New Xeon CPU Hot and Underpowered

Kasracer writes "Web Sites The Register and GamePC received several of the new dual-core processors from Intel, dubbed 'Paxville', and ran a battery of tests on them. What did they find? From the article: 'There's no doubt about it, Intel's dual-core Xeons are their most power hungry Xeons to date ... Even when idling, two dual-core Xeons consume nearly 400W of power at any given time, which is amazingly high, even by Intel's standards ... their new dual-core chips (while powerful in their own right) simply are bested across the board by AMD's dual-core Opteron processors. Even worse, the Opterons typically perform much better while running at slower clock speeds and only having half the amount of on-die L2 cache to utilize.'"

Comments

oh my god.

[imsabbel]

The _last_ intel CPU a few days ago was already found hot and underpowered by gamePc.
That means this _new_ intel CPU really must be craptastic...

Re:oh my god.

[PhotoBoy]

Unless I'm missing some sarcasm here, the article is referring to the same GamePC article posted on /. the other day. The only new thing this time round is the review by the Register.

While I expect the benchmark results to be fairly accurate, news sites and games sites don't really strike me as the best for reviewing chips for the server market.

Re:oh my god.

[1u3hr]

The only new thing this time round is the review by the Register.

Which isn't new, it just quotes the GamePC article.

Re:oh my god.

[Tatarize]

RTFA.

It's not a new review. It's the Register commenting on the GamePC article.

From Article: "GamePC got its hands on some of the Xeons - code-named "Paxville" - and put the chips through a battery of tests."

The original poster didn't RTFA either, and made the same incorrect claim as you did.

Re:oh my god.

[kjots]

> While I expect the benchmark results to be fairly accurate, news sites and games sites don't really strike me as the best for reviewing chips for the server market.

That's a nice sentiment, but I would prefer to believe that Intel has lost it (It's just more entertaining this way!).

The King (*cough*intel*chough*) is Dead, Long Live the King (*sniff*amd*cough*)!

Re:oh my god.

[grape jelly]

Oh my god indeed! Anyone else here notice that gamepc has been hacked, displaying our old friend goatse?

Safe link (start at page 2)

Re:oh my god.

[Mr Z]

Page #2 is hacked as well... move on to page 3. Of course that might get hacked too...

Re:oh my god.

[Bloater]

This is not a reply to the parent, but just to let people know early as they read these comments that the GamePC site has been hacked and now shows the inside of some bloke's anus, so don't click.

Re:oh my god.

[MarkRose]

DANGER: The GamePC link has been replaced with goatse!

oh well

[Janek Kozicki]

Dell wants dualcore to sell, so they have it.

how long before they realize that it was a fatal move?

Re:oh well

[Spacejock]

When all their 350w power supplies melt.

Re:oh well

[wwwillem]

gives a new meaning to "burn in Dell hell" :)

AMD's new ad campaign

[Anonymous Coward]

Paris Hilton looking an Intel Inside sticker and saying "That's hot."

Re:AMD's new ad campaign

[Entropius]

Price/performance benchmarks?

Re:AMD's new ad campaign

[rm69990]

This would be targetted at normal consumers. Trust me, Paris Hilton saying "That's Hot" would mean way more to them than benchmarks.

Re:AMD's new ad campaign

[blake182]

Paris Hilton looking an Intel Inside sticker and saying "That's hot."

"Hot and underpowered" -- sounds like a good campaign for Paris herself...

My apologies to Paris, who I presume only "plays underpowered" for TV ratings. Snicker.

400W?

[danharan]

That's going to be the SUV of its category. Big, ugly, inefficient design catering to those who desperately need the ego boost.

It will hopefully die (yeah, bad pun) a very prompt death.

Re:400W?

[gringer]

All it needs is to be powered by gasoline, and it'll be sold quicker than a 6th generation iPod.

Re:400W?

[zerojoker]

Even though it might be a little offtopic (But hey, this is slashdot ;-) :
Unfortunately that's not correct. Cars in Germany are quite expensive. If you compare the prices in Germany with those in the US for typical german brands like BMW, Mercedes, Audi you will notice that in fact those cars are usually cheaper in the US.

Concerning SUVs: I never understood why those cars with heavily overpowered engines are so popular in the US. I mean a typcial car with a 1.6 16V engine will have around 110 hp. That's enough to drive speeds up to ~200km/h (I think around 125mp/h?!) so it's sufficient to drive on the autobahns. And you can actually drive these speeds!
In the US, what's the usual speed limit? Like 90 mp/h? Why do you need such a strong engine, then? You can't use it anyway!

Re:400W?

[mysticwhiskey]

SUV: It's not what you do with it, it's how big it appears!

Re:400W?

[FlopEJoe]

I think it would be cheeper to buy a t-shirt that says, "I have a small penis."

Re:400W?

[Dr. Evil]

People spend a long time on the roads commuting. SUV's are high off the ground and comfortable. This is important, many people spend more time in their SUV's than they do with their families. SUV's are also particularly comfortable if you're overweight.

The engines are good for acceleration, low and high end. It makes up in part for the automatic transmission. You need the automatic transmission so that you can eat your breakfast or speak on the cell phone while in traffic. It's also less frustrating in a traffic jam. You need the acceleration to be able to cut people off on the roads, and the height lets you see over sedans.

Finally, relatively speaking, gas is cheap in the U.S.

Re:400W?

[rocjoe71]

In the US, what's the usual speed limit? Like 90 mp/h?

It's 65mp/h! Please try not to laugh your head off the next time you're driving an autobahn.

The real irony is they put it the low speed limit to save lives and gasoline, then they go and build SUVs that guzzle gasoline and are prone to rolling over at high speeds.

Re:400W?

[the eric conspiracy]

SUV's are high off the ground and comfortable.

That high off the ground also makes SUVs very unsafe - the death rate in SUVs is higher than in any other 4 wheel vehicle type because of the high center of gravity leads to a lot of rollovers. Even a simple tire failure or striking a guard rail can cause these things to flip.

Then there is of course the toll on the occupants of other vehicles these things hit. Car occupants are 50% more likely to be killed in a SUV-car accident than in a car-car accident.

The problem is so bad that the decline in highway deaths in the US has hit a plateau. Death rates in SUVs is actually rising.

Analysis of the statistics gives estimates that SUVs lead to an additional 6000 deaths on US highways every year.

Re:400W?

[kabz]

I think everyone needs to have a play in a V8 Expedition, or a leather-lined Lincoln Towncar.

I rent on business and both these ended up costing about $280 for five days, or about $55/day (30 GBP).

Mmmm, V8 waftiness.

Re:400W?

[jo42]

> SUV's are also particularly comfortable if you're overweight.

Which covers like over 80% of Americans. This is the #1 reason why SUVs, pickup trucks and minivans are so popular - they need large vehicles to haul their fat arses around...

Re:400W?

[Dr. Evil]

It may be cyclic too... poor public transit --> more vehicles --> less walking to trains and busses --> less exercise --> more stress --> traffic congestion --> expanding cities --> worse public transit.

I noticed this cycle when I started driving... I read less, exercise less and have a higher base level of stress. I didn't realize that a base level of stress existed until I started driving and noticed how it affected my personality. I don't think anyone who started driving before they started working full time would notice this.

Increase fuel prices, housing prices and improve public transit and everyone might get thinner, more well read and more relaxed. There won't be riots because.... there isn't enough parking!

Re:400W?

[Anonymous Coward]

What type of car you have and drive is a big ego booster here in the US. This has been happening so long and has become such a normal part of life here that people do not even think about the actual purpose and real utility of a specific vehicle. People try to rationalize such vehicle purposes and honestly believe they can justify it to themselves. It really is comical to listen to educated people ramble on about needing a 350HP 9 passenger urban assult vehicle for their four person family that may travel to the beach in the summer and the four wheel drive because their area gets some snow. That same person will then laugh at the thought of getting a less powered minivan or a station wagon and try to justify why those vehicles would not be sufficent. Even if gas mileage and internal space were equal (which they typcially are not), there is the ego buster that the neighbors might laugh at your minivan and it is not cool. I get the biggest laugh from the huge 3/4 ton diesel trucks (Ford F350/GM 3500) with dual rear tires and bed liner that have NEVER had anyhting in the bed, never towed anything, and look like the came off the showroom floor even after being in use for years around town. Yes it can pull 17000 lbs and carry another 3000 lbs but it will NEVER happen by that person.
My commuter car has a 1.3 liter 60HP engine. I can drive and maintain myself in highway traffic up to about 80mph without revving to high. During light load traffic (which does not happen often at all in my area) I do get passed by some but I am still within the average bulk of traffic flow.

Re:400W?

[manno]

Thats not fair I've never owned an SUV, but there havw been times my honda accord couldn't cut the mustard and I wish I had one. This new P4, has no use what so ever.

Re:400W?

[grazzy]

I took the liberty to translate some of your sentences.

People spend a long time on the roads commuting. SUV's are high off the ground and comfortable. This is important, many people spend more time in their SUV's than they do with their families.

Some people should get a divorce.

SUV's are also particularly comfortable if you're overweight.

So fat people can fit their fat asses in the car.

The engines are good for acceleration, low and high end. It makes up in part for the automatic transmission.

Some people cant drive.

You need the automatic transmission so that you can eat your breakfast or speak on the cell phone while in traffic. It's also less frustrating in a traffic jam. You need the acceleration to be able to cut people off on the roads, and the height lets you see over sedans.

So the fat people can get fatter.

Finally, relatively speaking, gas is cheap in the U.S.

Our president has bombed enough arab states for the oil to be cheap, relatively speaking to the number of dead arabs.

Re:400W?

[WuphonsReach]

What type of car you have and drive is a big ego booster here in the US. This has been happening so long and has become such a normal part of life here that people do not even think about the actual purpose and real utility of a specific vehicle. People try to rationalize such vehicle purposes and honestly believe they can justify it to themselves.

/chuckle

I know the type.

My compromise was to get a hatchback (Ford Focus ZX3 or the like). You get a small car, but since it's a hatchback you can fit in some quite decent large-scale items that wouldn't be possible in a sedan.

The only bad part about a hatchback is that you often don't get the luxury of a trunk (for storing things like first aid kits, emergency roadside kit, jumper cables, etc.).

Re:400W?

[bhtooefr]

You're only counting gassers.

VW also has the Lupo 3L (3L/100km - about 78.4 US MPG), which has a 1.2L TDI diesel engine.

Now, if only we could get something like that here in the US... I've read that it sucks to drive, but considering that it beats even the most efficient hybrids on fuel economy... (although, heck, the 1.9L TDIs in the Golf and Jetta DO beat some hybrids - even the Prius has trouble keeping up...)

Re:400W?

[Jeremi]

but there havw been times my honda accord couldn't cut the mustard and I wish I had one

You're in luck -- with an Honda, there's no need to buy a more powerful car. Instead, just add decals until it's fast enough.

