance of commercial success? (1)

renoX (11677) | more than 6 years ago | (#22778368)

>Transmetta had radically better power consumption for a while

Thet had low power consumption yes, but the review also said that their performances weren't that good..
So was their performance/power good enough or not?

I don't know: does someone have figures?

Re:Any chance of commercial success? (1)

ekstrom (941853) | more than 6 years ago | (#22778970)

Their ancestor is already a commercial success. The processor used to demonstrate this technology is the MSP430 line of low-power microcontrollers from TI. It is 16-bit Von Neumann RISC with an address space stretched to 1mB that with present technology runs at 3V using about 500uA/mHz. It will run up to 14MHz at 3V and does register-to-register operations in one clock tick, to and/or from memory in 2 to 6 depending. It idles at 0.1uA waiting for an interrupt, or at 0.6uA when keeping time with a 32KHz on-chip crystal oscillator. The family includes models with hardware multiplier, lots of handy embedded peripherals including 12-bit ADC, DAC, timers, serial ports of various types and so on. They will never power a general purpose computer, but they are a serious competitor in the market now served by PIC. I use them routinely in extremely low-power embedded applications and like the family a lot.

Cutting to the chase (4, Informative)

objekt (232270) | more than 6 years ago | (#22776352)

Just like all these articles on breakthroughs in energy efficient technology, there's only one thing I'm interested in.

from TFA:

So far the new chip is at the proof of concept stage. Commercial applications could become available "in five years, maybe even sooner, in a number of exciting areas," Chandrakasan says.

Re:Cutting to the chase (2, Insightful)

Firehed (942385) | more than 6 years ago | (#22776494)

So as usual, something ten times better than we have now is going to be available in five years. Since these breakthroughs happen all the time, we continue our remarkably linear trend by continually filling in the gaps.

Re:Cutting to the chase (1)

sucker_muts (776572) | more than 6 years ago | (#22776618)

So as usual, something ten times better than we have now is going to be available in five years.

Sure, in five years the available chips will be a lot better than the stuff that's here now. But when this technique has matured enough, it could be applied to the chips in 5 years and we'll still get a 10 fold improvement! (Or something like that :-P)

This seems to be a complete other kind of advancement than regular chip evolution we've seen so far.

Re:Cutting to the chase (4, Informative)

Jerry Coffin (824726) | more than 6 years ago | (#22776968)

This seems to be a complete other kind of advancement than regular chip evolution we've seen so far.


There's not enough in TFA to say for sure, but I'd guess rather the opposite. The main thing they mention is a lower power supply voltage. Power supply voltages have been dropping steadily for a long time. Once upon a time, the most common logic family was the 7400 series, which all used 5 volt power supplies. Somewhat later 3.3 volt CMOS logic was introduced. Most CPUs, memory, etc., now use somewhere between 1 and 2 volts.

For the most part, you get a trade-off between voltage and speed -- with a higher voltage, you can charge up a more reactive load more quickly, giving faster rise and fall times. That translates directly to higher bus speed.

At the same time, the power you use is the product of the voltage and the current, so as you raise the voltage you raise the power usage. Worse, the current you drive through a given impedance also rises linearly with the voltage -- so the power usage is proportional to the square of the voltage.

That (probably) explains to a large degree how/why they've reduced the power usage by a ration of 10:1 by reducing the voltage by a ratio of something like 4:1 (in theory, a 10:1 power reduction should imply a voltage reduction by the square root of 10, roughly 3.16).

In any case, however, nothing in the article really suggests that they've departed a great deal from the path everybody's been following for quite a while. Of course, they may have done something truly radical here -- but based on what they've said, that isn't necessarily the case.

Re:Cutting to the chase (1)

crgrace (220738) | more than 6 years ago | (#22777374)

They seem to have departed from standard practice in two ways. First, they use an on-chip DC-DC converter to dynamically scale the voltage down to the minimum required to meet whatever performance metric is specified. Second, they intend to power it using some kind of energy-harvesting technique (namely human body heat). I agree the reduction of power is just first-order physics, but how they do it is quite interesting.

Re:Cutting to the chase (1)

Jerry Coffin (824726) | more than 6 years ago | (#22780180)

They seem to have departed from standard practice in two ways. First, they use an on-chip DC-DC converter to dynamically scale the voltage down to the minimum required to meet whatever performance metric is specified.


