AMD Desktops Outsell Intel

prostoalex writes "For the week ending August 21st AMD managed to capture 54% market share among new desktops sold. Intel's share during the week was 45%. While Intel leads the U.S. CPU market with 82.7% market share, folks from AMD are proud to announce this is the second week this year - they also outsold Intel on the desktop market one time in April 2004."

Comments

Including businesses?

[Alowishus]

I'm not sure I buy this. Maybe for enthusiast and home gaming PCs, but if you include business desktops I'd venture to say that Intel still carries somewhere around 75%. Go look at the business-oriented desktop lines from HP, IBM and Dell and you'll see very few AMDs in there.

Re:Including businesses?

[Hawkxor]

Its for desktops selling within the retail channel. And Intel does have around 80 market share overall - its just that this past week AMD machines outsold Intel for some reason.

Re:Including businesses?

[mentalflossboy]

Agreed. AMD seems to appeal more to individual consumers/gamers. Businesses all still buy the generic Pentiums from Dell.

Re:Including businesses?

[rokzy]

you don't even have to RTwholeFA, just load the page and there's a big graph saying "retail dektop sales".

Re:Including businesses?

[bigman2003]

The lead-in paragraph mentioned that Intel has like 82% of the market in the U.S.. I would guess that the rest of the world does not just automatically call Dell/HP when they need new computers.

The more work a person is willing to do to buy a computer, the greater chance they will purchase AMD. Someone who is just picking up a box with 'everything in it' might be more likely to see the 'Intel Inside' sticker on that new computer stacked 10 high at Best Buy.

Then again, my purchasing department doesn't seem to understand that there are computer makers other than Dell.

But what if I was in Italy- and buying from Dell was a pain in the ass? The chance of purchasing AMD just went up about 200 times.

Re:Including businesses?

[Anonymous Coward]

I'm not sure I buy this.

You don't have to. 54% of your tech-loving friends have done so already.

More seriously, AMD makes a good chip at a reasonable price, and all other things being roughly equal why not support the underdog?

Re:Including businesses?

[JoshMooney]

Wasn't it Intel who decided that they were too good for nVidia's nForce mobo? And didn't people say that nVidia would go down for choosing AMD over Intel in this market? (correct me if I'm wrong, as I probably am)

Re:Including businesses?

[ricotest]

Maybe for enthusiast and home gaming PCs, but if you include business desktops I'd venture to say that Intel still carries somewhere around 75%.

The blurb itself says that despite AMD's share of new CPUs, Intel have 82.7% of the US market. Which is close enough to 72%.

The article itself admits that AMD's market is 'constrained' such that these results are very impressive. Intel indeed makes AMD a clear underdog for businesses and (at least up until very recently) notebooks.

Re:Including businesses?

[VoidWraith]

HPs have a trend of using AMD chips in their lower-cost models. I can't speak of Dell though. But not everyone's custom-built PC uses an AMD. Intels still run faster, even if they don't crunch the big numbers all in one cycle.

Re:Including businesses?

[rcamans]

AMD numbers are based on (mostly?) retail sales.
All of Dell sales are direct.
Most of HP sales are direct.
Most of IBM sales are direct.
Most of Intel sales are direct.
I am referring to desktops in the gov, and corp market, as well as direct to customer sales.
So yes, AMD sells more retail.
Retail sales overall are a decreasing percentage of the desktop sales figures.
Makes for a great headline, but it is not true at all, not even close.
AMD does not have anywhere near the production capacity Intel has, and both are cranking out full steam ahead.
So do the math yourself.
if AMD has 20% of the capacity of Intel, and both are maxed out, who sells more?

Re:Including businesses?

[Gherald]

> Intels still run faster, even if they don't crunch the big numbers all in one cycle.

s/run\ faster/have\ higher\ clockspeeds

Re:Including businesses?

[HFXPro]

s/run\ faster/have\ higher\ clockspeeds/ you forgot the last / My perl chokes on your subsitution regular expression.

Re:Including businesses?

[Silverlancer]

...

You've been brainwashed by Intel. In almost all applications, a similarly priced Athlon 64, without 64-bit, wipes the floor against Intel. And in 64-bit compiles in Linux 64-bit, the Athlon 64 gets an extra 30-40% boost. Now obviously we won't get that in Windows, as most companies won't come out with 64-bit compiled versions. But hey... who uses Windows anyways? ;)

Re:Including businesses?

[VoidWraith]

That's exactly what I meant. [sarcasm]Thank you for clarifying.[/sarcasm]

Re:Including businesses?

[mercuryresearch]

Actually, the lead paragraph mistakenly says the 82% figure is US -- it's not, it's ~82% worldwide.

The AMD > 50% figures are specific to US Retail sales, so they are totally uncomparable numbers.

Re:Including businesses?

[nadamsieee]

Intels still run faster, even if they don't crunch the big numbers all in one cycle.

Raw CPU speeds are fairly meaningless.

Its like the RPM guage on your car. Lets say that a Corvette has a lower RPM per mile per hour than a Porche and it also costs less. Now lets pretend that they both top out at 165 mph. If all you're worried about is how fast you get from point A to point B (and what else is there when talking about CPUs?), then the Corvette obviously gets you more bang per buck. Who cares if the Porche has higher RPM per MPH (its actually a bad thing!).

Re:Including businesses?

[MonkeyCookie]

Then again, my purchasing department doesn't seem to understand that there are computer makers other than Dell.

It's quite likely that your company has some kind of contract with Dell where they purchase exclusively from Dell in exchange for a better deal on those purchases

I believe it's relatively common for companies to do such a thing

Re:Including businesses?

[brentl]

I would guess that the rest of the world does not just automatically call Dell/HP when they need new computers.

Unfortunately, I think they do. I'm in Australia, and pretty much every company here buys brand name PCs. It's the same in other countries I've been to aswell. Good to see AMD doing well though.

Aaahhhhh!!!
/realises that I'm typing this on a Dell Laptop with Intel processor :-/

Re:Including businesses?

[griff199]

More likely is that they wanted to have just one company to worry about calling when something shits the bed. And the day they made that decision, that "Dell, Dude" guy was looking like the right solution.

Computers are a commodity. Sure, you can hire competent people to fix them when things go wrong, but isn't it cheaper to hire someone just good enough to talk to Dell tech support? Hell they don't even have to be smart enough to ascertain the brand; just tell them "Call this number when their computer screws up. If everyone's is doing it - email to "All Corporate Offices" saying the we are experiencing 'opportunities with our network'. "

At least that what happens when our Dells start freaking out. And are you really going to spend your job doing IT's job for them? Not worth your time.

Re:Including businesses?

[Anonymous Coward]

Most of HP sales are direct!?!?!? That is news to me.

Re:Including businesses?

[Anonymous Coward]

Intel is skewing their numbers, they're counting dual-core cpus as two processors!

Re:Including businesses?

[Anonymous Coward]

Do you have a source for AMD's manufacturing capacity compared to Intel's, or are you just pulling that out of your ass? I don't think the difference is nearly as large as you claim.

Re:Including businesses?

[rcamans]

look at the following, for example: http://www.theregister.co.uk/2004/03/23/ibm cpu sales rocket/ (replace the spaces with _)
Intel worldwide cpu sales 80% of the market.
By far the majority of Intel sales, and AMD's as well, are the desktop sales.
It costs 2 billion dolars to add another plant to make them, so as long as Intel makes more profits than AMD, they can keep ahead in adding capacity as well.
Sorry bout that.

Re:Including businesses?

[silverfuck]

both are cranking out full steam ahead

What makes you say that? Especially given recent price cuts to "Reduce excess inventories". In fact, just google for Intel excess inventories!

Also, see the register article (from five days ago I might add), which agrees with TFA in that this is the fifth week so far this year, not the second as the submitter says.

Re:Including businesses?

[mrchaotica]

Yeah, but high RPM is cooler (which is why Honda advertised it for the S2000).

Of course, being the uncool geek that I am, I prefer AMD. : )

Re:Including businesses?

[Deliveranc3]

http://news.com.com/2100-1001-206773.html?legacy=c net Fortunatly intel has a huge friggin chip surplus biting them on teh ass. So um it's not a comparison of production

Re:Including businesses?

[glitch23]

So do the math yourself. if AMD has 20% of the capacity of Intel, and both are maxed out, who sells more?

I hate math. Don't make me think.

Re:Including businesses?

[Grishnakh]

The problem with this comparison is that car engines are not like CPUs.

The advantage of higher-revving engines is that you can generate the same power (not torque) with a smaller sized engine. Notice that Formula 1 cars have extremely high-revving engines--usually over 10,000 rpm. In race cars like that, engine size and weight is extremely important. In F1, in fact, the engine size is actually critical in determining the shape of the body, and severely affects aerodynamics.

Higher-revving engines, properly engineered, also allow better fuel economy because of their smaller frictional loss. This is important if you only use the peak power output capability of your engine very rarely, and spend most of your time cruising at a normal speed in high gear. With variable valve timing and variable cam phasing technologies, you can build a smaller engine that gets excellent fuel economy at low rpms, and very high power output at high rpms; the best of both
worlds.

The only big disadvantage to higher rpms is, of course, durability, but modern mechanical engineering, metallurgical, and manufacturing practices easily make up for this.

With CPU speeds, however, the raw clockspeed is only one variable in how well the chip performs, but is also directly proportional to how much power the chip consumes. So if you engineer a crappy CPU which has a high clockspeed, but doesn't use those cycles effectively, you'll only succeed in wasting more electricity than the lower-MHz competitor which has a more efficient architecture.

Re:Including businesses?

[servoled]

Makes for a great headline, but it is not true at all, not even close.

No, it is true. However, it is also highly misleading, but that doens't make it false.

Re:Including businesses?

[RzUpAnmsCwrds]

"But what if I was in Italy- and buying from Dell was a pain in the ass?"

It's not. That's why Dell is so successful. No matter who you are or what you need, you call them up, and they ship it to your doorstop.

Dell is very good at what they do: taking Intel parts, slapping them in a box, and shipping them out the door. Dell was the first company that realized that succeeding in the PC business had nothing to do with having the best PCs. Succeeding in the PC business means undercutting everyone else in overhead. That's what their entire business is designed to do, and that's why they succeed.

Re:Including businesses?

[maddskillz]

Dell PC's use Intel Pentium 4 processors. It's in their ads. I have been brainwashed

Re:Including businesses?

[alienw]

Intels still run faster, even if they don't crunch the big numbers all in one cycle.

Bullshit. A 1.6GHz Athlon will slaughter a 2.2GHz pentium 4 at most applications. The top-end Intel chip and the top-end AMD chip have roughly the same performance.

Re:Including businesses?

[Anarke_Incarnate]

easy, they know something you don't know. Mhz is just frequency and not speed. Marketing aside (which is what intel's overinflated mhz rating comparative to actual performance), the AMD K7 and K8 series CPUs do a lot more per clock cycle so a lower clocked chip competes with intel's higher clocked chips. This is much the same way that the Pentium M (Banias/Dothan) compete favorably to the Pentium 4. A 1.8Ghz Pentium M beats the 2.8Ghz P4 down.

Re:Including businesses?

[Trogre]

In the South Pacific, Dells are only now beginning to make an appearance, and a pretty weak one at that.

HP are only good for their printers and medical instruments. Down here their computer line is a joke.

The vast, vast, VAST majority of desktop PCs are made up by local computer wholesalers. Vast.

Re:Including businesses?

[servognome]

I think CPU comparison vs car engines is a valid one. Engines just like CPUs have a number of different features each of which are important.
You can't just choose an engine based on one measurement, there are an assorment of features: horsepower, max RPM, torque, gear ratios, fuel economy; same with CPU you have GHz, cache size, memory controller, power efficiency.
A well engineered product will maximize all these features, or emphasize certain features over another depending on application. For example Pentium-M power consumption, or torque in a truck engine.
Higher-revving engines, properly engineered, also allow better fuel economy because of their smaller frictional loss.
Wouldn't higher revving = higher frictional loss, and less fuel economy? If you are moving something faster against a surface, and increasing the number of times it rubs against I would assume the friction would be higher. Isn't why cars are more fuel efficient at higher gears, since the revs are lower, more power/stroke. I would think in F1 the engines are high revving for faster acceleration
I'm not a car guy so was just curious about that.

Re:Including businesses?

[obeythefist]

Part right. nVidia didn't bother getting a license to do intel. We know the technology worked - just look at XBox. I don't think they revealed any particular reason for not pushing intel harder for a license, although it may have been some strange crosslicensing issues with Xbox and Microsoft. Also, it's possible that nVidia wanted to test the water with AMD's CPUs first, and found that market successful enough. Anyone with AMD would know that nVidia dominate the chipset market for AMD - and for good reason, the performance and stability are unmatched.

Re:Including businesses?

[nolife]

Maybe the fact they used AMD made them lower cost models? Yes there is definatley a really low priced computer in every add that cuts back on some things (normally a Celeron). I think overall, the AMD and Intel models from the mass producers like HP and E-Machine/Gateway have very similar features with the only exception being the processor. Pull up www.salescircular.com, pick a state and browse computer/monitor/printer packages ads from your local stores, for simplicity, here is a link to the VA ads.

Re:Including businesses?

[rcamans]

They said reduce excess inventories.
That does not mean they actually have excess inventories. They just said that. If you believe everything their PR men say, you are indeed a fool.
Actually, they do have excess inventories of laptop cpus. Not huge numbers. No big issue.
A recent slowdown in laptop sales. I do not know why.
Intel has a regular price cut schedule.
It assists them in inventory control.
They do it every three months, with a few extra dates thrown in for fun.
Every three months they roll out new speed steps, and those usually get the top price slot, so everything else has to move down to make room. Current Analysis is basing their numbers on retail sales, but most of "Intel" PC are direct sales, whereas most "AMD" PC sales are retail.
Current Analysis is not the only Market watch / analysis firm saying this stuff, without giving the whole story in big print. It gives them a lot of free publicity, since many big and little magazines and other news portals report that they said that. I have written Fortune about their bad reporting, but they do not care. Headlines mean sales. Not accurate headlines, just headlines. Ever see the national enquirer and other mags at the supermarket checkout? Same thing. Or as I say, same-o same-o. Notice the title of the graph in the article, though? Retail Desktop sales. Oh. now I see. It was in front of me all the time. Doh. heh heh. So they are not lying. Technically. If you look at the headline without reading the whole article, you jump to the wrong conclusion. THey did that on purpose.
It actually means the opposite, since retail sales is steadily declining as a percentage of total sales. Direct sales are the more profitable approach. Just ask Dell. HP does direct sales as well. And IBM. And Intel. Hmmm. The top three firms do direct? Most firms do. Of course HP does sell AMDs as well as Intel CPUs.
But you will have a very hard time finding any article reporting that retail sales of PCs are steadily decreasing as a percentage of overall PC sales. The easy way to figure this out is look at Dell. Each quarter their market share increases. And all their sales are direct. Hmmm. And they are number one in all PC categories in sales. The other big firms are not increasing market share, and they are not 100% direct sales. Ahh, now I see.

Re:Including businesses?

[rcamans]

3 million laptop processors is an annoyance to Intel, not a huge deal.
And market analysts are amazingly clueless. Just look at what they have said in the past. They are so often wrong in hindsight, but that is not something they kie to admit.

Re:Including businesses?

[rcamans]

good point. Although the 20% includes IBM power PCs, HP PA-Risc, Sun RISC, Compaq alphas, most of the 20% is AMD.
There, I did the math for you.
Now take two aspirin and call me in the morning. heh heh.

Re:Including businesses?

[rcamans]

See the title of the graph in the register article? AMD has a leading share of RETAIL desktop sales. Retail sales is declining as a percentage of overall sales. THere is not enough info in their data to tell if AMD is even gaining market share against Intel overall.

Re:Including businesses?

[nolife]

Honda advertised that point only for that specific car (engine/gear combination) because it was by design.

Considering that car peaks at only roughly 146ft/lbs of torque to the tires, high RPM's are the only thing getting it going ;)
Based on its peak torque figure being at ~6300 RPM, high RPM's are good for this car. That is NOT the general rule in all engine/gear combinations. Cars with torque that peak at high RPM's, can take advantage of gearing to get going once that get into that RPM range.

(Torque/5252)* RPM = HP

Re:Including businesses?

[j-turkey]

It's quite likely that your company has some kind of contract with Dell where they purchase exclusively from Dell in exchange for a better deal on those purchases

I believe it's relatively common for companies to do such a thing

I haven't actually seen an exclusive contract -- perhaps that's how it happens at larger companies, where I have less experience (and certainly less management experience). The way I've seen it work at smaller companies is your Dell rep calls you (or your IT Manager/Director/CTO) and asks you how much you have budgeted on hardware for the current FY, they usually ask you to break it down to client machines, servers, and misc hardware. Then they offer you a fixed-price volume-discount deal. Usually, they'll throw in enough stuff to sweeten the deal so that you'll take it. The last deal I worked out with Dell didn't involve signing anything. I could, however, understand how a PC manufacturer would insist on a contract for a larger company since there is far more at stake.

In my experience, more often than not, finance departments like to simplify things. They don't like to have to keep track of 30 different accounts with 30 different vendors just for a small IT group's purchasing (this can be extended to just about any kind of purchasing). Sometimes, they'll have the IT group come out with an RFP to select a vendor (with standard configurations for desktops, laptops, 1U servers, storage, etc), and go with support options and the bottom line. Other groups will just select their favorite vendor and try to finagle a volume deal based on what they have budgeted.

In the end, you have a limited pool of vendors. If your company buys Dells...Dude, you're gettin' a Dell -- that is, unless you're at a high enough level to make the requisite stink to get what you want.

Re:Including businesses?

[Anonymous Coward]

Sure, you can hire competent people to fix them when things go wrong, but isn't it cheaper to hire someone just good enough to talk to Dell tech support?

It might even be better, particularly if your cheap help speaks Hindi.

Re:Including businesses?

[mrchaotica]

Oh, I realize that it's good engineering; I was just commenting on the "Highest redline of any street car! Pretend you're a F-1 driver!" sound bite. Normal people don't consider whether high RPM is good or bad, they just think "ooh, big number!"

Re:Including businesses?

[nolife]

The 2004 S2000 redlines at 8200, previous years had 8900?. According to what I've read, the 2004 RX8 redlines at 9000. Any or all of my figures could be wrong.

Re:Including businesses?

[jazman_777]

Its like the RPM guage on your car.

Look both ways before you cross. Don't play with matches. Buy low, sell high. Car analogies on /. will fail every time.

Re:Including businesses?

[Viceice]

You are right about the high rev part. take a mptprcycle. Most sport bikes redline above 10,000 rpm. Mine does at 11,000 and there are bigger ones that go up to 15,0000+.

Now as an example, the BMW K1200s has a 1.2 litre engine, puny by car standards, but it does the 0-100 in 2.8s and has a top speed somewhere near 300kph, but nobody really knows coz nobody's heroic enough to find out.