Re:400W?

[Cromac]

Concerning SUVs: I never understood why those cars with heavily overpowered engines are so popular in the US. I mean a typcial car with a 1.6 16V engine will have around 110 hp. That's enough to drive speeds up to ~200km/h (I think around 125mp/h?!) so it's sufficient to drive on the autobahns. And you can actually drive these speeds! In the US, what's the usual speed limit? Like 90 mp/h? Why do you need such a strong engine, then? You can't use it anyway!

Can you tow a camper, boat or trailer with your little 1.6L engine? How about hauling 2,000 lbs worth of gear? Can you fit 8 people comfortably in your VW? Believe it or not some of us actually use a full size SUV for something other than commuting in on a regular basis.

Just because you don't use your car for anything other than hauling yourself to work and back doesn't mean everyone else uses their vehicles the same way.

The max speed limit on the interstates in the US is 75 mph, generally 60 mph on highways/interstates that run through cities.

Re:400W?

[Cromac]

I think it would be cheeper to buy a t-shirt that says, "I have a small penis."

You know what's funny, the only people who bring up dick size are the ones whining about other people driving SUVs. Makes you wonder who is really the one trying to compensate for something.

Re:400W?

[Jeremi]

Our president has bombed enough arab states for the oil to be cheap

And the perverse part is that he uses the general tax funds to do the bombing, instead of raising gas taxes. So even if you don't needlessly drive a gas-guzzler, your tax dollars are subsidizing fuel-delivery costs for the people who do.

(actually, the above isn't quite true: in fact he used neither fund, but instead simply charged it to the national debt, while simultaneously lowering taxes. Try that strategy with your own credit cards some time and see where it gets you... but that's a different discussion)

Re:400W?

[Jeremi]

It's 65mp/h

Of course, on most (non-congested) highways 65 mp/h is considered the lower bound, not the upper...

Re:400W?

[Glonoinha]

There exists a point of diminishing returns.
At normal driving mileage (1000 miles per month is what all leased cars give you, so I will use that) the difference between 50mpg and 60mpg is about $100 a year in gas savings. That's about three gallons (eight dollars) a month.
The difference between 60mpg and 75mpg, for 1000 miles per month, is exactly the same (about three gallons (eight dollars) per month, or about $100 per year.)
Even going from 50mpg to 75mpg, which sounds Earth saving, is only worth the sum of those two (about 6 gallons a month, or $200 per year at today's outrageous prices.)

Re:400W?

[sznupi]

Funny, there are videos of crash test of two Renault vehicles. One is Megane (you'd have hard time seeing such small thing in US probably...), the other is...truck comparable to your 18 wheelers. Both head on head, 60 kmph (contrary to what most people think, most of crashes happens at around that speed - drivers usually brake a substantiall bit).
Result: the truck barely notices the crash, the engine compartment of Megane dissapears...BUT the safety cage of Megane is basically untouched!

Only one detail: the truck was newest model, with very, very low bumper, designed so the small car won't drive under the truck, but...well, bump off.

It's not the issue of weight. It's the issue of bad, unsafe design.

And that issue is also present in SUVs. Look how heigh their bumper is (oh, but it must look cool and bad...)

Re:400W?

[Jeremi]

If you took the time to research the facts and accepted truths that are different than your opinion, you'd know that the Fed has actually received MORE money because of the tax cuts. It's called the Laffer curve, google it.

You realize how arrogant and condescending you sound, don't you? I'm aware of the Laffer curve, and I'm also aware that it has two sides: the left side, where raising taxes increases revenue, and the right side, where raising taxes further decreases it. You seem pretty convinced that we stand on the right-hand side, but what makes you so sure of that? Increased revenues can just as easily be due to the normal economic cycle. Not to mention that it is an oversimplification: it's not just the total amount of taxation that affects GDP, but also who gets taxes and how. Some taxes will hurt the economy more than others, even if they cost their targets the same amount in dollar terms. In any case, I think it's just common sense that we shouldn't spend money we don't have -- so if the government thinks it is really important to spend more money on Iraq or whatnot, they need to raises taxes to cover that spending. If the public agrees, then the public will tolerate the extra taxes. If not, the public will elect different leaders. Hiding the costs by passing them on to future generations to pay back with interest is a lousy scam.

Re:400W?

[Yartrebo]

In my area (New York City), highways are typically 50-55 mph limit, with people doing about 75 when the roads are open (a rarity, bumper to bumper is more typical). Doesn't stop people from buying overweight SUVs. The speed limit on all other roads is 30 mph within city limits.

Re:400W?

[pyrrhonist]

The max speed limit on the interstates in the US is 75 mph, generally 60 mph on highways/interstates that run through cities.

Where do you drive? In the northeast the speed limits are usually between 55 and 65 mph on the interstate and between 45 and 55 mph through cities.

Re:400W?

[Yartrebo]

Yes, yes, and no.

I've seen campers and boats towed by subcompacts (1.2 - 1.6l, not exactly sure since US cars are typically measured by horsepower). While you cannot fit 2,000 lbs of stuff in the car itself, you could probably throw 2,000 lbs onto a trailer or into a camper (even if you cannot or it's too much trouble to have a trailer, you can do two trips with 1,000 lbs each time, for those rare cases - personally the most I've had to transport is about 1,000 lbs at any one time).

You can fit about 5 properly sized (not overweight) people into a subcompact, though not all that comfortably. If one is a kid, then you can increase that number to 6 (though it is illegal, not that we've ever been ticketed). That said, if your household has more than one vehicle (as most US households do), then you can use both vehicles for those rare cases the whole family goes out together.

Re:400W?

[bhtooefr]

There's also the fact that diesel engines can run on biodiesel without modifications.

How many "Future Fuel Vehicles" that can run on ethanol w/o modifications are there? How many of those are gas/electric hybrids?

Also, I've read that biodiesel gives much more energy per unit of energy put in obtaining it than ethanol. Take a look at this: http://www.mda.state.mn.us/ethanol/balance.html

So, yes, there's a point of diminishing returns if you're running on fossil fuels. However, on biofuels, the environmental impact is a lot better, going for diesel instead of a gas/electric hybrid. (And, yes, I know that a diesel/electric hybrid is possible.)

Re:400W?

[Quino]

You may actually have a use for your super-sized SUV, so don't get offended as this doesn't apply to you. But, I have to say, that I agree that of all the peple I know with an SUV don't actually need it. They don't off-road, they don't need to haul hay or even own a boat.

The fact is that most (not all,but an overwhelming majority) of people who buy SUVs buy them for other reasons: vanity, image, or whatever. This silliness is where this criticism (I think rightly) comes from.

I actually heard numbers from Jeep, where they were touting that buyers of Jeeps are much more likely to actually go off-road than any other SUV owner -- implying that it was the other guys who were fake poseurs. Of course, the numbers for Jeep owners was pathetic, just not as pathetic (like 7% of Jeep owners off-road, compared to 5% of other SUVs -- or something in that range).

I don't think that this criticism is completely ungrounded -- fact is most people simply don't *need* an SUV, they rationalize it. They are bought more for vanity than utility.

Besides, if you actually need cargo room and off-road capability, you'd be better served by a Subaru than just about any SUV -- they're horribly inneficient use of space (big outside, little inside). Car and driver some time ago actually tested several models of cars -- they ended up using the Subaru to pull an SUV that got stuck in snow, and the Subaru had more useful cargo room than all but one or two monster SUVs.

Re:400W?

[jonbryce]

Ford Focus a small car?

Try the Ford Ka. Or the Smart Car. They are small cars.

Re:400W?

[sznupi]

Sorry, it was from an analogue medium.

Re:400W?

[Glonoinha]

Biodiesel is great, until you run out of cats.

Re:400W?

[Arthur Dent]

The Fed has recd more money? You're joking right? Why don't you take your own advice and google it? Go over to the treasury web site at look at the summary of monthly receipts. Here's a brief summary:

Federal govt receipts:
1998 : 1.747 B
1999 : 1.857 B
2000 : 2.043 B (Last year of Clinton's presidency)
2001 : 1.994 B
2002 : 1.814 B
2003 : 1.795 B
2004 : 1.926 B

IANAE, but it looks to me that if the tax cuts had not been implemented, there wouldn't have been such a dramatic fall in revenues.... Laffer curve? I'm laughing so hard, I'm begining to cry....

Re:400W?

[bhtooefr]

Hey, you haven't seen how fast the little things multiply.

We won't run out of cats.

Besides, what's wrong with hemp biodiesel?

Re:400W?

[Fulcrum of Evil]

Can you tow a camper, boat or trailer with your little 1.6L engine? How about hauling 2,000 lbs worth of gear? Can you fit 8 people comfortably in your VW? Believe it or not some of us actually use a full size SUV for something other than commuting in on a regular basis.

And the point is that most people don't do any of that, or else do it rarely. How often do you drag 8 people around? Maybe that's in your weekly routine, but most people aren't like that. A lot of them are happy to drive 2 cars once a week instead of driving a huge SUV every day.

Re:400W?

[evilviper]

In the US, what's the usual speed limit? Like 90 mp/h? Why do you need such a strong engine, then? You can't use it anyway!

SUVs aren't cars, they are pick-up trucks with a built-in camper-shell and extra seats.

While the powerful engines may be able to get up to 125MPH under perfect conditions, you should try to do that while towing a car or other type of heavy trailer. Or perhaps with 2,000lb load of cement in the back.

Horse power is not just a measure of what speed it can get the frame up to, it's a measure of how much weight it can haul at speed.

Of course most SUVs will probably never see loads like that. The real reason most people want engines with serious power is for the acceleration it gives. Mostly it's people that don't know how to drive, and insist on going from 0-60 in 4 seconds from a stop-light (only to be forced to slam on their breaks as they reach the next red-light or backed-up line of cars, but I digress).

That said, there is a legitimate argument for good acceleration. Merging onto a road with a 75MPH speed limit, from a very short on-ramp, uphill, etc., can be rather dangerous if you don't have the ability to accelerate quickly when you need to. One man was killed around here about a week ago partly because of that.

Of course, the only proper way to end any conversation with an autobahn devotee is to mention Montana...

Re:400W?

[mikefe]

Analysis of the statistics gives estimates that SUVs lead to an additional 6000 deaths on US highways every year.

Low IQ tax.

Re:400W?

[Grishnakh]

Well I certainly don't see any subcompact drivers attempting to intimidate other people with their vehicles, or run them off the road, like I do with SUV drivers.

Re:400W?

[level_headed_midwest]

It's more about torque and acceleration than top speed. Yes, you can go very fast with the right transmission and aero work with a small engine. But it has no "grunt." Small engines are more efficient but to acheive power numbers of larger engines, they must really wind up their engines. You have to make the tradeoff in cam profile and cylinder bore and stroke for low-end torque or high-end power. And you'll NEVER make up the torque difference if you have a naturally-aspirated engine.

