Intel Developing Ultra-Low Power Chips

ErikPeterson wrote to mention a C|Net article discussing Intel's development of low-power chips for mobile applications. From the article: "The chipmaking giant announced on Monday a new technique that it said could help cut back on wasted battery power in cell phones and mobile devices by as much as 1,000 times current levels. Active computing accounts for only half the power Intel processors use. The other half is gobbled up by a leakage current in transistors that exists when a machine is in a low-level sleep state, Intel said. The new version of the company's 65-nanometer wafer-making process, internally known as P1265, is better than Intel's current process at helping prevent the extra power from being sapped from the battery, the chipmaker said. "

Comments

1/1000th?

[biryokumaru]

Random quotes:
"1,000 times current levels."
"The other half [of the energy] is gobbled up by a leakage current in transistors"
"designed to consume a tenth of the power"
"about a tenth the demand"

"About two years ago, the Intel process and development groups decided to find out if they could expand the space or the scope that 65-nanometer technology could serve and make adjustments so it could make a chip with extra-low leakage."

Um, so, wait, making chips with extra low transistor leakage, where leakage is only half of the chips power consumption, can result in one tenth power consumption? And even, 1/1000th what some chips use? 1/1000th the power consumption of what, a penny in the circuit breaker?

Re:1/1000th?

[merreborn]

It's not that confusing. A current intel chip spends half it's power consumption on computing, and wastes the other half. This new process reduces that waste to 1/1000th of what it was -- if a chip used to consume 2 watts, 1 on computing, 1 on waste, now it will consume only 1.001 watts, 1 on computing, .001 on waste. The "designed to consume a tenth of the power" is about a completely unrelated processor: the next generation of Pentium M is supposed to consume 1/10th the power it currently conusmes. Score one for reading comprehension.

Re:1/1000th?

[Gactaculon]

So, this power "used for computing". Where does it go?

Re:1/1000th?

[John Hasler]

The only power that could reasonably be said to be "used for computing" is that required for the concomitant entropy change. That is a miniscule fraction of the power consumed by any cpu.

Re:1/1000th?

[merreborn]

So, this power "used for computing". Where does it go?

Ah, I finally understand your misunderstanding. When they say half the power goes to active computing and half to waste currently, they mean half of the power consumption occurs while your CPU is crunching numbers, and half while it's just idling.

In both cases, power is dissipated as heat. All they've done is something along the lines of turning the processor off when it's not actually crunching numbers.

They haven't magically done away with resistance, or anything like that.

Re:1/1000th?

[Suidae]

Into heat as the charge is dumped to ground, generally. Basicly you charge up a cap to indicate a logical 1 (or zero, if you prefer). When you want to make it a zero, you dump the charge to ground and you lose the energy, increasing the local entropy.

There are designs and a few working prototypes that recycle some of that energy, but they are more complex than regular chips. Basicly the idea is that in a given processor you'll have a bunch of gates turning on and off at any given time, so you can save some power by dumping charges from gates going from 1 to 0 to gates going from 0 to 1. Its really a heck of a lot more complicated than that though.

The field is called reversable computingand has a lot of potential to reduce power requirements of logic devices.

Re:1/1000th?

[TigerNut]

In a CMOS chip, the power consumption is a function of two major items: The leakage current, and the switching current. The leakage current is a function of the operating voltage and the device geometry size. Going smaller and lower voltage has increased the leakage current to the point where it's a roadblock to further development, so Intel is now apparently addressing it... that's good. The switching current comes about because the CMOS logic state for any node is controlled by either a transistor connecting the node to the supply rail, or a different transistor connecting that node to the ground rail. When the node is switched from one rail to the other, there is a brief period where both transistors partially conduct, and the current goes up dramatically. Hence, the more switching that goes on, the more rapidly, the more current is used, and that's the "power used for computing". The switching current is reduced by, among other things, lowering the operating voltage, which puts it at odds with the reduction of leakage current.

Re:1/1000th?

[carsamba]

It'll run with your blood sugar.

Re:1/1000th?

[Suidae]

Here's another link, a paper about clock-powered CMOS. Most of the university research I've seen on this kind of stuff is this sort of clock-powered CMOS.

Re:1/1000th?

[Gactaculon]

Well, kudos for the informative reply I guess, but the truth is that I already knew most of this. That all sounds to me like waste due to various inefficiencies in the CMOS device, not power that is necessarily required for any type of computation. Thus the talk about "power used for computation" wasn't making any sense to me. The OP clarified that they were talking about idle power vs. active power, which makes things much more sensible.

Re:1/1000th?

[InvalidError]

An idle P4 wastes 10-15W of static power (maintaining its current state, like when the computer is in standby/sleep mode... this is why we have suspend-to-RAM and hibernate), 15-30W in clock distribution and uses 30-60W more for switching transistors while doing useful work.