Now, with that kind of power it's feul consumption varies alot, at top gear, crusing at 90 kph which means the engine is reving VERY low it gets 49.8 mpg. At speeds above 120kph, the average consumption drops to 30+mpg. A very drastic diffrence.

Re:Including businesses?

[Viceice]

He is right. The AthlonXP 2200+ chips are only clocked at 1.5Ghz.

Re:Including businesses?

[rainman_bc]

First of all RTFA. No where does it say that Dell, IBM, or HP aren't a retail channel. Furthermore, I'd bet that HP and IBM sell a tonne of corporate desktops through VARs, and a tonne more through retail channels like Best Buy. Personally, I think you're full of shit.

Tell me how you figure most of Intel sales are direct? Last I checked, I can't go and buy a chip from Intel. Show me an invoice from Intel for one processor, I'd love to see it. Again, your full of shit.

Re:Including businesses?

[Anonymous Coward]

The Athlon XP 1800+ is clocked at 1.53ghz.

Where'd you pull that mhz metric out of your ass? The actual 2200+ runs at 1667MHz.

Re:Including businesses?

[Anonymous Coward]

seems like you missed the point then...

Re:Including businesses?

[Gentlewhisper]

"Where'd you pull that mhz metric out of your ass? The actual 2200+ runs at 1667MHz."

You forgot that 1Ghz is really 1024Mhz..
So 1.53*1024=... oh wait..

Re:Including businesses?

[Grishnakh]

Wouldn't higher revving = higher frictional loss, and less fuel economy?

Do you drive your car around at redline all the time?

There's a reason most new advanced engine designs are attempting to optimize for both low-rpm fuel economy and high-rpm power, instead of just maximizing power at the expense of all else.

Re:Including businesses?

[Anonymous Coward]

Uhh, no 1GHz is really 1000MHz ......

Re:Including businesses?

[mrchaotica]

Oh, right -- these commercials were before the RX-8 came out.

Re:Including businesses?

[Scarblac]

Intels still run faster, even if they don't crunch the big numbers all in one cycle.

For the same price, there is just no contest. AMD slaughters Intel.

Re:Including businesses?

[Jugalator]

First you say one will slaughter another, then you say the top end chips have roughly the same performance?? I guess one of the top end AMD chips is an Athlon 64 3700+; which Intel processor are you comparing with?

Re:Including businesses?

[Mia'cova]

I wonder if this reason would have anything to do with back to school. The week ending at Aug 21st sounds about right. Are kids deciding more and more on AMD boxes whereas older users are sticking with Intel? I know a few people who've gone purely Intel after the days of the cyrix or K6 chips.... I can't really blame them either. Those were horrible gaming chips.

In any case I'm seeing a lot of shiny new AMD boxes around my campus and it's alright by me.

Re:Including businesses?

[fr0dicus]

Have you seen how many utterly pointless upgrades have been done over the years because of it?

Re:Including businesses?

[mikrorechner]

But what if I was in Italy- and buying from Dell was a pain in the ass?

You might want to have a look at www.dell.it.

All the major vendors have European outposts; they wouldn't be major if they had not.

Re:Including businesses?

[Lord_Dweomer]

"HP are only good for their printers and medical instruments. Down here their computer line is a joke. "

Well, looks like our countries aren't so different after all.

Re:Including businesses?

[aurelian]

HP sells a lot of AMD boxes. Check it out on their site.

Re:Including businesses?

[froggysan]

Actually, well over 10,000 rpm.

Contemporary Formula1 engines rev more like 18,000 or 19,000 rpm at full throttle. Revs may have gone done slightly this season with new rules mandating engine longevity taking effect, but not by much.

Re:Including businesses?

[Wudbaer]

Not only very likely but also extremely attractive if you are a large company. My last employer (a large US life science corp) standardized their PC procurement on Dell getting obszene discounts that way. You can also define so-called "standard configurations" for your company that are even more discounted.

Re:Including businesses?

[severoon]

I'm sure AMD just managed to land a few big accounts that bought a boatload of machines all at once during a week that Intel didn't have a big customer make a bulk purchase. Big deal. If you break down the granularity to a short enough time, I'm sure there's a lot of seconds that go by where AMD sold 0.00953 more machines than Intel did.

Not as a slam against AMD--I like options.

Re:Including businesses?

[Anonymous Coward]

"But what if I was in Italy- and buying from Dell was a pain in the ass? The chance of purchasing AMD just went up about 200 times."

I live in Italy; and buying Dell computers here is about as painless as it is everywhere else.

Re:Including businesses?

[evilviper]

My perl chokes on your subsitution regular expression.

Wow! The first proof-of-concept DoS on the human brain, posted to slashdot...

Re:Including businesses?

[AhBeeDoi]

Trouble is that AMD's run much hotter than their Intel counterparts at higher clockspeeds.

Re:Including businesses?

[zerocool^]

I will tell you a couple of things, though.

I'm a big AMD fanboy. But, my last purchase was Intel, for a few reasons.

1.) It is vastly quieter in the default CPU cooling situation. I have an Antec Sonata and P-4 2.8GHz Socket-T, and when it's not doing 3D acceleration, it's damn near quiet.

2.) Hyperthreading is cool. Without having to spend the money on two physical items, I get some of the performance benifits of dual-procs. My system is more responsive while doing background tasks, like printing or burning a DVD - I can comfortably surf the web with no slowdown.

3.) Forward thinking technology. Origionally, I was going to buy the AMD Athlon 64 3200+, but after I started looking at it, I realized that the socket is at a dead end - if I ever wanted to upgrade, it would mean new motherboard. Also, no AMD boards that I'm aware of have PCI Express. And since the Memory controller is on the processor die, you're tied in to DDR memory forever with the chip. My new setup has DDR2 and PCI-express, so it's much more future proof than what I was looking at from the AMD camp.

~Will

Re:Including businesses?

[Anonymous Coward]

In the benchmarks I have seen, Intel's Xeons 'wipe the floor' with comparable AMD processors. Perhaps you have been brainwashed by AMD.

Re:Including businesses?

[satchboogie]

Exactly! What matters most is how the pipelining is implemented and how many microinstructions can be executed simultaneously.

Although if you have a 200MHz processor that has the best pipelining configuration and acceptional hit rates on cache and can execute many microinstructions at once, and you compare it to an Intel P4 3.06GHz processor, I am certain the P4 will run faster.

There are multiple layers to the P4's design, a complex instruction system over top of the RISK system. You can go to their website and view the history of the Intel chip design. It is quite interesting, especially the pictures.

Re:Including businesses?

[Zorilla]

Yes, but the RX-8 has a rotary engine. Those have always had ridiculously high redlines.

Disclaimer: My POS '96 Trueno has an 8000 rpm redline...doesn't really mean much.

Re:Including businesses?

[paulzeye]

But what if I was in Italy- and buying from Dell was a pain in the ass?

The company I work for uses Dell because they can get the same standard machine anywhere in the world. Whe have offices in Europe, North American, Asia, and Austrailia.

Re:Including businesses?

[Zorilla]

Damn, I have to get myself a 150,000+ redline mptprcycle now.

Re:Including businesses?

[Carewolf]

You mean the one from Anandtech they later corrected because it was wrong?

Re:Including businesses?

[Carewolf]

Actually it is only the performance that is any good (it isn't really that good either). The stability of the platform is not good, even if the drivers in windows does a good job with hidding. For instance NVidia haven't even managed to make a working APIC in their third iteration of Nforce (nforce3 which I have), even though they admitted to having a defective APIC in Nforce2.

Re:Including businesses?

[Blitzenn]

Yes, that's total sales. And I don't know where you have been hiding, but it's not all intel in business anymore either. Most places I have been in are concerned about the cost and AMD machines have been moving in enforce. Companies who are concerned about their bottom line are not goign to pay more money for a PC that does the same work. You can't use that compatability isse anymore either. The XP core was written for and AMD CPU and then horn swaggled to run on an Intel CPU. Thems the facts, swallow 'em how ya like dude.

Re:Including businesses?

[Blitzenn]

First off, yOu are quoting an article that is 6 months old. Second the article is noting the previous years sales numbers making the numbers at least 18 months out of date. You are trying to compare that to last weeks numbers? As far as production capacity. You have to look at it differently than you are now. I have worked at both places building FABS. Sure intel has more production capability, but not to produce the latest chips. Both companies can only produce their latest chips in the newest FABS. That levels the playing field for both of them. When you take into account that intel is on a smaller die size, that is gennerally going to mean that they are producing a lower percentage of 'good' chips. ones that pass the QC checks. You make things smaller, they are more prone to error until you iron out all of the kinks in the process. So FAB per FAB on Latest processor runs, AMD probably can produce more QC passing chips than intel right now. Sorry friend, but your wrong again.

Re:Including businesses?

[Anonymous Coward]

You know the difference between clock speed and a model number, don't you?

Re:Including businesses?

[rcamans]

All of Dell's sales are through their website. A few of those sales are to VARs. but a kot of VAR sales are direct as well.
Like I said before, HP, IBM, and Intel all sell through their web sites. Just goe there amd see.
Intel went to 100% of their sales through the web. All their big customers enter their orders on the web as well.
You can order one cpu or one complete motherboard on Intel's website. That has been going one for years. 25 bi;;ion folars of sales. So apparently you are not paying attention.
I worked for Intel as an EE designing some of those motherboards you can buy from Intel, for 6.5 years.
Intel is actually one of the largest motherboard manufacturers, with more than 10 million a year.
Dell is the largest. I have been working for dell as an EE designing servers for 1.7 years now.
I actually know what I am talking about.
So get a clue.

Re:Including businesses?

[harrkev]

It is vastly quieter in the default CPU cooling situation. I have an Antec Sonata and P-4 2.8GHz Socket-T, and when it's not doing 3D acceleration, it's damn near quiet.

The average /.er does not stick with stock HSFs. Quiet solutions exist for the Athlon 64 as well. Athlons produce less heat than the latest P4s.

Hyperthreading is cool. Without having to spend the money on two physical items, I get some of the performance benifits of dual-procs. My system is more responsive while doing background tasks, like printing or burning a DVD - I can comfortably surf the web with no slowdown.

This IS one area where Intel can legitimately crow. AMD should do this.

Origionally, I was going to buy the AMD Athlon 64 3200+, but after I started looking at it, I realized that the socket is at a dead end - if I ever wanted to upgrade, it would mean new motherboard.

Not if you go for the socket 939. But even the older socket is still OK. The new sempron line will also be advancing. If you are not a "latest and greatest" type of guy, in two years you should be able to get a nice speed bump for well under $200 using the older socket 734, even if it cannot keep up with the Athlon-128 12000+.

Also, no AMD boards that I'm aware of have PCI Express.

You say that like it is a bad thing ;)
Seriously, the only thing right now that needs that kind of bandwidth is gigabit ethernet (which is already built into the mobo anyways), and video (for which AGP is already doing just fine). In maybe two or three years it might be hard to find an AGP card, but the short-term outlook for AGP is just fine. Right now, the average computer user (even gamers) do not really need PCI Express.

And since the Memory controller is on the processor die, you're tied in to DDR memory forever with the chip.

Darn. I can't pay more for the same performance. I am heartbroken. DDR2 promises higher clock speeds in the future. OK. There is some darn fast DDR2 out right now, but I sure cannot afford it. That stuff is expensive.

But if you are talking about replacing the processor (as mentioned above) and getting faster DDR2, then getting a new mobo is really not much extra cost compared with what you are already planning to spend.

My new setup has DDR2 and PCI-express, so it's much more future proof than what I was looking at from the AMD camp.

So how much extra did you pay? I bet that if you had built an Athlon 64 system, not only could you run 64-bit software, you could pocket the difference and use it to do some serious upgrades in a year or two.

But I will admit that if you are the type of guy who is buying new hardware every month, then you probably made the right decision. For most people, though, the Athlon 64 makes more sense.

Re:Including businesses?

[Viceice]

Yes, i'm guilty of horrib

le typing.

Re:Including businesses?

[rainman_bc]

Do you not know what the fucking word retail means?

Here

I even copied it for you so you don't have to click: The sale of goods or commodities in small quantities directly to consumers.

No where does the definition say a website doesn't qualify. Just because you were an EE for Intel doesn't mean you understand english or understand business worth beans.

Re:Including businesses?

[Silverlancer]

No :). I mean the ones from countless review sites across the internet using 64-bit compiled versions of Linux games and programs.

Re:Including businesses?

[rcamans]

Retail means sold by retailers. The business definition of retailers is stores. Buildings that you go into to buy things.

Web sales are not retailers. Direct sales is manufacturers selling directly to the end user.

Dell does this, as well as HP, IBM, and Gateway.
A few stores buy from Dell and HP, and sell to the customers. That would be retail.

"Merchants selling tangible goods in a face-to-face environment who normally use conventional terminals and swipe transactions."

"Trade in which a client buys or sells an over-the-counter stock through a broker-dealer."

"In Retail Sales (formerly cash and carry sales), prices are marked before the store is open, and the customers pick what they want, pay and leave."
"the selling of products, usually in a shop."

So we have three categories:

1) Store sales. Retail.
2) Internet, or web sales by sellers to end users, not sold by the manufacturers.
3) Direct sales. by the manufacturers to the end users, whether over the web or at the manufacturers.

If you understood business worth beans, or at least took the time to research before you commented, you would know this. Do not take such a narrow view that just one info source is the complete answer, even if it calls itself a dictionary or definition. Just because it calls itself a dictionary or authority does not mean it is one.

So if you read the business news, there are constantly articles about how retail sales are being hurt by the internet sales businesses.

Do you remember Egghead going out of business?
Gateway stores closing?
That is because retail stores have much higher costs than web sellers, who have higher costs than direct sales.
This is why many retailers are going out of business, with only a few big guys still in business.
This is why many big store chains are going through chapter 11 or whatever.
Notice that the biggest web seller, Amazon, just started to make a profit after many years in business.

There is quite a turnover in web business as well.
The neat thing about taxes and business is you can run a business for years without making a profit. But you still gave yourself a paycheck.
What happens towards the end is that the employees (not the owners or big guys) stop getting paychecks. The companies supplying stuff on 30, 60, or 90 day till pay terms stop getting paid. Then the business declares bankrupcy. The owners, founders, and big guys did not lose out. They just go off and start up another business, with a different name. Rip off more employees and vendors. Pocket more paychecks.
Greed.

Re:Including businesses?

[Anonymous Coward]

The only area that is really true is video encoding, due to a good SSE2 implementation, and you'll find the P4 3.6 goes faster than the Xeon for that.

Opterons spank Xeons, even the Noconas, for anything else, and the Itanium 2 has but one true use - pure and simple low-bandwidth high-CPU-load number crunching... and it's still nowhere near cost-effective for a job which parallelises well enough that you could cluster it up.

Intel is not doing so well at the moment. Only time will tell how well it fares with the arrival of 45 and dualcores, but it looks like AMD will beat it to the punch again.

Re:Including businesses?

[Chicks_Hate_Me]

I'm glad someone brought it up. Who really cares if your car redlines at 9000rpm if there's no power after 5000rpm?

Re:Including businesses?

[rainman_bc]

Uhm, TFA didn't define what a retailer is. You took it upon yourself to define it.

All I'm saying is that on online business is no different than brick & mortar.

Remember IBM had a chain of home computing stores in malls? I'd call that a retail store. Ergo I'd define IBM's online business as a retail store too. It's no different than mail order. Sears and (in Canada) The Hudsons Bay company had catalog, mail order sales. Is Sears and The Bay then a direct to consumer instead of retailer?

If you read the articles carefully, they refer to in-store sales vs. online sales, not your bastardized definition of what retail is.

I provided you one source, you provided none. You're blowing smoke out your ass.

Re:Including businesses?

[Some_Llama]

"But you will have a very hard time finding any article reporting that retail sales of PCs are steadily decreasing as a percentage of overall PC sales. The easy way to figure this out is look at Dell. Each quarter their market share increases."

market share of what? Direct sales? (you say yes) since this is the only market they are in it has no correlation to retail sales of PCs and their decline.

Re:Including businesses?

[Some_Llama]

"Web sales are not retailers. Direct sales is manufacturers selling directly to the end user."

In the context of AMD selling more processors than intel in "retail" sales then web sites certainly do count.

Re:Including businesses?

[alienw]

1.) It is vastly quieter in the default CPU cooling situation.

The Athlon 64/Opteron is cool and dead-quiet even with the stock heatsink. Besides, there are hundreds of 3rd party coolers available.

Hyperthreading is cool. Without having to spend the money on two physical items, I get some of the performance benifits of dual-procs.

True. But you also lose performance on single-threaded apps. Besides, printing and burning CDs/DVDs doesn't cause any noticeable slowdown on any modern system. The processor doesn't need to do much for either of these tasks.

Origionally, I was going to buy the AMD Athlon 64 3200+, but after I started looking at it, I realized that the socket is at a dead end - if I ever wanted to upgrade, it would mean new motherboard.

These days, it doesn't pay to upgrade just the CPU. When you have ever-increasing memory and bus speeds, you don't want to put a new processor into an old mobo -- you'll always take a big performance hit.

And since the Memory controller is on the processor die, you're tied in to DDR memory forever with the chip.

The integrated memory controller gives a huge speed boost. Athlon 64s haul ass as far as memory bandwidth and latency go. In my subjective tests, an AMD64 system runs Mozilla (which is very memory-intensive) significantly faster than an intel chip.

Re:Including businesses?

[alienw]

Oops, forgot to take care of this one.

Also, no AMD boards that I'm aware of have PCI Express.

Try a Google search for "athlon 64 pci express" and let me know if you still can't find any.

Re:Including businesses?

[zerocool^]

Actually, and you might think i'm crazy, but the P-4 ended up cheaper.

My first choice was going to be AMD 3400+, msi motherboard, 1GB DDR-400, ATI X800 pro, SB Audigy 2.

I ended up with a P-4 2.8Ghz socket-T, 1GB DDR-2 533, PCI-Express ATI X600, and no sound card. And its actually cheaper. What I gave up is I went with the X600 graphics card instead of the X800. So, like, I get the 5th fastest graphics card on the market instead of the 2nd. And the motherboard I got, with the Intel 915G chipset, has onboard digital 5 channel audio built in.

My thought was this: I upgrade every two years or so. In two years, a socket 734 is going to be hard to find stuff for, not to mention, at this moment, there's no motherboards that I know of that can handle more than 2 sticks of ram in DDR-400 (some amd64 boards can handle 3 sticks of 333ddr). Also, I suspect that in 2 years, AGP will be on its way out, and may be relegated to budget cheapie video cards. With this configuration, I give myself leeway - I can still go for the next generation video card in a few years, the Socket-T was just introduced and will carry me onward for a while, and I have DDR-2 slots on my motherboard for future developments. The DDR-2 sticks (2x512mb) were about $110 a piece, btw. Could have been worse.

~Will

Re:Including businesses?

[zerocool^]

See my reply to the guy above you, but as far as putting new parts into an old motherboard, I kind of got a motherboard that has future expandability in mind.

