Actually, leakage power surpassed switching power during the shrink from 130nm ot 90nm as the number one for power consumption.
Your formula is correct, but it now accounts for a much smaler fraction of the total power.
Intel has abandoned high k dielectrics at 65 nm, and gone for air gap, "the best k is no k at all", an extremely expensive process, which is an indicator of how fundamentally extrordinary a problem leakage has become.
Intel's very business model relies on cheap processes, so the move to air gap is telling.
Physics has caught up with no one. Transistors are still getting smaller, but heat is on the rise, as any 2.5 Ghz water cooled G5 owner knows.
Think of it this way: work costs watts.
No matter if you do a given amount of work using a narrow speedracer architecture like P4 or PPC970, Or a wide architecture like G4, Athlon64/Opteron, Itanium or Pentium M, the work done costs watts, and generally the speedracer designs start paying more in work per watt.
The real current limitation is architecture complexity, where no one has a big enough brain to fit more than 150 million or so useful, non-cache transistors into their heads to debug the chip when there is a problem. Bob Colwell, former chief architect for Intel for the Pentium Pro/II/III/4, has spoken and written at length about this.
When he left Intel, there were perhaps 2 people left that could debug the Pentium 4.
Tejas was cancelled for this reason, as it was an even more complex version of the P4, certainly with AMD64 instructions included, possibly some EPIC (Itanium) compatibility, and a sort of SSE4 called at the time TNI or Tejas New Instructions, that were supposed to be the last straw in bringing complete vector processing to the x86 world, which Apple of course calls Altivec.
This complexity limit has caused architecture advancement to virtualy stagnate, while Moore's law marches on. 200 million transistors last year. 400 million in 2005 a billion in 2007. What to do with the transistors? Add more cores, since individual cores can not get any more complex and cache has a limited effect after 1mb, as Itanium and the G4 show. Cache is a poor substitute for a good memory bus, and after 2mb it's all crutches to keep poor architectures competitive with the better architecures out there.
Why the stagnation at 3 Ghz, or more specificly 3.06? Because that is all the northwood architecture could do, and Prescott, its replacement, was starting to hit that complexity limit and was delayed 8 months.
When Prescott arrived, it was hot, almost 175w per cm^2. This was not the process, 90nm, that caused the heat, because the Dothan (Pentium M centrino) was only 27 watts on the same process, and no one could figure out why it was so hot, so Intel got stuck, ramping clockspeed only 533 mhz in two an a half years, after doubling clockspeed to 3.06 from 1.5 in the previous 2 years.
AMD changed horses from 2.0 Ghz Athlon to 2.0 Ghz Athlon 64 and jumped 25% to 100% better perfromance, depending on the benchmark, mostly due to the integrated memory controller, not it's 64 bitness. It would take a 3.2 to 4 Ghz Athlon to match a 2.8 Ghz Athlon 64, and a 4.2 to 5.4 Pentium 4 to match it.
There is a performance race on, and a marketing bullshit race for clockspeed which may or may not mean a processor performs better..
Sounds like you have only been following the marketing bullshit race..
With AMDs hypertransport and integrated northbridge, every processor you add adds another memory bus. It's call NUMA, for non uniform memory architecture, supported in Server 2003, XP Pro since sp2 and Linux since 2.4, perhaps earlier.
NUMA was first used by SGI with their late 90s MIPS machines.
Intel uses a shared bus, with the exactly the limitations you describe, except with their Itanium in 8 way+ configuration.
"The term was coined by René Descartes in 1637 in his La Géométrie and was meant to be derogatory: obviously, such numbers were thought not to exist."
This statement does not give him enough credit.
The word has been misstranslated.
Imaginaire
Of the mind, or
Image-less Invisible Vision less
He described them this way because they could not be plotted using his cartesian coordinate system, not because they obviously didn't exist. He used them, and new better that that.
"Imaginary numbers were so named because no one figured they had real world uses: today, they're taught as a practical matter for electrical engineers to use in their electronics calculations."
That is very wrong.
Gauss, (hmm.. where have I heard that name before) invented imaginagry numbers because in electricity, negative values are as real as positive values, and math regains symetry through imaginary numbers so that we can find the root of -1, a very real value in gee, magnetism perhaps?
And about the name?
Gauss was french, and while imaginary has stuck, the proper translation is image-less. While imaginary has connotations of being of the mind or made up, image-less means invisible, or directly as it was described, as having no image on the "real" number line.
Better might be visible and invisible, instead of real and imaginary. Best is certainly Image-less, as it is a direct translation of what Gauss was referring to.
They are both absolute base systems that share clockspeed, memory and price.