Since a CPU is not operating at full-speed and full-load 100% of the time, reducing the average quiescent+clock power by 10X could already extend battery life by a substantial amount - how many people run SETI (or comparable non-essential extensive computational load) on their laptops while operating on battery power?

So, if static power and idle clock power are reduced to practically nothing, even if that power accounts for only half of the chip's budget it can decrease the CPU's average power by 10X, assuming the CPU spends ~90% of its time idling on average.

BTW, if you look at typical battery discharge curves, you will see that the effective AH rating depends on load current... so a 50% reduction in system power would come with a ~10% bonus in usable battery capacity. (Batteries are usually rated for 20H discharge and 12AH batteries typically have an effective rating of 7-9AH when drained at ~50A.)

Re:1/1000th?

[JourneyExpertApe]

So where did they get this 1/2 figure? Wouldn't this depend strongly on how you're using your computer? If you're playing a FPS, it would be more like > 95% used for active computing and 5% wasted. Maybe that 1/2 figure is for very light usage like loading a small web page every 5 minutes.

Re:1/1000th?

[Anonymous Coward]

Hmm, would research into superconducting semiconductors be feasible? that might be an interesting career choice.

Re:1/1000th?

[Anonymous Coward]

Um, so, wait, making chips with extra low transistor leakage, where leakage is only half of the chips power consumption, can result in one tenth power consumption? And even, 1/1000th what some chips use? 1/1000th the power consumption of what, a penny in the circuit breaker?

More than a penny, a pretty penny. Your average P4 burns like a light bulb not including monitor and hard drive. And anyone knows, an easy bake oven is powered by a light bulb.

So with the next version we can't bake cakes... guess we will have to settle for toast. Oh well.

Re:1/1000th?

[Profane MuthaFucka]

It's not related to processor usage, it related to representation of zeros and ones. The ones will still take just as much power. The zeroes won't take as much. You see, leakage current is the current that flows through transistors that are turned off. They are making those transistors NOT waste the energy that they leak.

Re:1/1000th?

[slazar]

So there is a physics entropy term for when a transistor flips a bit? Damn I am in the wrong field! :)

Re:1/1000th?

[Taladar]

And how do they know how many bits are ones and how many are zeroes? I don't think the numbers are exactly equal at any given time.

Re:1/1000th?

[mtibbitts]

It would be nice to see Moore's Law unstalled. Industry makes faster and faster chips..but they don't seem to be getting out to consumers. By the simplest of measures my 3 year old 2.4ghz p4 should be replaced by a 9.6 ghz monster today. It seems the real push over the last several years has been towards battery life and power consumption, rather than raw power. this of course makes sense with laptops making up more than 50% of new consumer computers being sold. This sounds like a light at the end of the tunnel. Martin Tibbitts

Re:1/1000th?

[Profane MuthaFucka]

You don't know, but there's a 0 and there's a 1. Without a very expensive and detailed study across multiple systems involving complicated statistics and intensive real-world testing, you'll never know.

Besides that's too complicated for marketing people anyway. They just say 50% because it's intuitive. It's probably close.

Re:1/1000th?

[Rambo]

When the node is switched from one rail to the other, there is a brief period where both transistors partially conduct, and the current goes up dramatically...

Ah, no. What you're referring to is called shoot-through and is more of a problem in power switching circuits (half/full-bridge/etc.). Shoot-through in this type of circuit can potentially cause destruction of the transistors due to the large amount of current passed by typical low-resistance MOSFETs tied to each supply rail.
The switching losses in a chip are related to the capacitance of the gates and interconnect-- each time a transition from one state to another occurs energy is used, either by charging the gate/wire up or discharging it to ground. So while the power usage scales with the frequency of switching, I don't believe it has much to do with simultaneous gate drive of transistors in a "totem pole" configuration.

Re:1/1000th?

[TigerNut]

I was remiss in not mentioning the gate capacitance, since that is the major contributor to the dynamic power consumption in a large CPU, but 'shoot-through' is a significant factor in any high speed CMOS design, even when the process has been tweaked to minimize its occurrence. As an example, the TI MSP430 has a feature where the I/Os that can be configured as either analog or digital inputs, can have the digital inputs disconnected in order to eliminate the extra power consumption that would occur due to having an analog signal near mid-rail.

Re:1/1000th?

[JourneyExpertApe]

If that's true, the the article is seriously misleading:

"Active computing accounts for only half the power Intel processors use. The other half is gobbled up by a leakage current in transistors that exists when a machine is in a low-level sleep state, Intel said."

I've never heard of the operation x = 1 described as "active computing" and x = 0 as "a low-level sleep state". What gives?

Re:1/1000th?

[Profane MuthaFucka]

Remember back in college when you had to interact with a journalism student? Were you struck by their superior intellect and ability to clearly communicate complex ideas about things which they knew nothing about?