Re:Including businesses?

[zerocool^]

And appearantly you yourself didn't do this:

The first hit contains this news story, published last week...

-----quote-----

Athlon 64 PCI-E chipset here by end of year 'for sure'
By Tony Smith
Published Friday 10th September 2004 12:05 GMT

The first PCI Express chipset supporting AMD's Athlon 64 processor line will have surfaced by the end of the year, an industry source situated not a million miles away from the integrated chipset business told us today.

"There'll be an AMD [Athlon 64] PCI Express product this year for sure," our knowledgeable source said. "At least in the integrated graphics space," he added.

The contrasts with what AMD itself is apparently saying. The CPU maker reckons PCI Express parts won't arrive until Q1 2005, though it's not clear whether it's referring to integrated or discrete chipsets.

Current speculation on the first integrated PCI Express chipset for Athlon 64 centres on Nvidia, which has been offering Athlon 64 chipsets for some time now, and ATI, whose RS480 chipset is said to support the AMD processor family.

Nvidia has already said "next-generation PCI Express [chipsets] are scheduled to be available later this year in time for AMD Sempron market availability", an admission made when AMD launched Sempron last July. Nvidia said its chipset is a Socket 939 part, and AMD's Socket 939 Sempron, the 3000+, is due to ship in Q4, according to AMD.

It's likely that our source was referring to the Nvidia part, but it's not beyond the bounds of possibility that he had an ATI chipset in mind. According to a Hexus report, citing sources familiar with ATI's plans, testing work on RS480 is "developing well, and a solid Q4 launch date is imminent".

VIA is also said to be working on an Athlon 64 PCI Express chipset, the K8T890, but it's release schedule isn't known. ®

----end quote----

Re:Including businesses?

[HuguesT]

I may be wrong but these sort of "I can upgrade later" argument don't really work anymore. An entirely new PC without the monitor is worth what? US$800 at present? In two years time you'll find that populating your currently new shiny with significantly upgraded components will be more expensive and less worthwhile than buying a new system entirely.

Very likely the PCI-express component you'll want to buy in 2 years time will not be compatible with your present motherboard, neither will the memory or the CPUs. This is the first iteration of PCI-express, expect major changes coming.

Just look at the AGP 2x vs. 4x example. Same name, different voltages, incompatible.

I just bought an AMD64 system for a very low price. I don't even plan to reuse the case when I upgrade this machine. I do plan to use this machine as a secondary system until it dies. All I plan to do is double the memory in it in a year's time or so when memory prices have fallen a bit.

Re:Including businesses?

[Psyrg]

I would have thought that producing the peak torque (for a given throttle setting) would producte the highest efficiency. Given that there will be a set charge of energy per inlet stroke (assuming a perfect stoichiometric mix) and assuming the following axioms:

efficency = usable energy out / energy in
usable energy out = force * storke (constant distance)
force = torque / crank size (constant radius)
energy in = roughly constant

then:
efficnency = constant * torque

where the constant is roughly equal to stroke length over crank radius times input energy.

Since torque is variant on angular velocity (RPMs), then wouldnt it be best to use the angular velocity that produces peak torque to gain peak efficnency?

Re:Including businesses?

[Psyrg]

It seems I forgot to mention in my other post that it has been my observation that higher revving engines have shorter strokes. This basically means the frictional force has less distance to work over. This reduction in work can be expressed as energy and as such, you will loose less energy to friction per stroke. Increasing the angular velocity and reducing the stroke may cancel each other out if carefully balanced.

Re:Including businesses?

[jo42]

> HPs have a trend of using AMD chips in their lower-cost models

And I wish that I could give every single one of those AMD-based low cost models back to HP. They have all been POS'es in my experience...

Re:Including businesses?

[obeythefist]

Compared to what, exactly? Via? I have had no end of problems with every Via chipset I've handled. Sometimes a Via chipset will outperform an nVidia one, although nVidia usually fixes that as quickly as they can.

Are there any other chipset manufacturers for AMD that are even worth considering?

Amazing.

[Anonymous Coward]

Better go whip them Euro-Dell's, no Intel? First the stick, _then_ the carrot.

I'm actually amazed that quality can win over brand. Maybe there's hope, after all.

Re:Amazing.

[kfergos]

Whether quality or no, the major thing AMD has going for it is price. For those of us non-gamers, who are fine with a pretty darn good machine, even sacrificing a little quality for a major difference in cost is worth it.

Re:Amazing.

[spuzzzzzzz]

I'm actually amazed that quality can win over brand. Maybe there's hope, after all.

While AMD's x86 stuff is better than Intel's x86, its a bit sad that Itanium has lagged so far behind Opteron. Itanium's architecture is vastly superior to Opteron's, as it marks a break from the 20-year accumulation of old designs and legacy crap. It would be nice to see people embracing a new architecture for once.

Or course I can't claim superiority, having purchased Opterons myself. I guess software availability will always win out over good chip designs. Just ask those poor Alpha designers over at Compaq/HP.

Re:Amazing.

[qopax]

do gamers actually buy intel cpus? gamers that know about hardware that is?

I'm a pretty dedicated gamer (it kills me in life hehe) and I always buy AMD for my system, mostly cause I'm only 15 and a barton 2500+ for 80 bux damn easily overclocked to 3200+ provides a shitload of performance relative to the price.

Looking at the other end of the spectrum, where gamners have enough money to buy powerful stuff, they would just buy an athlon 64 or even an FX to power their gaming machines, because I'm pretty sure it's well known those processors do better in games than intel as well...

I guess there's a middle of about 200 dollars where it's probably better to buy an intel processor, but why not just save a bit more and buy an athlon 64 processor anyways?

This is all regarding a new system purchase, cause if you're talking about upgrading a processor on an already a very good motherboard, the gamer doesn't have much of a choice...

Re:Amazing.

[Grishnakh]

Alpha also was a clean design, not a 20-year accumulation of legacy crap. Just because Itanium shares this does not mean it's "vastly superior". The proof is in the numbers.

The question is, how does the Itanium actually perform, using software generated by currently-available compilers? Lots of people talk about the Itanium in theoretical terms, but this processor has been in the marketplace for some time now, but isn't doing too well.

If I recall correctly, the Alpha performed extremely well, but was somewhat expensive (especially compared to x86 chips). The Itanium doesn't have the same excuses as the Alpha did for being expensive. It's made by Intel, which has by far the largest fab capacity of any company which could be used for making a chip like this; this is the big reason why Intel's x86 offerings are as inexpensive as they are. They could be even cheaper, but Intel wants to keep a high profit margin. So as far as manufacturing capacity goes, there's no excuse for Itanium to be expensive, compared to other 64-bit chips like the Opteron and the upcoming Prescott. It should only be a function of the die size. So if the Itanium is not able to outperform the Opteron, in real-world tests, on a dollar-per-benchmark-unit basis, then the Itanium is a failure, plain and simple.

Re:Amazing.

[spuzzzzzzz]

Alpha also was a clean design, not a 20-year accumulation of legacy crap.

Perhaps I should have explained myself better. I meant that Alpha was a good design that never got the market share it deserved. Similarly, I believe that Itanium has a good design. I feel it is a shame that the high price has kept Itanium from being widespread. I agree that Intel has no reason to keep the price so high.

All the benchmarks I have seen test the Itanium on workstation-style or single-task-sever jobs. The (theoretical) strength of the Itanium is in its multitasking. Its simpler design should reduce the overhead of switching tasks significantly. This makes benchmarking it kind of hard, since benchmarks generally only test one thing at a time.

Re:Amazing.

[Anonymous Coward]

since when would the common market embrace something stupidly expensive?

Re:Amazing.

[Glock27]

All the benchmarks I have seen test the Itanium on workstation-style or single-task-sever jobs. The (theoretical) strength of the Itanium is in its multitasking. Its simpler design should reduce the overhead of switching tasks significantly. This makes benchmarking it kind of hard, since benchmarks generally only test one thing at a time.

Single tasks are certainly an important measure of performance, especially in scientific computing (one of Itaniums obvious markets, since it has strong FP performance).

I'm not sure why you think it's so good at multitasking. There's a reason it requires such large cache sizes, and that's not a good sign for multitasking performance. Personally I'm pretty sure a similarly priced dual-Opteron will outperform current Itaniums, even heavily multitasked... ;-)

and with pointless time-wasters like EPIC

[Anonymous Coward]

...Intel will begin to lag behind in the server market too. What exactly does EPIC provide, apart from a lot of work for compiler writers and a (theoretical) maximum 4-fold speed increase with current designs?

Re:and with pointless time-wasters like EPIC

[ndykman]

Uh, a theoretical 4-fold speed increase isn't anything to laugh at. Ask AMD, Intel, and IBM if they'd like to have a new idea in microprocessor design that had a theoretical 4-fold speed increase, and I'm sure they'd all say: "Yes."

I feel bad for Intel and HP though. Sunk quite a bit into making sure Linux could run on the Itanium from the very start, getting little for it. Really, the Itanium and Opteron are like apples and oranges.

One major problem is gcc. GCC just can't handle EPIC stuff yet. The compiler from Microsoft, Intel and HP are quite a bit much better than gcc. But gcc is the defacto complier in Linuxland, like it or not. (Even though Intel's x86 compiler for linux can do better too).

Re:and with pointless time-wasters like EPIC

[MostlyHarmless]

What exactly does EPIC provide, apart from a lot of work for compiler writers and a (theoretical) maximum 4-fold speed increase with current designs?

"All right, but apart from the sanitation, the medicine, education, wine, public order, irrigation, roads, a fresh water system, and public health, what have the Romans ever done for us?"

Re:and with pointless time-wasters like EPIC

[Anonymous Coward]

Uh, a theoretical 4-fold speed increase isn't anything to laugh at. Ask AMD, Intel, and IBM if they'd like to have a new idea in microprocessor design that had a theoretical 4-fold speed increase, and I'm sure they'd all say: "Yes."

Why not pick an old idea instead? Ever heard of vector processors? If Itanium is bitch to compile code for, you haven't seen yet anything..

Re:and with pointless time-wasters like EPIC

[Anonymous Coward]

I sum up EPIC like this: a human can tap their head while rubbing their belly at the same time. It's quite hard for (most) people to do this successfully, and most of the time it's quite pointless, but on the rare occasion you need to do these things simultaneously, you will save time by having mastered these two skills together.

I'd say that analogy is closer to reflecting the likelihood that you're going to achieve the theoretical maximum speed increase than the combination of Roman sanitation, medicine, education, &c.

Re:and with pointless time-wasters like EPIC

[Grishnakh]

One major problem is gcc. GCC just can't handle EPIC stuff yet. The compiler from Microsoft, Intel and HP are quite a bit much better than gcc. But gcc is the defacto complier in Linuxland, like it or not. (Even though Intel's x86 compiler for linux can do better too).

Maybe if Intel would stick to selling processors, and not compilers, they wouldn't have this problem. If they were smart, they'd donate their compiler technologies to gcc, or work with them to use plug-ins.

As for this theoretical 4-fold speed increase, has this actually been realized? Itanium processors have been around for many years now; they're not a recent offering. If they're so great, why aren't we seeing actual benchmarks showing how great they are? Why don't we see them being used in all these supercomputing clusters that are popping up everywhere?

Re:and with pointless time-wasters like EPIC

[Billly Gates]

Personally I think Itanium is an electrical engineering disaster.

HP has no concept of sunk costs. It would be best to dump it.

Intel as the alpha and its trying to morph it into the Itanium.

Intel/HP should just recontinue the alpha instead.

Risc is in. VLIW stuff and moving things in software = bad.

Re:and with pointless time-wasters like EPIC

[Anonymous Coward]

They have given us Gina Lollabrigida.

Re:and with pointless time-wasters like EPIC

[Anonymous Coward]

and a (theoretical) maximum 4-fold speed increase with current designs?

I take you mean the VLIW design -- the instructions arranged as 4-element words and the processor arranged as 4 parallel engines -- right?

But current x86 or PowerPC processors aren't simple 1-pipe designs. They have been superscalar for many moons.

For example, Athlon "Classics" already had multiple execution units: they can do both a FMUL and a FMADD during one cycle. Current Pentiums and Athlons have SSE2. Motorola G4+ introduced the AltiVec for four 32-bit or eight 16-bit ops per cycle for vectorized code (albeit the crunching was limited by poor system bandwidth). Not sure about the PPC970 "G5" with 64-bit ops, probably not doable with Altivec; but it still has *two* 64-bit scalar floating point engines (and no such bandwidth problem).

Provided good compilers, I guess Itanium can get the 4-way goodness going on in way more places, if not in all of the code. But the competition isn't strictly "1-way everywhere", so... ... I humbly challenge your "maximum 4-fold speed increase". :)

Re:and with pointless time-wasters like EPIC

[Anonymous Coward]

aiigh I mean FADD not FMADD

Re:and with pointless time-wasters like EPIC

[10Ghz]

Brought peace?

Re:and with pointless time-wasters like EPIC

[ndykman]

Actually, VLIW ideas did borrow from vector processors a bit. Was it a good thing, hard to say.

Notebook sales

[erick99]

It seems that AMD's success on these occasions are due to notebook sales:

Duboise continues: "promotions continue to be the driving force behind retail PC sales and AMD's successes. In fact, $699 notebook promotions have been the driving force behind three incidents this year when notebook sales were able to overcome desktop sales. As long as Intel continues to place more emphasis on the more lucrative and successful notebook market, it leaves the door open for AMD's desktop wins."

I wonder if they believe that they can eventually drive notebook sales upward to the point that they outsell Intel more often than a handful of times a year?

Cheers,

Erick

Re:Notebook sales

[imbert]

I am VERY happy with my new Compaq notebook R3000 Good price! Excelent speed AMD64 3.4, etc!! Best warranty (e years full coverage) :) The only problem is wireless from Broadcom...doesn't have 64 bits driver since the only way to use that card is by ndiswrapper. AMD Rocks on my Notebook! Very fast!

Re:Notebook sales

[Anonymous Coward]

i'd agree.. hell, i'm a poor college student and even i'm tempted to get a notebook now that the prices are down so far.

not sure how that really affects desktop sales tho :\


______________
Free Ipods
It actually works!

Re:Notebook sales

[zaxios]

There's a fucking header with your fucking user name and a link to your fucking personal page where your fucking name can be listed if you want to.

Yeah, totally... except on yours. It would have been helpful if you provided your fucking name at the end of your fucking Slashdot messages so we know who the fuck you are next time we fucking see you. Fuck?

Re:Notebook sales

[erick99]

What in the world happened to you in your childhood that made you aversive to the polite custom of attaching your name to the end of any type of communication? Get a life - then work on being nice.

Cheers,

Erick

Re:Notebook sales

[Anonymous Coward]

Best warranty (e years full coverage) :)

2 years 8.6 months is good, but I've seen better.

Re:Notebook sales

[NotoriousQ]

(e years full coverage)

Sweet. Irrational term of warranty. But then again, no one will ever be annoyed that something failed in the last possible moment.

However, e years is not the best warranty, as I am certain I have seen 3 years at least.

Re:Notebook sales

[Anonymous Coward]

haha, that was the better reply of the two, thanks for making me LOL

Re:Notebook sales

[Fjandr]

Hmm, make mine a c years warranty. :)

Re:Notebook sales

[Behrooz]

Eh, I can beat that. I want a C grade warranty. :)

Re:Notebook sales

[Anonymous Coward]

I think his point is that it's completely redundant. Your name is already signed for you at the top of your post. We know who you are. No need to throw your identity in our face repeatedly.

At the very least, if you're going to write something at the end of every post you make, put it in a sig or something so the sig haters can block it. That's what sigs are for anyway.

Re:Notebook sales

[pete-classic]

They're running a promotion this week, mention Pythagoras and they upgrade you to pi years at no extra charge!

-Peter

Re:Notebook sales

[Bert64]

16 year warrantee? assuming a hexadecimal form.. that's not so bad

Good to hear!

[TheKubrix]

First ATI outsold Nvidia on desktops, and now this! Good to see theres not a monopoly on core hardware components! now if only software were the same way, :\

Re:Good to hear!

[Gilesx]

ATI mainly outsold Nvidia because of Nvidia's shoddy manufacturing of early Geforce 5 series cards - poor drivers, drivers that lied, and late to market hardware that looked distinctly weak by the time it was public. This was a direct mirror of the emergence of Nvidia over 3DFX as a major graphics card force a few years ealier, with the exception that this time around, Nvidia had a lot greater cash reserve than 3DFX ever did, so could actually afford to make the mistake.

As it is, I'd be very surprised to find out that the ATI share was more than 55/45 in their favor (remember - a LOT of people outside of hardcord gamer circles are still using early Geforce / TNT cards - I have even seen Geforce 2 *MX* cards still being sold as low cost no frills acceleration) and with the new 6600 cards coming out, this is going to be a firm kick to the nether regions of ATI. There just isn't a card on the market that can hold a candle to it, and when you combine this with Nvidia's far superior Doom 3 performance, I'd certainly not bet against Nvidia becoming a dominant 3D acceleration force over the next few years.

Re:Good to hear!

[shfted!]

It's not as if RedHat is the only OS you can install. Mandrake, Debian, Gentoo, and SuSe are also quite popular. Then there are NetBSD, OpenBSD, and FreeBSD. What is this monopoly you are talking about?

Re:Good to hear!

[nbowman]

Isnt the hardware business more of an Oligopoly situation? Not very many players, and high entrance cost makes it a hard business to get into.
and all software definitely isnt a monopoly market, the OS market is pretty close though.

Re:Good to hear!

[VoidWraith]

Even two way competition is good. In these markets, its not friendly, its bareknuckle. As it is, most of the big players are evenly matched, keeping our prices lower. More would be better, but its hardly a problem.

Re:Good to hear!

[Bri3D]

But ATI has quality control issues with Radeon 9800 series cards. I have seen two non-OC'ed cards with polygon loss(RAM Issues) and mine does not let my system boot(the BIOS freezes). They were all replaced with working ones, after the annoying return process, but...remember...these are ones I have seen personally. I hear there are more :-).

Re:Good to hear!

[kayak334]

What is this monopoly you are talking about?

I think he's talking about Windows.

Re:Good to hear!

[spuzzzzzzz]

and with the new 6600 cards coming out, this is going to be a firm kick to the nether regions of ATI. There just isn't a card on the market that can hold a candle to it

I'm not so sure about that. In DooM3, yes, nvidia kicks ass. And it will be my only choice until ATI comes out with some halfway decent linux drivers. However, the x800 and x600 series don't look half bad. When Half-Life 2 comes out, I think ATI's offerings will start to look more attractive.

I am of the impression that macs only use ATI and that the xbox2 will also use a graphics processor from ATI. So don't be so quick to dismiss them.

Re:Good to hear!

[shfted!]

X Windows? Naw, X.org is out now. No monopoly :)

Re:Good to hear!

