AMD Phenom II Overclocked To 6.5GHz

An anonymous reader writes "During CES a group of overclockers with access to liquid nitrogen and liquid helium for the extra boost of coldness cooled an AMD Phenom II X4 chip to -232 degrees Celsius. Once they got the chip cooled to this frigid temperature, they pushed the clock speed all the way up to 6.5GHz, which is a world record for a quad-core CPU, and then dished out an astonishing 45,474 3DMark05 score!"

Comments

The things you have to go through..

[DigitAl56K]

.. to get a decent score in 3DMark ..

Re:The things you have to go through..

[shootlessjoe]

And the question is, Why are they still using the 3DMark 05? Technology has advanced people!!

Re:The things you have to go through..

[Anonymous Coward]

Now Vista can run decently.... That's a shame, they want to replace it with Win7...

Re:The things you have to go through..

[Anonymous Coward]

It's good to know that I can spend a few thousand on cooling supplies now and get a machine that can run Crysis.

I was worried I'd have to wait a year or two for those kinds of numbers to reach the few hundred dollar range.

Re:The things you have to go through..

[twowoot4u]

Now if only they had set the cpuID bis to 'Genuine Intel' http://arstechnica.com/reviews/hardware/atom-nano-review.ars/6 they probably could have reached 70k!

Re:The things you have to go through..

[zippthorne]

Atom uses 60W???

Why not just get core 2 mobile, then?

Re:The things you have to go through..

[Anonymous Coward]

Atom itself uses 1 to 4 watts. It's the chipset that sucks.

Re:The things you have to go through..

[jopsen]

Now Vista can run decently....

No, they'll still be IO bound... :)

Re:The things you have to go through..

[slimjim8094]

Yeah. The sequel is called Cryo.

Re:The things you have to go through..

[Chabo]

You got modded funny, but I think you were probably being serious.

I think the reason is that the newer 3DMark suites advanced so much in the realm of GPU-intensiveness, that to overclock a CPU and get a higher score without being GPU-bound, you have to go back to 2005.

Re:The things you have to go through..

[sharkey]

Decently? Just because the 500 pound naked guy can run the 40 under 5 seconds, does NOT make him decent.

Re:The things you have to go through..

[Clay Pigeon -TPF-VS-]

A ultralow wattage athlon64x2 with a 780g chipset uses less power than an atom machine last I checked. the 945 chipset is shirts.

Re:The things you have to go through..

[earthforce_1]

Yeah, but if you watched the video, you only have about an hour to play before the liquid helium runs out, and your CPU turns into a slag heap. Think you can finish the game in time?

Re:The things you have to go through..

[tzot]

Atom itself uses 1 to 4 watts. It's the chipset that sucks power.

Fixed it.

Re:The things you have to go through..

[pascal.ch]

and it's not even the most recent edition!! ... it's the '05 edition ... damn I give up

Crunchy

[x1050us]

Numbers must be really crunchy at that temperature

Re:Crunchy

[thatskinnyguy]

I think spit goes *clunk* at those temperatures.

Re:Crunchy

[hclewk]

That's -40 degrees Celsius.

http://xkcd.com/526/

Re:Crunchy

[Kayden]

Before it even leaves your mouth.

Re:Crunchy

[NoobixCube]

Spit is more likely to go *click* at those temperatures. The sound of frozen muscles in your throat fighting against other frozen throat muscles (assuming any of the muscles can move at all) and fracturing into two or more pieces.

from TFA

[caffeinemessiah]

which is a world record for a quad core CPU and they dished out and astonishing 45,474 3DMark05 score! Watch the video below to see how it was done and how history was made:

Truly PHENOMenal, but I can't help but (cynically, I admit) think about how history inevitably mocks overclockers. Cue back to the 90s and a headline might have read "486 overclocked to 500Mhz -- history has been made!". Like Ozymandias, nothing beside remains...

Re:from TFA

[Anonymous Coward]

It is one thing to know that in two years you can regularly buy a system twice as fast as today's fastest system

I doubt it. CPU speeds haven't really increased that much in the past few years. We're reaching the limits of what we can do
with a CPU as far as speed goes. Even shifting to 32nm wont increase the speeds that much... mostly just lower power usage.
They wont be running too much faster but we'll have CPUs with many more cores and it'll take more than a couple of years to shift
to properly distributed programming which will be needed to properly use those cores.

Re:from TFA

[Spatial]

You don't know what you're talking about.

Compare a 3Ghz Pentium 4 with a 3Ghz X2 6000+. The X2 is dramatically faster. A 3Ghz Core 2 is faster than that again by a significant margin. A 3Ghz Core i7 is even faster again by a big margin. Where's this limit of what we can do? Performance has never stopped increasing by large amounts with each architecture. And I'm not even taking into account the increase in cores, this is just per-core performance.

Re:from TFA

[ahoehn]

Like Ozymandias, nothing beside remains...

"My name is G1G4BY73_PU5H3R, Overclocker of Overclockers:
Look on my 1337 benchmarks, ye n00bs, and despair!"

Re:from TFA

[Endo13]

What you're talking about isn't really "speed" or "faster". The term you are looking for is "more powerful". The GHz is how many hertz per second the CPU runs at... or in other words, its speed. Newer CPUs get a lot more done per cycle than the first 2-3GHz CPUs. And because they get more things done per cycle, they get things done faster, but that's not the actual CPU speed.

OP is probably right - 3GHz is probably about the practical limit of what CPUs can run at for everyday use. Speeds higher than that so far seem to increase heat too much to be useful for most applications. Think of GHz like RPMs for a car engine. Most car engines run somewhere between 2000-6000 RPM at driving speed, however some get a whole lot more horsepower at the same RPMs and therefore make the car go faster... but the engine is still running at the same speed. There WAS a time though when they were getting more horsepower by increasing RPMs.

Re:from TFA

[novakyu]

OP is probably right - 3GHz is probably about the practical limit of what CPUs can run at for everyday use. Speeds higher than that so far seem to increase heat too much to be useful for most applications.

Even if heating were not a factor (i.e. if you could build most efficient cooling system suitable for every day use), those pesky laws of physics, such as speed of light limit comes into play. Even at 3 GHz, light travels only 10 cm in one cycle, and that's not too much larger than physical size of a single core. One could probably shrink those by some factor, but not much more than factor of 10; then you have to start worrying about the size of atoms (which is one of the reasons why they are building multi-core CPUs in the first place, I think.)

Re:from TFA

[Spatial]

Even at 3 GHz, light travels only 10 cm in one cycle, and that's not too much larger than physical size of a single core.

It's far larger. In a C2D E8600 the physical die containing both cores is 107mm squared. Even the whole chip package is way smaller.

Re:from TFA

[Carbon016]

45nm Wolfdales are quite commonly overclocked to 4ghz on stock cooling and show maybe an increase of 10-15C (given that the chip usually runs at around 30C and has tolerance for up to 60-65C before you should start getting worried, this is a drop in the bucket).

Re:from TFA

[Spatial]

Come on, give me some credit. Thanks for explaining and all, but a car analogy? These CPU doohickeys aren't new to me you know.

Anyway, why do you think he was talking about clock speed? He never even mentioned that, he just said speed as a synonym for performance like almost everyone else does. As in, "The speed at which it gets shit done." On top of that, he talks about properly using multiple cores as if core performance isn't improving with time, and he's responding to someone who was explicitly talking about performance and not clock speed.

OP is probably right - 3GHz is probably about the practical limit of what CPUs can run at for everyday use

No. There are 3.5Ghz chips available right now with a TDP of 65 watts (C2D E8700). Voltage is more important than frequency with regards temperature and power consumption.

Re:from TFA

[Mr Z]

Hmmm... if you're going to go with 486 overclocking, that probably tops out more like 200MHz.