I wouldn't be surprised to learn that the quote attributed to "Intel" was hopelessly munged beyond any resemblance to what "Intel" actually said.

Hope that helps.

Re:1/1000th?

[Anonymous Coward]

So you don't do silicon. The scoop is all transistors leak current. As design rules shrink, the leakage current goes up, whether the switch is at one or zero, the leakage current is there. Whether the transistor is actively changing states or setting at one state, the leakage current is there.

What they are saying is, at the current ultra fine design rules, leakage current accounts for 50% of all the power used by the chip. Back when chip features where larger than 200nm, leakage current was insignificant, but at 90, then 65nm, the insulation layers have gotten so thin that the leakage current is a big issue. Intel apparently has come up with a way to reduce the leakage at these small feature sizes.

A very good thing, and very big news.

By the way, all the semiconductor companies are dealing with this same issue, it is not just an Intel problem.

Re:It was about time!

[AKAImBatman]

It was about time that Intel started worrying about their chips power consumption and heating.

You think they haven't been? Ever hear of the Centrino architecture? It finally knocked Apple off the top battery performer pedastal a year or two ago.

What you're thinking of are their high end Pentium IV chips, which are quickly approaching the per-centimeter thermal dissapation rates of a nuclear powerplant. (I say as the fans on my Dell case spin up quite loudly...)

Re:It was about time!

[Anonymous Coward]

Somehow I suspect this knowledge has something to do with Jobs's decision to go with Intel processors.

Yeah Right

[Anonymous Coward]

I'll believe it when I see it.

Re:It was about time!

[Anonymous Coward]

My P-M laptop would disagree with you.

Power???

[digital-madman]

How about reducing the heat first?? Or does Intel want to enter the space heater market?

-Digital Madman

Re:Power???

[GigsVT]

Power dissipated=heat. Hint: Both are measured in watts!

Re:Power???

[Anonymous Coward]

Ok, Mr Douche Pants, the heat comes from current. Reducing the leaked current reduces the heat.

Re:Power???

[Rakishi]

Yes the magic heat which has absolutely nothing to do with the power consumption. Its not like the power consumed is converted to heat or anything like that...

I'm sure they'll be worrying about it as soon as they find where that darn magic is coming from.

Re:Power???

[failure-man]

Thermo anyone? Every watt you put in that chip goes somewhere. What do you think most of it leaves as?
 
Less power == Less heat.

Re:Power???

[Anonymous Coward]

Waste energy = heat.

Jesus...

Re:Power???

[Anonymous Coward]

Why thank you, i bought them from the GAP ;)

Re:Power???

[dusanv]

No, heat (or energy) is measured in J (Joules) or BTUs. Power is measured in W (Watts). Power = Energy / Time.

Re:Power???

[Monkelectric]

Allow me to politely correct you. Joules is a discreet measurement of energy. 0.2389 Calories, or energy required to lift 1kg 0.1 meters.

A watt is a joule/second. Processors do not emit energy discretly, but rather over time, therefore the appropriate measurement is a rate, thus, heat dissipation is measured in watts.

Joule is only an appropriate measurement for discreet things ... such as, "the chemical reaction consumed 30 joules of energy."

A good analogy would be, how many miles of gas did you use to goto work? 10 gallons ... How much fuel does your car use? 10 miles per gallon, which again, is a RATE. The answer to "how much fuel does you car use?" is not 10 gallons. Its 10 miles per gallon.

In the same sense, a processor doesn't dissipate 30 joules of energy, it dissipates 30 joules per second, and the word for "joules per second" is WATT :)

Re:Power???

[GigsVT]

it dissipates 30 joules per second, and the word for "joules per second" is WATT :)

Yes of course, I meant heat dissipated (or rate of heat dissipation if that satisifes your nitpicking) and power consumed are both measured in watts, because they are the same thing in something that doesn't do work like a CPU.

Re:Power???

[GigsVT]

What do you think most of it leaves as?

s/most/all

Unless your case is plastic and leaks a lot of RF energy or something.

Re:Power???

[failure-man]

There is current going through the traces in the data busses, so technically not all of it leaves as heat.

Re:Power???

[lee1026]

energy required to lift 1kg 0.1 meters = 0.98 J not 1 J

Re:Power???

[GigsVT]

Yeah you are right there. That does in turn eventually end up as heat too though.

Re:Power???

[dj245]

Allow me to politely correct you. Joules is a discreet measurement of energy. 0.2389 Calories, or energy required to lift 1kg 0.1 meters.

Allow me to politely correct you. A calorie is a a discrete measure of energy. A Calorie (big C) is also known as a "food Calorie" and is 1000 calories. To avoid confusion calorie the base unit is always written with a lower case c. .0002389 Calories (big c) are required to lift 1kg 0.1 meters.

Re:Power???