Sure, variable valve timing systems that vary cam profile and timing can help to widen the powerband. But, the physics still remain- more air and more fuel make more power. Why do you think so many small engines blown, either with a supercharger or turbo(s)? That's just a way to get more air and fuel into the cylinders. Having a larger displacement does that too.

American cars tend to be a bunch larger than European ones. A Ford Focus in the U.S. is the same one as in Europe. Here, it is considered a compact and weak with a 2.0L 130-hp engine. In Europe, it is considered a regular-sized sedan with okay power. A normal-sized car here would be a 240-hp Accord or a 300-hp Mustang GT. Certainly a difference! And the midsize American cars weigh about 500-700 pounds more than the Focus. This requires more engine- cylinders and horsepower- just to get the same performance. But the Focus doesn't get the nickname "Slow-cus" here for no good reason- that Accord will eat its pants and the Mustang will eat BOTH for lunch. More mass + faster = much more engine needed.

BTW, the speed limits on regular 2-lane highways is 55 mph, which is 88 kph. 90 miles an hour in a 55 zone is about a $400 ticket, exc. in Illinois, where it is a grand. The interstates are 70 mph here (112 kph) in MO and 65-75 mph (about 105-120 kph) elsewhere.

Re:400W?

[Beyond_GoodandEvil]

You're kidding right? Looking at that data seems like as the market went down, govt. revenue went down. Oh my! Tax revenue falling as economic activity slows.

Re:400W?

[level_headed_midwest]

I drive an '02 Escape, which I believe is sold in Europe, too. Mine is a V6 with a four-speed auto- no manual with the V6 because the engine is a tight fit as is and an extra gear wouldn't help. It sits a little higher than a car but weighs no more than a midsize car (~3300 lbs.) and gets about 23 mpg highway, 18-19 or so stop-and-go. The V6 has a decent low-end and a good top-end. I wish it was a stick, though :( I can and do haul a bunch of stuff around in the back. It isn't that big of a vehicle, but with the seats flat, it holds a dorm room's worth of stuff. I have had to make numerous runs back and forth with it full to not only move me, but people who can't fit more than a laptop and a laundry basket in their Civics. Sure, they get a little over 30 mpg, but a handful of bucks a tank isn't all that much when the insurance on ANY car costs thousands a year (if you're a male under 25, perfect record notwithstanding!) It's about right for what I need, although a Miata would be a lot of fun. But it's kind of hard to stick too much in one of those...

Re:400W?

[Gwyn_232]

Until you eventually have one of these (Honda NSX)

Re:400W?

[Lucractius]

Im Sorry. What is this....Montana You Speak of?

Re:400W?

[buraianto]

Ah yes. 2000. The year of the bubble. If you want to credit clinton with the increase in taxes, credit him with the .Com bubble, too.

Re:400W?

[Dr. Evil]

I drove a Civic hatchback for a few years until I was rear-ended by an SUV.

I could carry 8' lengths of lumber in that car by folding the passenger seat back and still close the hatch, then load in a floor sander and strap two bicyles on the back... and if I needed more storage... I'd rent.

No doubt there are people who need the SUVs. I'm just saying that the Civic is wonderfully designed and immensely practical. I was shocked once to realize that a friend's SUV couldn't seat more than 4 people... My Civic could seat 5.

Re:400W?

[metallic]

And that issue is also present in SUVs. Look how heigh their bumper is (oh, but it must look cool and bad...)

If you have ever been on a deer lease in an SUV scouting out the property, you'd quickly realize that ground clearance at the front and rear of the vehicle is a must.

Yes, contrary to what most people on /. might think, some of us actually use these vehicles for what they were designed for.

Re:400W?

[sznupi]

Exactly...some.

And also I'd suspect that if someone really uses those vehicles how they should be used, he isn't as likelly to have accident as the typical person who buys SUV.

Re:400W?

[pipingguy]

Just because you don't use your car for anything other than hauling yourself to work and back doesn't mean everyone else uses their vehicles the same way.

I think his/her point is that the whole automobile horsepower race (one Hyundai now offers 235HP, for example) is bumping up the minimum power required and most of this "need" is based on perception of status.

It seems to me that the overall trend lately is towards more engine power that cannot even be used on non-Autobahn roads. It's just posing or maybe winning the stoplight-to-stoplight races.

And I'm not immune to this love of cars or performance either. Yesterday I saw a Ferrari 348 and was transfixed by it's design and sound, it's a beautiful machine.

Re:400W?

[orderb13]

Not many vans have 4 wheel drive and also they don't handle as well off road, since they aren't designed for it.

Re:400W?

[Cromac]

The max speed limit on the interstates in the US is 75 mph, generally 60 mph on highways/interstates that run through cities.

Where do you drive? In the northeast the speed limits are usually between 55 and 65 mph on the interstate and between 45 and 55 mph through cities.

In the NW, specifically Seattle area, it's 60 on the freeways in town and 70 just outside. In Idaho the speedlimit is 75 on the Interstate ouside of urban areas.

Re:400W?

[manno]

man I wish I read this sooner, damn funny.

Heat

[titla1k]

So they're a little bit hot, I see that as a good thing. Now you can cook breakfast on your case, without ever having to move!

Re:Heat

[Velox_SwiftFox]

So they're a little bit hot, I see that as a good thing. Now you can cook breakfast on your case, without ever having to move!

That's nice, Intel could use reasons their power consumption might be considered a good thing.

If they get a good enough set of them, perhaps they can adopt a more descriptive name for the CPU, like "Oxyacetylon".

Re:Heat

[VolciMaster]

Now you can cook breakfast on your case, without ever having to move

Yeah, the original Pentiums did require you to remove the side first. :)

Kinda makes you wonder about the whole 'low-power' thing that supposedly attracted Apple, though, doesn't it?

Re:Heat

[reboskyz24]

Realizing the folly in this decision, the majority of remaining 64-bit Optys will wind up in laptops marketed for college kids who will need both a computer and a hot plate in their dorms.

Re:Heat

[odaiwai]

Apple are going to go with Pentium-M chips. Probably dual core, 64 bit Pentium-M chips. At 35W per core.

Those things will make AMD look sick. *IF* Intel can deliver.

And if Intel can't deliver, Oh Boy are we going to see a mega-pissed off Steve Jobs.

Re:Heat

[pantherace]

70W per chip then? Nothing to be impressed about.

AMD will not have trouble keeping up with that, because their current DESKTOP chips are using 78-86W, under one of the worst cases actually observed, and 68W when running Doom3 (Athlon 64 X2 4800) Also note that the rating is 108W, but that AMD's and Intel's TDPs are calculated differently.

Source1: http://www.lostcircuits.com/cpu/amd_x2/12.shtml (Next page is the Doom3 result.)

Re:Heat

[mikefe]

I'm sure at least one service center has received a computer that "fried".

Food that is...

Correction

[Anonymous Coward]

The Register didn't test anything. They're just providing a short and more to the point commentary on the GamePC review, who appear to have adopted a more diplomatic tone so as to receive Free Crap To Review in the future as well.

These things are apparent when you Read The Fucking Article, you know? The thing you're supposed to do before submitting? Gah.

Re:Correction

[hayden]

The Register has a long history of telling it how it is. They called complete crap on WAP. They called complete crap on Itanium (Itanic as they called it). Also, they're British. As Winston Churchill said "They are the only people who like to be told how bad things are, who like to be told the worst."*

That makes their opinion on what GamePC said valid. GamePC has a vested interest in not pissing off Intel too much. They publish the numbers and then say "but it's all good" so Intel will send them their next generation of procesors.

* Except when they fluke a win in the cricket.

This should change

[ottffssent]

The situation (Intel's, not necessarily the difference) should change RSN when Intel's 65nm process comes online. Looking at the huge lead AMD has right now, I don't see how Intel can beat them in both power and performance anytime soon unless they're willing to add a few hundred more pins to their sockets to accomodate on-die memory controller(s). I doubt Intel will do that. I also doubt they'll come back a bit from the 150W/CPU these Paxville chips crank out, so they'll be using the process headroom for higher clocks and/or larger caches.

Ultimately an on-die memory controller is the only way to bridge the increasingly large gap between the CPU and RAM. Intel's managed amazingly low latencies to RAM given that they've got an entire extra bus and chip to run through, but they're still ~50% higher than AMD's. The netburst architecture was supposed to be insensitive to RAM latencies but Intel is not keeping up in the bandwidth department either, and it's clear that these CPUs are suffering from a lack of RAM bandwidth (twice the processing power per chip, but no increase in bandwidth).

Re:This should change

[ErikZ]

The situation (Intel's, not necessarily the difference) should change RSN when Intel's 65nm process comes online.

Yeah, but Intel has been depending on technical solutions to cover for their designing flaws. AMD can use yesterday's cheaper fab technology to achieve the same performance through superior design.

Eventually, Intel won't be able to make a smaller die, and then they'll have to go back to the drawing board and start over to make something that can compete in the marketplace.

I'm assuming that lithographic CPUs are not going to be replaced though some revolutionary new processor design. Organic CPUs, Quantum tech, or whatever.
Because then Intel would end up being top dog again because only they would have the money to throw at R&D and the new fabs.

Re:This should change

[PingPongBoy]

AMD developed their CPUs to beat Intel on a clock cycle for clock cycle basis. But isn't it about time both companies show new architectural wonders? Of course, they don't want to quit the market immediately after all the investment to make dual core processors running almost 4 GHz but Intel is reaching the limits of power consumption so what else is coming?

Re:This should change

[Anonymous Coward]

I'll give you a quick hint: the K8 (I.e. Opterons and Athlon 64 and X2 series) is a freaking marvel of technology. The dual-core variants even more so.

Intel's dual core chips are just two chips pasted on the same die. Take a look at a proper description of an X2 some time -- there's cache sharing, an inter-cache on-chip hypertransport bus and all that nice shit that Intel just doesn't have. It's not far-fetched to say that at the moment, speaking from a microarchitecture point of view, AMD is a generation ahead of Intel.

Re:This should change

[ottffssent]

It's less a matter of design flaws as marketing flaws, I think. The P4 sucks in a lot of ways, but it's also very well-done in a lot of ways. The big problem is not so much netburst as it is Intel's inertia. Intel's working on a way to transition their product matrix from netburst on the desktop and in server space to something based on the pentium-M. The faster pentium-Ms beat the fastest P4s hands down in most benchmarks, but there are no good desktop chipsets for it yet, etc. Eventually Intel will release a dual-core P-M-based design, hopefully with an on-die memory controller, and then the Opteron will finally have some real competition.