It's not like we are comparing G4s to P4s here.
As for power consumption, had he bothered to actually dive into the whitepapers, he would find that particular Turion at 35 w while the Pentium M is 27.
No mention is made of the Turions available at 25w.
But he does mention the 9 watt Pentium M that runs at half the clockspeed taking it completely out of this class.
And unless you want to be charged as an accessory after the fact or evidence tampering, you will get far, far away from that woman, even if the sex is good.
No, really.
Apparently an important factor is security...
on
IE7 Details Emerge
·
· Score: 1
Wasn't it always?
Every time the same thing: most secure ever, complete security audit, SP2 is all about security, blah blah blah.
Well they certainly know that word in the marketing department.
http://slashdot.org/interviews/02/01/10/1245221. sh tml
All bs.
Makes money?
Iron fisted grip.
Looses money?
Open source, twist the knife in MS's ribs
No better than Apple or SGI.
Sun:
The 15bn dollar market cap company that was able to parlay Java into a 3bn dollar settlement with MS over virtual machine.
1/30 the size of Microsoft, 1/20 the size of their primary competition, IBM.
"More linux desktops" than any company in the world (in china).
Supports the open source community at their convenience. So much for "write once, run anywhere. If it ever was the case with effectively 5 forks in their own code.
Well, BSD is dieing anyway right?
And their java web server runs better in a non-SPARC, non Solaris enviroment anyway.
I Know! We could use "Nucular" Zepplins! They could stay aloft for as long as a submarine can stay under water. They could launch cruise missles and control battlefield information.
We could put squads of HILO jumper Marines on them to hit trouble spots and drop pallets of food on indigenous people to win hearts and minds. We only crushed a few when we did that..
BTW, it's the MiG-25 FOXBAT that can reach those altitudes, not the MiG-29 and, oh yeah, we loose more Americans in LA county each year than we loose in Bagdad.
So, fighting insurgency starts at home and it is just an insurgency. No need to spend any more money there than we do on the LAPD.
Having the joy of participating in bushwarforoil 1.0, I would have to say that it is a good thing.
But call it for what it is.
bushwarforoil 1.0 liberated Kuwait, and the number one consumer of Kuwait oil was japan, getting 90% of their oil from there. At the time, Japan was our number 2 trading partner, after Canada of course.
bushwarforoil 2.0 is for China, which has reaped the benefits of our resolve by being able to increase it's imports. China has of course moved into Japan's position of number 2 trading partner, after Canada.
I personally can't wait for bushwarforoil 3.0 when in 2012, after Jeb wins the presidency, we roll into Iran, in order to secure oil for India, which will have moved into the number 2 position, after Canada of course.
Why don't we ever go to war for Canada?
They have their own oil.
Note to Corporations:
When outsourcing manufacturing and establishing trade with foreign nations, make sure they have their own oil. That way, we don't have to go to war to make sure our trading partners have enough oil.
They are used all over the country to store power in preperation for peak usage times.
If spot market megawatts are normally $400, peak usage megawatts cost $20,000, and a pumped storage dam can produce 1000 megawatts, you can save or make (depending on if you are a buyer or seller) almost 20 million dollars a day.
Slashdot Mirror
Bad ass, for about 6 inches while traveling through the air.
.0000000000000044 calories, our friend beta is what? 800 calories shy of the "10 mile hike" that a good performance consumes.
My performance may not be rated in Megaelectron Volts, but at least I have more than 6 inches.
And at
Actually, leakage power surpassed switching power during the shrink from 130nm ot 90nm as the number one for power consumption.
Your formula is correct, but it now accounts for a much smaler fraction of the total power.
Intel has abandoned high k dielectrics at 65 nm, and gone for air gap, "the best k is no k at all", an extremely expensive process, which is an indicator of how fundamentally extrordinary a problem leakage has become.
Intel's very business model relies on cheap processes, so the move to air gap is telling.
Physics has caught up with no one. Transistors are still getting smaller, but heat is on the rise, as any 2.5 Ghz water cooled G5 owner knows.
Think of it this way: work costs watts.
No matter if you do a given amount of work using a narrow speedracer architecture like P4 or PPC970, Or a wide architecture like G4, Athlon64/Opteron, Itanium or Pentium M, the work done costs watts, and generally the speedracer designs start paying more in work per watt.
The real current limitation is architecture complexity, where no one has a big enough brain to fit more than 150 million or so useful, non-cache transistors into their heads to debug the chip when there is a problem. Bob Colwell, former chief architect for Intel for the Pentium Pro/II/III/4, has spoken and written at length about this.
When he left Intel, there were perhaps 2 people left that could debug the Pentium 4.