[Monkelectric]

I think its a universl rule that its impossible to be 100% correct when correcting someone else. I tried REALLY hard to keep it together, but missed that one... thanks :)

Re:Power???

[glitch23]

Joule is only an appropriate measurement for discreet things ... such as, "the chemical reaction consumed 30 joules of energy."

Sssshhh, don't tell everyone.

Re:Power???

[Anonymous Coward]

A recent episode for CSI had someone state that two thirds of the power used in a server room was converted to heat.

The remaining third leaves as:
- Laser beams (fibre optics).
- Moving air from the AC.
- Sound waves from the fans/drives.
- Ones written to magnetic tape which is then removed from the room.
- Energy leeches in another dimension.

/K

Re:Power???

[ErroneousBee]

10 miles per gallon isnt a rate, its an environmental crime!

Re:Power???

[Monkelectric]

Yes and no. In reality, Yes. but the "standard" gravatational constant at which a Joule is calculated at 10m/s ... So technically, no. I didn't have time to write a physics textbook for my original post :)

Re:Power???

[dusanv]

I was just pointing out to the grand parent that power and heat aren't the same. There is noting to correct so relax please...

Re:Power???

[japhmi]

I'm sure they'll be worrying about it as soon as they find where that darn magic is coming from.

It comes from the Magic Smoke of course!

I know an easy way to reduce battery usage..

[Anonymous Coward]

Make a cell phone a cell phone and keep it at just that.

Re:I know an easy way to reduce battery usage..

[MikeFM]

Or just hold a rock to your head and talk into it as a little birdy memorizes your every word before flying off to tell Wilma.

Re:I know an easy way to reduce battery usage..

[Chosen Reject]

I have an LG VX3200 or some such. I love it. It is a phone. It has a contact list and a calculator and that's about it. But, alas, I think it's discontinued. Here's one link.

Alright

[42Penguins]

Just keep them out of my graphing calculator.

I have enough problems with math on my own, without the "pentium problem" giving me bad numbers.

(yeah, yeah, times have changed, but have they REALLY changed?)

Apple

[LennyDotCom]

I'm sure this was a big part of Steves decision to switch to Intel.

Re:Apple

[karnal]

Durrr. This has only been speculated on for the past I don't know how many weeks.

Bad Karma Whore. Bad.

Re:Apple

[Wesley Felter]

Apple doesn't use XScale.

Re:Apple

[Jozer99]

The what runs OS "X" on the "X" Serve? You gotta admit it makes some sense. ;)

I'll bet Apple knew about this...

[PapayaSF]

...when they decided to switch to Intel. When the switch was announced, my question was: "Hmmm, I wonder what Apple knows about Intel's plans that they can't or won't talk about?" This certainly looks like something that would fit with Apple's future plans regarding iPods and other mobile devices.

Re:I'll bet Apple knew about this...

[NETHED]

NEWTON part 2??!?!

iNewton?

Now that would turn some heads.

Re:I'll bet Apple knew about this...

[soft_guy]

This certainly looks like something that would fit with Apple's future plans regarding iPods and other mobile devices.

Yeah. They have plans to keep making mobile devices and would like them to run as long as possible on a single charge. Thus, pleasing their customers.

Re:I'll bet Apple knew about this...

[forkazoo]

Huh? They said on the day of the announcement that Intel's roadmap had the best MIPS/watt. We've known for a while that they realised the P4 was embarassing them, and that they needed a better direction.

Re:Reduce power should reduce heat

[technoextreme]

Power???
How about reducing the heat first?? Or does Intel want to enter the space heater market?

You reduce the power you reduce the heat.

Misleading

[fembots]

Ultra-Low Power != Ultra Low-Power

Re:Misleading

[Anonymous Coward]

Explain how. Give examples. Show your work.

Re:Misleading

[Dahan]

Simple--using the distributive law,

Ultra-Low Power = (Ultra Power) - (Low Power)

whereas

Ultra Low-Power = (Ultra Low) - (Ultra Power)

Leakage?

[kko]

Are these those famous Olestra chips everybody talks about?

Re:Leakage?

[RapmasterT]

I didn't want to let this go by without acknowledging that someoen DID get the joke, and found it both topical and amusing.

Re:Leakage?

[msdschris]

Oh yeah, I also "got" it... From Orville Reddenbacher Smart Pop. Worst experience EVER.

To summarize the story...

[slashname3]

So the article basicly says they plan to make chips that don't suck so much.....



power that is. :)

Re:To summarize the story...

[Anonymous Coward]

if you'd left that second line out i'd have modded you funny. a good joke relies partly on the wit of the audience... spoiling your own joke doesn't do you any favors.

Re:To summarize the story...

[Spodlink05]

if you'd left that second line out i'd have modded you funny. a good joke relies partly on the wit of the audience... spoiling your own joke doesn't do you any favors.

You haven't watched many US "comedies", have you?

;)

Re:To summarize the story...