That's not really new by itself. Just for an obvious example, most Flash ROM chips need a relatively high voltage for writing. Early Flash ROM chips used dual power supplies to support that, but most current Flash ROM chips use a single external power supply and an on-board DC-DC converter to produce the write voltage.

It's also not clear (at least to me) whether they're really scaling the supply voltage down to the point needed, or possibly scaling it up as needed. As noted above, scaling up has been common (at least for some kinds of chips) for quite a while. Scaling down isn't nearly as common, but I'm pretty sure some kinds of chips (for some reason, A->D converters come to mind) have done that as well.

Second, they intend to power it using some kind of energy-harvesting technique (namely human body heat). I agree the reduction of power is just first-order physics, but how they do it is quite interesting.


Energy harvesting is mostly a matter of reducing the power usage to the point that it becomes practical. RFID tags (for one obvious example) are powered entirely by the transmitter that reads them. That is, to read one, you put the reader close to the tag. The tag has an antenna that picks up the radio signal from the reader, and uses it to charge a capacitor. The power from the capacitor is used to transmit a signal back to the reader.

Simply using heat to generate electricity is well known as well. Thermocouples, which generate electricity directly from a temperature difference have been known for a long time. Most electrical power plants start by producing heat, use that to boil water, then use the steam to run a turbine.

Re:Cutting to the chase (1)

tsotha (720379) | more than 6 years ago | (#22779316)

For the most part, you get a trade-off between voltage and speed -- with a higher voltage, you can charge up a more reactive load more quickly, giving faster rise and fall times. That translates directly to higher bus speed.


I've been out of it for awhile, but yeah, that's what I was thinking. Unless they've really pulled out some whizbangery, they've just made a really slow processor that doesn't take much power. Meh. How much different is this, really from making a CMOS processor with a low-voltage external power converter? Sure, they put the converter on the chip. Again... meh.

Re:Cutting to the chase (1)

fluffy99 (870997) | more than 6 years ago | (#22779772)

It wasn't clear whether the chip circuits actually operated at the 0.3 volts, or that's whats fed to the on-board dc-dc converter and it's stepping it up or down as needed. If they really do have transistors that reliably work at 0.3 volts that's a significant breakthrough that I'm sure the heavy duty CPU guys would love to know about. I'm thinking this might simply be a matter of better battery use by integrating a high-quality dc-dc converter on the chip. It seems like most consumer grade electronics have crappy converters that drop off long before the battery is full depleted. How many gadgets, flashlights, etc do you have that drop off when your AA batterys hit 1 volt? The joule-thief is a great example of how a decent converter can extract a lot more energy from "dead" batteries.

Re:Cutting to the chase (1)

owlnation (858981) | more than 6 years ago | (#22776638)

Commercial applications could become available "in five years, maybe even sooner, in a number of exciting areas," Chandrakasan says.
I was about to post something witty about these guys having to run and hide from the Moore's Police...

And then I read that quote. Yep... just another aspirational "news" story. Tag under: "flyingcars", "dukenukemforever" and "robotbulter".

Re:Cutting to the chase (1)

cc22dd (1174901) | more than 6 years ago | (#22777934)

Although the study quoted by the OP got a lot of media attention because of MIT involvement, what is more interesting is this actual product that has been released last month: "One AAA battery! The boss must be kidding..." [embedded-computing.com]

This company (Silicon Labs) has managed to put a DC-DC converter in a microcontroller and have managed to do this on an actual product that you can buy now (not just a research project!). They claim to be able to run for years (even >15 years) on typical low-power applications such as data loggers that wake up for a short while take a measurement and go to sleep. This is also the first microchip that can run on one battery... if you think that adding an external DC-DC converter would do the same trick, you have to remember that the external DC-DC converter needs to be ON even during sleep mode so the micro can wake up again, which burns quite a bit of power. More articles on this product:
http://www.eetimes.com/news/semi/showArticle.jhtml?articleID=206801775 [eetimes.com] http://www.electronicsweekly.com/Articles/2008/02/26/43201/silicon-labs-microcontroller-features-integrated-dc-dc-for-portable-uses.htm [electronicsweekly.com]

I wonder where it will be built at? (0, Insightful)

Anonymous Coward | more than 6 years ago | (#22776466)

It seems like all of American know-how goes into designing things like this, then companies move the jobs overseas. Now, if they moved it to countries that are friendly to here, I would not care. But where do they send it to? China. Not even Taiwan anymore. Plain ole china. And yet, China is on a real spending jag to build their military up. The amazing part is that many of the CEOs at these companies are the same ones gripping about American's losing their edge, while they send the high-end jobs elsewhere.