Re:from TFA

[Shark]

Keep in mind that information has to travel out of that die at a significant clock speed to other components too. At 1Ghz, that leaves you 30cm of an ideal conductor to play with, and they can't all go in a straight line to your components.

Re:from TFA

[treeves]

"The GHz is how many hertz per second the CPU runs at"

ORLY? Hz = 1/second so you're saying that GHz = 1/sec/sec = 1/sec^2. Sounds like acceleration. GHz is like a measure of acceleration. Neat.

Re:from TFA

[SBrach]

Bad analogy. Horsepower is a function of torque and RPM. You almost always get more HP with more RPM for a given displacement size (of which torque is a function of). This is why Formula 1 cars turn 23,000rpm. When your are limited to a fixed displacement, other than forced induction, you reach a point where the only way to make more power is to spin it faster because you can't get any more power (efficiency) by doing more each rev(cycle). When we get to the point where no CPU cycles are wasted, that is, they are 100% efficient and no cycles are wasted, then the only way to make them faster will be to clock them higher. We are not there yet, though.

Re:from TFA

[toddestan]

OP is probably right - 3GHz is probably about the practical limit of what CPUs can run at for everyday use. Speeds higher than that so far seem to increase heat too much to be useful for most applications.

You can buy dual-core Wolfdales right now that have a TDP of 65W and clock in at over 3Ghz. That's very reasonable. If they were to make a single-core version it would run at around 30-35W and would still get more done than yesterday's 3 Ghz P4's. Ditto for the 3Ghz mobile Core 2 processors.

Re:from TFA

[earthforce_1]

My old Pentium 4 which I just retired ran at 3.6 GHz. My slightly overclocked i7 runs at 3.65 GHz. Obviously a 3.6 GHz P4 != 3.6 GHz i7 or Phenom 2, but the raw clock rate hasn't changed much in about 4 years. There is a practical 4 GHz ceiling. We just do more on each clock.

Re:from TFA

[Endo13]

Yes, I realized the error as soon as I posted. Thanks to no edit button and a 5-minute wait between posts, I didn't get it 'corrected'.

Re:from TFA

[MadMidnightBomber]

Two vast and shapeless puffs of smoke
Fill the basement. Near them on the floor,
Half black, a shatter'd hard disk lies,...

Re:from TFA

[default luser]

What you're referring to is IPC (Instructions Per Clock). CPUs of yesterday increased IPC by pipelining instruction execution (prest ICP for that is one per clock), and CPUs of today do so by re-ordering x86 instructions so they can be executed in-parallel. Increasing the IPC has been the holy grail for CPUs.

Compare a 3Ghz Pentium 4 with a 3Ghz X2 6000+. The X2 is dramatically faster.

This is not because architecturers had major efficiency gains at the time, it's because Intel took a huge step backwards in IPC with the Pentium 4. They did this because their engineers assumed they could continue clocking CPUs faster and faster. Their proposed goal was to have the Pentium 4 at 10 GHz by the end of the decade, and in hindsight it wasn't a good idea.

A 3Ghz Core 2 is faster than that again by a significant margin. A 3Ghz Core i7 is even faster again by a big margin.

The Core2 was a major leap, which is a rare thing to see. Core2 was great because it not-only produced impressive improvements on pure processing-limited code, it also improved I/O-limited processing with a slew of new caching systems and the new out-of-order memory accesses.

i7 is not a major leap - it sees performance improvements on some tasks, but for most it keeps parity with Core2 Quad.

Why is i7 not the amazing leap that we saw in Core2? It's because every improvement in architecture does less and less to improve performance, because every "improvement" adds more overhead than the last, and pushes the edges of architectural limits.

You can only go so far for two major reasons: one, adding another pipeline inside of a chip adds an exponentially increasing number of routing wires and support circuitry, so each improvement costs you more-and-more. Also, there is a finite limit to how "parallelizeable" common x86 code is; as you approach this limit, the pipelines youi add-on will sit there unused.

The Core2 is already approaching said limits, so I wouldn't expect anything amazing to come out anytime soon. Really, x86 IPC has only seen TWO major jumps in the last 15 years, and those have been the Pentium Pro, and the Core2. You can only do so much to parallellize x86 code.

Re:from TFA

[novakyu]

It's far larger. In a C2D E8600 the physical die containing both cores is 107mm squared. Even the whole chip package is way smaller.

So, for the physical die is about 1 cm on a side, and as a rough estimate, each core is about 0.5 cm on a side. And how many times does the "wire" have to go around in that chip to make a complete circuit?

10 cm is, on the order of magnitude (i.e. within a factor of 10, or probably even 5), how long the signal path is in a single core.

Re:Zomg

[sakdoctor]

Yeah, only 6.5GHz too.
Call me when it goes up to 11

first post!

[CaptainNerdCave]

whoo! and i _still_ get the first post with my q6600!

Re:first post!

[ByOhTek]

I'm fairly sure 'epic fail' doesn't even begin to describe your post

Re:first post!

[TheLink]

It would if he had said he was using an Itanic CPU.

Vista?

[omarabas]

can it run Vista?

Re:Vista?

[Nathrael]

Obviously, we've got some mod who is running Vista on his machine here...

And I'm Guessing

[Cnik70]

Windows still seems to run slow at -232 degrees

Re:And I'm Guessing

[dkh2]

Your Windows EULA prohibits operation in such an environment.

Re:And I'm Guessing

[CarpetShark]

Windows still seems to run slow at -232 degrees

Actually, it won't heat up that much.

Hosting their website on it

[ksd1337]

Then they installed it in their server and waited to see if it would pass the Slashdot test.

But...

[LWATCDR]

Can you run FSX and Cryis at 60FPS?

Re:But...

[Sabz5150]

Can you run FSX and Cryis at 60FPS?

No.

Re:But...

[A. B3ttik]

How about SupCom with 6 AIs with 1000 units each?

Re:But...

[interval1066]

The question, like Mt. Everest, is not "Would you?", but "Can we?"

Re:But...

[Spatial]

It'll run for a while. Then you'll run into a memory limitation and the game will crash. It's happened to me a few times.

New York weather...

[VinylRecords]

I'm not sure if it's quite -232 Celsius in my apartment but it's pretty close. They probably could have achieved 6.0GHz overclocking using an air-cooled system in my living room alone.

Re:New York weather...

[somenickname]

You must be married.

Re:New York weather...

[hattig]

You can't be married. Any woman would have turned on the central heating to max by now, and filled every room with electric heaters on top.

And they'd still complain about it being cold, even as you sat there sweating like a pig, wearing a wife-beater, with your feet in cold water, and a cold can of wife-beater in your hand.

Re:New York weather...

[Skye16]

What I found this go-round was to start her out cold. As in, I was freezing my god damned balls off when we first started dating, trying to keep the house cold enough. This set the calibrated cold properly for her.

Now I can keep the house at 63 in the winter and she doesn't bat an eyelash, because it's a far sight better than the 52 it was before.

The downside is that now she whines twice as hard as I do when it hit 78 in the spring. Neither of us move much in the summer months.

Long story short: if you start out right, you can train them easily.

Re:New York weather...

[Cornflake917]

And they'd still complain about it being cold, even as you sat there sweating like a pig, wearing a wife-beater, with your feet in cold water, and a cold can of wife-beater in your hand.

Maybe he wouldn't feel so hot if he wasn't beating his wife all the time?

Re:New York weather...

[Chabo]

My girlfriend and I moved from New England to California recently; it's hilarious to see the locals shivering in their wool coats and hats when it gets down to 45 or 50F, while we're walking around comfortably in short sleeves. We were walking outside at 8AM a few weeks ago (in "the dead of winter"...). It was right about 32F, and I was like "ya know, I'm slightly uncomfortable right now." :)

Then again, we're gonna be dying this summer, when it gets to 100F or more... :/

Re:New York weather...