[Anonymous Coward]

It does when you're posting a joke in a place that will get modded (-1, Flamebait) or (-1, Troll) when a mod doesn't get the joke. Because of this, even with a (+5, Funny), you'll still end up with negative Karma.

Re:It was about time!

[chez69]

he can't let something like the facts get in the way

Apple Powerbook with lower power consumption?

[lightningrod220]

I would really be interested to see what Apple can do with this. However, they need to make sure that the heat is as minimal as possible. I'm getting tired of even the G3 iBooks getting wayyyy too warm, let alone the G4 'books. We need to get those temperatures down, power consumption down (seems like these new chips will do that), and make the Lithium Ion batteries last longer, so we don't have guys putting "Powerbook batteries last only 18 months!" everywhere.

Re:Apple Powerbook with lower power consumption?

[jo_ham]

The G3 in the iBook (an IBM 750FX or something like that) draws 6 watts at 900Mhz.

That's six.

The heat's not really coming from the CPU in an iBook. In fact, the fan barely comes on - only in the most extreme temperatures (say using the laptop on a soft durface like a sofa or a bed when the air tempertaure is up in the "Baghdad August" range).

Re:Apple Powerbook with lower power consumption?

[MsGeek]

I have found that the hard drive is the bigger "heat-up" culprit in G3 iBooks/PowerBooks.

When I upgraded the hard drive on my Clamshell I purposely picked one known for low running temperature. Ergo, no "hot leg syndrome." Maybe "warm leg" but not hot.

However: the CPU starts becoming an issue with G4 lappies. As it does with even the Pentium M. And a P4 lappie does dual duty as a hotplate. Handy for those living on-campus. ;-)

Re:Apple Powerbook with lower power consumption?

[QuaZar666]

Sorry to burst your bubble hear, but portable's are not designed for laps, are compared to intel notebooks it produces far less heat. Even the G4s produce far less heat than intel boxes, but it still would be nice to see a notebook where the fan never kicks on yet runs at a decent speed (like a 1.67 G4 or a 2.4 P4M or equivalently rated).

Advice

[ackthpt]

Random quotes: "1,000 times current levels."
"The other half [of the energy] is gobbled up by a leakage current in transistors"
"designed to consume a tenth of the power"
"about a tenth the demand"

Re:1/1000th?
Um, so, wait, making chips with extra low transistor leakage, where leakage is only half of the chips power consumption, can result in one tenth power consumption? And even, 1/1000th what some chips use? 1/1000th the power consumption of what, a penny in the circuit breaker?

Don't bother to emphasize your math skillz when applying for a job at Intel

Thanks

[mnemonic_]

I'll bet none of suspected this. I have heard that Apple has also developed an Intel-compatible version of OS X, can you tell us more about this too?

Re:Thanks

[Anonymous Coward]

There is in fact a build of OS X that runs on x86 processors, and it has been leaked. Presently, it only runs on AMD64s or P4s, but otherwise it works like a charm.

Other heat source

[www.sorehands.com]

Power leakage is not the only heatsource. The sarcasm detector emits heat when the sarcasm reaches the limit of the detector.

ULTRA LOW!!!

[Wazukkithemaster]

AMD soon follows up with the patented and trade marked...

MUH MUH MUH MONSTER LOW Low low chips.

I'd stick with low power instead of ultra low its just sounds less like sensationalism (perhaps because it is less like sensationalism)

Re:ULTRA LOW!!!

[cerelib]

I think the naming scheme works. Regular chips in all notebooks. Low-Voltage chips in your ultraportables. And the Ultra-Low-Voltage in you palmtop and other small devices. It breaks up pretty nice.

Low voltage == less heat == higher clock speed?

[grumpyman]

Is that possible? The chase for lower voltage -> less heat -> speed up the clock?

Re:Low voltage == less heat == higher clock speed?

[Crusader7]

If they've determined a more efficient way to run their instructions, yes. There are more ways to make the processor perform better other than just speeding it up. A case study is the various AMD processors that, despite being technically slower, perform better.

Another valid option is to specialize the processor and focus all the resources on a specific task, but I don't think Intel is doing that.

Re:Low voltage == less heat == higher clock speed?

[Anonymous Coward]

clock speeds mean nothing...if for instance an AMD chip can do a similar instruction in half the number of cycles, then the slower speed doesn't matter.

So yeah, I think you're going to see a whole new ISA which will allow for slower speeds or increased parallelism.

Re:Low voltage == less heat == higher clock speed?

[Technician]

Is that possible? The chase for lower voltage -> less heat -> speed up the clock?

Yes. Many processors are limited by heat. Smaller transistors mean smaller connections means less capacatance and shorter connections. Less voltage most times means slower chips. Smaller low voltage parts means less power draw per transistor, faster transistors, and together they mean lower voltage and faster parts. Lower voltage does not make a faster part. Smaller parts using less power means faster parts for the same power in the same space. The overall result is lower voltage, faster part. Don't leave out the smaller parts portion of the triangle. Lowering the voltage alone does not make a faster part.