Heh. (0)

Anonymous Coward | more than 6 years ago | (#22776532)

Just try and tell that one to Bill Gates.

Re:Heh. (0, Offtopic)

WindBourne (631190) | more than 6 years ago | (#22776608)

While MS is expanding, the question is how much of their research did the United States Gov (or any other western nation) fund? Well, Gates has stolen a lot of tech, but he steals from all over. In the end, the answer is that little is paid for by American or EU nations. But on the hardware side, that is a very different matter. We fund a great deal of HP, Suns, Intels', AMDs, etc. work. Then we allow the real money to go overseas. It is bad enough that boards are built over there. But now chips are going. Hell, if we are going to do this, then I think that we ought to just ship our Tank, Ships, lasers and even nuke building on over to China. Heck, we could go even lower costs and ship all that to Iran or even North Korea.

Re:I wonder where it will be built at? (4, Insightful)

Bob-taro (996889) | more than 6 years ago | (#22776624)

It seems like all of American know-how goes into designing things like this, then companies move the jobs overseas ...

The researchers are: "graduate students Yogesh Ramadass, Naveen Verma, and Joyce Kwong, along with Professor Anantha Chandrakasan". While they may very well all be U.S. citizens, it makes me want to ask for a precise definition of "American know-how".

Re:I wonder where it will be built at? (1)

jcnnghm (538570) | more than 6 years ago | (#22776750)

I take it you haven't been inside an engineering college lately.

Re:I wonder where it will be built at? (4, Funny)

Bob-taro (996889) | more than 6 years ago | (#22777072)

I take it you haven't been inside an engineering college lately.

Even when I was in engineering school, the majority of graduate students were foreign. I forget where, but I once read a quote that went something like this: "American universities are the best in the world. In fact, they are so good that American high school graduates can't compete in them".

Re:I wonder where it will be built at? (2, Insightful)

AvitarX (172628) | more than 6 years ago | (#22776780)

The know how to get them all into one institution doing work for one of our companies at one of our schools has to count for something.

Re:I wonder where it will be built at? (1)

WindBourne (631190) | more than 6 years ago | (#22776804)

Who paid for it? Who exactly is Anantha Chandrakasan? Trained at Berkley and MIT. All the current work took place at MIT in association with a DARPA grant. Yeah, I would say that is America know-how as well as funding.

Re:I wonder where it will be built at? (0)

Anonymous Coward | more than 6 years ago | (#22776870)

Yah, no Indian parent would never name their US born kids with Indian names. Only guys named Steve or Bob qualify as American.

So do 4th generation Americans named Mario or Jose qualify as American?

Re:I wonder where it will be built at? (4, Funny)

gnick (1211984) | more than 6 years ago | (#22777076)

The researchers are: "graduate students Yogesh Ramadass, Naveen Verma, and Joyce Kwong, along with Professor Anantha Chandrakasan". While they may very well all be U.S. citizens, it makes me want to ask for a precise definition of "American know-how".
You were expecting to see "graduate students Geronimo, Running Bear, and Pocahontas, along with Professor Hithawea"?

Re:I wonder where it will be built at? (0)

Anonymous Coward | more than 6 years ago | (#22777394)

You worry about China's military build up and how it could damage the US? Maybe you should focus a little closer to home. How much is the Iraq war costing? What about world oil prices? And consider the dollar figure one might put on the damage done to the US international reputation?

Lacking details (1)

vsage3 (718267) | more than 6 years ago | (#22776526)

I am very interested to see how they managed to reliably demonstrate on/off states of individual transistors at the 0.3V level given that standard logic families use between 1-5V for individual transistors. Of course the article wouldn't have these details considering the article was entitled "Ten times more energy-efficient microchip recharges itself". I suppose whoever wrote the article drew that conclusion from the CONJECTURE posed in MIT's press release.