[Exitthree]

Apple offers about half of their computers with either an ATI or nVidia card respectively (or a choice betweeen either). The top of the line (BTO) card in the PowerMac G5 is currently an nVidia card, but all the Powerbooks are ATI-based. It goes back and forth depending on current tech fairly quickly.

Re:Good to hear!

[RyuuzakiTetsuya]

This was a direct mirror of the emergence of Nvidia over 3DFX as a major graphics card force a few years ealier, with the exception that this time around, Nvidia had a lot greater cash reserve than 3DFX ever did, so could actually afford to make the mistake.

not just that, but nVidia also has more than that going for them.

They diversified in the industry. They're not just manufacturing video cards, they're also making mainboard chipsets as well as other multimedia pursuits. Not exactly a wide, sweeping diversification, but given the mistake, they atleast have another business to fall back on. I will swear on my nForce2 based mobo and recommend it to anyone who's looking for a solid AMD board. I used to be Intel-only, until I decided to take the gamble with AMD. I haven't looked back yet.

Re:Good to hear!

[mrchaotica]

all the Powerbooks are ATI-based

Actually, the 12" comes with an nVidia GeForce FX Go 5200. But you're right that they're pretty graphics-vendor agnostic. However, considering that ATi is better at DirectX and nVidia is better at OpenGL (at least right now), you'd think they would use nVidia in all their computers.

Re:Good to hear!

[Barlo_Mung_42]

"Exactly my point"
-BillG

Re:Good to hear!

[The Kow]

I fail to understand why posts like this get modded up. How is this an interesting statement? Is there a (-1 Boring) or (-1 Uninspiring) mod in the works? Please, can we at least try to quit pandering to karma whores?

Re:Good to hear!

[Calroth]

"However, considering that ATi is better at DirectX and nVidia is better at OpenGL (at least right now), you'd think they would use nVidia in all their computers."

That's more a driver issue than anything else. I wouldn't infer anything about Mac OpenGL performance from Windows OpenGL or DirectX performance.

Re:Good to hear!

[Fjandr]

Umm, it's a joke. :)

Re:Good to hear!

[mrchaotica]

That's a good point. I wonder if, being a corporation, Apple can influence them more to make a good driver than the conglomeration of geeks that use Linux?

Re:Good to hear!

[StarsAreAlsoFire]

considering that ATi is better at DirectX and nVidia

You have that backwords, right? Or does ATI just suck at both? I suppose it *was* a year or so ago, so the 'right now' modifier might come into play....

I certainly have never heard of anyone that had to return a laptop because NVidia couldn't write to specs: link.

After dealing with the issue for a month (favorite advice from laptop vendor: 'just disable hardware acceleration') I will most certainly never buy a laptop from said dealer, nor I will ever own another ATI card.

Dealer wasn't IBM (as per the link); it was Dell. What the hell is it about the crooked 'E' that makes a company turn to the dark side?

Re:Good to hear!

[Anonymous Coward]

Nvidia's far superior Doom 3 performance

You act like doom 3 sold more copies than deer hunter. Doom 3 was a total blowout, nobody bought it and nobody is buying it now. HL2 performance is what will matter.

Re:Good to hear!

[aurelian]

And it will be my only choice until ATI comes out with some halfway decent linux drivers.

Or ANY 64-bit linux drivers, which, considering that we're talking about AMD sales and Athlon64 is what most of the new boxes contain, is a real problem.

Re:Good to hear!

[zonker]

something interesting to note is that apple doesn't do directx, they do opengl. nvidia tends to do opengl stuff a far bit better/faster than ati (at least in the past anyway)...

Re:Good to hear!

[PantsWearer]

Or ANY 64-bit linux drivers...

This is a niche in a niche market. Yes, a relatively huge number of 64-bit capable chips are being sold by AMD currently, but they're at best only half of the recent chips sold, so overall still a very small portion of the market. And what number of those chips actually run a 64-bit OS? That's even small compared to the overall linux market, which is already pretty damn small.

I expect this will change very quickly once a 64-bit Windows OS actually goes final.

They may have won a battle......

[ARRRLovin]

....ergo the cliche.

A little competition never hurt anybody.

OK, what was the specific info last time?

[Anonymous Coward]

Last time this info got erroneously foisted on us here by a submitter who didnt bother to RTFA (oh gee, big surprize for Slashdot), it turned out that it was number of solo CPUs sold that AMD won the share for (for like 3 hours during the middle of night), not desktops (like Dell, Gateway, etc.) sold. Is this more of the same?

Re:Intel secret plan

[Anonymous Coward]

I doubt that 10.000 MHz will kick AMD's ass.
Come to think of it, 10.000 MHz would barely outperform my 7.314 MHz Amiga 500 :-}

I like it

[Aggrajag]

I've been using AMD chips since K6 and haven't been disappointed yet. The criteria I've used when buying a new cpu has always been the price. The next one I'm going to buy will be the Sempron for Socket A, 64 bit cpus can wait a year or two.

Thanks AMD,
Your Faithful Customer

Re:I like it

[PedanticSpellingTrol]

The sempron is a real performance dog, have you considered one of the athlon mobile line instead? Unlocked multiplier and minimal power consumption = yummy. ~$75

Re:I like it

[HermanAB]

Well, if price is your main concern, how about VIA?

Their processors are tiny, use almost no power and only costs about $50.

Re:I like it

[Anonymous Coward]

You are on crack, aren't you? Check the reviews (in abundance) on Sempron. Apart from the idiotic name, it's very attractive: better than lukewarm performance and clearly the best bang for the buck AMD offers.

I agree that Athlon mobiles are attractive for overclockers and silent system builders.

Re:Intel secret plan

[Steve Cowan]

Nobody could ever need more than 7.314 MHz of speed.

whole AMD desktop?

[thhamm]

whats an AMD|INTEL|... desktop anyway?! my "desktop" is made of cherrytree.

Re:whole AMD desktop?

[thhamm]

seriously, buy what suits you best. and dont argue bout it. hugh. and where have my qoutes gone?!

Re:whole AMD desktop?

[Anonymous Coward]

im sure they meant as opposed to a beowulf cluster. i know i know i only have super computers in MY house but others.. well lets just say their mother sniffs paint.

Re:Who would buy intel?

[ecc0]

Anyone who cares about using their computer for professional audio applications, for one. Sorry, I used to be like you, but experience has led me to realize that if you're going to do any serious A/V work, AMD -- or more specifically, AMD motherboards with cheap, ultra-crappy VIA south/northbridges and the like -- just won't cut it. You will get crackles in the audio, all around shitty performance and will have to apply about 200 patches to get things working at all.

If you're going to do anything more than play Quake ]|[ or browse the WWW and write Word documents, getting the CPU and chipset which is considered the reference implementation (meaning Intel chipsets on Intel motherboards with Intel CPU's) is worth it.

Re:Intel secret plan

[ricotest]

Intel is planning to release a 10.000 MHZ cpu and kick AMD poor lame ass.

Watch it, in several European countries (and a bunch of non-European ones) that truly does mean 10,000 MHz :)

Chickens finally coming home to roost. :-)

[Anonymous Coward]

Enjoy, Intel. Through years of doing little but ramping up your clock speeds and calling it "innovation," you made your bed.

Now it's time to lie in it.

Eat shit and die.

lower tha prices

[djxploit]

hopefully with this bit of healthy competition intel will lower their prices over amd and then amd being able to post correct clock speeds not at what the chips called.

One monopoly down

[Anonymous Coward]

One more to go! ;-D

Congrats to AMD.

Re:One monopoly down

[Gilesx]

The thing about monopolies is that, as in a business sense, it's the equivilent of launching the space shuttle at the end of Civilisation - you win, you achieved the ultimate goal, and every company wants to be in that place.

Therefore, behind a monopoly, there's always another monopoly watching and waiting in the shadows, looking to take over from the popular and dominant market force. For example:

Microsoft (Apple)
Intel (AMD)
Nvidia (ATI)
iTunes (Napster)
IBM (Sun)
Gnome (KDE)

They'll tell you that they wouldn't ever want to be a monopoly, but it's poppycock - the whole principles of business are based on monopoly.

Re:One monopoly down

[Anonymous Coward]

Actually, you dare not kill off all the competition. In that case, they take you to court and take you apart. Apple still exists because Microsoft had to show courts that it actually had competition.

Re:One monopoly down

[The Lost Supertone]

Business is based on monopoly? Dang... fisher price is gona start going all SCO and making businesses pay licensing fees... if it's based on their technology... or board game...

Re:One monopoly down

[VoidWraith]

Intel hasn't been a "monopoly" for many many years.

Re:One monopoly down

[Anonymous Coward]

Nvidia (ATI)
IBM (Sun)
Gnome (KDE)

these three seem suspect....none of the three represent monopolies and the Nvidia (ATI) and Gnome (KDE) might not even represent market share dominance....

stendec@gmail.com

Re:One monopoly down

[mrchaotica]

Gnome (KDE)

Regardless of your personal opinion*, KDE is more widely used, so that's just a little backwards.

*I don't like KDE all that much either, but then again I'm also not a big fan of GNOME

Re:One monopoly down

[Anonymous Coward]

Heh, I'm with you. That comparison should not have even been made.

And yeah, both KDE and GNOME suck for various reasons. If you could combine them and take the best from each then you would have one kick-ass system.

Re:One monopoly down

[mrchaotica]

I don't really hate GNOME and KDE; I hate the fact that there are two of them (well, plus XFCE, etc). I really wish there was One True GUI Program Toolkit. Until there is, I'll have a problem with ALL "Desktop Environments."

Re:One monopoly down

[Anonymous Coward]

IBM (Sun)

Really it is the other way around :) but now Sun is already going down so it does not matter.

iTunes (Napster)

but Napster sucks.

Re:One monopoly down

[Anonymous Coward]

There is. The One True GUI Program Toolkit is called MFC. To a first approximation, no other toolkits have any substantial existence.

A Long Way from "AMD is Dead"

[dragon_imp]

We've come a long way from the "AMD is Dead" and "Intel Rules" days.

Intel let its marketing people get caught napping. Intel pushed the Itanium and said it will never make a 64-bit chip that is x86 compatible.

AMD came out with the 64 bit chip that was compatible with the x86, and it got rave reviews. And, it gets sales!

Now, AMD outsells Intel again. Did you see that -- the article said "again."

Not bad for a company that was being written off a couple years ago.

Re:A Long Way from "AMD is Dead"

[melted]

>> Intel let its marketing people get caught napping.

To the contrary. Intel let its marketing people tell the engineers what to do. So they basically said, "we want a 3GHz chip, because consumers are stupid and they only look at GHz figures". P4 is a result of this. It's only real feature is that it can be clocked insanely high. Clock for clock it's not only dumber than AMD chips, it's also dumber than some of Intel's own processors (Pentium M for example).

Re:A Long Way from "AMD is Dead"

[Anonymous Coward]

but... but.... :( Netcraft confirmed it.... how can this be!?

Re:A Long Way from "AMD is Dead"

[Anonymous Coward]

Heck, isn't it slower clock-for-clock than the PIII?

Re:A Long Way from "AMD is Dead"

[Jeff DeMaagd]

Heck, isn't it slower clock-for-clock than the PIII?

The grandparent post mentioned Pentium M, but Pentium III counts as Pentium M is true successor to the P6 (Pentium Pro, II & III) heritage, but with a FSB speed (400MHz vs 133). They have larger caches available as affordable consumer chips, 1MB & 2MB are available in Pentium M when those were available in PII, PIII Xeon, but not cheap.

All iterations of Pentium IV (Willamette, Northwood, Prescott) have a lower IPC than Pentium III.

Not that I think IPC is a useful measure on its own, much like GHz is not a useful figure on its own. I really didn't see that lower IPC was necessarily a negative when the increase GHz more than made up for it, which it did for two or three years, except for a brief period when 1GHz was the latest, it wasn't until Athlon64 did AMD match or beat Intel's performance at the top end. Up until then, AMD's selling point was mostly the more affordabe midrange. Now, AMD has the top end, the midrange and price.

Re:A Long Way from "AMD is Dead"

[Anonymous Coward]

You'd better hurry- you might be late for your freshman EE class!

Re:A Long Way from "AMD is Dead"

[jo42]

> Willamette, Northwood, Prescott

Then you have the Prescott, which has twice the cache of the Northwood, sucks more power (generates more heat) than a Northwood, and benchmarks about the same. Talk about an engineering fubar...

Figures

[chaffed]

I was an ardent fan of intel until the Athlon 64 came out. My brothers new PC has an Athlon 64 along with other goodies (1gig ram, dual layer dvd writer...) for a very reasonable $1,000 USD.

There is no way I could have done that with an intel chip and motherboard and still get the same performance.

Re:Figures

[here4fun]

The last home made PC I built was an Intel, and it was very inexpensive, an abit motherboard, cd-rw, etc for under $500. And I am happy with the Celron chip. The only reason I have not built an AMD system is because back in college I knew someone with an AMD and it overheated. Another guy saved a few bucks buying a cyrix, and windows crashed on it all the time. The systems that worked were intel systems.

But the past two years I have been hearing a buzz surrounding AMD. I might have to check it out. I just don't want a $1000 mistake sitting on the floor because I didn't apply enough thermal gel or because I got the wrong heat sink. And I don't want any system instablilty.

Re:Figures

[chaffed]

I was really impressed by the OEM heatsink and mounting system. As long as the system runs at its rated clock speed the OEM gear works fine.

I highly recommend AMD chips on custom built machines. Plus you don't need to spend $1000 dollars. My brother does audio production so the audio processor I put in was $270+. Nor do most people need a dual layer DVD RW drive.

Re:Figures

[optikSmoke]

Um, ya. Since you said "back in college" I can only assume you are talking in years. Get with the times, man. On the other hand, if you're "happy with a Celeron" you obviously don't need much performance anyway..........

Re:Figures

[qopax]

hehe, if you put "the wrong heatsink" (i.e. cat), on your pentium processor, it wouldn't last much long either. But thank god for those amd-only nforce motherboards... I'm sure other motherboards have cpu overheating protection, but can they really beat the nforces :)?

and like someone else stated, if you're running celeron, you obviously don't need that much performance...

hell I'm running an AMD k-6 166mhz at 450mhz with windows xp installed, and 128mb ram. Runs pretty well with all the visual effects enabled, hehehe

Re:Figures

[Fjandr]

Yeah, the k5's were horrible back in the Pentium classic era. That was a while ago. AMD has been through three chip families since then.

Ah, that explains Intel pimping a new Internet.

[Anonymous Coward]

If you can't beat 'em, change games.

Re:Who would buy intel?

[Anonymous Coward]

The VIA KT600 chipset is definitely slower than the exceptional NForce 2 chipset.

It is also cheaper.

Anything Intel blows goats compared to a decent Nforce 2 board.

Re:Who would buy intel?

[Anonymous Coward]

.. ever heard of the nForce chipset?

Great to hear!

[Donoho]

Competition means better products at lower prices. The intertwined/alternating point on the graph in the article makes me happy about the state of hardware.

Nothing like competition to make a good company's products better.

Re:Great to hear!

[Jeff DeMaagd]

Competition means better products at lower prices

Almost always, competition means lower prices. It does not always mean better products, unless you mean better product for the dollar. If a product were a third less reliable for half the cost, then I suppose that is a win, if you don't consider environmental impact of each respective product.

As for better product, it doesn't always seem to be the case, at least as often. Sometimes there is some corner cutting on the part of all competitors, note the quality or availability of support lines, flimsier material or the missed testing step in a rush to get the product out to beat the next company. Many consumer electronics seem less reliable these days.

Re:Great to hear!

[Fjandr]

Almost always, competition means lower prices. It does not always mean better products

Very true. Product quality depends on either an educated consumer willing to pay comensurately, or a company with market presence choosing to build a quality product. Since the former doesn't exist except as a tiny minority, the only time competition produces quality and low prices is in a dynamic market that has not (yet) stagnated. Generally, that stagnation will occur sooner or later, unless the market has a major player that is A) privately held AND B) run by someone with integrity.

Re:Great to hear!

[Spoons]

Almost always, competition means lower prices. It does not always mean better products...

That's right! And as we all know, monopolies lead to great products!

Hey, Dell !!!

[Nom du Keyboard]

Hey, Dell,

Are you listening?

Re:Hey, Dell !!!

[MerlynEmrys67]

Are you listening?

Loud and clear I would assume - Lets see AMD has a 50% share of the retail (Desktop) market and a 16% share of the overall market. Guess where Intel is selling ?

Re:Hey, Dell !!!

[Anonymous Coward]

Are you listening?

Intel will rape us of our special prices if we break our exclusivity agreement.

Besides, I like the under-the-table kickbacks for being intel-only, and I own a shitload of INTC!

Re:Hey, Dell !!!

[scrod]

They'll catch on eventually. Once enough of their competitors begin selling cheaper, better performing PCs they won't have any choice but to switch. The king of commoditization will eventually be pulled kicking and screaming by the same forces of capitalism that it had (ab)used previously to obliterate existing markets.

Re:Who would buy intel?

[Anonymous Coward]

200 patches just to work?

the BS-o-meter is going crazy... yes, we have a troll.

Re:Who would buy intel?

[Anonymous Coward]

That was total bullshit, not insightful at all. Anybody doing serious audio work on an X86 PC, even at the low end of the market, is using a breakout box for I/O and conversion.

Congratulations.

[Junior J. Junior III]

I really wish I had bought stock a two years ago when it was at $5/share...

HT

[Lux]

Does AMD have any Hyper-Threading-like technologies in their chips or in the works?

It's such a high bang-for-your-buck optimization that I'd feel a lot more comfortable buying AMD chips if I knew they weren't hobbled by not exploiting it.

I'd be really stoked if they started maintaining an extra execution context internally, but use it for speculative execution. Sair got some pretty impressive results doing that, and single-threaded apps actually stand to benefit from that application of the technology.

-Scott

Re:HT

[ARRRLovin]

Yes, it's called Dual Core. They're foregoing the "emulation" of a second chip and just puting it on the same core as the first.

Re:HT

[Anonymous Coward]

Considering AMD doesnt need the HT crutch to make up for an extra long pipeline, no they probably wont be implementing it. Not to mention the Dual core chips that are going to be out before Intel's Dual core chips.

Re:HT

[spuzzzzzzz]

The purpose of HT is to make up for Intel's crappy super-long pipeline (something like 32 stages!? Someone correct me if I'm wrong). Whenever it does a jump, many instructions are wasted. AMD's pipeline on Barton was something like 12 stages, so there's much less wastage going on. All HT does is allow the wasted cycles to be used for another thread. Since AMD's processors don't waste so many instructions, HT wouldn't really help that much.

Re:HT

[Anonymous Coward]

Whenever it does a jump, many instructions are wasted

No. Branches generally do not cause a pipeline flush. (This is why branch prediction is a hot topic.)