Re:Low voltage == less heat == higher clock speed?

[Guy Harris]

So yeah, I think you're going to see a whole new ISA

Perhaps you will, but Intel's already got 3 ISA's they're implementing (x86^H^H^HIA-32, IA-64, and ARM with XScale - 4, if you count x86-64, and then there might be some other microcontrollers they have); they might not be interested in doing Yet Another ISA. (That's not what the new "architecture" they announced recently is; it's another internal architecture implementing the x86 and perhaps x86-64 ISA's.)

Re:Low voltage == less heat == higher clock speed?

[grumpyman]

Gosh, wtf is with you moderator(s)? This is a genuine question. I am no chip expert and wonder if this is a 'nice' by-product. Ok, mod this OT. Gees.

About time...

[Anonymous Coward]

Just sold my P4/3GHz to buy a leet Athlon64 because requiring a 550 Watt power supply for stable operation is hot, noisy and sad.

does that come with MMX?

[sl4shd0rk]

Just re-release the P66.

Re:does that come with MMX?

[Slashcrap]

Just re-release the P66.

You mean the original Pentium 66, one of the first Pentiums to be released, right?

The one that had to be recalled, as it had a nasty habit of burning out due to excessive power consumption.

I'd love to believe that you knew that and were being ironic. But this is Slashdot.

Typical Intel Un-Announcement

[dsginter]

This is the typical Intel announcement:

- Zero details
- How much performance penalty?
- What technology is utilized?
- Are we smoking crack?

Nobody knows (even us)! We just like to put out press releases. Read about our next generation architecture. What is it? We haven't a fucking clue!

But we like to issue unannouncements! We're Intel!

Don't think I know that one

[Anonymous Coward]

P-1265? Anyone know the gate address?

Re:Don't think I know that one

[BoydWaters]

P1265 =
12-inch silicon wafer
65-nm process

Re:Don't think I know that one

[Anonymous Coward]

Interesting. Perhaps you could elaborate on why P1264 isn't a 64-nm process? Thanks.

Re:Don't think I know that one

[Anonymous Coward]

well, half right.... it is a 12 inch process but the 65 is just a coincidence. P1265 follows P1264 (another 65 nm process), P1263 and P1262 (90 nm), and P861 and P860 (eight inch wafers on a 130nm process).... I think you get the idea

Good but...

[coopaq]

Everytime I here this I think:

Intel Developing Ultra-Low Power Chips= Intel Developing Ultra-sLow Power Chips

Re:Good but...

[2nd Post!]

Wear are you hereing this?

Besides, isn't it IBM with the Ultra sLow Power Chips?

Re:Good but...

[coopaq]

Haha :)

Now eye hayt ewe.

Re:It was about time!

[rgmoore]

What you're thinking of are their high end Pentium IV chips, which are quickly approaching the per-centimeter thermal dissapation rates of a nuclear powerplant.

It's not just dissipation, either; there's also a problem with power connections. Modern chips operate at low voltages but still consume huge amounts of power, which means that they draw very high currents. Since future chips would supposedly have even lower voltages and even higher powers, their current requirements would get truly outrageous. At some point, the chips would get to the point that they'd need to use their whole surface to conduct in all the current they need, and I've heard that they'd reach that point before they got to the point of being impossible to cool. Cutting power consumption obviously attacks both excessive current and excessive heat simulataneously, so it's the smartest solution.

ObHomer

[sharkey]

I've got to do a little "spring cleaning".

Re:It was about time!

[corvair2k1]

The summary specifically states that this is dealing with cell phones and the like. Although I'm sure that some of the technology is going to come back to the mobile chip division, Intel already had IBM and AMD won with their current offerings.

How does 'half' = 1000 times less

[craznar]

If half of the current power is used by the computing, and half by heat.

1000 times less means 500 times slower ?

Or am I missing something ?

Re:How does 'half' = 1000 times less

[dancpsu]

Okay, a little chip power management 101.

First, chip power can roughly be divided into two components:

1) Switching - When the transistor is going from high to low, or low to high

2) Leakage - When the transistor is "off" but still letting through a little current

Since CMOS was first put into play, two transistors per state have made things as low power as possible. The line between power and ground is controlled by two opposite mosfet trasistors, one that switches high, and the other that switches low. Since they are opposite, one of them is always "off" so that it doesn't allow current through.

*HOWEVER* even with at least one transistor off, a little current always gets through. This is called leakage. While larger transistors only let through a little current, smaller and smaller transistors became leakier and leakier. So while earlier processors had only a little power used for leakage, according to the intel report, this has risen to 50%.