Even if Intel eventually hits a process wall, they'll still be able to rest on their huge manufacturing capacity. For the past 5ish years, Intel has been building bigger chips than AMD, mostly in terms of L2 cache. It's not necessarily the best way to improve performance, but it's fairly easy, and leverages Intel's manufacturing strengths. Intel can afford to crank out dies 50-100% larger than an equivalent AMD die, and still make money at it, and still not run out of capacity. The reason Intel has been having shortages of late in the chipset arena is not lack of capacity but bad capacity management. They mis-read the market 3ish months ago and are paying for it now.

Anyway, here's hoping AMD's 65nm transition goes as smoothly as their 90nm one.

Re:This should change

[Anonymous Coward]

It's less a matter of design flaws as marketing flaws, I think. The P4 sucks in a lot of ways, but it's also very well-done in a lot of ways. The big problem is not so much netburst as it is Intel's inertia. Intel's working on a way to transition their product matrix from netburst on the desktop and in server space to something based on the pentium-M.

The P4 succeeded because of marketing. It's the "more MHz is better" myth, which was eventually debunked by AMD. My hunch is that the engineers knew better, but management wanted to create their own reality. A big chunck of the market bought into it.

Technically, AMD's design is much better than Intel's. The P4 itself was an aberration. Why else did Intel look to older P3/P-M architecture?

Re:This should change

[KDR_11k]

Rate != latency. And with latency, the less the better.

Re:This should change

[timeOday]

so they'll be using the process headroom for higher clocks and/or larger caches.

Haven't we already hit the point of diminishing returns on bigger caches? It's just a simple way to use up transistors. I think I rather have more cores instead.

Re:This should change

[I Like Pudding]

No. That's like saying we've hit the point of diminishing returns on RAM.

Re:This should change

[bhtooefr]

They HAVE admitted that it failed.

That's why they're replacing NetBurst (and P6, the old Pentium Pro design that the Pentium M is a variant of, for that matter) RSN, with Merom, Conroe, and (crap, forgot the server core).

The new designs are going to be inspired by the Pentium M, though.

I just hope that AMD has something up their sleeve, like K9, because Intel's Fred architecture (hey, that's what Tech Report is calling it) looks like it'll kick K8's ass. And remember when AMD was only for budget PCs, and the company wasn't doing well? It meant that Intel's stuff was EXPENSIVE, because they didn't have competition at the high-end.

Re:This should change

[Chandon Seldon]

Current IBM server CPUs have 36 megs of cache. Some desktop applications are significantly more cache limited than webserving (for example) is.

Re:This should change

[timeOday]

No. That's like saying we've hit the point of diminishing returns on RAM.

Then why are P4EE benchmarks always so unimpressive?

Re:This should change

[AvitarX]

For desktop we certainly have.

the return on 256MB when you have 256MB ram is huge,

the return when you have 512MB ram is not near as large, but still big.

On database servers that point is largly dependant on the database size, but for many uses has been hit too.

an example of design vs size in cache is the overclocked Celeron 300A vs the competing pentiums (speed being part of design).

for many application 512KB cache to 1MB is a small improvement, 0KB-512KB is tremendous.

Re:This should change

[waferhead]

Paxville _IS_ 65nM, it's still a space heater.

A 8 core box would almost hit the legal limit for electric space heater size in the US
(1500W)

I'm a longtime AMD fanboi, but I expected better from Intel, having worked there for awhile as a vendor.

Intel has no shortage of VERY bright folk, so IMHO management should all be shot.

Re:This should change

[I Like Pudding]

Yes, you can fill up your machine with RAM to the point where RAM isn't an issue...today. A few years ago 128MB was the standard and 256 was considered good. Currently, I wouldn't even build a computer with less than 512. 1 gig min if you are running games. 1 gig if you are running OSX. The machine I'm writing this on has 2gigs. My development laptop from work has 768 - I am constantly swapping on that. The amount of RAM you will find useful will always increase as time goes on, and will never, ever go down. Same with cache.

If I could replace all my main memory with cache, I'd do it in a second, because the cache is stupidly faster and performance would increase by a factor of stupid. The problem there is that your die will end up the size of a pizza, and the chips will cost 20k each. It's not that you've hit a point of diminishing returns with cache, it's that you can't fit the entire memory footprint of an app inside 2m, 4m, or even 8m, which is the point at which everything REALLY gets sped up. That is the point where there is no additional value from cache.

Random thought: Additional unused memory can serve as a page cache for the hard drive. I sure don't mind that at all.

Re:This should change

[I Like Pudding]

P4EEs outperformed their non-EE bretheren, especially with heavily cache-constained apps. Also, the early EEs had L3 cache, which is slower than L2. I think the current high end one has 2MB of L2 and no L3. Anyways, the EE came about because the Athlons were kicking their asses on the high end. Simple solution: add more cache to speed things up. Just a case of Intel using their vast production capacity to plug holes in their crap architecture.

Re:This should change

[corngrower]

I'm not sure if it's the case right now, but back 5 or 6 years ago, AMD's chip development engineers were using design automation tools substantially superior to those that Intel's engineers were using, allowing AMD's engineers to be much more productive. Intel's management, of course, didn't envision a 64 bit future for the x86 architecture.

Re:This should change

[Grishnakh]

Exactly which tools were these?

Re:This should change

[Anonymous Coward]

The P4 succeeded because of marketing. It's the "more MHz is better" myth, which was eventually debunked by AMD. My hunch is that the engineers knew better, but management wanted to create their own reality.

As an engineer at Intel, I have to disagree with this. I don't know any engineers in the processors division, but I do know some in chipsets, and they fully bought into the MHz myth. "We're winning the Megahertz wars!" was a common line heard in business meetings back in the P4 heyday, and when I talked to chipset engineers about AMD vs. Intel architecture, they believed it all too. If I pointed out how AMD's CPUs outperformed ours at a much lower clockspeed, they'd just say how the P4 could potentially be clocked so much higher, so AMD would just hit a brick wall where they couldn't clock any higher, and Intel's chips could be clocked up to 10 GHz or so.

Of course, we all know now that was pretty stupid, since they didn't even hit 4 GHz before running into a thermal barrier.

Don't fool yourselves about Intel engineers. They're not that smart or special. Most of them just sit around every day writing HDL code, writing unit tests, and other boring stuff. The high-level architects are just these same dull engineers who have stuck around long enough to get promoted to that level; they're shooting from the hip too. Really smart engineers don't generally work at places like Intel. Considering how bad pay is for engineers in general, really smart people don't stick around at big companies for very long, and go start their own businesses, become consultants, or go into a different field altogether.

Re:This should change

[timeOday]

It's not that you've hit a point of diminishing returns with cache, it's that you can't fit the entire memory footprint of an app inside 2m, 4m, or even 8m, which is the point at which everything REALLY gets sped up. That is the point where there is no additional value from cache.

Diminishing returns doesn't mean no returns. But the fact is that the first meg of cache helps A LOT, the second quite a bit, the third somewhat... and so on. Bumping from 8 to 10 would help only a little, yet those 2 megs would cost just as much as the first two (which made such a huge difference). That's what diminishing returns means.

As for system RAM, if you haven't hit the point of diminishing returns, you should add more, because it isn't that expensive.

Re:This should change

[william_w_bush]

Ultimately an on-die memory controller is the only way to bridge the increasingly large gap between the CPU and RAM. Intel's managed amazingly low latencies to RAM given that they've got an entire extra bus and chip to run through, but they're still ~50% higher than AMD's. The netburst architecture was supposed to be insensitive to RAM latencies but Intel is not keeping up in the bandwidth department either, and it's clear that these CPUs are suffering from a lack of RAM bandwidth (twice the processing power per chip, but no increase in bandwidth).

Good call, but there is one thing missing. FBDRAM. The limiting factor in cpu/ram communication for intels is the fact that there are essentially 2 busses between the cpu and the actual chips. With fbd, you can have an 8x pcie link to each dimm, give the processors a 16-32x link, which is low pin and easy to implement (this is easier in hyperxport, as shown by amd btw), and the dram handles its own refresh, timings, read/writes, commands, as an autonomous unit. Even if next year we go DDR3 or QDR or MRAM or RDRAM, you don't really have to change the sockets, and nobody outside of the ram chip knows the difference. Ram will finally be as easy to upgrade as video cards (ever hear someone ask for ram in a computer store? its like getting a diagnosis for brain cancer, takes a lot of tests).

AMD using ht for the chip-chip bus was pure genius, moreso because they managed to pull it off. Now all support logic can be made to a standard, and configured by the motherboard manufacturer, instead of prescribed from on high by intel. Now AMD will have to pull the same rabbit out of a hat with PCI express, for their next socket, and however they plan to treat FBD. Should be fun.

Re:This should change

[Lucractius]

noooooooo.... No K9 cpus damn it... Dog hair and sensitive electronics dont mix.

Hopefuly AMD skip the K9 series and rebrand somehting snazzy with a title like the " K-X " series

Re:This should change

[corngrower]

Which tools?

AMD was using tools from Synopsis. A hdl simulator along with a graphical front end called VirSim. Intel had their own in-house developed tool set. (They may have been using a commercial simulator, I don't know.) I'm sure the tools that both companies are using have advanced quite a bit since then.

The VirSim front end allowed you to run multiple simulators at the same time. (Some simulators were faster, others would have been more 'true' to the actual hardware.) VirSim also let you define sets of break points, so you could automatically set a whole bunch of break points with a click of the mouse. If you've used something like Model Sim, VirSim was similar (but more powerful). You could set a breakpoint on a certain instance (of an object) executing particular source statement, or when some expression evaluated true.

You can collect data from a long simulation run, and then use VirSim to analyze the data 'offline' as well. The data files generated tend to be rather huge (we're talking hundreds of megabytes or event gigabytes) and VirSim would allow engineers to quickly see what they needed to see from this mountain of data.

You might think of VirSim as something like your software debugger on steroids.

400w?!

[Anonymous Coward]

Sweet zombie Jesus, the heatsink must have fins like a '57 DeSoto! At 400w and using it 24/7 will make the cops think you're running a grow-op.

Re:400w?!

[Midnight Thunder]

Now I understand why my local grow-op is labelled with "Xeon inside".

Re:And Apple wants to replace PPCs with *THAT* ...

[FidelCatsro]

Apple will most likely be using Pentium M's, as they are currently planning on phasing in the Intel chips from the ground up, whilst leaving the Power Macs to run on the G5's for the moment IIRC.
So there is a good chance they already know that Intel has something far better in store for them to use in 2007/2008

This new xeon Chip is sucking down more juice than the three macs I have In my house , It is 100W less than my server and PC

Note to self

[ChrisMroz]

Rules of successful businesses. 1.) Listen to consumer 2.) Don't lie to consumer 3.) Make decisions based on companies largest comodity, not companies largest stock holders 4.) Don't ignore competition

Re:Note to self

[neomunk]

You're apparently not American (well, United Statesian), cause around here, that's just not the way things are done. Mind you, it SHOULD be the way things are done (well put and concise BTW). It's not though, especially that number 2 point. If American buisinesses started not lying to people, the whole world would rumble, and if the American government followed suit, well, processor comparison would be the last thing we'd be talking about for a few decades. Just MHO.