HT does not exist to operate only in pipeline stalls. HT exists because analysis demonstrates that most x86 programs do not exhibit enough parallelism to fully utilitize all of the multiple execution units in a modern Pentium. You've got a lot of silicon devoted to peak performance that isn't used all the time, because you don't happen to have (for example) a bunch of full-width add instructions going on at the same time. HT allows a second thread to use those chip resources.

HT is cheaper than building two processor cores, as lots of the instruction fetch and decode logic is shared. Putting two complete cores on the same die does not increase the efficiency of utilization of the resources in either core. Dual core is much more of a brute force solution to the problem (a complaint AMD fans usually lodge against Intel). In this case, execution units in both cores will often be idle, as neither thread alone happens to need the full capability of a single core.

Since you've spent more silicon on the problem, dual core can have performance advantages -- specifically whereever you actually need that duplicate logic that would be shared with a HT design. Often, however, that extra fetch/decode logic is going to waste as well.

HT is an elegant optimization for a modern superscalar processor. It is not, however, the same thing as a dual processor, nor does it solve exactly the same problem.

Re:HT

[Lux]

Yeah, that sounds a lot more like what I learned in architecture than the "AMD doesn't need HT" dogma I'm hearing. Efficient use of silicon is good no matter how long your pipeline is.

Thanks for your insightful, if anonymous reply. :)

So it sounds like AMD is going without HT.

Damn.

Re:HT

[MrNemesis]

HT (Simultaneous Multi-Threading) won't benefit AMD's chips as much as it does Intel's because of the way they are constructed.

Due to the P4's incredibly long pipeline (30-odd stages?) and very high clockspeed, if the branch prediction goes wrong, the chip will stall

HyperThreading is a clever hack that runs two simultaneous threads on the same die. In this way, if one thread stalls, the other can execute in it's place while the other thread waits for the pipeline to redo itself, hence being a very clever way of making up for the design "faults". AMD's typically run at a lower clockspeed, and have a much shorter pipeline, so even when their piplines stall, the chip does not waste as many cycles - in short, they;re not really designed to take advantage of SMT. Hence AMD not having SMT support is a bit of a non-issue.

(Disclaimer: I'm not much of a buff on chip architecture, this is just stuff I've picked up from reading /.)

Re:HT

[Lux]

Yeah, HT works better on deep pipelines, but it's still a win for shallower ones. I couldn't find any more recent figures on AMD's pipeline depth, but duron was using a 20-stage pipeline.

That's not shallow.

I saw a graph based on HT simulations, it was CPU utilization versus degree of hyperthreading. Utilization didn't start to level off until you have something like 6-8 execution contexts on the chip (that's assuming you have work for all of them, of course.)

That's probably geared towards an Intel-style pipeline, but surely AMD would benefit from adopting the technology. Though since they're already going with dual-core, I'd rather see them apply it towards speculation to absorb cache misses than expose those contexts. That way single-threaded apps could get some love.

Re:HT

[MrNemesis]

I've just googled the info, apparently the Opteron has a 12 stage integer pipeline and a 15 stage FPU pipeline, half the length of the P4's. I'm assuming the 1P AMD64 chips are pretty similar.

And don't worry, I too think AMD will benefit from introducing SMT, it's just that the processing loss without it isn't as catastrophic as it is on the P4. I'd rather they go all guns on the dual core thing too before they tackle something as "soooooooo 2002" as HyperThreading ;)

Re:HT

[MrNemesis]

D'oh, preview! I mistyped. I should have said 17 stage FPU.

Just to stop this post being totally worthless, the P4 Northwood had a 20 stage pipeline, and Prescott has a 30 stage pipeline, although google doesn't give away if that's int or float, or both, and I'm too tired to figure out the rest of it :)

Re:HT

[Lux]

HT may be less sexy, but it's cheaper. I'd rather pay +20% for an 80% speedup than +40% for a 100% speedup.

There's something to be said for playing catchup when you're behind, walking before you run, et cetera. :)

Re:HT

[MrNemesis]

There's no way in hell SMT gives you an 80% speedup; on the P4's it gives 10-15% at best. Thankfully, it comes at little-to-no-cost in terms of silicon, IIRC. SMT is nowhere near actually having two physical processors, it just utilises the power of the P4 more efficiently by just filling in the gaps in the execution stages. Some apps actually show a performance decrease under SMT.

Dual cores won't give you 200% either, even with the Opteron arch it'll still be 190% at the highest.

Re:HT

[Ryvar]

So it sounds like AMD is going without HT.

Damn.

They may not have a choice (patents).

Re:HT

[Calroth]

"There's no way in hell SMT gives you an 80% speedup; on the P4's it gives 10-15% at best. Thankfully, it comes at little-to-no-cost in terms of silicon, IIRC."

Well, there you go. I'd rather pay 2% more for a 10-15% speedup, that stacks with other speedups.

Re:HT

[Sique]

AMD and Intel have a cross license agreement for their patents. So THAT's not the issue here.

Re:HT

[Anonymous Coward]

Due to the P4's incredibly long pipeline (30-odd stages?) and very high clockspeed,

High clockspeed has nothing to do with the penalties from stalling.

Unless it has to go outside the (on-die) L2 cache for data, a higher clocked processor also recovers faster from pipeline stalls.

It's *only* a matter of pipeline length.

(The ideal would be a short pipeline clocked extremely high, which is of course impossible. Pentium-M has really smart branch prediction, really big caches, and really beefy execution units to get the high performance it gets, in addition to its separate low power and heat advantages. Much like AMD's K8 cores. They would be even better if they could clock the same cores higher at the same power draw -- high clockspeed in itself doesn't bring any disadvantages.)

***

You say are not a buff, but look who's at "4, Insightful" anyway. Welcome to Slashdot :P

Okay, that was a valid and pleasantly understandable explanation of HT, you earned it.

***

Slightly off topic: It's fun to see the "Just a different route to the same goal" Pentium 4 apologists finally admit that the design wasn't engineer driven, but marketeer driven from the start. It's also funny how quickly Moore's "Law" got cancelled by real life (okay, quantum physics) -- a year ago I didn't see any talk about that! Supposing Itanium keeps not conquering the server/mainframe marketplace, Intel's near future depends on the remote Israeli team's Pentium 3 hack coupled to AMD's 64-bit extensions, and that's genuinely funny.

Re:HT

[Anonymous Coward]

and i'd rather pay 80% less for a 15-20% speed advantage in the first place.

Re:HT

[Carewolf]

I think the cross license only covers ISAs (instruction sets). This is the reason AMD had to invent Socket A when Intel changed to the patented Socket 1, and the reason AMD can not make an Itanic processor (Intel has patents on the implementation-details of the ISA).

Re:HT

[10Ghz]

It's such a high bang-for-your-buck optimization that I'd feel a lot more comfortable buying AMD chips if I knew they weren't hobbled by not exploiting it.

Well, Intel-CPU's don't have Hypertransport, are you uncomfortable when buying Intel-CPU's since they don't support such a kick-ass feature? They also lack integrated mem-controller, does that make you "uncomfortabe"?

Fact is that some CPU's have features that others may not have. P4 has HyperThreading. Athlon64 has HyperTransport and integrated mem-controller.

Re:HT

[10Ghz]

AMD has SMT (AKA HyperThreading) related patents as well, so they could do it.

Re:HT

[Amorpheus_MMS]

Intel's crappy super-long pipeline (something like 32 stages!? Someone correct me if I'm wrong)

The Pentium III had a ten-stage pipeline.
The Pentium IV doubled that to a whooping 20. But that wasn't long enough. The latest "Prescott" P4 has 31 stages.

If it were the other way around...

[Anonymous Coward]

...AMD would be hailed for coming out with a CPU that didn't have the baggage of an old design. How many times has this been the song praised when talking about PowerPC? Or let's take this into software, and this is the crowning glory when Linux is being touted over Windows? why is coming out with a clean design that isn't held back by the past always good, except when it's Intel we're talking about?

Fuck legacy, give me something that utilizes all it's resources for up to date design. I'll run an emulator if I need to use an app from 1976.

Re:If it were the other way around...

[Anonymous Coward]

The problem is you'll have problems running apps from 2004. And the cost of Itanium, Jesus Christ. If Itanium cost as little as PowerPC, I would have given it a shot. But at the time I bought my dual Opterons, Itanium cost 500 more for a single proc and the motherboard. Fuck that.

In case Intel hasn't noticed, this isn't a boom time anymore. Delivering extremely expensive technology is a real hard sell when alot of people are out of work and those who have it, have often had to settle for lower pay. And as for corporate IT departments, who the fuck wants to try to convince their CIO that it's a good idea to ditch the cheaper and stable X86 platform for the unknown Itanium platform, when there's no guarantees that the organisations apps will run at better, or even just the same speeds?

I'd hate to be the dumb schmuck who moved his organisation in that direction and had it blow up on him.

Re:HT -- MultiCore

[Nom du Keyboard]

Does AMD have any Hyper-Threading-like technologies in their chips or in the works?

AMD is going HT one better by putting two Athlon 64 cores on one die next year. Much better performance bump than HT provides in a single processor core.

64-bit CPUs

[phorm]

One of the major complaints I have about my XP2500+ is that the thing runs hot, like really hot. We hit a heatwave locally and temperatures were up to about 40 celcius at peak. My CPU actually hit 95 celcius (for those that use Fahrenheit, 100 celcius is boiling temperature).

I have a bigass thermaltake fan in there now, which I can turn down when the weather is cooler. The computer is still rather noisy.

My point to all this is not AMD bashing however. Apparently the 64-bit CPUs do much better for heat dissipation. The CPU die is much larger (the actual die is small on an 32-bit Athlon), so heat dissipates much more nicely into the heatsink due to the increased surface contact area. When I do upgrade, I'll be going AMD64... more power (in 'nix anyhow) and cooler running than my current CPU.

Re:64-bit CPUs

[nbowman]

I have the same processor, OCed to 2200 MHz (thats real MHz, not a + rating) and it seems to run about 20 degrees C over ambient temp at full on processor usage, and it will knock a good 3-5 degrees off of that if I use a loud fan, but that is quite annoying and I have never seen it over 55 or so ( any more than that and the ambient is high enough that I turn on the Air Conditioning :P ) Obviously taking much of any conclusion from such a small sample size would likely be erroneous, but I think mine is closer to average from what I have seen.

Re:64-bit CPUs

[aldoman]

I think you have a real problem with cooling there.

I have a XP2500 and it's constant at 40*C in the UK, where obviously it's not very hot but not as cold as Iceland ;). I'm using the worst and cheapest cooler I could find and it runs great.

Check that their is some thermal paste, because sometimes it dries up if you have a heatwave.

Re:64-bit CPUs

[MonkeyCookie]

Whoa, it sounds like something is not right there. CPUs don't normally reach 95 C unless there is something wrong with the heatsink or fan. Since the fan is noisy, it's probably working just fine. I'm guessing there's a problem transferring heat from the CPU to the heatsink.

I have an XP2600+ with a normal heatsink and fan. I live in Sacramento, California where it gets pretty hot in summer (including inside my apartment), and I've never seen the CPU temp exceed 40 C.

Re:64-bit CPUs

[qopax]

something tells me if your cpu hit 95 degrees celcius, your computer would not be able to detect the processor anymore, much less than function...

maybe you used a temp. gun to measure the small fire inside your pc?

Re:64-bit CPUs

[Fweeky]

AMD64's aren't just better at dissipating heat, they're better at not generating it in the first place; for most simple tasks, they'll stick to a low power mode that pumps out at most around 35W of heat (about the same as a 500MHz Celeron).

AMD Cool'n'Quiet.

Re:64-bit CPUs

[MrNemesis]

Definitely a problem with your cooling setup there! Either that or your motherboard temp sensors are out of whack. My 2800 never goes above 50*C (temperature in my room here in the UK has been about 35 thanks to all my rackmounts). When all my fans are maxxed out, it does the 15*C above room temp on the CPU (on low it's more like 20-25*C above).

There's a decent little chart here http://www.cybercpu.net/howto/other/amdpr.asp that shows the power output of the Athlon chips; the 2500 is quite low down at ~55W, and it's only rated to live up until 85*C.

Re:64-bit CPUs

[phorm]

Actually, I put the heatsink on with arctic silver. Processor isn't running overclocked either. I would suspect a defect except that for all other intents and purposes it runs perfectly fine.

It might not be uncommon either. We had a dual Athlon MP 2200 at work, one of two CPUs was constantly overheating. Replaced CPU, no problemo...

But for now I'll put up with the mega-fan, since the processor itself runs fine with proper cooling, and I'd hate to be without my main machine. What's the warantee on AMD, perhaps I'll send for replacement once I upgrade to a 64-bit rig.

Re:64-bit CPUs

[phorm]

heatsink came with the fan, and it was hotter with the previous. And I've never had an AMD stay at 40c... right now it's happy at 49, but even the ones at work are usually close to that at normal temperature.

Here it's dry, perhaps more humid there?

Re:64-bit CPUs

[phorm]

No, I checked the hardware diagnostic in the BIOS. After about 80/85 it used to start resetting... tripping the internal auto-shutoff, etc

Re:64-bit CPUs

[phorm]

No she's definately hot... it used to crash when it got that hot. Case is side-vented... but the back is against the wall so perhaps not too much venting there.

I'm thinking on replacing the PSU to see if that helps - perhaps it's underpowered - but it doesn't seem to be struggling... the overheating generally only happened when I was doing things like encoding video for DVD (maxing the CPU as much as it will go).

Re:64-bit CPUs

[qopax]

it only started tripping internal auto-shutoff at 80-85? As far as I know, intel and AMD's processors both officially overheat and die... at around 80 degress celcius, so by the time auto-shutoff seems a bit unnecessary... i know there's a chance the processor might live, but whats the point of having auto shutoff at 80 degrees celcius? maybe it wasn't auto shutoff, maybe it just stopped operating due to overheating thus reseting the system?

Re:64-bit CPUs

[phorm]

The actual alarm on the board was in that range. I've wondered about it myself, but being that I'm using that very machine - about 6 months post to overheating, and no more issues - I'd say it survived nicely

I think I've set the record amongst my friends for hottest a CPU has been though :-)

Re:64-bit CPUs

[MrNemesis]

Hehe, I do lots of video encoding too...! Those temperatures I quoted were taken at high load.

Granted, I've got a heatsink the size of Delaware (Alpha 8045) sitting on top of my chip along with a not-underpowered 80mm fan, but even so your temps are way outta whack. If the CPU really was at 95*C you'd burn your finger if you touched the heatsink (which would be at about 60*C). The Barton cores aren't hot chips (the Athlon T-Bird and Palomino cores *were* hot chips - the two 2000MP's in one office server run about 15*C higher than the 2800MP Bartons); something has gone wrong somewhere!

Re:64-bit CPUs

[phorm]

Oh trust me... touching the heatsink for more than a few secs and I definately agree I'd probably be burned... it is hot.

Re:64-bit CPUs

[qopax]

bah, why do you get modded up? argh, i might as well say I have the record of having the longest penis among my friends. Well, it could be false if I actually had any... or seen their penises for that matter.

im just gonna stop talking now.

Re:64-bit CPUs

[Anonymous Coward]

Whoa, it sounds like something is not right there. CPUs don't normally reach 95 C unless there is something wrong with the heatsink or fan. Since the fan is noisy, it's probably working just fine. I'm guessing there's a problem transferring heat from the CPU to the heatsink.

I have an XP2600+ with a normal heatsink and fan. I live in Sacramento, California where it gets pretty hot in summer (including inside my apartment), and I've never seen the CPU temp exceed 40 C.

I also live in Sacramento, with an XP2600+, and I have never seen my CPU temperature go above 144 F. I also have BOINC or Folding@HOME running 24/7. I can think of a few problems that grandparent is having... They might be having some issues with their heatsink/fan (cheap, not mounted properly, or something to that effect)... Something is wrong with the computer case (although this is rare, maybe the power supply is mounted directly above the CPU/Heatsink?)... or maybe they have a really fast video card that is helping heat up the CPU. The video card can particularly be a problem (if its a high-end model) if they aren't using any extra fans to vent air in/out of the system.

Re:64-bit CPUs

[mabinogi]

The technical specs for the maximum operating temperature of the Athlon say 90 degrees. I looked it up when I was concerned about a Thunderbird Athlon running at a constant 60 - 70 degrees.
Of course having a CPU running at that sort of temperature for any length of time starts to have a negative effect on the nearby components...

The XPs seem to be much cooler though...my 1700+ hardly ever reaches 40 even in the middle of an Australian summer.
Of course the 1700+ is at the low end of the XP range, so it's possible that the faster XPs are quite a bit hotter...

Re:64-bit CPUs

[qopax]

all the athlons have that max? and most likely as the temp goes above 60 degrees celcius the cpu life decreases considereably. In engineering(according to my dad...), anything below 40C is grade A. below 60 is grade B. and below 80 grade C. This seems to apply to processors quite well.

Re:64-bit CPUs

[shfted!]

With a regular heatsink/fan (for the time), my Athlon 1.2 GHz TBird runs at 56 to 60 C under full load with an ambient temperature of around 21 C.

Re:64-bit CPUs

[Anonymous Coward]

I have an XP 2600+. I run at 45 degrees C with the stock cooler and an ambient temp. of 79 degrees F.

Even with a "heatwave" as you say.. What, maybe another 20 F tops?.. It shouldn' go above 60 C.

Perhaps you're just an idiot? mmm yes. Let's have some tea.

Re:64-bit CPUs

[GISGEOLOGYGEEK]

damn!

go spend a couple weeks allowance and build yourself a water cooling system!

Buy the water block for the processor, make the rest of the setup from cheap parts ...

car transmission radiator, rubbermaid resevoir, plastic tubes and clamps ... bought at Canadian Tire (I'm sure you americans have some equivalent hardware store)

300gph water pump bought at the local pet store.

during a heatwave this summer here with ~33celcius temps, my computer was running at only 42celcius under full load 24/7 running Prime95. ... in a small cramped room with poor ventilation.

and it is QUIET with just the rear case fans (not the power fan) blowing through the radiator.

Re:64-bit CPUs

[davidsyes]

Whoaa, Keep the cooler going and the heatsink attached. Otherwise tis could be the seas'n to be weas'n.

Re:64-bit CPUs

[Anonymous Coward]

Are you sure you aren't measuring the motherboard temperature? It may be that your BIOS is reporting motherboard temperature as CPU temperature.

Typical operating temperatures for modern Athlon's are around 40-50 C idle and 50-60 C under load. You're correct that 95 C indicates a problem.

But I've never seen an Athlon (and I own 4) running at a CPU temperature less than 40 C (except when underclocked - this may also be the problem, check your BIOS to make sure your clock multiplier is correct).

Re:64-bit CPUs

[Anonymous Coward]

The builtin thermal diodes can have really big variances. So if yours shows +25C and someone else's shows -15C...

Re:64-bit CPUs

[incabulos]

My XP2000+ clocked at 1.667 GHz is sitting at 27 degrees Celcius right now, after being running for about 50 minutes. I cant believe your CPU is running so hot, even if it isnt kernel-panic unstable now, it will be after a few months of running at that temperature, the whole silicon substrate basically degrades and its operational stability will get worse and worse.