But, you can still make those larger transistors, so you can still prevent the leakage, you just need to have them stop leakage current when you're sure that certain parts of the chip will be "off" for a while. Apparently, intel has found a way to use some architectural method to put these larger transistors in place to reduce leakage current to 1/1000th the amount it would be otherwise.

So now the power can be reduced by nearly the entire amount of the previous leakage current, or nearly the entire 50% that leakage was taking up. In all of this, you shouldn't lose any processing power, because the frequencies are all still very fast, and even though the larger transistors take more time to switch, these will not be the ones performing your actual calculations.

Re:How does 'half' = 1000 times less

[dancpsu]

Guess I should have read the article first. They are changing the transistor fabrication process--not putting in larger gating transistors--to make the leakage current less. The new transistor process has 1/1000th the leakage, but switches at half the rate. Intel thinks it's a good compromise for low-power situations.

Re:How does 'half' = 1000 times less

[Surt]

1000 times less by heat, leaves you with the computing half (plus a marginal fraction of the heating part which they didn't bother to count).

Re:How does 'half' = 1000 times less

[craznar]

"So now the power can be reduced by nearly the entire amount of the previous leakage current, or nearly the entire 50% that leakage was taking up."

Yes - which is half ... where does the other 500 times come in :)

i.e. start at 100%, aiming for 0.1% ... you have us at 50%.

"to reduce leakage current to 1/1000th "

Not exactly the wording that I see :)

BTW - it was meant to be tongue in cheek at Intel ... but lets see where it goes :)

Re:How does 'half' = 1000 times less

[craznar]

Yeah - the wording leaves a bit to be desired.

So they are aiming at 50.1%, not 0.1% of the original.

That's fine then.

Re:How does 'half' = 1000 times less

[19thNervousBreakdown]

Easy, it's a binary fraction with the trailing zeros left on to show the high precision, but they left out the point. Like so:

.1000(base 2) == .50(base 10) == 1/2

Re:How does 'half' = 1000 times less

[MightyMartian]

Guess I should have read the article first.

This is Slashdot. It's not even a requirement that you actually be able to read.

They are changing the transistor fabrication process--not putting in larger gating transistors--to make the leakage current less. The new transistor process has 1/1000th the leakage, but switches at half the rate. Intel thinks it's a good compromise for low-power situations.

I think Intel is finally starting to make sense again. Yes, this is a very good plan.

Re:How does 'half' = 1000 times less

[craznar]

"It's not even a requirement that you actually be able to read."

The ./ head article refers to wasted power .... on an intel that's not exactly the same as wasted heat power :)

Re:How does 'half' = 1000 times less

[default luser]

Actually, this has more to do with Intel pushing a . They expect leakage to be reduced by over 100 times.

Basically, the article is simply parroting what the Intel marketing department has been told by their engineers: that leakage current will be reduced by allmost 1000 times. The Intel marketing department is simply palming it off as some newfangled "ultra low power" process, when it's been on the roadmap the whole time.

That is, if the reality about high-K comes anywhere near Intel's claims. I recall quite clearly Intel skipping out on SOI, and claiming quite calmly to the world that strained silicon would solve all their problems in the move to 90nm. That didn't happen. Ultra-aggressive dynamic power-saving modes and the Pentium M are the only things that saved Intel's 90nm process.

Will be impressive if they actually get results in the same ballpark as their estimates, though.

What The Hell Is That Supposed To Mean?

[John Hasler]

> cut back on ... power ... by as much as 1,000
> times current levels.

"Cut back by 1000 times"? Are these people native speakers of english?

Re:What The Hell Is That Supposed To Mean?

[Nasarius]

Yeah, "1000 times less" and similar expressions are nonsense, but they're commonly understood to mean 1/1000.

You forgot

[Ironsides]

Is that possible? The chase for lower voltage -> less heat -> speed up the clock?

I think you missed something in there.
The chase for lower voltage -> less heat -> speed up the clock -> higher power consumption -> The chase for lower voltage

Other coverage

[CoderJoe]

An article from The Register from this morning, also covering the new process.

Apple *DOES* Use XScale

[Khyber]

Or at least an Intel-licensed derivative of it.

http://news.com.com/Apple+has+Intel+inside--sort+o f/2100-1042_3-5598783.html

Re:Apple *DOES* Use XScale

[Wesley Felter]

Oops, I knew I forgot something. But I don't think Xserve RAID was a very big factor in the Intel switch.

1/2 Active, 1/2 Leakage?

[Israfels]

That's funny, I would have thought that having a giant color screen would have been much more of a burden for the battery life than either of these.

Already Done

[Anonymous Coward]

Easy, just switch off the damn computer and read a book.

Re:It was about time!

[dfghjk]

Apple never made processors. They do make the loud-as-hell Powermac G5's though.

Watt-seconds

[MrYotsuya]

1 Joule = 1 Watt * 1 Second, so wattage is related to waste heat.

Re:It was about time!