Re:Not terrible...

[strider44]

So you're saing having a chip that uses 50% more power than its direct competitor and doesn't even beat it in a single benchmark isn't terrible? The Opterons beat the Paxvilles in every single test. Are you reading the same articles I am?

Prescott?

[Dogtanian]

(Disclaimer: I haven't read the entire article in depth yet, and I don't know **** about chip design)

I assume that this chip is basically derived from the Prescott P4, which (in spite of all the hype) has been considered by some commentators to be a dead-end in chip design and a mistake on the part of Intel.

There are rumours that Intel are now using the Pentium M (ironically, a chip which supposedly owes at least as much to the Pentium III design as to the Pentium 4(*)) as their "reference" design instead of Prescott. Bearing in mind that the Pentium M has been praised for providing performance approaching that of the P4s with *much* smaller power requirements and minimal cooling needed, this wouldn't surprise me. In fact, I've read several articles (including one via /.) promoting the Pentium M as a desktop chip (IIRC passive cooling was possible with a reasonably-sized desktop heatsink).

(*) And I don't know if that was the previous-gen P4; it's been said by some that Prescott was different enough to warrant the "Pentium 5" name. At any rate, the Pentium M isn't simply a power-efficient Pentium 4.

Re:Prescott?

[cr@ckwhore]

"Pentium 5"?? HAHAHAHA

Seriously, Intel should start thinking about a new name before the next version bump.

Re:Prescott?

[Dogtanian]

"Pentium 5"?? HAHAHAHA

Seriously, Intel should start thinking about a new name before the next version bump.


Kinda silly when you think about the latin root; but you knew what I meant. Anyhow, is the use of the name "Pentium D" replacing that of "Pentium 4"? (Obviously they won't use this for the single cores, but in the not-so-distant future all Intel's x86 chips will likely be multiple core anyway).

Strikes me that after they get past a certain point, most companies seem to like to move away from version numbers (as in films). "Pentium D" may indicate that Intel are doing this.

I don't think they'll ditch the Pentium name altogether though, whether you think it's justified or not. It's pretty well known, and if they change it there's always the risk of people thinking it's a new and incompatible chip.

Okay; perhaps not, most people don't know enough about CPUs to *mistakenly* demand a Pentium because they think it's x86 compatible and refuse a "Bogium" or whatever because they think it's not (a la "Itanium"). In fact, most people will simply want a "Pentium" if they ask by name, or something that'll run their spyware-infested smiley apps if not.

Re:Prescott?

[I Like Pudding]

Pentium 3 would be closer to the truth.

Re:Prescott?

[Lonewolf666]

Okay; perhaps not, most people don't know enough about CPUs to *mistakenly* demand a Pentium because they think it's x86 compatible and refuse a "Bogium" or whatever because they think it's not (a la "Itanium").
Actually, the people who refused to buy Itaniums on that basis were well informed. The native instruction set if the Itanium is NOT x86 compatible. And while the Itanium could emulate the x86, its performance in that mode was inferior even to that of a (far cheaper) Pentium III.

Re:Prescott?

[Dogtanian]

Yeah, *I* know that the Itanium isn't x86 compatible. Anyone likely to be buying one is going to know that.

I was saying that Joe Public not buying a (hypothetical) next-gen x86 CPU from Intel called "Bogium" because they mistakenly think it's not x86-compatible (in the same way that the Itanium genuinely *isn't*) is an unlikely situation.

People are likely to buy positively on the "Pentium" name; but if it isn't called a Pentium, all they're going to care about is whether it runs Windows and their favourite apps or not.

It was a crap example anyway.

Re:And Apple wants to replace PPCs with *THAT* ...

[hector_uk]

you could run 20 mac mini's and consume about the same amount of power, now if it were utilised what would be faster 20x 1.42GHz G4's or four 2.8GHz xeons?

Intel catching up with 65 nm?

[miffo.swe]

I cant see how Intel is going to catch up without a redraw of the whole CPU structure and mem controller. Simply going for smaller die wont solve the performance problem. I really think Michael Dell was wrong in thinking that 65 nm is going to get Intel in par with AMD. This is really and interesting moment, will dell stick to Intel and deliver less performance than HP/Intel/Sun etc? At some stage they will have to start selling AMD and i suspect it will be a horrific moment for Dell. To date they always state that they use Intel because they are better but really how long can they put that charade up?

Re:Intel catching up with 65 nm?

[PingPongBoy]

I wonder what the problem is with Dell and AMD. A Dell system with AMD inside might be referred to in short as D-AMD.

Re:Intel catching up with 65 nm?

[miffo.swe]

I dont think the problem is between AMD and DELL. Id rather believe the incentive for DELL is big juicy rebates on Intel CPU's. Its really something the DOJ and EU should take a look at.

Re:Intel catching up with 65 nm?

[akihabara]

If I were AMD I'd tell Dell to go fuck themselves; AMD isn't short of customers; Dell is long shitty products.

Re:Intel catching up with 65 nm?

[Bishop]

I suspect the opposite: AMD won't be bullied by Dell. In the past AMD may have been willing to cut a deal, but Dell would have been correct that the volume and performance weren't there. HP is Intel's Itanium partner and the biggest (only?) supplier of Itanic systems and they don't seem to have a problem with selling AMD systems. Given HP's recently deposed CEO, if Intel was offering killer deals I would expect HP to be Intel only. It could also be that Michael Dell is a stubborn bastard. He wouldn't be the first CEO to say something stupid and not admit to being wrong.

Re:Intel catching up with 65 nm?

[John Hasler]

> Id rather believe the incentive for DELL is big juicy rebates on
> Intel CPU's.

Rebates which would vanish were Dell to buy any AMD parts.

Re:Not terrible...

[masklinn]

Yup, power consumption figures are for the whole system.

But using roughtly equivalent systems (same power supply, an Antec TruePower so you're looking at an "at or above" 75% efficiency power supply when drawing 200+W from it, a single Raptor 74Gb and a Plextor DVD-RW drive), AMD's Opteron system top at 235W idle (for the 2.8GHz Opteron box) without using PowerNow's power management system (GamePC reports that the total power consumption @idle fell to around 170W using PowerNow) while Intel's 2.8GHz Xeon system chews through 390W idle...

Re:Nonsense

[Anonymous Coward]

You're right thta they didn't measure the power of just the CPU, but they also didn't benchmark just the CPU, because you can't run the benchmarks with out memory, buses, IO etc. Duh. Besides, CPU power is not an apples to apples comparison - one CPU integrates memory controllers and high-speed CPU-CPU connections, the CPUs require different types of RAM...

The power the system draws is the more relevant to the system owner than the CPU power is. They used comparible systems.

400 Watts idling?

[Bueller_007]

I realize that 400 W while idling isn't typical of Intel processors, but still, didn't Apple choose Intel because of their supposedly low power consumption?

Re:400 Watts idling?

[sl4shd0rk]

> didn't Apple choose Intel because of their supposedly low power consumption?

No, apple chose intel because they sold their soul to Wintel (tm).
It's not even a choice really, but more of a business requirement.

http://www.gcn.com/16_24/news/32421-1.html
http://www.forbes.com/2003/01/07/cx_ah_0107apple.h tml

Re:400 Watts idling?

[rlthomps-1]

You might have a point if this was a laptop chip, but it's not. You should have stopped after "isn't typical of Intel processors"

Re:400 Watts idling?

[Jozer99]

Isn't the 400W while idling for TWO Xeons (4 cores) and the rest of the system? That works out to 100W per core, which isn't bad for a modern CPU under load.

Re:400 Watts idling?

[Bloater]

> Isn't the 400W while idling for TWO Xeons (4 cores) and the rest of the system?

"while idling"

> ... which isn't bad for a modern CPU under load.

"under load"

What kinda crack you smokin'? I want some!

BTW, when you look at what an Opteron can do, yeah, that's bad for a modern CPU. For a ten year old computer with that processing capability it is very very good indeed but this is 2k5 here and we expect a little better. I know I do, but I've got used to AMD64.

Re:400 Watts idling?

[Jozer99]

I didn't read the article. Is this ONE Xeon using 400W? That would be rediculous. How can they do that? I have a dual core Pentium 4 820, which UNDER LOAD uses about 130W. The Xeon has a few extra features, but is really not that different. How is it so much less efficient?

Re:400 Watts idling?

[Mattintosh]

He does have a point, even if this was about a laptop chip. The PPC970FX has a full-load (the spec sheet says they couldn't find a program more strenuous than the one they used to test it) heat dissipation of 24.5W. The Pentium M has a "real life usage" heat dissipation of 27W and a theoretical maximum of 34W. So, yes, it is typical of Intel processors, and no, Apple didn't choose Intel for their low-power-consumption chips.

Re:400 Watts idling?

[Gothmolly]

Cache. Cache sucks an amazing amount of power.

Re:400 Watts idling?

[WMD_88]

Uh...that's 24.5W @ 1.25GHz, last I checked. If you're talking about the newer 970FXs, that's 1.6GHz @ 24.5W, tops. The P-M goes much higher (2GHz+) at 27W.

Re:400 Watts idling?

[iluvcapra]

didn't Apple choose Intel because of their supposedly low power consumption

Mmmm, Apple has not announced which CPUs they would be installing in their 06-07 machines, but the speculation was that they would be using Pentium-Ms or Pentium-M-derived CPUs, which come off a different lineage from this stellerator^h^h^h6h Xeon chip they were testing in TFA.

OTOH, a single-core G5 only consumes ~50W, so who knows what they were thinking.

Re:400 Watts idling?

[Jozer99]

My Pentium D has 2MB L2 cache, 1MB per processor, how much does the Xeon have. It better have like 512 MB for the amount of power it sucks, and if it has that much, how come its performance sucks so much?

that is not a surprise

[kylie69]

actually intel processors are known to consume *lots* of power for quite a time now, so it is no wonder that the new Xeons don't perform better. I wonder how intel could drop the new Xeon on the market. It will be a huge blow for them.

Didn't Apple choose Intel comments....

[TheSkepticalOptimist]

Yes they did, and if Intel ONLY made Xeon processors, then Apple may be quick to regret their decision. But Intel makes other products, and their Pentium M line of processors which have much better power/performance ratio is what will eventually make it into the Apple Macs.

So, be a little more open minded.

Asside from that, this is a big blow against Intel especially after having realized they can't keep releasing power hungry and underwelming processors anymore. Coming out with a CPU that consumes more power idling then most home desktops PSU can provide is greatly discouraging for the Xeon lineup. I can't imaging anyone wanting these processors, in either a server or workstation environment, these kind of stats are uncalled for.

Re:Didn't Apple choose Intel comments....