This is simply an experiment in voltage scaling! (4, Informative)

Terje Mathisen (128806) | more than 6 years ago | (#22776560)

As the MIT report states, the key was to make the chip operate at 0.3 V instead of ca 1.0V

Since power usage is (roughly!) proportional to voltage squared, getting the chip to run at less than one third the usual voltage will indeed give an order of magnitude reduction in power usage.

From the report:

One of the biggest problems the team had to overcome was the variability that occurs in typical chip manufacturing. At lower voltage levels, variations and imperfections in the silicon chip become more problematic. "Designing the chip to minimize its vulnerability to such variations is a big part of our strategy," Chandrakasan says.
I.e. current state of the art transistors does not work reliably at such voltage levels, I'm guessing that they have to give up significant parts of the theoretical power reduction in order to make it work at all.

Terje

Re:This is simply an experiment in voltage scaling (1)

avandesande (143899) | more than 6 years ago | (#22776742)

Without RTF I would guess that you could run every voltage as needed on a CPU instead of a single voltage. The MMX processor would need a higher voltage than the pipeline units (just making an example for illustration)

Perhaps memory chips may hold data at a much lower voltage and only need a boost during a write operation.

Re:This is simply an experiment in voltage scaling (1)

MemoryAid (675811) | more than 6 years ago | (#22777316)

So it sounds like "10 times the efficiency" means 1/10 the power. I read the article specifically to see how they defined a tenfold increase in efficiency. I imagined that an increase from, say, 9% to 90% was not reasonable to expect. Maybe it was energy converted to waste heat that was reduced.

Anyway, I didn't find an explanation in the article. So what is a theoretical 100% efficiency with respect to logic circuits? Every electron turned into a bit of information? Every pair of electrons? It seems impractical to define efficiency in this case. For instance, I don't think coulombs per megaflop could be construed as efficiency, per se. If there was a standard out there, maybe a theoretical minimum number of electrons to perform a logical operation on a standard circuit, we could compare other circuits with that...

On the other hand, I could just cut journalists some slack when they use technical terms in a mass-media context. Now where did I leave my soapbox?

Re:This is simply an experiment in voltage scaling (1)

TigerNut (718742) | more than 6 years ago | (#22778452)

Simple: You assume that every bit of power going in is wasted. Therefore, there is no such thing as absolute "conversion efficiency", but using 10 times less power does make you 10 times as efficient, since you get the same bits coming out (assuming the same program and input) for 1/10 the watts going in.

P proportional to V^3 for chips, not V^2 (1)

AySz88 (1151141) | more than 6 years ago | (#22778208)

Since power usage is (roughly!) proportional to voltage squared
Actually, max clock speed also goes up/down proportionally to voltage. If we make the reasonable assumption that you clock the chip as high as it allows at a certain voltage, then power grows as the voltage cubed.

Body Heat powered....Brain Implants.... (1)

jameskojiro (705701) | more than 6 years ago | (#22776662)

Would a body heat powered computer in your head have enough processing power to serve the same function as a "Cyberbrain"?

If they are in the head, the head is pretty warm, in fact it is one of the warmest areas of the body.

I think that powering protectics, at least the control systems via body heat would simplfy the power requirements for prosthetics.

Re:Body Heat powered....Brain Implants.... (0)

Anonymous Coward | more than 6 years ago | (#22776722)

Can I then overclock my brain?

Will people then bow to me as their new brain overclocked overlord?

Re:Body Heat powered....Brain Implants.... (1, Informative)

Anonymous Coward | more than 6 years ago | (#22776834)

What kills me is that people talk about "body heat powered" and "implantable". You need a temperature differential to generate power - I suspect that there is the same differential inside the body as there is in a glass of room-temperature water. You'd need an external radiator (like implants in your ears), which isn't nearly as attractive as something that just goes in your body.

Re:Body Heat powered....Brain Implants.... (2, Funny)

jameskojiro (705701) | more than 6 years ago | (#22776930)

Cute anime style chobits type ear implants????

Re:Body Heat powered....Brain Implants.... (0)

Anonymous Coward | more than 6 years ago | (#22777484)

Well, running on the difference between your inner head and your earlobe or something isn't such a bad idea. And if you don't want to mess with your head, I hear your testicles are always kept a few degrees below body temperature...