Tejas was cancelled for this reason, as it was an even more complex version of the P4, certainly with AMD64 instructions included, possibly some EPIC (Itanium) compatibility, and a sort of SSE4 called at the time TNI or Tejas New Instructions, that were supposed to be the last straw in bringing complete vector processing to the x86 world, which Apple of course calls Altivec.
This complexity limit has caused architecture advancement to virtualy stagnate, while Moore's law marches on. 200 million transistors last year. 400 million in 2005 a billion in 2007. What to do with the transistors? Add more cores, since individual cores can not get any more complex and cache has a limited effect after 1mb, as Itanium and the G4 show. Cache is a poor substitute for a good memory bus, and after 2mb it's all crutches to keep poor architectures competitive with the better architecures out there.
Why the stagnation at 3 Ghz, or more specificly 3.06? Because that is all the northwood architecture could do, and Prescott, its replacement, was starting to hit that complexity limit and was delayed 8 months.
When Prescott arrived, it was hot, almost 175w per cm^2. This was not the process, 90nm, that caused the heat, because the Dothan (Pentium M centrino) was only 27 watts on the same process, and no one could figure out why it was so hot, so Intel got stuck, ramping clockspeed only 533 mhz in two an a half years, after doubling clockspeed to 3.06 from 1.5 in the previous 2 years.
AMD changed horses from 2.0 Ghz Athlon to 2.0 Ghz Athlon 64 and jumped 25% to 100% better perfromance, depending on the benchmark, mostly due to the integrated memory controller, not it's 64 bitness. It would take a 3.2 to 4 Ghz Athlon to match a 2.8 Ghz Athlon 64, and a 4.2 to 5.4 Pentium 4 to match it.
There is a performance race on, and a marketing bullshit race for clockspeed which may or may not mean a processor performs better..
Sounds like you have only been following the marketing bullshit race..
But then, you are an Apple owner.
With AMDs hypertransport and integrated northbridge, every processor you add adds another memory bus. It's call NUMA, for non uniform memory architecture, supported in Server 2003, XP Pro since sp2 and Linux since 2.4, perhaps earlier.
NUMA was first used by SGI with their late 90s MIPS machines.
Intel uses a shared bus, with the exactly the limitations you describe, except with their Itanium in 8 way+ configuration.
It means you need an Intel P4 processor clocked somewhere between 4.2 and 5.4 Ghz, depending on the applications you like to use.
Or you need a Pentium M "centrino" style processor running at the same clock or better, again depending on the application.
You are correct, I'm terrible with names.
"The term was coined by René Descartes in 1637 in his La Géométrie and was meant to be derogatory: obviously, such numbers were thought not to exist."
This statement does not give him enough credit.
The word has been misstranslated.
Imaginaire
Of the mind, or
Image-less
Invisible
Vision less
He described them this way because they could not be plotted using his cartesian coordinate system, not because they obviously didn't exist. He used them, and new better that that.
"Imaginary numbers were so named because no one figured they had real world uses: today, they're taught as a practical matter for electrical engineers to use in their electronics calculations."
That is very wrong.
Gauss, (hmm.. where have I heard that name before) invented imaginagry numbers because in electricity, negative values are as real as positive values, and math regains symetry through imaginary numbers so that we can find the root of -1, a very real value in gee, magnetism perhaps?
And about the name?
Gauss was french, and while imaginary has stuck, the proper translation is image-less. While imaginary has connotations of being of the mind or made up, image-less means invisible, or directly as it was described, as having no image on the "real" number line.
Better might be visible and invisible, instead of real and imaginary. Best is certainly Image-less, as it is a direct translation of what Gauss was referring to.
Look at the system specs yourself.
They are both absolute base systems that share clockspeed, memory and price.
It's not like we are comparing G4s to P4s here.
As for power consumption, had he bothered to actually dive into the whitepapers, he would find that particular Turion at 35 w while the Pentium M is 27.
No mention is made of the Turions available at 25w.
But he does mention the 9 watt Pentium M that runs at half the clockspeed taking it completely out of this class.
Mentions the 35,27 and 9 but not 25.
No, no, that might be a favorable data point.
Might as well be a Fox Spinner.
And unless you want to be charged as an accessory after the fact or evidence tampering, you will get far, far away from that woman, even if the sex is good.
No, really.
Wasn't it always?
Every time the same thing: most secure ever, complete security audit, SP2 is all about security, blah blah blah.
Well they certainly know that word in the marketing department.
Do WOAC radii have a LOC conversion factor?