[operagost]

Old Wife Beater is piss. I only shotgun dry-ice cooled pounders of Child Neglect.

Re:New York weather...

[Xabraxas]

Come live out in small-town Saskatchewan, Canada. Average winter day is -25C... coldest being -46C (so far this year, anyway).

My sister lives out in Minnesota and I have an aunt and uncle and a couple of cousins living in North Dakota and its no picnic there in the winter. They get down to about -40F. The really shitty part is that the summers are just as extreme in the other direction. I was in ND a few years ago for my cousins wedding and it was over 100F every day I was there. I don't know how anyone deals with those kinds of extremes. As a New Englander myself we have more whacky weather than extreme weather. A couple of years ago on a trip to Boston the temperature was near 60F in January. I've also seen it snow in late April, early May. As a side note, living in the city during the winter always sucks because the wind is ten times worse in a city than it is in the burbs or the sticks. It's like walking around in a giant wind tunnel.

Re:New York weather...

[freedom_india]

Obviously you didn't get what the GP said.
Oh forget it. Only married idiots can get the GP.

wearing a wife-beater, with your feet in cold water, and a cold can of wife-beater in your hand.

Gem of a statement, although a bit crude...-) But hey, thanks.

Re:Zomg

[Yvan256]

Don't we already have CPUs running at 3GHz?

I was there

[Rinisari]

I was there, too. The coolest it got was approximately -242 degrees C; the warmest was approximately -218 degreesC, at least while I was watching.

The party was the XtremeSystems.org party at its LV headquarters, and it was sponsored primarily by AMD, DFI, Gigabyte, Cooler Master, and Thermaltake. It seems to me that Commodore had a presence there, too.

See ThinkComputers' blog for some more pictures (disclosure: my article).

Re:I was there

[horza]

Why doesn't -242C exist? -273C exists.
http://en.wikipedia.org/wiki/Absolute_zero

Phillip.

Re:I was there

[adriantam]

-273C is absolute zero, -242C is not.

Re:I was there

[bbn]

Absolute zero is -273.15C.

http://en.wikipedia.org/wiki/Kelvin

Re:I was there

[Rinisari]

Both I and Wikipedia disagree (edited to make the degrees and negatives display correctly):

By international agreement, absolute zero is defined as precisely 0 K on the Kelvin scale, which is a thermodynamic (absolute) temperature scale, and -273.15 degrees on the Celsius scale. Absolute zero is also precisely equivalent to 0 degrees R on the Rankine scale (also a thermodynamic temperature scale), and -459.67 degrees on the Fahrenheit scale.

Re:I was there

[berend botje]

Perhaps it is in your own little mental bubble, Bub.

Meanwhile, in the real world, absolute zero is -273.15 Celsius.

Your -242C might be considered balmy compared to that.

Re:I was there

[jimbo3123]

This is true; Absolute Zero is -237.15 C.

Re:I was there

[Hordeking]

-242C is a temperature that doesn't exist, unless your religion allows temperatures below absolute zero. All we need now is a campaign for Intelligent Cold. ;)

In this house, we obey the laws of thermodynamics!

Re:I was there

[Menkhaf]

I know that all of /. has commented on your apparent stupidity/shortsightedness, but I still wanted to kick you while you were down.

Reaching a negative reading on the Kelvin scale is possible, although I think my head needs some cooling to comprehend it.

Re:I was there

[Piranhaa]

I can vouch for that one. Just come take a trip up to Canada!

Re:I was there

[Too Much Noise]

Not only are you wrong about where 0K is, you're also wrong about negative temperatures. Temperature is a statistical measure. A positive temperature corresponds to an equilibrium population distribution across a bunch of energy levels, which will have occupancy probabilities decreasing exponentially with energy. A negative temperature is obtained when the population distribution is inverted, for instance in a 2-level system where an external energy source resonantly pumps up the occupancy of the higher level. Presto, $\exp(-\beta E)$ greater than 1 requires negative $\beta$.

Re:I was there

[dex22]

Yeah, and this was the mistake I made. 237, 273... Ah the price I pay to make a joke about "Intelligent Cold"! :)

Shoulda got me some "Intelligent Dumb!"

Re:I was there

[Daswolfen]

No.. not really...

http://www.commodoregaming.com/us-en/newHp.aspx

They make high end gaming PCs now.

Re:I was there

[mhall119]

And from this point on, you will double check the accuracy of every snarky comment you ever make on Slashdot.

It's like a right of passage, really.

Re:I was there

[Repossessed]

Seems pointless. If you overclock all that way and the temperature is still below zero, wouldn't a cheaper cooling system make more sense?

Re:I was there

[daybot]

In this house, we obey the laws of thermodynamics!

In Soviet Russia, the laws of thermodynamics obey YOU!

Re:I was there

[tellthepeople]

from this point on, you will double check the accuracy of every snarky comment you ever make on Slashdot.

You mean some people don't check the accuracy of their snarky comments. I would think any sane person who has seen the mountains of comments that are posted after an inaccuracy would never post again for fear of making a mistake.

Re:I was there

[mqduck]

Ow. Please don't do that again.

Reliability?

[Anonymous Coward]

How reliable is that thing?

Re:Reliability?

[__aardcx5948]

Who modded parent as interesting? It's self evident that an extremely overclocked chip isn't reliable, and reliability has nothing to do with this record attempt. Good stuff: it's the first AMD chip to not have a cold bug (afaik), it overclocks very well on air (3.5GHz on stock voltage - not bad) compared to its predecessors. It's an alternative to Core2Quad now, depending on what applications you are running. Core i7? Faster, very much so, but the mainboard and DDR3 are gonna cost you. The most bang for the buck seems to be the cheapest Core i7 today anyway.

Re:Reliability?

[Huntr]

Ah, it runs hot and cold. Mostly cold.

Re:A cat has gotten my tongue

[Thornburg]

An slightly overclocked Core i7 965 (Extreme Edition) in a similar rig (in terms of video cards, etc) scored about 26,000 in the same benchmark (3DMark05).

So, no, they didn't have to go to liquid helium to be competetive, but going to liquid helium did allow them to set a world record (although I don't see any Guiness Book or other "official" information about this).

FIRST POST

[A. B3ttik]

...with my lightning-fast 486!!!

Re:FIRST POST

[Anonymous Coward]

> ...with my lightning-fast 486!!!

Did you forget to press the "TURBO" button? :-)

And this is exciting exactly why?

[Giant Electronic Bra]

Last I checked neither my condo nor my office has a piped in supply of cryogenic gasses...

Re:And this is exciting exactly why?

[moderatorrater]

That's why I have the milk man deliver me some every week.

Re:And this is exciting exactly why?

[pipboy9999]

I agree...its neat. But why not spend all that time, money and effort into research on making faster chips that are actually usable by a normal person?

Re:And this is exciting exactly why?

[mdm-adph]

Why travel to the moon? The average person has no purpose there.

For that matter, why design the ICE? Walking or horseback is just enough for anyone.

Re:And this is exciting exactly why?

[Anonymous Coward]

(a) Some people just enjoy this sort of thing. I don't really pay attention to who is the fastest at driving a car around a track, but some other people do. So, one answer to your question: maybe it's not exciting at all, to you, and no one says it must be.

(b) Epic overclocks like this presage the chips to come. While I won't be using anything but stock air coolers in my own computers, I'm happy to know that the Phenom II overclocks this well, because that means there is headroom in the design and that AMD will be able to get the clock rate up over the next year or two. So, maybe it's not exciting, but it is interesting to know that the Phenom II isn't a chip that can just barely support its official clock.