[MerlinTheWizard]

You obviously don't know what Intel makes beside desktop Pentium chips... they have some excellent embedded processors. The XScale series, for instance, which is wildly used in PDAs nowadays...

Less leak, slower performance

[whovian]

But the technology does have some trade offs in performance. "Intel's ultra-low power process is a significant part of Intel's strategy to reduce platform power," he said in an interview. "But the transistor performance is lower by a factor of two" compared to the company's high-performance 65-nm process.

Source: www.eetimes.com

Ironic for Intel, no?

Been doing this for a while.

[freidog]

The 90nm process intel is using has a very similar thing, only not done near to the extent their talking about here.

Intel has two sets of transistors for 90nm, high voltage threshold and low voltage threshold.
High VT are fairly power efficient as it is, about 40nA/um leakage, Ion about 31 times greater than Ioff (NMOS)
Low VT (which were used extensivly in Prescott to get it to scale to the 4-5ghz range it was intended for), which are horribly inefficient, with a leakage of about 400nm/um, Ion around 3.5 times Ioff (NMOS)

Seems like this is largely a really really high VT transistor, with a few tweaks to the oxide thickness for good measure.
In any case, it should help out the ultra low power devices to an extent, but won't effect any of intel's 65nm desktop/laptop chips. (save maybe a chipset, but I doubt we'd see a 65nm chipset).

Re:Been doing this for a while.

[chipace]

I want to see mixed VTs on a process. The "controller" core uses the high VTs, and additional cores use low-VTs that leak like a son-of-a-gun, but are held in S3 suspend and only used for peak processing. The key is to remove the power rails so that the low-VTs don't cost you when you aren't using them.

I beleive Cell uses this method, and throttles them with temperature to ensure that the cooling and power supply design is never exceeded with long heavy loads.

Umm... So What?

[bbrack]

So intel is going to build low power dsp and microcontroller type devices on a more efficient version of their current process node

Big Deal

Everyone does this

It would be RETARDED to build a chip designed for a cell phone on the same process node as a chip designed for a server - you tailor your process to help your chips perform their job better, not build chips that were designed with opposite goals in mind all on the same process

When you want ultra-high performance, you will lose some efficiency - the opposite is also true - this is a perfect example of how everything in engineering esp. semiconductors is a trade-off

Currently the company I work for has 3-4 different process flavors per process node, ranging from ultra high performance to ultra high efficiency, and even within the larger process flavor, tweaks are always performed to ensure chips are getting maximum speed/yield possible out of that specific process

Once you get to 65nm, if you don't have some pretty novel ways of reducing leakage, your leakage power alone can easily eat up 75+% of your power budget.

Re:It was about time!

[QuaZar666]

you forgot to mention that the Powermac G5 was the first pre-made computer to use water-cooling (or better yet require water cooling)

Can't Get Better Than This

[Anonymous Coward]

My P4 PC consumes ZERO watts. You heard me. ZERO.

Of course I leave it turned off because otherwise my SLOWEST AMD system would KICK IT'S ASS!

Re:It was about time!

[WMD_88]

The G5 doesn't require water cooling per se, it just needs it to prevent the fans from becoming consistently too loud.

worm hole computing

[Anonymous Coward]

From the article,

"The chipmaking giant [...] said could help cut back on wasted battery power in cell phones and mobile devices by as much as 1,000 times current levels."

1 - 1000 = - 999

"The new chips shall suck power from a worm hole and pump it into your computer circuits," Intel added.

Further Optimization

[Doc Ruby]

If this tech reduces current in one transistor state, "zero" to 0.1% that of "one", from equal current, it can favor one state. Perhaps it's possible to process instructions to ensure that more possible states of the program are representable as zeroes than as ones. If the difference is large, then it could reduce X=2 power to Y=<1.

Don't they already have a couple?

[BestNicksRTaken]

"Intel Developing Ultra-Low Power Chips"

Erm, you mean like the StrongARM or XScale?

Dump NMOS

[quarkscat]

CISC (Complex Instruction Set Chip) processor technology has been replaced with RISC (Reduced Instruction Set Chip) processors, while maintaining backward compatability with the x86 architecture. Manufacturing tolerances have improved quite a bit, with 90nm lithography being replaced with 65nm. NMOS has been Intel's chosen cell technology for more than two decades. It offered lower power per MIPS than ECL technology, which had raw speed matched only by "toaster-level" heat generation.

With smaller die sizes due to improved manufacturing tolerances, and a new focus on really low power designs, perhaps it is time for Intel to drop NMOS technology for the original low power cell design -- 6 transistor CMOS technology.

Re:It was about time!

[QuaZar666]

ok true, but who wants there computer to sound like a rocket ship 90+% of the time?

Re:It was about time!

[WMD_88]

Nobody *wants* it I'm sure, but I've seen plenty of people who don't mind it. PC Gamers mostly.