If you are running Linux try using a power saving utility like Athcool, or LVCool if you are running Windows. The theory and downloads are here, it activates the native power saving mode of the Athlon/Chipset pair. It works great with my via KT266A board, the temp without it idles at about 39C, over a 10 degree difference. When busy compiling or encoding things heat up quickly, but no more so than it would with the power-saving mode completely disabled. Check your heatsink contact too, it does not sound like its making a good deal of contact with the die.

Re:64-bit CPUs

[Anonymous Coward]

my xp-m 2600+, clocked at 2.4 ghz, is sitting at 33 C right now. room temperature is 26 C.

the problem is your heatsink. thermaltakes can't cool worth shit. get yourself a thermalright.

Re:64-bit CPUs

[Penguinoflight]

Above 40 is easy, even with a good fan considering surrounding temperatures, but 95c is quite rediculous. Most MB's will shut off the whole system if the temperature exceeds 60C, and I don't think there's any point trying to run a system above 70C. I'd be more concerned about that guy's MB than his fan, although the old fan was probably either broken, or not designed for a xp2500.

Re:64-bit CPUs

[evilviper]

We hit a heatwave locally and temperatures were up to about 40 celcius at peak.

Those of us living here in the Desert, with 100F+ tempuratures more than half the year would collectively like to say... Screw you.

Hell, it gets up to 130F (in the shade) for several weeks every year, right in my back yard... I only have one small swamp cooler that I barely even use... And I have 3 computers running constantly. One of those happens to be an AMD 2000+, who's tempurature never rises about 130F degrees, despite being nearly silent. Oh, and I haven't yet mentioned it, but I have it overclocked to about 1950MHz, which is close to being a 3100+ in AMD notation.

My CPU actually hit 95 celcius

You either have experienced several independant failures, your software is mis-reading your tempurature sensor, or you are making this story up.

Not only would you have to be having a MASSIVE cooling problem, but your motherboard would also have to be seriously screwed up, because they all SHUTDOWN when your CPU gets to about 150-170F degrees (shutdown temp depends on mobo) which is well below the 200+ degrees you say it got up to.

I have a bigass thermaltake fan in there now, which I can turn down when the weather is cooler. The computer is still rather noisy.

Then you've done something wrong. First you should be using the termal sensor. Second, changing your CPU fan/heatsink can't possibly quiet down your power supply fan. Unless you've got an Enermax or Antec power supply, you'll get the biggest noise reduction by changing your PS fan (requires stripping wires).


Now... Assuming you aren't just making your story up, here are some tips for anyone looking at serious cooling, and near silence...

EVILVIPER'S TIPS FOR MASSIVE COOLING:

Remove the thermal tape, and replace it with thermal paste for at least a 30F degree drop in temp. $1

Buy a good heatsink, with copper base insert, and mounting for an 80mm fan. $10

Buy a handful of Enermax 80mm thermo fans. Replace all the fans in your system with them. $5/piece, 10 for $35 (newegg)

That's all you need, but with just a little bit of brain-power on your part, you can do much more. For instance, in my 2000+ system, the rear case exhaust fan is directly over my CPU fan, so I turned it around to blow air in, making my CPU another 20F degrees cooler.

Follow these rules, and your system will run cool, even if you poor molten lead over it...

Re:64-bit CPUs

[ManitobaMoose]

it really looks like that. i have an AMD64 and even after hours of gaming it never peaked over 45C with an ambient temp. of around 25-26C, while the motherboard temp. stayed fairly constant around 32C. for "normal" desktop operations the CPU temp. rarely exeeded 35C.

Re:64-bit CPUs

[10Ghz]

One of the major complaints I have about my XP2500+ is that the thing runs hot, like really hot.

News-flash: Comparable Intel-CPUs run just as hot or even hotter.

Re:64-bit CPUs

[WuphonsReach]

Even when it gets to be 32-35C here in my home office, my AthlonXP 2600+ never goes over 57C and internal case temperatures are in the 40-45C range. You've got something wrong if your CPU temp is over 60C under that sort of heat wave.

For the record, I have a large copper heatsink installed, with a teeny tiny dab of Arctic Silver (overkill), and a 120mm fan running at 2000rpm attached via an adapter collar.

I think your heatsink is installed incorrectly (or too much thermal paste).

Re:64-bit CPUs

[phorm]

450W Antec PSU, artic silver paste, rear cooling on the case, and a bigass thermaltake fan.

I think it might just be that the case in its desk cubby is too confined - that or the CPU is slightly defective. The only of the above that keep this puppy cool seems to be the thermaltake fan, and when I crank that down to 3500RPM the temperature goes up again.

Re:64-bit CPUs

[evilviper]

I think it might just be that the case in its desk cubby is too confined

Ah, you hadn't mentioned you were doing something so insane as keeping your system in a confined space.

No matter how good the system is, there's no way it can dissipate as much heat as modern systems need to, if it's enclosed, and the intake fans are sucking in the hot air the system just exhausted. I can safely say you'd be worse of with a P4, since they put-out more heat than AMD's chips. However, even moving to AMD64, you aren't going to see much improvement. You really must deal with your extreme heat problem.

You should also know that having a massive fan on the CPU only cures the symptom, not the problem. With such high internal case tempuratures as you must have, ALL the equipment in your system is overheating, and going to have a seriously decreased life-span because of it.

But besides all this, you have a serious problem if your motherboard isn't shutting down when the CPU reaches such high temps.

Re:64-bit CPUs

[phorm]

The shutdown option is.... well... an option. One that is now enabled. The desk isn't that confining though. Rear and front of the cubby are open to the room, with 1' clearance behind the desk (then the wall) and 3/4' to the side of the machine where the CPU fan would blow out.

I'm thinking of sticking a nice fan on the front though - as it's probably the lack of airflow in the rear that would be trapping heat. So a blow-in fan on the rear, blow-out on the front could help a lot...

Re:64-bit CPUs

[evilviper]

The shutdown option is.... well... an option. One that is now enabled.

All the systems I've ever seen that have any option in the BIOS, only allow you to set the tempurature (between about 150-170). I have never seen one that allows you to defeat the function entirely.

3/4' to the side of the machine where the CPU fan would blow out.

I don't understand what you mean. Your CPU fan doesn't blow air out of the side, or anywhere else... It just takes air inside the case, blows it over the heatsink, and it remains inside the case. It's the power supply fan that blows it out, and that doesn't go out the side. What are we talking about here?

So a blow-in fan on the rear, blow-out on the front could help a lot...

Not likely. If you've got poor airflow in the back, sucking in air isn't going to help much. Plus, because it's a confined space, you'll probably be sucking in heavily heated air that was just blown out from the power supply.

Why I love AMD

[ShatteredDream]

Their processors aren't as fast as Intel's but for the price, they're so much better. If a $500 AMD processor is almost as fast as a $800 Intel processor, that $300 buys an iPod. Most of the people I know share that view. So what if a 3.8Ghz Xeon performs better than a Athlon64 3800, the Athlon is $300 cheaper!

Re:Why I love AMD

[afidel]

Except that the person buying the 3.8Ghz Xeon is buying it for a $5000+ server with a couple gigs of ram and a RAID5 subsystem. They don't care about saving 12% of the system price on a pair of CPU's, if it runs the risk of causing them downtime. Hell I pay a lot more than 12% of the purchase price to upgrade the basic waranty to 24x7 4 hour support to avoid downtime! Now if Dell or HP offered a fully backed AMD system for significantly less than an equivilantly configured Intel one then I would care (HP only offers 2-way systems with 2 drive bays of 4-way with 4 drive bays, neither is apropriate for my clients).

Re:Why I love AMD

[jcr]

that $300 buys an iPod

Comment like that warm me from the bottom of my stock options. ;-)

-jcr

Re:Why I love AMD

[crabpeople]

what are you talking about? I havent seen a intel system beat a comperable AMD system in the last year. Not to mention that the intel on average runs 1ghz faster.

where are all those intel favourable benchmarks?*

lots of amd favourable ones

in my personal experience, Intel's always have a small lag that is quite noticeable. Although this is comming from the same person who can tell a 85hz refresh rate from a 75 so its probably not something most people have to worry about.

and THEN there is the huge price difference :)

*(i wouldn't personally count office benchmarks like word but i know intel has a weird history of doing well in those)

Re:Why I love AMD

[Anonymous Coward]

Except the Athlon64 3800 is $630, and Intel doesn't even have a 3.8GHz Xeon (or even a 3.8GHz P4 EE). The Athlon64 3800 is faster than the fastest P4 EE (3.2GHz at a cost of about $1k) at everything except Quake 3 (and just Quake 3).

Re:Why I love AMD

[Anonymous Coward]

Now if Dell or HP offered a fully backed AMD system for significantly less than an equivilantly configured Intel

Sun, HP and IBM sell Opteron wprstations and servers. A lot of them. Full support. Cost less and work better than Dell, especialy in multiproc configs.

Re:Why I love AMD

[Anonymous Coward]

Yeah but a 3.4 Ghz Xeon performs the same or better than a A64-3800. And guess what? It costs the same too. And the Intels have much better motherboard support (PCI-Express, high bandwidth SATA and GBe, DDR2, etc.).

Re:Why I love AMD

[afidel]

None of them offer a 2 way server with RAID5 which is what we almost always use for our clients. The Dell 2850 with dual Xeon EM64T and RAID-5 is much more capable for general server use than any of the 1-2U systems from those vendors. For some reason the Opteron offerings have either been rack optomized systems or four way monstrosities, they don't hit the midrange where all of our SMB client's needs lie.

Re:Why I love AMD

[10Ghz]

Their processors aren't as fast as Intel's

They are faster than Intel's.

A whole week?

[graveyardduckx]

They outsold Intel by 10% for a _whole week_??? How is this news? I'm sure AMD outsells Intel many days out of the year. Someone wake me when they outsell them per quarter.

Works great

[Anonymous Coward]

At work we have both AMD and Intel hardware, but I tell you, we will buy only AMD from now on. It works as good for us (servers & workstations), and they're cheaper.

Re:Who would buy intel?

[jcr]

Anyone who cares about using their computer for professional audio applications, for one.

Funny, we seem to have rather a lot of pro audio users on Macs, last I checked.

-jcr

I don't

[DogDude]

I tried setting up our web server several times with AMD chips, and for some reason, we could never get any kind of stability with AMD and W2K. We had so many kinds of flaky problems with 3 completely different AMD machines, that we had to go straight Intel. Don't know why, but just to be safe, I'm not going to try AMD for a long time. I don't have time or money to experiment like that. I don't know what the problem was, only that AMD didn't work and Intel did.

Re:I don't

[Anonymous Coward]

Well, it could have been anything; the video cards, the motherboards, the RAM, the power supplies, inadequate cooling or ventilation, who knows. Who put together the computers?

I am very doubtful that the actual CPU was the cause of any instabilities.

Re:I don't

[Anonymous Coward]

Same behavior i've seen. Usually afew bios updates will fix it.

I've yet to have any similar instability on Intel. They may cost more but atleast on a server uptime is better than saving afew bucks.

(also another hint, look to the SCSI card, after removing the SCSI on the AMD servers things usually got better. Hard to call it a server after that though.....)

Re:I don't

[Aggrajag]

For 32-bit servers I would buy Intel, for 64-bit my choice is Opteron.

Re:I don't

[DogDude]

Who put together the computers?

It was actually a very large co-location facility. They're not used to W2K (the vast majority of their boxes are FreeBSD), so they had not seen that before. They literally swapped out 3 different whole boxes before finally trying an Intel box. I lost about a week up uptime due to all of the switching and the problems. Oh yeah... identical hard drives.

Re:I don't

[Sven The Space Monke]

Oh god, I can't believe I'm about to make a post like this, but here goes...

I think you get what you deserve for having a windows machine exposed to the net. I know that has nothing to do with CPU stability issues beyond a karma thing, but still...

Anyway, now that my kneejerk linux zealot reflex is satiated, it could be the mobo chipset you were using - a Via chipset can be problematic. At my workplace, we HAVE to experiment, so I've seen a lot of configurations come through. I've found that an nForce chipset is worth it's weight in solar dust. Seriously, I'd rather have 1 Asus mobo with an nForce chipset than 10 gigabytes with a VIA.

Our primary internal (IE: no internet connection) fileserver used to have stability issues running win2k on a VIA chipset. About a month ago, I swapped it out with a spare ASUS I managed to grab. Presto - no more strange crashes.

The same thing can be said for winXP machines, too. nForce chipsets are a MUST if you want things to run smoothly.

*I'm a sysadmin for a 3 30-desktop gaming center chain - you think you have it bad with peple running MS Office and Seibel? Try it when you have a bunch of gamers running the latest bug-riddled POS to come out of EA (*coughBFVIETNAMcough*). And don't get me started on the heat issues.

Re:I don't

[karnal]

Similar things that almost turned me off of AMD... In their K7 days, the Via MVP3 chipsets needed the chipset drivers installed to get an ethernet card working in Win98se/2k (something with the PCI bus)... Of course, just something little, but kinda pissed me off that the ethernet card wouldn't work right away...

Next, I was heavy into TomsHardware, and bought a kt133 RAID (abit?) mobo. I used to have really weird issues with that board... Finally figured out that it's REALLY sensitive to the quality of memory (had to go all out and get the best stuff I could find) and it didn't like sharing IRQ's across the PCI bus.... so I couldn't use a SCSI card, TV tuner, ethernet and sound card at the same time (along with switching on the RAID portion of the mobo, which took slot 5 out of the picture completely...)

Now, given, these weren't AMD faults, but since I was having problems specifically with that platform, I really pined for an Intel solution.

Nowadays, myself and my fiancee are running on XP's with mine on an Nforce2 board (asus) and hers on a gigabyte (kt400?) board.... and neither of them has given us ANY problems.....

AMD looks to be finally back to rocking. Can't wait to get rid of my k6-2 file server, though..... darn spurious interrupts....

Re:I don't

[Anonymous Coward]

For 32-bit servers I would buy Intel, for 64-bit my choice is Opteron.

Yet why not choose Opteron every time? The price and performance are competive with the 32-bit Xeon range, so why not go for it?

Re:Who would buy intel?

[RichM]

In my experience, AMD Athlon CPUs tend to perform better at video editing which is why I chose an FX-53 over a P4EE - it has much better memory bandwidth due to the memory controller being on the CPU instead of the northbridge and the FPU performance is usually superior.
Of course, there's always the 64-bit thing to take into consideration (and the no-execute extensions in Windows XP SP2).
Intel themselves admit that the only way they could make 64-bit desktop chips was by copying the AMD 64-bit extensions.

Prices

[Anonymous Coward]

Could you imagine a market WITHOUT AMD?
Scary thought.......

Re:Who would buy intel? Who would use onboard...

[Nom du Keyboard]

AMD motherboards with cheap, ultra-crappy VIA south/northbridges and the like -- just won't cut it. You will get crackles in the audio,

Professional Audio applications aren't running with onboard AC97. You'll have added a high-end card, or two, to your system. The only way the N/S bridge chips could add crackles would be if they weren't exchanging data properly, in which case nothing would be running correctly on your computer.

Whether you like Intel or AMD or neither

[Anonymous Coward]

This is a 100% bonafide GOOD THING. Have you seen what these guys have done to each others margins? Have you seen how fast processor speeds have become these last 4 years? This is competition at its absolute finest.

Cheers to AMD for not giving up and dying. And cheers to that chairman of theirs who looks like he oughta be out selling chicken.

Re:Whether you like Intel or AMD or neither

[Entropius]

Indeed. Especial applause to AMD for always finding new things to do. They realized that there was a market for 64-bit, so they made the Athlon 64. Then they realized that their target market for the A64's was enthusiasts who live in the same room as their computers, so they introduced Cool and Quiet.

And, of course, they noticed the rising market share of laptops and realized that the same technology they use to make A64 machines not sound like leafblowers can also provide decent battery life on a laptop.

The Athlon 64 laptops don't have battery life like a Centrino, but they're much better than Intel's P4-based laptop line, and they blow the Pentium-M's out of the water in performance for hundreds less.

The Athlon 64 may have started out as a niche product, but now it's the preferred performance-processor for many enthusiasts and a decent processor for both performance and low-price laptops (you can get an Athlon 64 laptop for $1150).

Then they realize that Intel has been neglecting the low-end foreign markets: *poof*, Sempron.

The Athlon XP-M chips are still wonderful in laptops--they're Fast Enough for almost anyone, don't drain that much power, and are cheap.

AMD gets credit for doing marketing the old-fashioned way: find an area that Intel's not up to par in, and design something that beats Intel's current offering in that area. This is the sort of marketing I benefit from, the sort of marketing that gives me cheap, fast hardware. I like that.

Re:Whether you like Intel or AMD or neither

[Anonymous Coward]

Cheers to AMD for not giving up and dying. And cheers to that chairman of theirs who looks like he oughta be out selling chicken.

Jerry Sanders....Colonel Harland Sanders....OMG it all fits now.

Re:Who would buy intel?

[neildiamond]

Ummm...

Last time I checked I used AMD for pro audio applications. In fact, I've won more awards than many other audio producers I know (not that it has anything to do with AMD... or doesn't or whatever). My 700 mhz PC worked fine. I still use a 1.2 GHZ AMD PC as well too. I've done some corporate video work on both PCs as well.

Unless Intel gets better and cheaper, I'll be sticking with AMD.

Re:HT -- MultiCore

[chez69]

the Power 5 is even better. HT done right with muli core design.

Makes Sense To Me

[Anonymous Coward]

Honestly, I just built a desktop and didn't even consider Intel processors. It seemed that all the boards I wanted were AMD.

Coincidence?

Re:Makes Sense To Me

[celeritas_2]

I'm very impressed with AMD and how it has evolved from a basic clone to an innovative competetor, and in my opinion, the better of Intel. My next system will be AMD, but really what are the benefits of Intel? Before it was the cool 'brand name' but now it just seems to be the bloated expensive version.

Re:Makes Sense To Me

[MonkeyCookie]

I'm also impressed with how far AMD has come. Part of me wants to see AMD clobber Intel, which had a monopoly for a long time, but it's probably best for each to have about 50% market share. It will keep both companies from getting fat and lazy, meaning more research and lower prices. Competition without a particular company dominating the market is generally a good thing for the consumer.

Re:Makes Sense To Me

[Cthefuture]

It seemed that all the boards I wanted were AMD.

Uh, what?

Lets see... We have Intel with PCI-Express, DDR2, high bandwidth low-CPU usage SATA, full speed GBe, and all in a MicroATX form... At AMD I get, uh, crap like PCI and AGP from the 90's era.

Good CPU's for sure, but the motherboards are not a reason I would pick AMD.

I've been holding off on buying a new system for several years because of the situation. I want a 64-bit system that can run generically compiled applications well (eg. Linux) and AMD has the right CPU for that. However, all the available motherboards for AMD suck.