[Nossie]

then who did? I might have bought one if they went the AMD route...

but then I'm just as happy over powering my G4 cube with dual cpu's... :-$

heh.

after buying my cube 2nd hand I really thought about my next computer being a dual G5 dual core monster... then they announced their partnership with intel and I laughed my ass off..

history ?

BeOS changed to x86 architecture and died (their change was their 'last hope for peace')
NeXT changed to x86 architecture and died (their change was their 'last hope for peace' but by shear fluke apple wanted jobs back)

Apple changes to x86 compatible architecture and. ..... ? Mikey Dell wanting to sell dell/apple branded intels? apple to extinction ? long live the ipod?

Without the inherently impressive PowerPC arch inside an apple I'm back to my AMD worshipping days... Steve you should listen to me because the reason most people bought apples was because they were fed up eating MS compatible pears...

I really don't want x86 arch to take over the world like Microsoft did :-| Alas, I feel Apple may have hit the last nail in its coffin by histories regards.

Re:Didn't Apple choose Intel comments....

[toddestan]

Yes they did, and if Intel ONLY made Xeon processors, then Apple may be quick to regret their decision. But Intel makes other products, and their Pentium M line of processors which have much better power/performance ratio is what will eventually make it into the Apple Macs.

The Pentium M (or Celeron M) will be a fine chip for Apple's laptops, and for the Mini. But do you really think that the Pentium M is appropiate for the PowerMac? As terrible as these new Xeons and P4's are, they are more powerful than the Pentium M. If Apple introduced a Pentium M based PowerMac, it would easily be the slowest computer at the price range Apple likes to sell their PowerMacs at. Of course, Apple is used to selling underperforming computers at premium prices - but I don't think it's going to work after the switchover to x86.

Re:Didn't Apple choose Intel comments....

[Anonymous Coward]

It could be worse than you think... SGI was once the king of the high end graphic workstation market. (Really cool UNIX environment running on MIPS RISC with really cool apps and kick ass 3D hardware.) At some point they got this really bad idea of switching to x86 and where are they today? No loger the graphic workstation king but not quite dead yet either.

Watching Apple I was thinking they were becoming the new SGI. They have a really cool UNIX environment running on PowerPC RISC. They have really cool apps running in this environment. I didnt want to see them go down the x86 path of certain death.

Who knows what their actual plan is. It could be they want x86 to better compete in certain (read consumer) markets. It could be they intend to keep their high end workstation products on whatever offers the most performance. Who knows...

Re:Didn't Apple choose Intel comments....

[Nossie]

oops .. sorry :D add silicon graphics inc to that list of 'what happened after x86' hehe ;)

its a shame... I've always loved sgi...

*sighs deeply*

Becoming a consumer issue?

[cliffski]

How long till CPU power requirements become a buying factor? Like many people I leave my main PC on most of the day, and also use a laptop which a distressingly short battery life (sony vaio). These days, apart from the old game of battlefield 2, I rarely find any need tor a maxxed out CPU. I'd be much more interested in a PC that would consume noticeably less power. 400W is like having 4 brightly lit rooms all day long, its just wasteful.
For laptops its already a big issue, as anyting that can stretch out the meager battery lfie is good, but even for desktop PCs now, we should be hitting the point where people start asking how much it costs to run a new PC all day.

Re:Becoming a consumer issue?

[Bishop]

It is slowly happening. One the adds for Sun's Opteron servers lists "cooler" as a selling feature. About six months ago Intel stated that the next generation of cpu would emphasis operations per watt. Intel has known for a while that the power requirements of the P4 architechture was not sustainable, but it takes a while to develop new chips. These Xeons were released as a stop gap to keep investors and Dell happy.

Re:Becoming a consumer issue?

[WuphonsReach]

Yeah, it's already started. I'd say about 2 years ago.

The first wave of it was folks starting to harp about the amount of noise that most PCs make. While the Mini-ITX folks were already using fanless systems to reduce noise in specialized application, I think the fanless iMac gave momentum to the movement. At least, that little acryllic cube was cool enough to generate lots of press about being nearly silent.

Then there were the folks that started buying laptops instead of desktops due to the noise / form-factor / lifestyle choices. Laptops are easier to hide away, can be setup anywhere in the house on a whim, and laptops used to be pretty quiet.

Hard drive manufacturers switched over to fluid bearing drives a few years ago. At first, it was a marketing thing that allowed them to differentiate themselves from the pack. Now, I don't think you can find too many drive makers who are still making noisy ball bearing drives.

Energy costs have also gone up in the past few years. My electricty bill has gone from ~$40/mo up to ~$120/mo and I'm starting to consider whether I really want 4 servers, 3 desktops and a laptop running all the time. (And whether I can pack more storage into fewer watts.) Plus the heat issues that all of those systems cause.

PCs are also continuing to move further and further out of the office. As you moving into the living room / kitchen, folks start paying closer attention to noise / power / heat issues. And PCs have gotten powerful enough over the years, that they are "fast enough" for a lot of tasks. They're starting to compete against things like VCRs / DVD players / DVR & PVR (devices which are typically dead-quiet, low-power, and low-heat).

Dell even started selling "quiet" PCs a few years ago (most quiet PCs are also lower energy and lower temperature). Those are nice in an office setting for lowering the ambient noise level (which leads to a less-stressful environment).

Holy Smokes!

[acid_zebra]

Whatever happened to their new mantra 'Performance/Watt over Clock Speed'?

Re:Holy Smokes!

[cloudmaster]

That doesn't get a "dual core" CPU that everyone seems to feel is the be-all, end-all in preformance now. Intel needed something for that particular hole in the market, and they needed it quick. They stumbled a bit on this one. Of course, no oher company who's led the game for decades has ever made a single sub-par product, eh? Look at The early 80's Ford Mustang, the Cadilac Catera, the Mercedes C230 Kompressor, etc for automotive examples, or look at most of Microsoft's products (maybe focus on Bob if a MS fanboy) for software example. Eh. Whoops.

Re:Holy Smokes!

[despisethesun]

Actually, when you consider the other new cars that were coming out at the time, the 1982 Mustang GT was one hell of a performance bargain. We might laugh at 157HP now, especially when we compare it to the models that came before (in the 60's) or after (the new Cobras), but when high performance was considered dead in anything but high-end Italian sports cars, it was a pretty mean machine. The 80-81 cars with the 4.2L V8 were jokes, though.

Well, that was off-topic. :/

Re:Holy Smokes!

[cloudmaster]

I believe I laughed at 157HP out of a 2-bbl carb'd 302 cubic inches then, too. Just as a point of reference. :) Oh, and the turbo four-cylinder mustang GT in '83 and '84 really sucked, as well...

Of course, in 1982, you could also get the newly redesigned Camaro with a whopping 165HP, but it was also fuel injected (something the performance mustangs didn't get until the late 80's). So yeah, American cars generally were still suck-ass after the big gas crunch. Unless you count the turbo Buick T-type, which laid the ground work for the '84 to '87 Grand Nationals (which certainly *didn't* suck). Even in '82 the turbo T-type was pretty decent - and had a useful back seat. ;)

Cool chips.

[MoogMan]

FFS. I've got enough speed in my P4 2.6GHz. What I need is a lower (by today's metric, but still damn fast) clock speed with much less power consumption and heat production. Where is it?

Re:Cool chips.

[Dogtanian]

I've got enough speed in my P4 2.6GHz. What I need is a lower (by today's metric, but still damn fast) clock speed with much less power consumption and heat production. Where is it?

It's in the shop that sells "Pentium M" CPUs.

Re:Cool chips.

[sharkey]

Over here.

Re:Cool chips.

[metamatic]

Pentium M. AMD.

Or if you're really hardcore about low power consumption, VIA C3.

Re:Cool chips.

[wirelessbuzzers]

Or if you're really hardcore about low power consumption, VIA C3.

Or Via C7: 100 milliwatts while idle, and not so bloody slow as a C3. But good luck getting your hands on one.

Bring on the winter!

[WidescreenFreak]

I don't understand what all of the fuss is about. With systems that generate this much heat, I want to buy one just because it will probably be cheaper to heat my house with a dual Xeon system than with heating oil, considering current fuel prices! I can simultaneously have the benefit of running one hell of a Battlefield 2, Half-Life 2, or SETI@Home system! Then when summer arrives I can switch back to my cooler AMD X2 system, thus saving money on cooling bills as well!

AND after playing BF2 for an hour or so the top of the case will be ready to cook up some chicken and steak tips! Introducting the new Intel, dual-Zeon, charcoal-less hibachi system!

Come on! You guys need to look at the positives of having such hot CPUs!

Re:Bring on the winter!

[Random Hacker]

Electricity is almost always the most costly way to heat a home.

Re:Nonsense

[yfkar]

A dual-core CPU is going to draw twice the power. Duh.

So? The AMD Opterons were also dual-core. And they used 60-70% of the power used by Xeon systems.

Ok so this version of the Xeon is a power whore...

[NoMercy]

But Intel released details of there future plans for much lower power processors accross the board earlier in the year, so... what's the big deal?

Re:wasting the planets resources - powered by inte

[cr@ckwhore]

Are you kidding me? I bet you aren't even aware of your own ego (which is so large it's coming out of your ass). Do you actually think that we humans could destroy the planet in a lifetime? The earth has been here for billions of years, and I'm pretty confident that it will get rid of us long before we humans could get rid of it.

(BTW, it wasn't the crab people, I broke the dam)

Re:Nonsense

[Bloater]

The level of technical accuracy in most of these reviews is really really lacking: * They really didnt measure the CPU's power draw. They measured what the power supply draws from the wall. A whole different ball of wax altogether.

I don't know if they had the same PSU, hard drive, etc. but if they did, then measuring the power at the wall is the most realistic and reliable number. The Opterons and Xeon's have different relationships with the components on the mobo, and the mobos and memory are different. So to reflect the difference that will be experienced between choosing an Opteron system versus a Xeon system the *actual* power consumed to do the tasks is required - otherwise you are measuring the power required to perform two or three tasks on the Opteron and the power required to perform one task on the Xeon.

* A dual-core CPU is going to draw twice the power. Duh.

The Opteron was a dual-core too, so the Xeon shouldn't draw twice the power (and it didn't, it drew about 1.5 times the power, which is almost as bad).

Sextium (with added alcohol)?

[mysticwhiskey]

It provokes the desire, but takes away the performance!

Re:Ok so this version of the Xeon is a power whore

[ctid]

The big deal is that Intel, the market leader by a country mile, is way behind AMD on power consumption and performance. They may be ahead in vapourware, but you can't buy (or run software) that.

Re:Rumors?

[Dogtanian]

Those aren't rumors, they're confirmed fact by intel itself. Did you miss all their marketing on future low-power cpus all across the board?

Yep. So their marketing can't have been that good. As I said, I'm not an expert in this area, and I'm not planning on replacing my 3-year old 1.8GHz P4 system in the next six months.

