Pentium 4 Overclocked to 7.1GHz, Sets World Record

Netmonger writes "This Japanese guy overclocked a Pentium 4 to 7.132GHz!! The system managed to calculate pi to 1 million decimal places in 18.516 seconds, setting the world's record." The article notes that a Pentium 4 had been overclocked faster earlier this year, but at that speed it was not possible for the machine to function beyond BIOS. Of course, they'd yet to try diverting power from the dilthium crystal reactor to the deflector array.

Comments

World record?

[bcmm]

World record for the P4 or for a single x86 processor?

Re:World record?

[Anonymous Coward]

Yes.

He overclocked the Intel Pentium 4 670 processor with stock speed of 3.80GHz.

The processor system bus was overclocked to 1520MHz.

processor's voltage was pumped up to 1.70V, significantly higher than default setting; memory latency settings were CL4 3-3-4, memory voltage was set to 2.3V.

Still no word on what his 3dmark2005 score was! (CPU and Total, of course)

Re:World record?

[68K]

Exactly. I'm sure there are many supercomputers out there that can smoke this P4.

Re:World record?

[PsychicX]

My question is this. Tom's Hardware put a P4 under liquid nitrogen a while ago, put the northbridge chipset under a phase change compressor, and replaced the motherboard power converter in order to supply enough power to the chips, and they were only able to achieve 5.25GHz max. What did this japanese guy do different that gained him another 2.5 GHz? Is it entirely a result of using newer chips with new manufacturing technologies like Silicon-On-Insulator (SOI)? Or did this japan guy do something that Tom's didn't?

Re:World record?

[magarity]

Calculating pi is a series of mathematical operations where you can't do the next one without the prior because you need the remainders. Supercomputers are super due to a heck of a lot of CPUs all working on different parts of a problem that can be broken into chunks. How exactly do you break a series of operations that depend on the priors into chunks for a supercomputer to rip through? So anyway, it looks like this calculating pi is a record in general, not for just a PC. It's a speed job for a single CPU.

Re:World record?

[keesh]

Not true. You can get any digit in base 16 of pi without needing to know any other digit. This means you can get groups of digits in any base by only calculating a relatively small (maximum 16 * base) number of other digits in base 16 and converting.

Re:World record?

[pooly7]

That's not exactly true, you can compute the Nth binary digit of PI without the need of the previous one. Here is the guy who discover it : http://fabrice.bellard.free.fr/pi/

Re:World record?

[wfberg]

Calculating pi is a series of mathematical operations where you can't do the next one without the prior because you need the remainders. Supercomputers are super due to a heck of a lot of CPUs all working on different parts of a problem that can be broken into chunks. How exactly do you break a series of operations that depend on the priors into chunks for a supercomputer to rip through?

Use the BBP Formula. Pifast is just a benchark, like all benchmarks it's rather silly. The record is for PCs, the top 500 supercomputers are benchmarked using another silly benchmark (LINPACK).

Re:World record?

[MoonBuggy]

I'm probably being very dumb here, but how come many machines on the pi_css5 page are doing it significantly faster than this? Is the method used in that program innacurate in some way?

Re:World record?

[ArcticCelt]

"The system managed to calculate pi to 1 million decimal places in 18.516 seconds"

...but still took 25 to open an Adobe Acrobat document!

Re:World record?

[cperciva]

Calculating pi is a series of mathematical operations where you can't do the next one without the prior because you need the remainders.

Leaving aside the BBP algorithm which several other people have mentioned, you're mostly correct here.

How exactly do you break a series of operations that depend on the priors into chunks for a supercomputer to rip through?

But you're going a bit astray here. Large classical computations of Pi are exercises in performing big Fast Fourier Transforms; and there are very good algorithms for doing those in parallel. Using the AGM or a Borwein iteration, computing a million digits of Pi requires approximately 200 full-length FFTs plus some additional linear-time trivially parallelizable work.

So anyway, it looks like this calculating pi is a record in general, not for just a PC.

Give me a 4 processor 3.8GHz Pentium 4 system, and I can beat the reported time by a factor of two. If you can do parallel FFTs, you can do a parallel classical computation of Pi.

Re:World record?

[jmak]

Certainly not even for a single CPU. Performance of SuperPI (using Gauss-Legendre series) sucks by today's standards in comparison to the programs using Chudnovsky elliptic curve algorithm. QuickPI spits out 1 million digits in about 5 seconds on an Athlon-XP.

Re:World record?

[ikkonoishi]

I know you are joking, but try adobe liposuction

It worked for me. I can open a pdf in less than the time it takes a politician to go from idealistic young upstart to corporate whore.

Re:World record?

[kesuki]

or course it doesn't scale well.

quite frankly intel was supposed to be selling 6-7 ghz p4's around now... the fact that they can't but can be overclocked that fast is proof that the architecture was designed to run at those clock speeds, but that actually implementing that kind of clock speed would require insane cooling and power requirements that most sane people find unacceptable.