Re:HT -- MultiCore

[Anonymous Coward]

And so is Intel, only they'll *also* have HT support for each core.

In all honesty, though, HT was developed to get around the massive wastes of the extra-long pipeline on the Pentium 4; so AMD doesn't really *need* any HT tech.

Re:Who would buy intel?

[Anonymous Coward]

I wish he was trolling, but I find his assesment to be quite accurate. I'll never buy another VIA-based mobo due to all the problems I've had with audio. It's a constant battle between various 4-in-1 drivers and sound card drivers, amongst other things. Troll my ass, he should be modded informative. I'm certainly not going to swich to a Mac, but my next computer will have an Nforce sticker inside of it someplace.

Warning!

[Anonymous Coward]

nVidia fanboy making irrelevant reply to the parent post.

Proceed with caution.

Re:Warning!

[Anonymous Coward]

Nothing to do with being an nvidia fan boy, dude. A lot of people hate(d) nvidia for their blatant cheating on 3dmark and hailed ATI as a champion of goodness, forgetting/overlooking the fact that they have cheated just as much in the past.

As mentioned by the parent, it's part of the reason for ATI's success with the 9800's and so is of complete relevence to the topic.

Sad that people can't speak truth without being labeled a fan boy.

No wonder

[melted]

It's embarrassing when a company makes a mobile chip, that at 1.6GHz outperforms their "desktop" 2.8GHz chip, especially if 1.6GHz chip uses their previous generation core with bits and pieces of P4 core thrown in for compatibility. I just hope they pull their heads out of their butts and release a replacement for "fast but dumb" P4. And for the love of god, someone tell them to make this replacement 64 bit. There's no point in denying that 4GB of address space is no longer "enough for everybody".

Re:No wonder

[Anonymous Coward]

> There's no point in denying that 4GB of address space is no longer "enough for everybody".

This is the grammar police. Drop your negatives and don't come out without your hands not down.

Bad comparison...

[Chordonblue]

I wouldn't say that Intel has everything wrapped up. comparing a Xeon to a 3800+ is hardly fair as you are comparing a server processor to a desktop one. Now if you compared it to say, an Opteron (a much more fair comparison), well then you'd see AMD still wins or pulls up even.

What's more, the more processors, the better. Hypertransport gives each processor it's own bus.

That said, comparing an FX-53 to a 3.8 GHz Intel would also be a more fair comparison. And while it's true that the Intel wins it's share of benchmarks, keep in mind: You are comparing a 3.8 GHz Intel chip to a lousy 2.6 GHz processor (the FX-53). Theoretically, the Intel should totally kick it's ass - but it doesn't. That's some good chip design there my friend!

I just got a 3800+ last week. All I can say is: WOW!

Re:Bad comparison...

[Anonymous Coward]

Intel doesn't have any 3.8GHz parts to compare in the first place.

Re:Bad comparison...

[MrNemesis]

The FX-53 is actually clocked at 2.4GHz

But like you say... wow! I'm of the opinion that AMD have totally pulled the rug out from Intel with the AMD64 line (HyperTransport, embedded memory controller), and Intel has yet to retaliate.

Re:Bad comparison...

[silverfuck]

That's some good chip design there my friend!

Or more accurately, very very bad chip design on the part of Intel: They designed the Pentium 4 with clockspeed, and clockspeed alone, in mind. Nevermind how efficient it is, let's make the pipeline as long as possible, because clockspeed sells! After all, to somebody who doesn't know about such things, a 3GHz processor is faster than a 2GHz processor every day, right?

Intel is now realising the folly of this (thanks in no small part to AMD), and Tejas (which was to be the successor to the Prescott and become the Pentium 5) was cancelled back in May. They are concentrating on the Pentium-M (Dothan) architecture and are adding 64-bit extensions and multiple cores (sound like any other chip company you know?), which can be traced back through the Tualatin to the good old Coppermine Pentium 3 and further back still to the debut of the P6 architecture in the Pentium Pro!

Re:Bad comparison...

[Jahf]

adding 64-bit extensions and multiple cores (sound like any other chip company you know?)

Sun?

Re:Bad comparison...

[smithmc]

Now if you compared it to say, an Opteron (a much more fair comparison), well then you'd see AMD still wins or pulls up even.

Well, here you go... AMD's fastest Opteron runs neck and neck against Intel's fastest Xeon, and guess which one is cheaper? (Only slightly though...)

Re:Bad comparison...

[Anonymous Coward]

umm no.. Sun's been 64-bit for years.

you are talking utter shit

[RMH101]

presumably to appear knowledgeable. Having built upwards of 150 AMD XP systems, tens of which were for professional audio use, you're talking out of your arse. Any nforce2 board's audio system is usable: whether soundstorm compliant or not, they're *all* good. 2% cpu usagle onboard sound for free: at very high quality indeed. what *exactly* are you insinuating? care to detail your *exact* problems? onboard sound is no longer a poor substitute for a second card: it's now great - and if the chipset bundles it then you *know* that it's been tested to within an inch of it's life and will be stable.
put up, or shut up.

Re:you are talking utter shit

[suckmysav]

"onboard sound is no longer a poor substitute for a second card: it's now great"

Unfortunately, nVidia is dropping SoundStorm on the next nForce chipset.

It's a shame really, coz it is great.

Re:you are talking utter shit

[Jedi Alec]

heh, he's likely referring to the sound problems back in the days when using an SB Live on a VIA KT133. In fact, the pc I'm typing this on right now has that configuration and yes, sound quality isn't exactly spectacular. Then again, we're talking cheapass junk hardware here, so hardly cause for a fuss...

Re:you are talking utter shit

[RMH101]

this could be because the OEMs don't want to pay soundstorm royalties for the use of the name: the hardware *might* be identical - this is what happened with nforce2, and is why some are badged "soundstorm" and some realtek: to get the former name applied they had to supply optical connectors (i think) and pay for the branding - same internal soundcard hardware though.

Re:Intel secret plan

[VoidWraith]

With that many digits after the dot, what number system are you using? That could be either 10 GHz or 10 Mhz, depending on what kind of audience is viewing...

Huh?

[zaxios]

I wonder if they believe that they can eventually drive notebook sales upward to the point that they outsell Intel more often than a handful of times a year?

The article says that AMD's desktop successes are partially a result of Intel's tendency to emphasize notebooks. If "they" (Intel, I hope you mean) drive notebook sales upward, and assuming that damages desktop sales, Intel's sales would increase because of their notebook dominance and AMD's would decrease because of their desktop interests. Overall the desktop market would shrink (or grow less), while AMD's share of it might grow marginally as a result of the notebook market distracting Intel from pushing its desktop CPUs as aggressively. We might then have more "AMD Desktops Outsell Intel" stories, but it would definitely not be good news for AMD.

Re:Who would buy intel?

[FatherOfONe]

So you make up around less than 1% of the desktop market. The example you gave is around 99.999% of the desktop world.

Yet Intel still holds a lions share of the market. Weird.

Also if you do video encoding Intel beats AMD.

MOD PARENT DOWN

[Anonymous Coward]

he obviously didn't even look at the article

Re:Who would buy intel?

[suckmysav]

" I'll never buy another VIA-based mobo due to all the problems I've had with audio."

Soooooo, , don't buy another Via based mobo then. It's quite simple really.

What has this to do with AMD CPU's again?

only if you need to...

[fontkick]

I hate to support what this person is saying but I've had a hard time with my Athlon XP (2000+) recording audio. I am using a low end setup (Santa Cruz card), and I've had several recordings that were no good due to crackles and static which were fine when done with a low end Mac (running at only 466mHz) and a similar recording setup.

I love it for everything else though. Unless you must buy Intel, I have no idea why you wouldn't get an AMD and pay a lot less for the same performance.

Re:only if you need to...

[MysteriousMystery]

It depends on the chipset, the Nvidia, and AMD proper based chipsets are usually a lot less buggy, but it seems like 75% of people using AMD based systems are running the cheaper VIA chipsets which have generally have problems.

what I'd like to see from Intel

[Tumbleweed]

I'd like them to make a dual core Pentium M, add the latest SSE stuff (SSE3?), an on-die, dual channel memory controller, a HyperTransport bus, and sell it for the desktop crowd.

Is that too much to ask? *sigh*

Re:what I'd like to see from Intel

[AngryParsley]

Yeah, AMD makes something like that. I think it's called the Athlon 64.

Re:what I'd like to see from Intel

[Tumbleweed]

Have you checked the power consumption of an Athlon 64 lately?

Re:what I'd like to see from Intel

[Fallen Andy]

No (but they'll wait for a competing product from Via before they launch it ).

If Intel isn't *really* really careful then AMD = Zilog. (Which would make Itanic = iAPX432 if my
history circuits still function huh?)

Oh wait. It was supposed to be an object orientated
processor - maybe when they fab at 13nm it will reappear...

Re:what I'd like to see from Intel

[Vulture101]

" I'd like them to make a dual core Pentium M, add the latest SSE stuff (SSE3?), an on-die, dual channel memory controller, a HyperTransport bus, and sell it for the desktop crowd.

Is that too much to ask? *sigh* "

wouldnt that be a DC-A64 ?

Re:what I'd like to see from Intel

[Tumbleweed]

Think power requirements, then you'll see why I'm not thinking Athlon 64. The Pentium M is really quite amazing along those lines. The only thing that competes with the Pentium M at the clockrates it runs at are the PowerPC970 chips, and they only beat out Intel & AMD in vector operations.

Re:what I'd like to see from Intel

[True Grit]

What about power consumption?

Get an MSI K8T Neo motherboard with an Athlon64. It can automatically vary its CPU speed from 800Mhz to its full rated speed (2Ghz+). So if you have its throttling control turned on, you don't worry about the Athlon64's maximum power consumption because it rarely runs at max speed. Best of all, although you turn this on or off in the bios, its controlled on the motherboard, so it works in either Windows or Linux (no software drivers).

Re:what I'd like to see from Intel

[Vulture101]

it would be much more easy for amd to implement some kind of strict power control ( speedstep + hardware hacks )in its cpus than intel add on dye memory controler, support for DC and hypertransport, me thinks

Re:what I'd like to see from Intel

[Tumbleweed]

Almost certainly, and about as likely, it seems. :(

I'd also really like a nice, fairly inexpensive PPC970FX workstation (or better yet, just a motherboard only) from IBM that I can pop Linux on. There's a developer's board available, but 'fairly inexpensive' doesn't quite apply. *shrug*

Re:what I'd like to see from Intel

[evilviper]

True, but unfortunately AMD still hasn't learned from their S2K problems on 32-bit chips, and the feature you describe (called "Cool n Quiet") has spotty availability. I've heard that something like 50% of AMD64 motherboards don't work with Cool n Quiet, so you have to really watch-out when buying a motherboard.

This is where Intel is still doing great. Buy any Intel CPU/motherboard, and you are guaranteed to get full and proper power management.

Re:what I'd like to see from Intel

[True Grit]

No.

There is a reason why I specifically mentioned that motherboard. That motherboard doesn't just support Cool-n-Quiet, it goes well beyond it with MSI's own motherboard technology called "Core Cell". Cool-n-Quiet is just the lower layer.

I've heard that something like 50% of AMD64 motherboards don't work with Cool n Quiet

Well since its a new kind of technology (or mobile technology showing up on the desktop for the first time), I'm not surprised some motherboard maker's didn't support it at first. That was NOT AMD's fault however. MSI's motherboards do show you how far you can take the Cool-n-Quiet idea, if the motherboard is designed to support and enhance the technology.

Buy any Intel CPU/motherboard, and you are guaranteed to get full and proper power management.

Maybe on their laptops and such, but not on desktop machines. AMD was the first to put this technology on their mainline CPUs (Athlon64), and not just a special mobile edition of the chip. MSI's motherboards are for desktops, not laptops.

I've had "full and proper power management" on my desktop machine for nearly a year now, the CPU automatically controls its own speed based on its load, and not just throttling down, but when the load is high it will automatically overclock itself too, based on bios settings (to what degree overclocking works, of course, is dependent on the individual CPU - so you have to do a little experimenting to find out how far you can safely overclock your CPU).

The whole thing is automatic, controlled entirely on the silicon, from throttling down to 800Mhz to overclocking the CPU by as much as 15%, and I'm pretty sure no Intel chip or chip/motherboard combo (for desktops) did this at the time I got the Athlon64 and this mobo (MSI claimed at the time no one else was doing this for desktop machines).

Re:what I'd like to see from Intel

[evilviper]

Will the stupidity never stop?

I'm not surprised some motherboard maker's didn't support it at first.

No. I'm not talking about "at first". I'm talking about TODAY. Even now, AMD64 mobos are being made, that, for one reason or another, do not support CnQ.

That was NOT AMD's fault however.

Yes it is. They don't require a motherboard to have CnQ to get certified, so mobo makers are lax about implimenting it.

As I said, AMD just doesn't whip their mobo/chipset makers into shape.

Maybe on their laptops and such, but not on desktop machines.

No, no, no, no, no. I'm not talking just about CPU frequency scaling... I'm talking about power management in general.

If you get a P4, you know the CPU will use less power and output less heat while it's idle. If you get an AMD XP, you've got a crap-shoot. Only very few boards support the S2K feature, and without that, AMD chips are running at full speed, and full power, even when they have nothing to do. That's why AMD was known for making hot chips. It's not that their max heat output was that bad, it's that they didn't idle properly, so they were outputing lots of heat, even when they didn't need to.

Even today, the AMD Athlon/XP situation is quite bad. Very few motherboards support S2K, and even if it is implimented, that board/chipset may still be using up nearly as much power, because implimentations of S2K vary.

This is, no question, AMD's fault. First because they should have implimented normal idle functionality like every other CPU in the world. Secondly, it's their fault because they didn't require chipsets to handle their messed-up method properly.

You don't need to give me a lecture on S2K. I know, in-detail, how it works.

Re:what I'd like to see from Intel

[True Grit]

Will the stupidity never stop?

Hopefully. Fortunately, stupidity is a correctable condition, whereas arrogance is usually a permanent flaw.

AMD just doesn't whip their mobo/chipset makers into shape.

We'll just have to agree to disagree on whether AMD should dictate what mobo makers do with their CPUs.

I'm not talking just about CPU frequency scaling... I'm talking about power management in general.

You said the AMD chips run hot, and I said that with an Athlon64 on a motherboard that supports CnQ, this isn't true. Even if you try to change the definition of what we're arguing about, what I said *still* isn't true.

If you get a P4, you know the CPU will use less power and output less heat while it's idle.

I also know this with an Athlon64 on a motherboard that supports CnQ. Which is all I said.

If you get an AMD XP

And I was talking about Athlon64s on mobos supporting CnQ.

Re:what I'd like to see from Intel

[True Grit]

what I said *still* isn't true

err, that should be "what I said is *still* true".

Re:what I'd like to see from Intel

[evilviper]

You said the AMD chips run hot

No, I didn't say that. You just pulled that out of your ass.

with an Athlon64 on a motherboard that supports CnQ, this isn't true.

And I pointed out that only about half of the motherboards support the function. Then you went off on completely unrelated crap.

I also know this with an Athlon64 on a motherboard that supports CnQ

No you don't, because you have no way to know that the motherboard supports CnQ. Even if it was a feature advertised by motherboard makers, it still might not work in certain configurations (ie. More than one stick of RAM. etc.).

And I was talking about Athlon64s on mobos supporting CnQ.

And I was talking about AMD screwing up in general. What the hell is your point? AMD is repeating their same mistakes with AMD64 that they had with their K-7s.

Intel says WWW will RIP

[otisg]

Then it's no wonder Intel Predicts Death of WWW, as seen earlier today on Slashdot.

Not of ALL Desktops!

[Socket Scientist]

Ashamed to admit it, but I did RTFA and have a look at their graph. Nowhere did I see a mention that the research was limited to x86 desktops, but clearly it was. I'm absolutely certain PPC (IBM + Freescale) has a larger than one percent market share.

Re:Not of ALL Desktops!

[Anonymous Coward]

You live in San Francisco, don't you, the only place where the Mac culture is still alive?

Newsflash: Mac is dead.

Re:Not of ALL Desktops!

[Socket Scientist]

Nope I live almost as far from there as you can get (and be on the same continent).

And last time I checked "Mac" didn't make processors.

You're 0 for 2 AP.

about time

[Anonymous Coward]

Finally, now I can stop hearing about how great Itaniums are and start seeing more amd64s on the market. (PS: don't tell me amd64's are pathetic and buggy. I run Gentoo Linux and 3 BSDs on an amd64 3400+, the rig is perfect for production programming.)

Re:Who would buy intel?

[Lord Kano]

I have an N2U400A from ECS. It is a cheap ass mobo, but when I get it installed and running I was surprised at how sweet the NForce 2 Ultra chipset is. Fast, and I do mean FAST and stable. I put some stress on my system. I'll run the distributed.net client in the background, while I have a VPN connection open to the database server, I'll be importing records into that database, A VNC session running, and all while while playing Counter Strike in the foreground. System stays smooth and responsive.

Via needs to change something, because the NForce chipset is kicking ass all over their offerings for only a few dollars more.

I was kind of wary when I heard that NVidia was releasing a mobo chipset, I thought that since they were a "video card" company that they wouldn't be able to make a good chipset. I am so glad I was wrong. Now, I'd like to see what ATI can do.

LK

So everyone here buys AMD?

[Anonymous Coward]

Either there is a serious case of hypocrisy or people on slashdot just do not buy computers. AMD has next to no market share but everyone here buys AMD? Kinda like how everyone here uses linux right but the desktop market share is less than 5%?

Re:So everyone here buys AMD?

[HermanAB]

Well, there are waaaaaaaaay more people in non technical jobs, who all loooooove their Windoze simplicity, screensavers and viruses.

According to IBM Marketing, there are about 30 million Linux desktops in the world, compared to about 600 million Windoze desktops. However, there are about 2 billion Linux embedded devices. Consequently, Linux systems outnumber Windows systems by a very wide margin.

It all just depends on how you count things.

See this article: http://www.pcmag.com/article2/0,4149,1210067,00.as p

Re:So everyone here buys AMD?

[Anonymous Coward]

This may be a case of lying to pollsters. Once they did a survey of how much beer do you drink. Then they compared the numbers against the sales figures.....

Huh.. Ain't that a corker...

[Chordonblue]

Funny, I've never had the problems you've described short of bad configs or bad boards. Shall we cite chapter and verse of Intel's mistakes lately? WE don't even have to talk about direct mistakes. The 810/815 series isn't exactly what I would call 'stable'. All things being equal there are ALWAYS cheaper alternatives that aren't quite as stable from Intel, VIA, SiS, and Nvidia. That's why you plan to get the better ones when stability is a must.

Last week I purchased a NEO2 board (NFORCE 3) and a 3800+ Athlon 64. It TOTALLY rules. Know what else? I do video and 3D rendering - a LOT of it. I have it OC'd about 5% so far too. No lockups. No 'crackles' from the Audigy. Everything is cool and I can't believe how frigging CHEAP it was!