Perhaps they only advertised in "'Gotta have the latest'-Obssessive CPU Freak" magazine :)

so

[akhomerun]

so what exactly is new here? i personally think this isn't newsworthy since that's all intel has been for the past 5+ years.

anybody who doesn't use an opteron for their servers or an athlon 64 for their computers has some serious issues - because intel's main competitor has faster, cooler, and cheaper processors.

Re:so

[Amouth]

personaly i run the p4 northwook chips they run a hell of alot cooler than the presscraps and honestly they are just as fast.. when intel move from .13 to .09 i think they messed up - instead of using less power and making less heat they went the other way - hell even my daul 2.4 xeons (the old 533) only consume about 180w - i don't know where they are going but i hope their mendra thing comes soon..

(and it isn't that i don't like AMD i love them .. i just have a hard time finding chipsets i like)

Bring on the burgers!

[Spokehedz]

Next up: The George Forman Lean Mean, super-processing, grilling machine!

It knocks out the fat, and your breaker switches!

"With my new machine, you can cook your food in half the time--and compile your kernel in no-time flat! My new Lean Mean, Super-Processing, Grilling Machine knocks out the fat, and the compile time!"

But seriously... What the hell is wrong with Intel? They have a perfectly good model (AMD) to base their business model off of... What do they do? They take one good part, implement it half-assed, and then forget all the rest of the improvements.

call this flamebait, but Intel doesn't seem to be making a lot of good judgment calls here lately. Basically, I see the same-old Intel, back when AMD and Cyrix chips were out on the market.

"Our chips are the fastest! See? Ours have 200 Mega-hertz! And MMX! Games have sound now, so were going to put some sound-bits on the chip! See? SEE?!?"

Sorry Intel. The cat is out of the bag: MHz don't mean much. Our computers are so overpowered now, why do we need more speed? Games? ok, I'll bite: Isn't most of the processing is done by the videocard these days... So what's your excuse now? Windows Vista? Again, that's going to use your videocard more than anything.

Intel, your overpriced and inefficient. It's time for a paradine shift. Innovate or get out.

Re:Nonsense

[Anonymous Coward]

As a small-time, low power switch-mode power-supply designer, I would like to object that it is very difficult to accurately measure the power draw of a modern desktop CPU, unless you design a whole motherboard specially dedicated to power measurements.

These beasts have peak current draws in the range of 100A @ 1.5V.

If you were to allow for a maximum of 1% voltage loss (15mV) across a measuring shunt resistor, this would mean 0,00015 Ohm Resistance - the equivalent of a piece of copper Wire with a cross-section of 1.0mm^2 and a length of 7mm. Good luck attaching a 1% accurate kelvin sense connection to this. You will also have to design a high performance multiplier circuit to make accurate RMS power measurements in the MHz Range.

But even then, and not considering mechanical difficulties like buried traces/planes, you cannot place this shunt betwwen the cpu and the Power supply (i.e. 1.5V switcher and bulk 1.5V capacitors), because the added inductance and skin effects would probably cause the CPU to malfunction.

So your best bet is to place shunts between the multiple switching coils and 1.5V Caps, which would probably work. But then you are only measuring the sum of the CPU current draw + the dissipation losses in the traces and the capacitors - which arent neglible, as a lot of people learned recently, as the overheated low-quality organic caps died on their motherboards.

But what is it worth ?

As a computer user, I care only for four things :

- The reliability
- The noise coming out of the box
- The cost of the power going into the box
- The cost for extra air-conditioning, or the savings on heating -
    depending on the season and where you live ;-)

For all of these, the amount of power drawn from the wall plug is a very suitable indicator.

hypertransport bus

[246o1]

I used to take the hypertransport bus to school, it's not all it's cracked up to be. Especially when you have two(!) hypertransfers.

Re:Ok so this version of the Xeon is a power whore

[the eric conspiracy]

future plans

Ahh - vaporware. We all know how stuff always gets delivered on time and lives up to all the marketing hype. And of course after issuance of the preannouncement the competitors are not allowed to improve their products.

Feh.

Re:Ok so this version of the Xeon is a power whore

[tcgroat]

The big deal is this: to build a computer you need parts, not press releases. Xeon is what Intel is selling, and Opteron is what AMD is selling, today. We'll see how next-generation parts (from both sides) measure up when they are in production. Many changes can happen between the marketing plan and production silicon.

They only have to keep business types happy

[jschottm]

As the article notes, it seems that the main reason they released this chip is so business oriented writers on the Wall Street Journal will stop complaining about Intel/Dell's lack of dual core CPUs. Those writers aren't all that likely to care that much about the power/heat issues so much as keeping up with the Joneses.

Dual Use for Dual Core

[Mynister]

Intel is missing the boat. They need to market this chip to students in dorm rooms, as a dual use product. It is a Hot Plate / set top Media Center PC.

You can prepare a gourmet dinner while you watch a Tivo'd Food Network.

That my friends is dual use and dual core!

65nm won't help much

[complexmath]

Read this article:

http://www.tomshardware.com/cpu/20051007/a_sneak_p eak_at_intels_65nm_pentium_4-11.html#power_consump tion

In short, decreasing the die size has allowed Intel to reduce power consumption a bit, but it's also increased thermal density to such a degree that they can't clock them as fast as even the current crop of 90nm processors. Why upgrade to these chips?

power issues foremost?

[YesIAmAScript]

I'm sure all of you that complain about this Intel system using 400W when an AMD system would use perhaps 200W are also energy conscious in other ways, right?

You turn on power saving nodes (Intel's Speedstep, AMD's Cool n Quiet)?
You use 55W fluorescent torchiere lamps instead of 300W halogen ones, right?

http://www.lbl.gov/Science-Articles/Archive/torchi ere-SMUD.html

You could easily save as much power on lighting in your house as you do on computing costs.

Re:power issues foremost?

[gentlemoose]

This is about large-scale computing, not your home machine. I run a 3600-processor Xeon cluster. Power consumption differences on that scale make a real difference.

Re:power issues foremost?

[RzUpAnmsCwrds]

"You turn on power saving nodes (Intel's Speedstep, AMD's Cool n Quiet)?"

Yes. My Athlon 64 Newcastle (CG stepping) idles at 800MHz/0.8V, consuming ~8-10W. My LCD blanks itself after five minutes. I do not use a power-intensive screensaver.

"You use 55W fluorescent torchiere lamps instead of 300W halogen ones, right?"

There is nothing other than flourescent lighting in my room - a 19W tube as a cieling light, a 19W compact-flourescent desk-lamp, and three 13W compact-flourescent lamps. Usually, I only use the 19W overhead lamp.

And, I ride the bus or walk. I don't have a car.

Re:Apple chose Intel

[the MaD HuNGaRIaN]

I was just going to post the same thing.
Anyone know what the new dual dual-core G5 PowerMac consumes power wise?

recheck the links

[stuartkahler]

It's time to drop the gamepc link. Apparently they don't like the extra page views that /. is sending them.

Re:And Apple wants to replace PPCs with *THAT* ...

[toddestan]

Depends on what you are running. Tasks like Seti@Home are well suited for large clusters of computers, but I sure wouldn't want to run HalfLife 2 on 20 Mac Mini's.

But AMDs are unreliable

[InnerParty]

As a PC technician that works on multi-cpu Opterons day in and day out, I can tell you that despite the performance and lower cost of AMD (which gets all the limelight), they are far less reliable than intel processors and the motherboards on the market tend to be unreliable as well. Anybody who has dealt with the Rioworks HDAMB or Tyan S2875 can tell you that. The 32-bit Athlon MP motherboards were god-awful as well and the trend continues. Instability issues are rampant because AMD CPUs are so picky about the type of RAM used. My company is constantly having to order more expensive single-ranked RAM modules to solve banking limitations of the CPU memory controller, machine check errors, and ECC corrections. This also makes RAM warranty replacement take longer than it would normally. Even when you are talking about consumer level AMD-based systems, such as 3000+ up, nForce-based motherboards (by far the most popular chipset) have been historically picky about what RAM is installed. Believe me, people don't love their AMDs so much when they are down, they are often referred to as pieces of sh** by customers that experience these all too common problems. As AMDs are becoming more popluar these days, more small businesses are buying servers based on AMD, they don't have backup equipment, and they get really pissed off when their Exchange server quits because an Opteron just suddenly gave up the ghost. Seriously people, you can champion performance and lower heat output all day, but both platforms offer more power than most people utilize, and I would think that less downtime is what equates into real dollars saved, not faster CPU operations. Most people would rather drive a Toyota that gets them from point A to point B everytime than rely on a Ferrari that drives faster but breaks frequently. In some ways, AMD has already demonstrated my analogy for me. Acer-Ferrari overheat notebook anybody? Reminds me of Fiat-Ferrari...:(

Re:But AMDs are unreliable

[Mashiki]

Anyone using Rioworks or Tyan boards should have their fingers broken in multiple places in order to stop them from making foolish purchaces again the future. Those boards could be replaced by a flavor of Gigabyte, Asus or ASRock 939Dual boards.

Re:But AMDs are unreliable

[InnerParty]

Give me a break! Asus...maybe,but they only have one Nforce-based dual board as far as I know. First one I installed had a bad onboard NIC, second one worked though. But ASRock or Gigabyte? I don't see how you can recommend tertierary brands as a solution, Giga-byte is more unreliable than either Rioworks or Tyan, hands-down. They are popular as enthusiast boards because they are cheap and pack more features, but break they do, all the time. And their customer support is pretty weak. ASRock? Ghetto....

Re:But AMDs are unreliable

[RightSaidFred99]

*cough*retard*cough* You seem to be confusing the "I build a machine and buy my motherboard from newegg" market with the "I run a multimillion dollar business and I buy real server hardware". ASRock, indeed.

Re:But AMDs are unreliable

[PlusFiveTroll]

nstability issues are rampant because AMD CPUs are so picky about the type of RAM used. My company is constantly having to order more expensive single-ranked RAM modules to solve banking limitations of the CPU memory controller, machine check errors, and ECC corrections.

Most mainboards have memory compatabilty charts, follow them and your life will be eaiser. Though I do recommend running a 24 hour burn in memtest on any new server, can find bad ram before it goes in to production.

I have had exactly 0 Opteron processor or mainboard failures. Maybe you should reevaluate the boards your using.

The Athlon-MP platform sucked, totally.

Re:But AMDs are unreliable

[5pp000]

The Athlon-MP platform sucked, totally.

Huh. I have a Tyan Tiger MPX with two Athlon MP 2800+ and 2.5GB -- it's been rock solid for two years -- great machine, as far as I'm concerned.