Re:Body Heat powered....Brain Implants.... (1)

Kattspya (994189) | more than 6 years ago | (#22779250)

I don't care if it looks attractive or not I just want something inside me.

Re:Body Heat powered....Brain Implants.... (1)

peragrin (659227) | more than 6 years ago | (#22777026)

I have actually wondered if an implanted version of a bluetooth headset would be possible if one could provide power from the heat transfer of the head.

Tap the skin behind your ear, turns the unit on for talking. a thin hollow cable to the front of the ear for listening, and a jawbone transducer for talking.

It would look a lot better than existing headsets(since everything is subdermal/cranial) but you would look even crazier talking in public without anything on your head. Also it couldn't be stolen or taken away from you.

Re:Body Heat powered....Brain Implants.... (1)

crgrace (220738) | more than 6 years ago | (#22777414)

I imagine you're joking... but a Bluetooth transceiver requires far more power than could be harvested from heat transfer of the head. Sorry.

It's from TI, and points from the article. (0)

Anonymous Coward | more than 6 years ago | (#22776714)

It is from TI, and the article mentions a variety of uses, so it is most likely going to be a DSP, which is what TI is very good at making. The article also mentions that typical chips use about 1 watt right now, so this is certainly not a large CPU unless they've been working on some AMAZING stuff with sun lately on the Sparcs.

So yeah, small little DSPs, running on a tiny amount of power -- pretty neat stuff with a wide variety of applications! (but not your CPU or heavy graphics work)

First Step to Matrix (0)

Anonymous Coward | more than 6 years ago | (#22776752)

"The power consumption in the new chip is so low that devices using them may even be able to be recharged by human body heat."

That's what we be now - human batteries. Just wait until AI gets ahold of this new source of energy...

wouldn't you want the voltage to be HIGHER? (1)

wattrlz (1162603) | more than 6 years ago | (#22776920)

If memory serves power dissipation has a formula on the order of I2R, I being current in amps and R being resistance in ohms. So, if you had a chip that ran on 1,000 volts at 30ma instead of the usual 1 volt and ~30 amps wouldn't it be just as efficient as a 0.3volt chip running at (and I'm guessing here because tfa doesn't mention current) 1 amp or maybe even less?

Re:wouldn't you want the voltage to be HIGHER? (1)

smolloy (1250188) | more than 6 years ago | (#22777032)

Yes, if the current and voltage scale in a way governed by Ohm's Law. However, since power is also equivalent to V^2/R, if you can build a chip that can operate at lower voltage AND the same current, then you win by the square of the voltage.

Re:wouldn't you want the voltage to be HIGHER? (1)

Overzeetop (214511) | more than 6 years ago | (#22777050)

I believe current is constant in this case, so that the power decreases with the square of V. By reducing the voltage to 1/3.3 the typcial usage, a theoretical power decrease to (1/3.3)^2 or 1/10 is achieved. Again , in theory, and provided that they can get the transistors to work at that voltage. Then again, IANAEE, so I wouldn't take that as genuine truth.

Re:wouldn't you want the voltage to be HIGHER? (2, Interesting)

LehiNephi (695428) | more than 6 years ago | (#22777086)

You're thinking about a physics-land purely resistive circuit, where we can arbitrarily control the resistance. Unfortunately, that's not quite the case with very-much-minified microprocessors. We can't arbitrarily force a couple hundred million transistors to use less current. And at the same time, transistors can be designed so that they don't require 1000V to operate.

Every transistor leaks current to some extent. And as those transistors get smaller, that amount of leakage likes to get bigger, because the "off" resistance of the transistor decreases. There has been an awful lot of research into how to reduce the leakage, with some remarkable successes, but the problem is becoming harder as the features shrink. So in order to limit the amount of leakage, the manufacturers are working to lower the voltage required for the transistors to operate. A lower voltage across that same off resistance means less current leaking (V=IR), and so less power is lost. The fact that the power is a function of the square of the voltage (or current) means that even a relatively small decrease in the voltage means a significant power savings.