Re:And this is exciting exactly why?

[hierophanta]

uhh chips are so fast right now (for normal people) that the industry has been looking at making them super low powered. case in point - the Atom. why on earth do we need faster chips for 85% of use [by hours of usage]?

im sorry to say it but i dont think normal people dont game on their PCs. Even if you and i do.

Re:And this is exciting exactly why?

[zippthorne]

You laugh, but I talked to the welding supply guy not too long ago, and two things were apparent:

1. The process is not all that dissimilar to milk delivery: you provide the bottle they put stuff in it. For small amounts, though, you have to leave the bottle at the shop and they fill it when the truck comes.
2. N2(l) is cheaper per liter than milk.

Re:And this is exciting exactly why?

[FishAdmin]

Last I checked neither my condo nor my office has a piped in supply of cryogenic gasses...

That's funny; mine does! Then again, I work in a chemical laboratory...

Re:And this is exciting exactly why?

[DMUTPeregrine]

Because overclocking enthusiasts and cooling specialists have the same skillset as chip designers. Yup. Exactly the same. That said, cryogenic gases aren't that hard to get. I know the Airgas (a wielding supply store) near me sells liquid nitrogen. I don't think they sell liquid helium, but the nitrogen's not that expensive IIRC. Another fun cooling technique is cascaded direct-die vapor phase-change cooling.

Re:A cat has gotten my tongue

[triffid_98]

Some of them run faster than that, actually.

Intel's chips run at 6.5 GHz?

There is no mockery.

[unity100]

at the time a 486 might have been overclocked to 500 mhz, it would have been a great deal. more precisely, at the time anything has been overclocked to phenomenonal mhz, it has been a great deal AT THAT TIME.

Apple IIe

[philspear]

That's pretty impressive, but right now I'm posting using my overclocked apple IIe.

Re:Apple IIe

[Nimey]

Yeah? Did you get it up to 10 MHz?

Re:Apple IIe

[philspear]

MHz? Try GHz.

Re:Apple IIe

[Shadyman]

Ok, so you got it up to 0.010 Ghz?

Re:Apple IIe

[ben0207]

GHz? Try THz....

[NO CARRIER]

Re:Apple IIe

[anss123]

Overclocing an Apple II from 1 MHz to 10 MHz is a bigger overclock than 3.4 Ghz to 6.4 GHz. So there!

Re:Apple IIe

[Nimey]

See: Zip Chip. One model did in fact go all the way up to 10 MHz, and Apple looked at using one in the Apple //c+, but went with something else instead.

nice and swolen..

[viper34j]

my nerd is swelling...

Re:A cat has gotten my tongue

[horza]

I think if you post something like this to Slashdot and nobody contests it, that's about as official as you can get :-)

Phillip.

Re:A cat has gotten my tongue

[Merovign]

AMD doesn't make any $1200 chips.

Like it or not, that's just not the market they're in. They're doing well at the $200 level, though.

I'm not particularly concerned that there's little competition in the segment I'd never pay for anyway. I mean, it's nice that there are Maybach Mercedes and McLaren F1's, but that doesn't mean I'm worried about competetiveness in the segment.

Whereas I'd be worried if there was only one mid-priced performance sedan, especially if it was sub-expectations in some way.

I don't think AMD is ashamed to have set a record with a $235 chip, in a world previously dominated by $1000+ chips.

Stability?

[SanityInAnarchy]

I know I once bought a specific CPU because I knew it would be good for overclocking. It wasn't a bad idea -- a 1.8 ghz CPU that I could get running at 2.4, at perhaps half or a third the price of a similar CPU at 2.4 ghz, and I'd overclock my RAM, also.

I learned two things:

First, you really have to know your stuff. The RAM I had wouldn't overclock very well, and RAM which would cost a bit more. I had the BIOS helping me out, and I still had to fiddle with timings and voltages.

And second, despite all the stress testing I did, it would still occasionally crash. I never tracked down these crashes until I clocked it back to spec. Once I got a job, I decided that shelling out another hundred dollars or so for a faster CPU was a better use of my time than trying to overclock one, and dealing with the instability once I did.

Now, that's probably a completely different area than overclocking to 6.5 ghz, but if I really needed that, I imagine it would be much more cost-effective to buy two or three of them. It won't really help rasterized games (that'd be video-card bound), and raytraced games should scale to multiple machines.

Re:Stability?

[anss123]

First, you really have to know your stuff. The RAM I had wouldn't overclock very well, and RAM which would cost a bit more. I had the BIOS helping me out, and I still had to fiddle with timings and voltages.

My experience has been similar. I've managed to overclock CPUs but to get them runing stable I had to fiddle with voltages and even then the overclock had to be pretty mild (2.4 Ghz to 2.6 GHz for instance).

Ignoring stability performance did increase, yay, but so did the heat and the noise of the CPU fan. Not worth it.

Re:Stability?

[__aardcx5948]

I have heard claims that people has been running Prime95 (stress test) on all cores for 24 hours, and still had problems while playing a game for example. Often it's because the PSU can't keep up, not many people stress test CPU and GPU at once, which is essential to make sure that the PSU isn't holding you back. You make it sound like OC is extremely difficult and not reliable. I'm willing to agree with you to some extent, but it's like everything; you have to know your stuff. If you know how to do it, or want to spend the hours learning, it could save you a couple of hundred bucks. I remember running my 1700+ tbred (1.4) at 2.4Ghz solid with watercooling. Those were the days...

Re:Stability?

[zippthorne]

I've always been more of a fan of underclocking, myself. Or as you say, regular-clocking.

ten or even thirty percent just isn't that much of a difference in performance to justify a stability headache OR paying an extra couple hundred bucks.

To be interesting, the performance improvement per dollar ought to be significantly better than linear, and at least double. Or you need an application that is CPU bound, time sensitive, and has large processing chunks. 30% isn't going to make much difference in UI performance. Spend that money on RAM.

Re:Stability?

[Belial6]

Overclocking is like tricking out your Honda Accord. It is a hobby in and of itself. It isn't a good idea for people who just want a computer that works well, just as constant modification to an Accord isn't a good idea for people that just want a reliable form of transportation. That doesn't mean that they are not perfectly reasonable hobbies. It just means that they are not hobbies for me, not hobbies for most people, and most people will think you are wasting your time because it is not their form of entertainment.

Personally, I have purchased a brand new homebrew Amiga clone within the last year, and have purchased 2 C64 clones within the last 5 years. I certainly know what it means to enjoy a hobby that the vast majority of people "don't get".

The biggest problem with overclocking for the masses is that if you don't enjoy the act of overclocking in and of itself, you can achieve better results through procrastination.

Re:Stability?

[__aaoyac5342]

Overclocking is more of a hobby for hardware enthusiasts. Even if you don't know much the biggest thing is buying a motherboard, ram and cpu that will overclock well together. Find someone else's posted specs and duplicate their setup.

Re:Stability?

[MBGMorden]

Same boat for me. When I was younger (and cheaper) I would always try a little bit of overclocking. Overclocked a Celeron 366 to 550 and it worked pretty good after some tweaking. After that I overclocked my next chip as a Celeron 566 to 850. Again though, took some tweaking. I used "slockets" back then and constantly was looking for high quality HSF's, slockets that offered voltage or FSB adjustments via jumpers, etc. Took a lot of work to get them stable. After that I tried several different chips and had various troubles overclocking to a stable speed.