Re:But AMDs are unreliable

[InnerParty]

I agree about memory compatibility charts, but it's usually a narrow selection of vendors/models, and there is a heck of alot of memory out there which say the same thing on the label. To most people, including the salespeople that spec the systems, a 1GB DDR400 Reg ECC is the same as another stick with the same label, because that is how RAM is marketed to the public. Sticking to the vendor charts was not as necessary in the past, and still doesn't seem as critical with the intel platform. With the special memory architecture of the AMD CPUs, it is definitely more sensitive to the RAM used, but it seems people have not been educated about AMD memory compatibility and are still playing the old "just cram in a stick" method of the past. I am surprised AMD and the motherboard manufacturers aren't extremely insistent that a builder follow the qualified memory list. As long as the 64 bit CPUS have been out, it should be common knowledge by now amongst all Opteron users, OEMs and enthusiasts alike, that the memory compatibility charts need to be heeded for a stable system. How hard would it be for a mainboard manufacturer to put an errata sheet in the box so that it is the first thing you see when you open it? If you don't educate your customers, how can you expect anyone but a professional PC technician to know what single-ranked RAM is (needed to fully populate slots with DDR400)? By the way, the on-die memory controller can go bad on the Opteron and look like memory errors. Some CPUs will appear to run fine until you push them really hard, such as compiling code for hours on end. Believe me, the customers my company sells these to push the systems very hard; there is no possible way you could achieve a zero failure rate in the same situation. Furthermore, a software-based memory tester is not a robust enough indication of system reliability as memory will not fail the test unless largely compromised. Only real life tests where you are pushing the cpus 100% for extended periods of time, using most or all of the memory space available, and stressing the disk I/O can give you reasonable confidence in a newly built system. MemTest or Windows Memory Test don't mean squat, they are just quick tools to confirm what you already suspect...that faulty memory is causing that IRQ_NOT_LESS_OR_EQUAL stop error you keep getting. I'd like to know what motherboards you are using as well as power supplies please, so I can try them out. To be fair, Tyan does have some boards that behave alot better than the S2875 I mentioned, namely the S2885 (dual) and the S4882 (quad). The S2895 looked promising at first, but apparently there have been some BIOS-level memory issues since the board has been on the market, and Tyan only got around to fixing it recently. Unless you are talking about first tier equipment such as an HP Opteron Quad, I don't think there are many other mobo choices beyond Tyan, Rioworks, Asus, Giga-byte, or MSI. The Asus K8N-DL seems ok, but I don't trust MSI or Giga-byte dual boards to last long before needing replacement. So who do you use? Tyan has really been the leader in Opteron board offerings and to have so many issues with their motherboards really puts a dent in the confidence of the Opteron platform as a whole in my opinion. I could build a bunch of Xeons with intel boards and be done with the headaches if were up to me.

Re:But AMDs are unreliable

[InnerParty]

I think you might have one of the boards that actually worked for the MP platform, but there were many others that suffered stability issues due to an immature chipset and also the K7D-Master from MSI was a commonly used board in MP systems that just loved to fail.

Re:But AMDs are unreliable

[Mashiki]

Very 'impressive' reply. Not only to you start out with a ad-hom attack you add nothing in the way of a response in reply. Let me know when you figure out what ASRock actually makes okay?

Re:But AMDs are unreliable

[Mashiki]

Rioworks are known for using non-branded componets, and Tyan has questionable reliability issues. ASRock is also not a 'parent' company, much like others these days.

Try adding 'constructive' arguements, I realize that it may be difficult for you however, giving the use of limited vocabulary and the use of the word 'ghetto' as a descriptor.

Re:But AMDs are unreliable

[InnerParty]

Sorry, that's just tech lingo these days.

!!!WARNING!!! goatse on first page !!!WARNING!!!

[donkeyoverlord]

The second page is safe

that's how intel gets...

[KillShill]

those pentium M cpus to perform on par with athlon64's.

their massive 2MEGABYTE l2 caches make up for the fact that they suck without them.

the turion64 and athlon64 cpus perform as well with 512KB cache, hence the much lower cost of those cpus in relation to pentium M's.

ever wonder why notebooks with "centrino" and PM cpus cost a whole lot more than equivalent systems?

Re:Most pro-Intel site on web!

[Sique]

We just didn't expect any better, so the disappointment was limited ;)

Towing capacity

[Yartrebo]

For most people's needs, a small car has enough towing capacity to pull a modest boat or camping trailer. I remember my grandfather towing his boat with his subcompact, and it seemed fine enough to me. True, you couldn't safely go at highway speed, but to tow the thing a few miles to the ramp, you don't need an overpriced SUV.

Even if you do have an SUV, to be safe, you should drive pretty slowly anyway on the highway, because you are very prone to jack-knifing. Your braking is limited by the tires on the trailer, not by the brakes on your car.

Re:Don't follow link in parent: *itself* is GOATSE

[donkeyoverlord]

GAH!!!! Looks like when linked from slashdot the page becomes goatse. Sorry :(

Re:RTFA for chrissakes!

[the MaD HuNGaRIaN]

Ugh....

Ok. Let me try to explain it to you.
Some of us have better things to do than scanning the internet for new content and reading each article posted.
We simply don't have idle time to read "gamepc" articles all day long.

So, we have sites like Slashdot, which are called "News Aggregators."

No, try to follow. The point of a new aggregator is to pull all news related to a certain field, area, or subculture into one convenient location and provide this thing called an abstract which allows busy people to get the gist without having to waste time actually reading the thing. The links are there for reference and for those who are intrigued enough by the abstract to actually care enough to read the whole piece.

Don't worry, you will learn more about this concept once you are at University.

And for the truly impatient---we use RSS!

And it works too, because I got the answer to my question delivered right to my inbox without ever having to visit gamepc.

Re:Nonsense

[mauriceh]

No, a PROPERLY designed dual core does NOT use twice the power, at the same clockspeeds.

For example, look at the Opteron 248 REV E4 versus the dual core Opteron 275REV E6
85W for single core 248
http://www.amdcompare.com/us-en/opteron/details.as px?opn=OSA248BLBOX

95W for dual core 275
http://www.amdcompare.com/us-en/opteron/details.as px?opn=OSA275CBBOX

Both share the same .09nmSOI technology.
AND remember the Opterons contain the memory controller on-chip, so simple CPU power comparison to the Intels does not reflect that the Intel chips require a chipset on the motherboard with a memory controller that is also consuming power, and making heat.
About another 25W in the case of the new Intel chipsets that support Paxville.

The difference is dramatic.
Total of 95W on the Opteron side, versus around 350W for the same setup on a Paxville, 2.8GHz, which is still a LOT slower than the 275 in most tests.

The only fan boys I see spouting here are Intel fanboys.

Re:Ok so this version of the Xeon is a power whore

[RightSaidFred99]

"vapourware" doesn't mean what you think it does. Vaorware is something that has been promised to be revolutionary, but which has missed schedule repeatedly to the point where people are starting to believe a.) it won't deliver its promises, and b.) it may never even be released. I love how it's now come to mean "anything that's been announced and not released yet".

Re:Nonsense

[vlm]

Isn't it easier to just put it in a styrofoam box, run it for 5 minutes, see how much the thermometer increases, then cool it down, insert a 100 watt lightbulb and measure how much the temperature increases, then calculate based on the ratio?

Re:You're a funny guy: article fine, but ur link n

[Mr Z]

Actually, if you click on the link up in the original Slashdot article, you'll discover that the first 3 pages of the article have been hacked w/ the Goatse guy. At the time I posted, only the first two pages had been hacked. I imagine the whole article might get hacked eventually unless the guys running GamePC fix whatever the hole is.

Re:Laughing At The Apple Loonies

[xouumalperxe]

Ok, since you seem to hace a serious case of clue deficit disorder, I'll feed it to you in small bite-sized chunks.

The Pentium M is actually a pretty decent chip.
Centrino is a pretty decent platform.
IBM hasn't done much regarding laptop processors in a while, intel has a vested interest in keeping that area going.
Laptops are now the majority of computer sales, unlike laptops
AMD does NOT have a decent platform to fight Centrino/Pentium M with. Their desktop CPUs are great (typing from an Athlon 64 3500+), but their Turion line sucks too much power.

So yeah, you have the big, studly, 4-core PowerMac. It's essentially a workstation machine. that, truth be told, doesn't add up to that big a part of Mac sales. iBooks and Powerbooks, however, are a different matter. Those, despite the increasingly apparent weakness of the G4, still sell like hot cakes. Simple logics would indicate that a stronger processor under the hood would sell even better. So yeah, the intel roadmap might actually look pretty decent for apple.

Re:Laughing At The Apple Loonies

[xouumalperxe]

brain fart, for which I ask forgiveness. That should be "Laptops are now the majority of computer sales, to the expense of desktops".

Re:Not terrible...

[evilviper]

So you're saing having a chip that uses 50% more power than its direct competitor and doesn't even beat it in a single benchmark isn't terrible?

Yes, I'm saying that.

Look, Pentium 4s have been peaking around double what equivalent Athlons were, and that has been the situation for years. It hasn't cut into Intel's lead all that much.

They may not be great specs, but these stories are vastly exaggerating the significance. These aren't at all out of line with previous Intel CPU specs, so the whole "OMG HOTTEST CPU EVER" crowd is making no sense. What's the big deal? Should we have a /. story every time Intel comes out with a new CPU that uses more power than an AMD processor? There's nothing here.

AMD Cool and Quiet

[Anonymous Coward]

Most Athlon64 motherboards have a feature called "Cool and Quiet" which if you install the proper software, will on the fly underclock your PC. To give an example my cpu is an AMD 3000 which runs at 1.8ghz. Most of the time it runs at only 1 ghz using Cool and Quiet and only steps up the clock speed when needed.

AMD is so far ahead of Intel in so many areas: Cool and Quiet, perforamce per watt, on die memory access, dual core, etc. that is not funny any more. If marketing did not matter so much Intel would be in deep.

Re:Not terrible...

[swillden]

Should we have a /. story every time Intel comes out with a new CPU that uses more power than an AMD processor?

Why not? It wasn't so long ago that we had a new /. story every time AMD came out with a CPU that used more power than an Intel processor. And we cared less about power consumption then than we do now, due to the increased focus on quieter and more environmentally-friendly PCs and servers.

This is worthwhile news.

Back in the 1980's

[Wytil]

I remember a large company capturing the heat from their IBM mainframes and using the heat to provide heat to the building. Aux heaters helped out. I guess Intel is planning on a servce called "Integrating your new computers into the company heating system." Cooling? Thats a problem...

Re:This should change - Pentium M descendants?

[Lonewolf666]

Intel is not only planning to switch to a 65nm process, they also are working on a new line of products that are said to be based on the Pentium M. With that technology, they should get a much better grip on their heat problems.
A problem that will remain, however, is their SMP design with only one memory bus, as opposed to AMD's where each processor has its own memory bus. In the tested configuration (see the GamePC article), the AMDs won bigtime on memory bandwith. In a 4-way or 8-way system, expect this to become even more pronounced.
Bottom line:
The next generation of Intel products will probably catch up on the desktop, but I don't see how they can keep up with the Opteron anytime soon.