Re:wouldn't you want the voltage to be HIGHER? (2, Interesting)

vsage3 (718267) | more than 6 years ago | (#22777438)

You're assuming R is independent of applied voltage, which is not true for any transistor. Resistance is a derived quantity that can be (for example) formulated in terms of the ratio of resulting current from an applied voltage. Ultimately, electronic devices require a certain current to operate, so it's not as simple as minimizing power by arbitrarily scaling down current. If you cannot supply enough current to a system, transistors may not have enough juice to produce those 1's and 0's quickly enough, causing unreliable operation. A more accurate way to look at the problem is that if you wish for a device to operate and it requires a given current, if you can find a way to deliver this current at a lower voltage then you will require less power to run the device.

Probably not. (1)

jd (1658) | more than 6 years ago | (#22778322)

Well, not if you wanted any kind of performance. A very crude model of electronics is to think of voltage as the difference in the amount of charge between two points and the current as the rate at which units of charge are being shovelled through the system. Think of it as lumps of coal and the size of shovel being used. Ten times the voltage with one tenth the current would mean it would take a hundred times as long to do the shovelling. Smaller voltages work with smaller currents, if the two are comparable, because the reductions cancel out.

This method, incidentally, has been five years away for several years. I think that other methods of reducing power output on chips (Gallium Arsonide, asychronous solid-state circuits, reduced resistivity in the interconnects) and on computers overall (better caching on drives, smarter offloading, etc) will prove to be the way to go.

Interesting but what about variability? (5, Interesting)

quo_vadis (889902) | more than 6 years ago | (#22776962)

Their work is definitely interesting, but I think some important questions remain unanswered, the main among them is the tradeoff between correctness of operation vs. performance because of variability. There is a paper in ISLPD 2006 which shows that for a 65nm circuit to operate at 0.3 V, the clock period must be scaled up by a factor of at least 230% to compensate for variability related issues. Additionally, there is a huge problem as far as tool support goes. This is not just mix-and-match style design. In order for this to have widespread use, it needs to work well in the EDA tool workflow. This means that libraries (and to some extent transistors) need to be characterized well at the subthreshold operating voltages. This causes a catch-22 situation. In order to design something using this subthreshold voltage technology, you need good transistor models, but the fabs have no interest in providing these models unless there is large customer demand. It is pretty expensive to get good models. The way this works is most fabs actually create transistors/gates at the given feature size, characterize them, including parameters for variation/process variability and give these to their customers, who design their chips based on these simulations. The reason these are so important is that for synchronous circuits, you have to base the design of the clock scheme on the worst/average case delay, and this you can get only by doing complete (usually Monte Carlo based) simulation of the chip using the transistor models that fab gives you. If you base the parameters solely on simulation based tools, you ignore all sorts of effects in the real world, causing a massive drop in yield(i.e. working chips made by fab).

See You In 5 Years (1)

ThreeGigs (239452) | more than 6 years ago | (#22777132)

Slightly O.T., but can I petition for a reusable "seeyouinfiveyears" tag that gets tacked onto any article where the technology will be out of the lab and in the factory in 'five years'?

Because 5 years from now, I'd really love to quickly and easily see just how accurate or inaccurate that industry standard five-year prediction really is.

Re:See You In 5 Years (1)

davburns (49244) | more than 6 years ago | (#22779234)

Well, some things are "expected" to take 3 or 7 years. So how about just tagging this 2013. If in 2012, there's an article predicting that Duke Nukem Forever will come out next year, then we can tag that 2013, too. Then at the end of 2013, we can check that tag, and see what was supposed to happen that year, and compare that to the reality.

I could use those converters. (3, Interesting)

jcr (53032) | more than 6 years ago | (#22777692)

If they have a substantially better DC-DC conversion technology, that's worth a lot of money to a lot of people already.

-jcr

Is my logic flawed? (0)

Anonymous Coward | more than 6 years ago | (#22777932)

If efficiency is a ratio of input to output does that mean that our current day microchips are less than 10% efficient? 100% efficiency means there is no heat dissipation and greater than 100% would mean that energy is being created from nothing.

Power more epensive (1)

Heembo (916647) | more than 6 years ago | (#22778246)

This is good cause my power bill will be 10x todays by the time this new chip goes live.

ten times more efficient? (0)

Anonymous Coward | more than 6 years ago | (#22778558)

So if the existing design is 1x efficient, does that mean the new chips are -9x the power of the current chip, meaning that you actually get energy from using them? They should put those into the priuses.
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?