Pentium M:
http://www.lippert-at.com/index.php?id=21
Best FPU per Clock of any current microarcitecture:
http://www.cpuid.com/PentiumM/index.php
At 2 Ghz it should be to OGG encoding what a GPU is to Graphics Rendering.
- expect to pay laptop prices for it -
How fast do you want to spend?
ISP charges? .mac?
$130/year for the OS upgrade?
That adds up to $500 a year if you have a $20/month ISP
$40 a month -vs- $13
Post a 65 mb file to slashdot without a .torrent?!?!
Comparable to VIA's Eden-N at 2.5w at half the size.
I'm pretty sure NetBSD has already got an x86 port too...
An extra $50 can buy a lot more technology elsewhere.
The biggest player in the market is not Oracle, its pawn Peoplesoft or Microsoft.
The biggest player is SAP, and they will be extracting their due.
Open source...
. sh tml
Solaris, Open Office, Danese
http://slashdot.org/interviews/02/01/10/1245221
All bs.
Makes money?
Iron fisted grip.
Looses money?
Open source, twist the knife in MS's ribs
No better than Apple or SGI.
Sun:
The 15bn dollar market cap company that was able to parlay Java into a 3bn dollar settlement with MS over virtual machine.
1/30 the size of Microsoft, 1/20 the size of their primary competition, IBM.
"More linux desktops" than any company in the world (in china).
Supports the open source community at their convenience. So much for "write once, run anywhere. If it ever was the case with effectively 5 forks in their own code.
Well, BSD is dieing anyway right?
And their java web server runs better in a non-SPARC, non Solaris enviroment anyway.
BSD is dieing?
Sun is dieing!
A Lazarus Long saying comes to mind:
Democracy is based on the assumption that a million men are wiser than one man.
Autocracy is based on the assumption that one man is wiser than a million men.
Hmm.. 99% perspiration and the job is not done yet. The 1% is necessary.
I can hire plenty of people that work hard, but I can replace them all with one person who works smart.
"86 bytes in body"
Yeah, more like 128227 bytes in body.
I Know! We could use "Nucular" Zepplins! They could stay aloft for as long as a submarine can stay under water. They could launch cruise missles and control battlefield information.
We could put squads of HILO jumper Marines on them to hit trouble spots and drop pallets of food on indigenous people to win hearts and minds. We only crushed a few when we did that..
BTW, it's the MiG-25 FOXBAT that can reach those altitudes, not the MiG-29 and, oh yeah, we loose more Americans in LA county each year than we loose in Bagdad.
So, fighting insurgency starts at home and it is just an insurgency. No need to spend any more money there than we do on the LAPD.
Having the joy of participating in bushwarforoil 1.0, I would have to say that it is a good thing.
But call it for what it is.
bushwarforoil 1.0 liberated Kuwait, and the number one consumer of Kuwait oil was japan, getting 90% of their oil from there. At the time, Japan was our number 2 trading partner, after Canada of course.
bushwarforoil 2.0 is for China, which has reaped the benefits of our resolve by being able to increase it's imports. China has of course moved into Japan's position of number 2 trading partner, after Canada.
I personally can't wait for bushwarforoil 3.0 when in 2012, after Jeb wins the presidency, we roll into Iran, in order to secure oil for India, which will have moved into the number 2 position, after Canada of course.
Why don't we ever go to war for Canada?
They have their own oil.
Note to Corporations:
When outsourcing manufacturing and establishing trade with foreign nations, make sure they have their own oil. That way, we don't have to go to war to make sure our trading partners have enough oil.
"When you saw people on CNN saying that their primary concern was 'moral values', that was just code for 'we REALLY don't like gay people.'"
"moral values" was number one among both democrats and republicans.
"moral" can mean religious right.
"moral" can mean we don't like being lied to.
Do not let republicans take this and run or "moral values" will mean "fundamentalist state" exclusively.
We all knew his spine was as squishy as jell-o.
Cheney prolly Bent the him over the matter...
http://www.sfgate.com/cgi-bin/article.cgi?file=/c/ a/2002/06/03/MN215596.DTL
Bush to Florida: "Welcome to adaptation b**ches!"
And you all thought Bush didn't belive in evolution.
He thinks the floridians will grow gills and armored exoskeletons.
Pumped dams *are* cost effective for storing electricity.
n fo /water-power/wp-pump.asp
:)
http://www.ferc.gov/industries/hydropower/gen-i
They are used all over the country to store power in preperation for peak usage times.
If spot market megawatts are normally $400, peak usage megawatts cost $20,000, and a pumped storage dam can produce 1000 megawatts, you can save or make (depending on if you are a buyer or seller) almost 20 million dollars a day.
The dam pays for itself in just one summer