But.,.

[Anonymous Coward]

Does it run... OSX86 ?

Retep Vosnul

Re:But.,.

[uberdave]

Ah! Witness the emergence of a new Slashdot catchphrase.

Prove it

[Danborg]

Paste all the numbers here for verification please.

That's cool but

[notque]

How long can the machine last at that sort of overclocking? How much experience have others had with lifetimes of chips once you overclock them by a lot.

Re:That's cool but

[alragh]

About 18.516 seconds by the looks of things

Jesse James' calculator??

[gearmonger]

Calculating Pi is good...but what FPS can it do in Battlefield 2?

In related news..

[leathered]

Overclocking experiment results in largest single release of thermal energy in Japan since 1945.

Casualty figures as yet unknown.

Re:In related news..

[pwroberts]

Some tasteful moderation there by whoever modded the parent "Flamebait"... ;-)

Re:In related news..

[Anonymous Coward]

Considering the two bombs dropped on Japan saved more lives, both American and Japanese, than they killed I find the comparisons to the holocaust and Indian slaughter to be simply ignorant of history. There were more lives lost in the fire bombings in Europe, but I don't see anybody complaining about that. It was an ugly war that was thankfully brought to an end.

You should really use your head for more than a hat rack.

Re:In related news..

[Xabraxas]

REALLY FUNNY considering we just had the anniversary of Hiroshima and Nagasaki in Japan. Perhaps you want to joke about the Jewish holocaust or the slaughter of the Native American indians.

It is really funny. Hell I even joke about the holocaust and Native American slaughter. Guess what, so do my Jewish and Native American friends. If you don't have a sense of humor then I pity you.

I'm about as liberal as can be but I absolutely hate political correctness. It doesn't help anyone, it just gives some jerkoffs the opportunity to complain about trivial shit. If you really want to complain about something try taking a stab at shit that is going on now and affecting us now like the criminals in office and their illegal war, illegal torture, and illegal detention.

Re:In related news..

[joNDoty]

One of my relatives died in the holocaust. It was a horrible, humiliating way to die -- he fell out of the guard tower.

So please, don't ever joke about the holocaust.

and people womder...

[C0vardeAn0nim0]

why siberia's permafrost is melting @_@

Re:and people womder...

[aktzin]

Maybe this is what Sir Arthur C. Clarke meant when he said that supernovae are probably industrial accidents.

More info and a pic or two

[erick99]

I'd like to see a pic of the machine, especially the liquid nitrogen cooling stuff. I would also like to know if this machine ran for five minutes, ten minutes, melted?

Re:More info and a pic or two

[The Hobo]

Here you go Clicky

Booting Windows XP

[kihjin]

The maximum speed at which he could boot Windows XP operating system and perform memory testing was 6.60GHz.

Funny, since, no matter how fast I OC a Windows XP box, the XP "loading bar" still moves the same old speed.

pi=42

[Spodlink05]

Good job they didn't try that on an original Pentium, what with all those decimal places...

actual link

[Anonymous Coward]

Link to the actual forum posting, complete with pics.

http://www.xtremesystems.org/forums/showthread.php ?t=70225&page=5&pp=25

Apple was right!

[Geckoman]

Now we know why Apple switched to Intel! I can't wait for a PowerBook running one of these, complete with the Ghostbusters-style backpack pumping liquid nitrogen to my laptop!

Diverting power from the dilthium crystal reactor?

[exley]

Yeah, maybe, but they could really get some serious overclocking done if they just reversed the polarity.

And people said

[Drew Curtis]

that x86 would never hit 100MHz!

Still slower than G4!

[porneL]

I'll be missing "nah, that's just megahertz myth".

Dry Ice Slot

[gregor-e]

Somebody needs to make an auto-overclocking system that has a slot for feeding in bricks of dry ice.

You want that render to finish before lunch? Just slide in a brick of dry ice and watch the steam come out the sides as your motherboard's temperature sensor gives the go-ahead to crank the clock up to 7 GHz.

Stardate 7234.1

[MightyMartian]

Spock: Our ship's computer is running too slow. I project we will collide with the Klingon cruiser in 2.3 minutes.

Bones: You green blooded sonofabitch.

Kirk: Bones... shut... up... Scotty, I need 7ghz... now.

Scotty: I canna do it, cap'n. It'll blow the mains, and Windows 9000 will crash.

Kirk: No... excuses... Scotty... just get me... 7ghz.

Scotty (resigned): Aye, cap'n.

Uruha: We're getting in a signal for Starfleet Command. It's Admiral Gates.

Kirk: Put... him... on...