These days, I've generally found that: for the most part, most overclocks are small, particularly compared to what you could once get. That 366 to 550 overclock was a huge percentage increase in clock speed, as was the 566 to to 850 overclock. Nowadays you have people still working just as hard to get another 200-300 mhz out of chips that are in the mutli-gigahertz range already. And for the most part, aside from video encoding, I can't generally tell a real world difference anyways. I've just become content with living with a processor that isn't on the bleeding edge. I make big upgrades. Previously I was running a Sempron 3400 single core. My last upgrade took me to an Athlon x2 4600 (double the cores, faster raw mhz, and more L2 cache). My next upgrade will probably be a ways down the road and will certainly involve jumping to quad cores.

It's just too easy to sit back, take the safe route, and just wait for the stuff that's fast at stock speed to just get cheaper.

Re:Stability?

[MBGMorden]

And just as an addendum: while I think overclocking is a fine hobby (like I said, I used to do it), and wish overclockers all the best, I must admit that sometimes they can be annoying with user-generated reviews. With processors, countless times when looking through user reviews I have to wade through a ton of reviews which rate the chip poorly simply because it didn't overclock as well as they'd hoped, or even worse, write that they are returning the product because it didn't overclock well.

Do as you want, but it's just bad ethics (and drives the prices up for everyone) to return a product that works within the spec it is sold at simply because you want to run it out of spec and it won't.

If a chip is sold as a 3ghz chip, then try your luck. If you want 3.6ghz and you get it, then great. If you only get 3Ghz though then that's all you paid for, don't return it expecting more, or bash the product.

Re:Stability?

[rwiggers]

That's really what it is (hobby), but I still choose CPUS easily overclockable. Just to run on specs and get a better stability for those of us that have ambient temperature over 30ÂC and high humidity.
Usually, easily overclockable means better design margins.
By the way, more than one report is a must, or it could be only luck. These things are designed for the worst production scenario, and not all pieces produced fall in these conditions.
--
{blank};

Re:Stability?

[jopsen]

I agree, things are IO bound today...

Re:Stability?

[Mex]

Yes! Thank you! It's also great for reducing the use of fans, and I suppose it even uses less electricity?

I have a Q6600,and I honestly never use the 4 chips at once, specially not in regular work... Even when playing, I don't think most games use more than 1 or 2 cores, right?

The video card on the other hand... noisy and overheats a lot when playing 3d games...

Re:Stability?

[SanityInAnarchy]

I've always been more of a fan of underclocking, myself. Or as you say, regular-clocking.

Well, never under spec -- after all, once you're no longer overclocking, you can let the normal CPU scaling kick in.

This laptop is dual-core, capable of 2.5 ghz, usually runs at 800 mhz. If I deliberately clocked it down further, I don't see what I would gain -- and sometimes you do have a CPU-bound application. Especially great if it's a CPU-bound command -- most of the time you run at 800 mhz, then you run a command that takes ten seconds instead of a minute because it slammed the CPU up to 2.5 ghz, then when it's over, drops to 800 mhz again.

30% isn't going to make much difference in UI performance. Spend that money on RAM.

This laptop came with 4 gigs of RAM, so I decided to spend more money on a faster CPU. I don't think it could fit more RAM.

Re:Stability?

[SanityInAnarchy]

The problem is, I, and many other overclockers, had a bit of an attitude about it. Why would you waste an extra $100 to change the clock speed, or open up a few more pipelines, on what is basically an identical chip?

Answer: Because you're also paying for that particular chip to have tested those pipelines, and to have tested and been insulated for that speed.

As long as it's made clear that this is a hobby, that's fine. But it should be made equally clear to anyone going in that this is not a way to save time or money, as it will cost both in the long run.

Re:Stability?

[machine321]

Personally, I have purchased a brand new homebrew Amiga clone within the last year, and have purchased 2 C64 clones within the last 5 years. I certainly know what it means to enjoy a hobby that the vast majority of people "don't get".

How fast were you able to clock the C64 clones?

Re:Stability?

[shaitand]

In a laptop you would have been better off spending it on a fast hard drive if you could find a laptop with a fast drive that didn't weight 10lbs+

Re:Stability?

[shaitand]

'The simple truth is that these chips come shackled...why not get a little more bang for your buck, and how often does anyone really 'burn out' a CPU before it is obsolete anyways?'

Often, very often. I see them with scorch marks and actual holes burned right in the cpu. I've also seen heat marks around the chipsets and nearby capacitors that are bulging and leaking because they have been operated past spec.

Like another poster in this thread I do watch which chips are overclockable. I might even play with a new chip, but I never run a chipped OC'd for any length of time. Being easily overclockable means the chip will perform extremely well at the shipped speed. I also never order a chip at the top clockspeed in a series BECAUSE it is essentially the same chip overclocked. Pay less and get a chip that is stable.

The truth is that the for most applications (particularly anything on the desktop) the CPU isn't the real bottleneck these days anyway.

Re:Stability?

[SanityInAnarchy]

Given that I was employed then, and this was to be my main work computer -- or at least, that was the justification...

I did get a fast drive -- SSD. They wouldn't ship with less than 4 gigs of RAM. It's a 2.5 ghz core 2 duo with 6 megs of cache.

The main regret was getting this model without checking my basic assumptions -- it does not come with gigabit. Had to get a gigabit PCI express card. I could also have used a newer, better video card, and ideally something for which XP drivers are readily available.

But yes, I do appreciate the speed when I need it -- both for gaming and for the less-optimized work applications. (As in, applications we've developed at work.) In fact, with a scripting language, RAM and CPU are what matter, disk is going to be way smaller, if anything.

Re:Stability?

[Jarik_Tentsu]

I bought my entire setup at the moment with the intention of overclocking.

I bought the 2 year old Q6600 because I knew I could take it from 2.4GHz to 3.6GHz on air cooling. So stock there were better, newer Penryn models, but overclocked, there weren't that many better. I got my mobo - the Gigabyte E38DQ6 because I knew it had good overclocking features to take my CPU there.

Considering at 3.6GHz, my RAM only had to run at 800MHz to keep 1:1 FSB ratios that wasn't a concern. But I know if I got the E8400/E3100 (another good overclocker), I could use that same RAM - which was good until 1200MHz at its native timings of 4-4-4-12.

But yeah, you're right. Choosing components is half of what overclocking - and you gotta know what to look for in different components. But then again, to make best use of your money, you *do* have to do a shitload of research. You can't just rock up at your local computer store and grab whatever parts they have in stock. I spent a month of in depth research into every part, how they would react to each other and whatnot until I decided on what I'd get. Then I had to spend a fair bit of time trying to track down those parts.

~Jarik

~Jarik

Re:Stability?

[shaitand]

I can't help but wonder why anyone who needs a system with those kind of specifications would purchase a laptop in the first place. If you need that kind of performance why not just get a desktop for work and another for home, move your working files between the two and have dramatically better performance for the same money?

Re:Stability?

[drsmithy]

If you need that kind of performance why not just get a desktop for work and another for home, move your working files between the two and have dramatically better performance for the same money?

Because keeping two environments synchronised properly is a massive PITA.

Thats Cool but the real question is

[jammindice]

Does it run Vista?

Re:Thats Cool but the real question is

[Locklin]

It sure does, and so well that it can do an infinite loop a second faster than Linux!

Re:Close to superconduction?

[CookieOfFortune]

It really depends on the type of substances used, usually an alloy with the correct proportions, but I doubt they would stumble upon a superconducting combination on the chip. (Not that it would matter anyways, since the speed limitation is caused by the switching speed of the transistors).

Re:Zomg

[rbanffy]

No x86s in this space. IBM has POWER6 running at 5 GHz.

Re:Zomg

[hierophanta]

yes, yes we do.

i cant tell but is there an incredibly large whoosh goin over my head? (or just your head?). 6.5Ghz is faster than 3. And in other news six is afraid of 7, because 7, 8, 9

Anyone know

[Anonymous Coward]

what would happen if you cool down a CPU to temperatures where the CPU becomes super conductive?
Or it that even possible with doped/diffused Si? Would it still work as a semiconductor?
Would it give you even better benchmarks? Did someone already try?
Someone should... ;)

Metric ?