Admiral Gates: Here at Microsoft we believe in innovation. Thus we are sending you SP23482378485847825727347198874741 which will allow Windows 9000 to interact with an overclocked Pentium without making rude sounds and the voice of the computer changing from Majel Barrett to Carrot Top. Let me demonstrate.

Spock: Admiral Gates, it appears that your demonstration computer has exploded, taking out a large chunk of the planet Earth.

Scotty: Cap'n, we've got 7ghz now!

Kirk: Good work, Scotty. Sulu... reboot... the... computer.

Spock: Captain, it appears there's a penguin on the main viewer.

Re:Stardate 7234.1

[thumperward]

Whoever modded this funny should take a long, hard look at themselves. And never tell jokes again.

  - Chris

athlon 2400+ using FASTPI 1M places in 4.4 secs

[t35t0r]

Comment taken from the website:

"18.516 must be wrong. My athlon 2400+ did
1 million places of PI using FASTPI in 4.4 secs.

Maybe the number should read 1.8516 secs.
That would be more in line with factors of
speed differences between my 2400 and
the P4 system."

Re:athlon 2400+ using FASTPI 1M places in 4.4 secs

[TheLink]

He was using super pi.

Some stats: http://www.pcstats.com/articleview.cfm?articleid=1 815&page=3

http://www.planetamd64.com/lofiversion/index.php/t 5459.html

I'm not sure about fastpi. But pifast appears to allow 2-4GHz computers to do 10 million digits of pi in about 30 seconds.

Now at 7285.1 MHz

[Ixalon]

Someone has already clocked up an extra 150MHz!

Calculating 1m decimal places of Pi now down to 18.093s...

http://www.xtremesystems.org/forums/showthread.php ?p=1001108#post1001108

Vista requirements

[Medieval_Thinker]

...and in related news, Microsoft has revised the minimum processor requirements for Vista.

Intel

[ziggyboy]

I'm sure some Intel chip designers have already tried all sorts of cool shit with their processors even prior to their release. I know I would if I had worked there.

Re:Does it really matter?

[databyss]

Hey, my 486dx computer had a turbo button on it!

I'm sure it would've smoked this P4 thing...

Besides... 486 is 121.5 times bigger than 4!

RTFA, please,

[hummassa]

it's just two paragraphs for $DEITY sake.
Ok, I'll tell you, lazy boy: besides cooling with liquid N2, they tweaked the processor and the memory voltages.

Re:Does it really matter?

[Tony Hoyle]

How how would they be able to do this without Scotty?!?!

Easy. They reversed the polarity. You can fix anything by reversing the polarity.

According to his calculation...

[TCQuad]

All the numbers were 1, 2, 3, 4, 5, 6, 7, 8, 9 or 0.

Re:wow...

[Dogtanian]

By my estimates, it's about as fast as a Cray-2, but I may be pulling that out of my ass.

You intend pulling a Cray-2 out of your ass?!

Are you a fan of the Goatse man by any chance?

My only question is....

[LordPhantom]

... will it meet the min. specs for Duke Nukem Forever?

Re:The "dangers" of overclocking.

[mOdQuArK!]

If your instructions take longer to execute than the number of cycles available to them, it just can't do the work and you get junk in your registers.

Your statement is true on its face, but do you really understand why this guy could get his chip to overclock so high? He's not cooling it in LN2 just to keep it from melting (although that is certainly very important).

At low temperatures, typical silicon transistors operate much faster, and wires have less impedance, thus allowing a properly-designed chip to operate correctly at a much higher frequency than it would normally be able to achieve.

It's certainly not useful for a user who wouldn't have a constant source of LN2 available, but the fact that it can be done makes some interesting engineering scenarios possible.

And some luck.

[btarval]

They also must have gotten lucky with the CPU. Back before Intel started adding their stupid locks to limit overclocking, it was painfully obvious that the production run played a factor. And even within the production run, there were always CPU's within any given batch that were better than others.

The last set of great overclocked CPU's were the Celeron 300's. Many of those went to 450-500 MHz with no problem. A very few could be made to hit 600 MHz, though it is questionable on how reliable they were at that point. Certainly reliable enough to calculate the value of PI quickly; but you wouldn't want one for reliable web server.

Granted, some of the one's which could do 450-500 MHz were made for that speed, and then sold as 300's. But certainly not all of them.

The bottom line is that cherry-picking your CPU's helps lead to a better chance of success with overclocking.

The future is asynchronous!

[Midnight Thunder]

Cooling and power requirements aren't the only issue, since at those types of frequencies you are likely to be interfering with radio frequencies, unless you have really good shielding.

The real future is asynchronous CPUs, that are actually clockless. They generate much less heat and consume much less power. The only reason that they aren't replacing the current batch of chips fast, is that all chip design and testing processes are built around clocked CPUs.

A few articles on the subject:
  - Will Self-timed Asynchronous Logic Rescue CPU Design?
  - Computer Chips Without Clocks