[DrYak]

I don't know what weird kind of units you are using in your part of the world. But the rest of the planet is using Celsius for everyday temperature measures and Kelvin for scientific measures (same step size, different zero).

And on our scale, absolute zero (0K) is -273C.

Thus -242C (aka 31K) is pretty legal and possible temperature. (Although maybe not a very common one outside university labs and mad overclocker's basements)

Now please stop using Réaumur scale and start using what everybody else is using around.

--

PS: I checked, -242Ré is indeed impossible on Réaumur scale - 0 K is -218Ré

Re:Metric ?

[wumpus188]

I don't know in what weird part of the world you are, but here on Earth -242C is not considered to be "everyday temperature".

Re:Metric ?

[Alastor187]

And on our scale, absolute zero (0K) is -273C.

You metricpeans would think -273C is cold. Our scale goes all the way down to -460F, now that is cold!

Re:Metric ?

[weirdcrashingnoises]

o yea, well a scale i just invented goes all the way down to -1,000,000 Y

what is this Y scale?

Y scale is a temperature scale defined as being -1,000,000 at 0 degrees K. and 0 degrees Y is defined as being 72 degrees F.

oo pwnd.

Re:Metric ?

[h4rm0ny]

My scale goes down to -11. -11 !

Re:Metric ?

[DrgnDancer]

He never said it was. In fact he mentioned that it was an extremely uncommon temperature here on Earth (not really possible outside of human engineered environments). He said Celsius was a common every day measurement and that -242C was a valid and legal temperature.

Re:Metric ?

[quintus_horatius]

Yeah, well MY scale goes up to 11.

200% is pretty decent

[rwa2]

Considering that chip is rated to run at 3Ghz and you can OC only around 5 - 15% at room temperature, I'm pretty impressed by >200%. Also that the chipset held up while the CPU was running that as well.

Wonder what kind of power requirements that would translate to... Current leak becomes a significant loss above 3Ghz (which is pretty much why no one really makes 4Ghz+ chips), do the low temperatures keep those leakages under control, or does it just keep the hemorrhaging from making the system unstable? Also would be interesting to see what kind of chiller you'd need to keep a constant supply of liquid N2 flowing...

Re:200% is pretty decent

[ChrisMaple]

Leakage is not a direct function of speed. Thin gates and short transistors leak, and they are necessary for high speeds. Higher voltage increases speed and dramatically increases leakage. Lower temperatures decrease leakage.

The primary effect of low temperature is to make FETs conduct better. This reduces switching time, so the CPU can be run faster.

Re:200% is pretty decent

[Pulzar]

Also that the chipset held up while the CPU was running that as well.

The chipset was running at its usual speed. You don't need to overclock the chipset to overclock the CPU -- that's only the case when the multiplier is locked and you have to get the FSB/HT running higher to get the CPU to run higher... and nobody uses those CPUs for serious overclocking.

Re:A cat has gotten my tongue

[hierophanta]

fair that, that chip went to 8Ghz (which is dam impressive). But it was Pentium 4 architecture, and we all know the P4's sacrificed actual throughput for silly clock speeds (as a marketing gimmick). i bet the 3dmark score on that wasnt more than 25k

3DMark?

[Sir_Lewk]

What the hell is a "3DMark"? It sounds about as objective as a bogomip.

On the summary and grammar....

[117]

During CES a group of overclocker's

a group of overclocker's what exactly? Is it just me or is the correct use of apostrophe's [sic] starting to become a lost art these days?

Re:On the summary and grammar....

[cffrost]

It's not just you; that shit drives me fucking crazy. I'd like to know why acronyms are so often subject to this type of abuse. Would any apostrophiliacs care to explain the logic/motivation behind this ? Is "DVDs" possessive, and "DVD's" plural? Please help me to understand why this textual equivalent of goatse is preferable.

DISCLAIMER: In order to avoid being moderated down, I hereby state the following, per Slashdot protocol: "I know I'll probably be modded down for this."

Light Distances

[SuperAndy]

What I think is really amazing about this is that at a clock speed of 6.5 GHz, each cycle takes around 15 nanoseconds (15 * 10^-9 seconds) to complete. In this time frame light can only travel around 5 cm. Electrical signals travel close to this speed themselves, so the limit of clock speeds is being reached, since the chip itself is on this same order of distance. It is around the point where one side of the chip will not be able to communicate with the other side in a single clock cycle.

Re:Light Distances

[diablovision]

You're off by two orders of magnitude. 6.5ghz is 153 picoseconds per cycle.

Re:Light Distances

[SuperAndy]

I am, my apologies. The end of a long day of physics. That gives a light-distance of 46 mm, or around 5cm. So I got the right final value, just dodgy working.

Re:Light Distances

[Orestesx]

With pipelining, a signal does not have to travel across the entire length of the chip in one clock cycle. In modern processors, there are always several (usually 10-20) instructions in process at any given moment.

Re:Light Distances

[Anonymous Coward]

Ummm, if you do the rest of the math, I think you'll see that the electrical signals propagate sufficiently slower than the speed of light so that they in fact can't cross the chip in a single cycle.

Also, I think if you run all the numbers, you'll find that in 45nm chips at 3GHz, a signal can't cross the entire chip in a single clock cycle. Or if it can, it can't actually be used (ie. it can't participate in logic or be stored in a flip-flop or latch).

This has been a well-known problem in micro-architecture for a few years now. And is a contributing factor for why we've gone to multiple cores instead of higher performing single cores.

Re:Light Distances

[diablovision]

I am not aware of any chips that have path lengths long enough to require multiple clock cycles to cross the entire chip. Clock distribution is a very complex problem, just getting a single cycle across the whole chip synchronized. "Pipelining" in the sense that you mean it doesn't make this problem any easier, since the pipeline stages aren't necessarily laid out sequentially anyway. Going superscalar means even more complexity.

Re:Light Distances

[drinkypoo]

OTOH the paths aren't direct either, you might have to make jumps even and even the lack of straightness of the paths has an effect on the time it takes signals to get from one part of the chip to another. The P4 has multiple "drive" stages in the pipeline to allow for signal propagation (but I would assume that this is based on the time it takes the gate to settle, and not the time it takes for the signal to get from point A to point B) which do nothing but wait around and increase the penalty for branch misprediction (which is why the P4 is awful.)

Real world results

[Enderandrew]

I'm more curious to see real world results. How well can you overclock this on air?

I just ordered the same proc, a 790GX mobo, and a 1 gig HD 4850 yesterday on the cheap. The cpu+mobo combo was $295, and the video card was $161.

Intel still has the top end market, but at these prices, I'm pretty happy with what AMD is offering.

Nobody knows

[Anonymous Coward]

"Although the local resistivity of semiconducting silicon in its standard crystalline form can be changed by many orders of magnitude by doping with elements, superconductivity has so far never been achieved. "

E. Bustarret et al, Nature 444, 465-468 (23 November 2006)

So it doesn't look like anyone will be trying any time soon.

cheaper than milk

[pbhj]

Liquid helium is alittle more expensive (about 2x AFAICT) per litre than Milk. Not sure if it has a full weeks shelf life under normal fridge temperatures though? How good are dewar flasks nowadays?

Re:A cat has gotten my tongue

[Sosarian]

Priced any high end Opterons lately?

Re:A cat has gotten my tongue

[binarylarry]

yeah but Intel's $200 chips (see Q6600) can compete with AMD's expensive stuff.

Normally, I'd be amazed that got marked troll... but this is Slashdot afterall.

I'm suprised it even worked

[w0mprat]

I'm highly surprised and intrigued the chip even worked at -242C (31K!) for a long time it was speculated in overclocking circles that weird things would happen to current silicon much below the temperature of liquid nitrogen. It does seem liquid helium has been tried a few times but this is the lowest reported temperature I have ever seen on a overclocked CPU. It might not mean much for people who don't care about overclocking but I think this is a significant achievement.

I'm also intrigued by the possibility this chip could have gone faster, it may have become bound by motherboard reference clock and multipliers at this speed. It's not uncommon for the motherboards ability to deliver current to become the limiting factor.
8ghz is reportedly the outright world record http://www.nordichardware.com/news,5505.html Although I think this was reset to 8.2ghz not long after.

Re:I'm suprised it even worked

[UDGags]

http://valid.canardpc.com/records.php

Re:I'm suprised it even worked

[Prof.Phreak]

I'm also curious whether (if) the chip generated -less- heat at those temperatures (?), due to some superconducting effect.

Re:A cat has gotten my tongue

[triffid_98]

Actually, the original plan for the P4 netburst architecture was to hit 8Ghz, but then we discovered the magic of Electromigration and why that was not such a great idea.

Re:A cat has gotten my tongue

[rezalas]

Ahem. You are wrong. http://i.i.com.com/cnwk.1d/i/bto/20090125/amd-opteron-3g-pricing-small-2.jpg

Celeron 300A

[Locke2005]

I made the same mistake. I bought 3 OEM Celeron 300A CPUs, based on widely published claims they could be overclocked to 450MHz. Only one of them could successfully run at 450MHz, and that required cranking the CPU voltage way up (that machine is still running and still crashing all the time, but then it is still running Windows 98 SE.) I suspect the dealer had gone through the batch and picked out all the good ones for himself. At this point, overclocking seems silly when you can wait a few months and get a CPU that will run the same speed without overclocking!

Re:Celeron 300A

[Bengie]

I bought a celeron 300a when they were popular. Dropped it on my new mobo, set fsb to 133, booted up the first time and it was 450mhz. never changed anything except thee fsb. You must've had back luck. I had 4 other friends do the same thing and all got 450mhz without tuning anything, BX chipset was teh sex

Re:Celeron 300A

[Airline_Sickness_Bag]

I bought two Celery 300A cpus - both overclocked to 450. Conversely, I had an Athlon 2800 that kept crashing until I underclocked it.

Re:Celeron 300A

[fava]

A lot of 300A's were actually 450's that passed as a 450, but due to demand were relabelled as 300's. So technically there were not being overclocked. They just were not being underclocked anymore.

Re:Celeron 300A

[operagost]

It was a 66 to 100 MHz bus speed adjustment, not 100 to 133. 4.5 x 66 to 4.5 x 100.

Re:Celeron 300A

[drinkypoo]

Overclocking has utility for a small slice of the population. It's like dd-wrt. If you're going to buy a device to run it, getting the right version can be a bitch (sometimes you end up only with -micro or something...) But if you just happen to have a device it will run on, or just happen across one at a flea market or something, then you can get dramatically more functionality out of the device without having to pay anything. Overclocking has pretty much the same utility; if you happened to get a processor which overclocks easily, then by all means overclock it. Otherwise, spend another twenty bucks. To be fair though, your last sentence is silly; overclockers are not people who want to wait.

Re:Celeron 300A

[toddestan]

That seems unlikely, since the 'Celeron-A' of the time was a different design than the Pentium II, and no model was ever released for a 100Mhz FSB. So the famous Celeron 300A wasn't a cut-down example of some other processor (unlike almost all other Celerons), it was just an unusually easy to overclock processor that happened to be a budget model.

Re:Celeron 300A

[default luser]

Yes, you're right. What he probably meant was the overclocking of Coppermine cores, which only required a bus increase from 100 to 133 MHz. A 600 or 650 MHz core was almost a guaranteed overclock to 133 Mhz bus, and hence they werre always sold-out.

Other great Intel overclockers of the time:

Deschutes core (.25 micron) Pentium II at 300 MHz. Since the first "official" Deschutes core release was at 333 MHz, it was a surprise when 300 Mhz versions of the core started appearing. Apparently, a lot of the new Deschutes 300 Mhz cores were marked-down 450 Mhz cores, and evev had the same 225 MHz L2 cache chips!

Celermine 533 or 566 overclocked to 800 or 850 Mhz. Sure, it wasn't as fast as a Pentium III at the same clock speed, but with the chip costing just $100, how could you say no?

Coming soon, to a mobile phone near you

[VorlonFog]

So you can watch all your postage-stamp-sized video and hear all your high-bitrate MP3 tracks.

Re:Close to superconduction?

[ChrisMaple]

Actually, at temperatures that low, speed-of-light limitations can be as significant as transistor switching speed.

Re:A cat has gotten my tongue

[Nimey]

They do have a $950 processor, the Opteron 2384.

Re:Zomg

[aliquis]

P4? Core2Duo E-series?

Re:A cat has gotten my tongue

[Nathrael]

I think someone here has to look up the definition of "troll"...just because you disagree does not mean somebody's a troll *eyeroll*. Oh, and feel free to cue the "Are you New Here" jokes...

Re:Zomg

[Anonymous Coward]

Yeah, only 6.5GHz too. Call me when it goes up to 11

Of course, only AMD's "Black Editions" permit overclocking in this way. As this is an uber-overclocked CPU, I guess you could ask "how much more black could this CPU be"?

And the answer is none. None more black.

Re:NASA Processors?

[ChrisMaple]

Deep space may be cold, but vacuum is a superb insulator. The chips can't be pushed hard without extensive and expensive heat sinks. Considerations on deep space probes are reliability and low power consumption, and there isn't a lot of need for speed. Reliability, radiation hardness, and low power consumption all have requirements that oppose speed.

Furthermore, since space probes take a long time to develop and use only very well established technology, they are using nearly-obsolete semiconductors by the time they're launched. They're really old when they get where they're going. It's not fast stuff by today's standards.

Re:Zomg

[ZarathustraDK]

People use liquid nitrogen to over clock a CPU, news at 11.

Yeah, someone should use bottled slashdot-sarcasm instead, the problem is avoiding turning the computer into Bose-Einstein condensate.

Re:Okay..

[cide1]

I generally don't let these types of things affect the CPU I use for work. I have found that in order for a system to be fast, all components much be equally matched. When the CPU is overclocked by a factor of 2, and the memory is not, the amount of time spent waiting on memory will increase significantly. If a designer knew the chip would be run at the higher speed, more cache would generally be included to make up for the disparity between CPU speed and memory speed. A good rule for buying new systems is to upgrade in two halves. I generally buy motherboard, RAM, CPU, and power supply at the same time for compatibility reasons. A year or two later, I will update my storage and video card. I buy a motherboard that supports the fastest memory made, I buy a lot of memory, and I buy a CPU that is at a point on the price to performance curve where spending more doesn't yield much more performance. In a year or two when software starts to actually use this capacity, Ill upgrade storage and video for a bit of a boast. Unfortunately, faster hard drives only make a bit of difference.

Obligatory

[smcdow]

Imagine a Beowul... oh, never mind.

Re:Obligatory

[The Grim Reefer2]

Imagine a Beowul... oh, never mind.

You'd probably need one to calculate the electric bill.

Re:A cat has gotten my tongue

[afidel]

And they are fast as hell as a DB server, and when the software running on them is in the range of $25K per 2 cores you'll gladly spend $1k per CPU for the fastest available processor =)

I heard

[mandark1967]

that it still only scored 4.7 on the Vista Performance Index...

Or in my offce...

[duragnulinux]

Thanks to Minnesota and the wonderful temperatures right now, really really bad windows, and being on the North-East side of the house and in the basement, I bet they could have got to 6.5GHz with air cooling in my office. The temp in there as we speak is 59.9F. It makes my server-cabinets interior temps run nice and cool anyway :)

Re:Zomg

[VagaStorm]

The liquid nitrogen looks cool, but sereously, is -230c the optimal runintemprature for a CPU?

Re:Zomg

[sFurbo]

I think you should be thinking in binary to get the joke...

Re:Zomg

[SleepingWaterBear]

Wow.

People use liquid nitrogen to over clock a CPU, news at 11.

Well, really they used liquid helium. When you use liquid helium (which has a boiling point of about 4.2K at 1 atmosphere), you're using the liquid nitrogen (boiling point of 77K) just to keep the liquid Helium cold longer. Using liquid nitrogen is sort of boring - you can store it in an insulated jug for a good long while even, but using liquid helium is, well, pretty damned cool!

Re:Zomg

[Cornflake917]

And in other news six is afraid of 7, because 7, 8, 9

Oh, that makes sense! I always thought 6 was just a big pussy.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Zomg

[rbanffy]

Sorry. I thought we were talking about the 5+ GHz space. It's not hard to imagine POWER7 going 6+ GHz without any liquid Helium just as much as POWER6 does 5 GHz.

3 GHz on x86 is routine, as far as you don't mind the extra heat output.

Four Ghz . . . . . .

[bogidu]

. . . . . ought to be enough for anybody! *ducks*

Re:Okay..

[beckerist]

Summary of the summary:
1. Chip cooled down to very cold temperatures.
2. Chip ran faster than it ever has before.
3. Intel creates a micro-fridge.
4. The world turns.

It's all relevant somehow, and news is news... *shrug*

no gravity - no convection

[junkgoof]

Outside the atmosphere you start to really need fans if you have air. Otherwise you get hot spots that just get hotter.

Re:no gravity - no convection

[raddan]

Convection is only one form of heat transfer. The other two are conduction and radition. Our sun's energy, for example, reaches us primarily via radiation.

I think the most interesting thing about NASA's use of computers is that they don't allow any dynamic memory allocation. That makes certain things easier (e.g., squashing bugs), and other things much, much harder (e.g., writing a useful program).

Re:no gravity - no convection

[junkgoof]

Agreed, but convection is used extensively for most other electronics. It leads to unusual problems (and different fan selection) when gravity is not available. Not necessarily a huge problem, but one that is easy to overlook until equipment is in place.

Re:Zomg

[Hurricane78]

Not without liquid nitrogenium.

Or do you mean all the cores together.

Well, I think GP did mean single-core speed without such tricks.

Ascii games

[IdahoEv]

Screw Crysis! I'm still waiting on the processor that can play the ascii-based Dwarf Fortress at a decent framerate.

Maybe all I needed is a little liquid helium...

"Really Bad Windows" isn't that Redundant?

[BigAssRat]

All you needed to say was "Windows" the "really really bad" is kind of redundant isn't it?

Re:"Really Bad Windows" isn't that Redundant?

[diqmay]

I believe he was referring to a framework of wood or metal that contains a glass windowpane and is built into a wall or roof to admit light or air. Also, I hear they afford nice vistas when placed properly.

Re:"Really Bad Windows" isn't that Redundant?

[duragnulinux]

Haha that's only too true...

Re:Zomg

[who knows my name]

using liquid helium is, well, pretty damned cool!

indeed, the liquid nitrogen works like an insulating jacket.

Re:Zomg

[Fumus]

Strangely, the result states that the CPU was running at 4481 MHz.

Re:Okay..

[amias]

and continues to consume more power

Re:Okay..

[TwistedSymmetry]

In a year or two when software starts to actually use this capacity, Ill upgrade storage and video for a bit of a boast. Unfortunately, faster hard drives only make a bit of difference.

Ah, so that's what it's actually for...

Toxic coolants

[Bazer]

I'm skeptical. Extreme overclocking requires copious amounts of dangerous coolants which are known to cause severe brain damage to the system's user. Side effects include: empty fridge, finding strangers sleeping in your tub and massive hangovers.

Re:Zomg

[bdraschk]

Go ask Baltar on that subject.

Re:A cat has gotten my tongue

[Nimey]

As of this afternoon, they /do/ make chips that expensive, and more:

http://techreport.com/discussions.x/16298

Their new top-of-the-line chip:
Opteron 8386 SE 8 sockets max 2.8GHz 105W $2,649

Re:Okay..

[shaitand]

'Unfortunately, faster hard drives only make a bit of difference.'

I suppose a 'faster' hard drive doesn't make a big difference compared with a 'fast' hard drive. But a fast hard drive compared with a slow hard drive makes a HUGE real world performance difference. Clunky and slow drives are the primary reason that laptops are so doggedly slow compared to desktops.

Of course 'speed' is defined by rpm's in this case, not throughput.

In Soviet Russia

[turkeyfeathers]

Every day is -232 degrees Celsius.

Re:Zomg

[aliquis]

http://en.wikipedia.org/wiki/Pentium_4
"Max CPU clock 1.3 GHz to 3.8 GHz"

http://www.slashgear.com/intel-35ghz-core-2-duo-e8700-quietly-introduced-2631907/
"Intel has quietly updated its Core 2 Duo range with a new processor, offering higher performance than most of its other dual-core CPUs. The 3.5GHz Core 2 Duo E8700 has 6MB of cache, a 1,333MHz front-side bus, and a 65W TDP, and is built using 45nm technology as with the other recent Wolfdale chips."

Re:Zomg

[aliquis]

Also AMD Phenom II 940 is 3 GHz.

Re:Zomg

[MrEd]

Yea, the poster must have gotten confused - they actually overclocked it to 6500+

Decisions, decisons.....

[earthforce_1]

Liquid helium cooling or should I go to the bother and expense of adding another CPU?

Re:Zomg

[dukeofurl01]

I believe so, as I write this from my Athlon 64X2 @3.2 GHz.

Re:Zomg

[Linknoid]

Q: Why don't jokes work in octal?
A: Because 7, 10, 11

Re:A cat has gotten my tongue

[et764]

You know, I'm not sure that's the model number I'd pick for modern CPU, since it's so similar to 80386.

That's where SSDs come in.

[__aailob1448]

Check out the X-25M. I predict at least one of your storage drives will be SSD within 365 days and all new systems, desktops and laptops will carry them within 3 with seek/read/write speeds that will put to shame today's top of the line 15RPM SCSI drives.

Now, all that remains is replacing Optical drives with new flash-based floppys (or USB keys) and the era of non-solid state devices in the computer will have ended.

Good riddance.

Re:Zomg

[Dolda2000]

That just seems so very far fetched, though, mixing number systems in one post without any hint. Isn't it more likely, in that case, that we should interpret "11" as "11 Hz"? I wish the OP could give his opinion on this matter, because I really don't get it either.

A lot of hype

[LS]

For an AMD advertisement... I cringe at the video editing and forced cheers throughout. Low temperature fluids have been used to do this before, big woop! When something qualitatively different happens, call me.

LS

AMD does not recommend...

[The Lord of Chaos]

AMD does not recommend pouring liquid nitrogen on your head or on your friend's head...

Unless, of course, your friend works at Intel.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Proper CPU-bound benchmarking

[MooglyGuy]

I ask this whenever I see a "such-and-such chip overclocked to this-and-that": WHY don't any of these groups ever use MAME, http://mamedev.org/ , as a test of CPU power? MAME is entirely reliant on your CPU's speed, and it can emulate quite a few games that bring even a decently overclocked Core 2 Duo to its knees. It is the perfect objective test of how fast a CPU is.