Re:Huh.. Ain't that a corker...

[Inda]

"I do video and 3D rendering - a LOT of it"

Me too! On my P3 450! Cool as you like! Quiet as you like!

Takes 2.5 days to RIP a DVD to SVCD.

Takes 20 minutes to render a 1000 surface model in Rhino (800x600 I might add!).

I'm deadly serious. I even bought a duel-layer DVD burner for it a month ago. People I work with are shocked that I can burn a DVD in 8 minutes.

"Don't you need a new P4 3Ghz for that sort of thing?" they say.

I laugh.

Re:Intel secret plan

[Anonymous Coward]

So you also had an Amiga 500? :-)
M.

Re:mmm smell the toasty CPUs

[Anonymous Coward]

Heh. I guess my several racks of AMD servers with 100% uptime and happy customers is a myth.

Re:Who would buy intel? Who would use onboard...

[afidel]

Uh, this is a VERY well known problem with Via 686B Southbridges (among others) and Creative labs soundcards. Neither company knows how to follow the PCI 2.0 or 2.1 spec and so burst data transfers done by the sound card are corrupted. Some Firewire cards also have problems transfering to the iPod for the same reason. The problems were enough to put me off VIA permenantly. I now use SiS chipsets for my AMD systems and have had no problems (I don't need any of the expensive integrated stuff from a NForce board).

Hot On Their Heels

[bgumm]

I just read this InfoWorld article, which had a pretty good account of how AMD is starting to make progress against Intel.

(hint: they're actually innovating)

article lacking details

[nbert]

Why doesn't TFA state whether they are talking about the US market or worldwide sales? I know that they are talking about worldwide sales, because timothy implies it and my sanity tells me that AMD hasn't outperformed Intel in the US. But it's not my job to guess the obvious - the author of the article should make sure that everybody knows what he's talking about.

However, I believe that this is good news. AMD offers more bang for the buck and they seem to focus on reasonable improvements while Intel concentrates on (expensive) ideas their marketing departement came up with (HT is an obvious example - it might make chips a little bit faster, but you obviously pay more for this speed gain than it's worth).
I know that AMD also came up with nifty technologies like the Athlon64, which doesn't suit any purpose for the average WinXP user right now, but quite a few people apparently think that it is a good idea to buy them and it won't hurt anybody in the long run if x86 computers become aware of 64 bits.
Since I only use OSS on my computer, which I usually compile on my own the Athlon64 even attracts me ;)

Re:Who would buy intel?

[bob beta]

Right. And there's no AMD processor in the Mac. The choice is: Macintosh or Intel.

Re:Who would buy intel?

[matyas47]

That's not the point. I have an M-Audio 1010 (with breakout box) in my 1.1 Ghz T-Bird. I've never been able to get rid of crackles - under several flavors of both Linux and Windows. There is a well-documented problem with Athlons and Via chipsets and many pro-audio cards. (Unfortunately it wasn't very well-documented when I bought the card!) However, I don't think that more recent mobos (at least the better ones) have this problem. I do know people successfully using AMD systems for audio. In general, I would tend to support AMD over Intel - but I've more or less abandoned X86 for audio in general.

In other news...

[mod_parent_down]

The X-Box has outsold PS2 on a weekly basis in the past. It's sort of a blip lost in the noise of the whole "100 million units sold" thing. Vaguely interesting, but not really meaningful.

Re:In other news...

[Stevyn]

Wait, they also have Intel chips in them.

Clarification on figures...

[mercuryresearch]

I posted in a reply that doesn't appear to be getting modded up, so:

The figures for Intel's total share are worldwide, not US. (I should know, my company is the source cited in the link.) Meanwhile the AMD weekly share data (from another company) is for US Retail system sales. So the two data points really aren't comparable on any basis.

I know the figures I cite are exclusive to x86 CPUs. Someone mentioned PowerPC in this thread, and Apple provides sales figures as part of their financials -- based on Q2 data, PowerPCs in Apples comprise about 1.8% of the market if you included them in the calculations.

One company placed a big order

[HermanAB]

Those weekly figures don't mean squat. I think AMD has good processors and actually I'm using VIA miniITX machines lately, simply because they are so nice and small, but the main market still belongs to Intel.

Re:One company placed a big order

[Anonymous Coward]

"but the main market still belongs to Intel."

Bullshit.

There is *no* market when something like a Dell is contractually obligated to buy Intel procs. Let Dell out of its contracts and AMD would fly.

Buy AMD. (And not just because I own some of their stock.)

Re:Who would buy intel? Who would use onboard...

[cbare]

As I understand it (which may be fulla holes) the N/S bridge chipset matters a lot. Professional audio apps are notorious for having problems on the PC platform and the problem is that the PC platform was not designed with realtime (or even psuedo realtime) constraints in mind.

Even if you have a pro audio card that does a/d conversion, the data still has to get from the card to HDD fast enough. The system is probably way more than fast enough on average, but you get pops anyway if some other process keeps the cpu busy long enough for a buffer somewhere to fill up.

The chipset is key because audio is much more i/o intensive than compute intensive. So, the bottlenecks are definitely on the i/o bus (or maybe memory bus? I dunno.). I would guess that any pro audio app will have code that's been hand tuned to work with the patterns of latency typical in intel hardware.

But still, cheers to AMD for kicking some flabby, complacent, celeron-crippling, market-segmenting, mhz-is-everything intel ass.

-chris

Re:Who would buy intel?

[soliptic]

Very true. Also, Intels seem to be more vulnerable to the dreaded P4 denormalisation behaviour (whereby many older plugins cause massive cpu spikes when not being used), which is a Very Bad Thing for audio work.

Yeah, but who's sold the most

[ClosedSource]

in the last 5 min?

Re:mmm smell the toasty CPUs

[HermanAB]

Heh, a week ago, I loaded Mandrake Linux with a multiprocessing kernel onto a friend's P4 and the friggen hyperthreading processor promptly burned out - fortunately the machine was still under warranty.

Shit happens, but Linux is probably more stressful on a processor than Windoze.

They also declared new dawn as the Itanium.

[Anonymous Coward]

so RIP WWW must be true!

Wrong

[Anonymous Coward]

Opterons beat Xeons at media encoding.

Most sales are in the low-end anyway (very few people buy the fastest CPUs). Where Intel really scores is corporate "office" PCs (usually Celerons or very low-end P4s), because it has long-term contracts with big companies. If you look at systems bought or built by "enthusiasts" (gaming PCs / workstations / servers), the Opteron has been sweeping the floor with Intel for quite some time now.

AMD Selling

[Deliveranc3]

I agree it's a good thing, perhaps people are actually making headway against retailers and doing some research.

Re:HT -- MultiCore

[Proc6]

Does AMD plan on pricing dual-core 64's in the "power-user / gamer" realm, or up into the high-dollar server realm?

Re:Who would buy intel?

[MrNemesis]

Agreed, nForce2's are the dogs danglies chipsets.

Just be warned: nVidia's IDE drivers are still a little on the dogy side. Unless you really need the speed, stick to the Microsoft/kernel supplied drivers.

global warming

[tasinet]

On another page, global warming is still rising at an alarming pace.

Coincidence? Naah...

Re:global warming

[Stevyn]

At work we had this Dell XPS running a 3.4 ghz p4. That thing ran hot as hell. We had problems with the machine when it was rendering for 3dsmax and when we opened the case the heatsink was very hot. Actually, it was a pretty crappy heatsink considering the cost and thermal needs of th 3.4 ghz p4. Anyway, I'm assuming your joke was to point out how hot AMD's can get, well Intel chips can get very hot themselves.

Re:global warming

[evilviper]

Intel chips, since at least the intro of the Pentium 4, have HIGHER max thermal output than AMD chips.

It's gotten so bad that the P4's performance is getting hurt... Some of the newer processors they introduced don't perform any better than the older/cheaper ones, because when they are at full-speed, they output too much heat, and the CPU has to slow down to keep from burning up.

Now, AMD has some power management issues, but even with that, they aren't any worse than Intel.

Re:HT -- MultiCore

[Anonymous Coward]

Does AMD plan on pricing dual-core 64's in the "power-user / gamer" realm, or up into the high-dollar server realm?

Both I believe. I've read they will have a 940 pin part (Opteron) and a 939 pin part (Athlon64). Currently there's a price premium for socket 939 boards and cpus, so it's adoption is relatively slow at the moment.

Socket 754 is a much better buy in price/performance terms. The upgradibility question is moot since no AMD boards have PCI-Express or DDR2, so you'll be buying a new board if you those technologies anyway.

Overview of the discussion so far:

[Anonymous Coward]

• Intel still sell more CPUs than AMD
• My friends' AMD blew up so I'm never using them again
• The Celeron is a 'fine' processor
• People keep mentioning that VIA make CPUs (chuckles ;-))
• No-one gets to the point!

Being that competition is a GOOD thing - why Intel is allowed to bribe Dell into supporting its massive monopoly is mind boggling. Another thing that I don't understand is this - if Dell were to swap CPU lines overnight, from Intel to AMD, no-one would care because its the Dell brand that matters not the CPU maker - I suppose its just about volume - you can bet if AMD could guaratee supplies to Dell they'd do it because it would be good for profits, whatever deal they are getting from Intel.

Re:Overview of the discussion so far:

[mikis]

Except that VIA *does* make CPUs -- ever since they bought Cytrix. See here:

VIA Antaur
VIA C3
VIA Eden

Re:Intel secret plan

[deathcow]

I remember building a Z80 single board computer when pursuing my electronics degree.

The Z80 CPU was so stable that we could actually hook a potentiometer up to the timing circuit and scroll the system clock speed up or down and it just went on about its business of running happily. As I recall, it maxed out about 1 megahertz, but you could reduce it way down without trouble.

Re:Who would buy intel?

[MrNemesis]

Firstly, parent is a troll... but then you knew that ;)

Secondly, I agree that some of VIA's chipsets for the Athlon range were rubbish. The KT800 series for the AMD64's however seem to be prtty solid performers and are giving nVidia's nForce3 a good run for it's money.

It's nice to see good chipset support for AMD platforms, as this has been one of their major failings (although not really their fault) - I just hope this extends into the server/workstationb arena, where the Opteron platform *really* shines. They've always launched kick-ass chipsets at the time of a chips launch (AMD7xxx with the Athlon MP, AMD81xx with the Opteron) with very open specs - then everyone forgets the workstation platform, and the AMD server chips get left behind because of their poorly performing old chipsets. I remember how everyone raved about the Athlon MP kicking Xeon's ass (well, dollar for dollar anyway) when it came out, but in languished with the 7xx chipset for the rest of it's life.

Although now that the CPU's themselves take care of the memory, chipset design has become alot simpler I imagine. I just hope nVidia and others get offof their asses and cook me up some sweet 2P and 4P goodness!

AMD == outsourcing hardware.

[Anonymous Coward]

You object to having your job shipped to India, but you don't mind shipping a fab worker's job to Taiwan, or a logic designer's work to Japan.

What's with that? Buy Intel. They're American. The job you save will, ultimately, be your own.

Re:AMD == outsourcing hardware.

[Anarke_Incarnate]

put down the crack pipe. AMD's first megafab was in texas at Austin. They now have a larger megafab in Dresden, Germany with one larger one as an add on to the same facility. IBM employees stateside help AMD with their research and technology. Intel has Fabs in Ireland, etc all over the world. The base corporation being in the US means nothing as to where the chips are fabricated. Intel also packages the chips in costa rica and asian facilities. Your argument makes no sense.

Re:AMD == outsourcing hardware.

[ricotest]

You object to having your job shipped to India ... The job you save will, ultimately, be your own.

I'm not American, I don't care if my job gets outsourced (because it's simple business, and I understand business) and finally IT isn't my field. So you are making gross assumptions to appeal to the general case.

I see no reason for companies to pay more and pass those costs to me if they use American labour. I only see problems if the outsourced labour does not perform its job adequately, or there is a breakdown in communication.

Intel Inside! Profits not.

[Mulletproof]

"For the week ending August 21st AMD managed to capture 54% market share among new desktops sold. Intel's share during the week was 45%."

I'm sure it's that bastard who composed the "Intel Inside" jingle, sapping company profits with his royalties each time they run it!

don't forget..

[devhen]

this doesn't include systems that people build for themselves, which more often than not are built around AMD processors.

Re:don't forget..

[BCW2]

The white box store I'm working at sells 9 AMD's for every P4. Bang for the buck is the rule for home and small business. AMD has Intel whipped by that standard.

Re:don't forget..

[Some_Llama]

I have 5 home computers and guess what they run ;)

all of them built from scratch, all of them with more memory and better video card than offered by a dell comparable product... save money on the CPU, get the same performance, buy more ram and better video card maybe even a bigger HD...

woot... competition is a good thing.

Re:don't forget..

[BCW2]

Thats it. If you can get the same performance for $100 less, thats an extra hard drive, or an upgraded video card, or More RAM. You could get a better motherboard that can be upgraded for years!

People will normaly go for the bargain if the quality is still there. That is more true when the economy is down. Intel just doesn't give enough bang for the buck.

What I thought it said

[Cyclone_TBW]

AMD desktop sellsout Intel. Gates: They may think it's selling out, I think it's buying in.

They are the big guys now.....

[phantomfive]

Uh oh.....time to start hating AMD.

Re:Who would buy intel?

[Anonymous Coward]

Then don't buy crappy motherboards - you get what you pay for, afterall. Pay a little more, and get a nVidia board with a better reputation.

Not a problem particular to Soundblasters

[Anonymous Coward]

not many people realised it, but it isn't just with Creative Labs souncards. The PCI implementation on early (as in pre 166MHz FSB days) Socket A VIA chipset is a total mess. PCI transfers cross over each other causing corruption and lock ups. File transfers sometimes corrupt, there can be screen corruption, audio becomes distorted, I could go on.

Plonk a PCI video card in a VIA motherboard, go into the bios and set the PCI to optimal settings, then boot and set the screen to 256 colours. Loading the same image twice will load the same picture with different colours; you can watch the PCI corruption in real time.

I was stung by two motherboards from Asus with VIA chipsets. Both were total trash. I will never touch either brand again. I can not say enough bad things about Asus chosing to release motherboards over a long period of time that should have been blatantly obviously faulty.

The SiS & Intel chipsets I've come across have always been solid, not so the VIA.

Re:Not a problem particular to Soundblasters

[Anonymous Coward]

I was stung by two motherboards from Asus with VIA chipsets. Both were total trash. I will never touch either brand again.

Your problem was most likely attributed to the chipset, not the motherboard manufacturer. I have a few Asus systems that work great. Of course mine are nForce 2 chipset, and I won't touch VIA with a 10 foot pole. The 2 models I have are Asus A7N8X Deluxe and A7N8X-E. The E model's second NIC port is 10/100/1000Mbps, but aside from that, they are essentially the same thing. Great boards though.

Re:Not a problem particular to Soundblasters

[Anonymous Coward]

I thought that at first too, but I found out that much of the problems were because the mo/bo companies ignored VIA's implementation notes (presumably in the quest to increase performance), and in doing so caused the PCI to be stuffed up more. That's why Asus's early VIA Socket A motherboards had massive data corruption issues when transfering large files.

Whatever extent it was or wasn't VIA's fault, it doesn't change that Asus sold boards that were glorified paperweights, and any Q.A. department with half a clue would've known it. Either they did no testing on their own products, or just sold them anyway.

Alright everyone, with me now!

[Chapium]

We hate AMD we hate AMD.. that darn chip monopoly. Get your pitchforks ready. ;-)

Re:Who would buy intel? Who would use onboard...

[real_smiff]

there is a patch for this as long as you are running Windows 5+ (i think). I can tell you it fixed my BSODs a treat (Abit KT7a, SBLive). Seems to be beta and discontinued unfortunately, but like I said it works. I figure if/when i change OSes, I'll change the sound card. I only post this here because it might help someone else who recognises this problem (I didn't even realise it was the sound card, that was the last thing i thought to check. i thought i had a cracked mainboard!). Shame the necessary adjustments aren't built into the mainboard BIOS, or like you say, people kept to the spec.. this was a real arsehole to track down. I knew about the infamous VIA/Live! issue, but it only affected me after i changed from a UDMA66 to UDMA hard drive. So like i said, real arsehole that one.

Re:And...

[LiMikeTnux]

actually alot of people buy emachines too, so joe-user is starting to see amd xp on the label and no intel. in fact i just set up someone's emachine for them, and inside was an amd xp 3000+ chip.

Re:Who would buy intel? Who would use onboard...

[wemgadge]

"AMD motherboards with cheap, ultra-crappy VIA south/northbridges and the like -- just won't cut it. You will get crackles in the audio, Professional Audio applications aren't running with onboard AC97." ya.. and my AC97 crackles quite nicely on my P4 based PC (with Via Chipset) If you're going to use it in a studio, pony up for a real soundcard. As for me, it works quite nicely for hobby recording..quick pass with an FFT filter clears up the static. Don't mistake this for an AMD problem. Via makes Intel boards too.

Friction is proportional to surface area.

[Anonymous Coward]

Friction is directly proportional to surface area.
Smaller engine=less area=less heat.
IIARTC, the OP is saying that the heat gained from speed is more than lost by reduced size.

Re:Friction is proportional to surface area.

[Anonymous Coward]

Smaller engine=less area=less heat.

That rule would seem to be inverted for CPUS.

Re:Friction is proportional to surface area.

[Grishnakh]

Smaller engine=less area=less heat.

That rule would seem to be inverted for CPUS.

Actually, it's not.

Power consumption in CPUs is dictated by clock speed, die size, and feature size (90nm, 130nm, 180nm, etc.).

If you were to take the old Pentium II design, and re-engineer it for the modern 90nm process (or better yet, the upcoming 65nm), you'd be able to shrink it down to a smaller die size. This would yield both a smaller physical chip (which would be cheaper to produce because it's using less silicon), and lower power consumption, assuming you ran it at the same 300 MHz or so that the old P2s ran at.

The problem is that chip companies and consumers don't care about lower power consumption; they want faster performance, or more precisely, they want bigger numbers so they can brag to their friends and feel like they're doing better than the Joneses. So while going to smaller feature sizes helps reduce power consumption, going to a higher clock speed more than makes up for it, so the actual power consumption is continually rising.

Even worse, with transistors becoming ever smaller, the heat they produce is being concentrated into smaller regions, which causes localized heat problems on the chip, necessitating more engineering solutions to keep those areas from overheating. If you look at a thermal map of a CPU in operation, you'll see that a very small part of the CPU is generating the majority of the heat--the ALU and execution units, which are constantly utilized, produce most of the heat, while the SRAM cache produces very little even though it probably accounts for a majority of the die real estate.

Re:Who would buy intel? Who would use onboard...

[afidel]

Doesn't work all the time, it doesn't fix the fundamental problem, and it shouldn't be necessary. I was really, really pissed when I found out what the problem was. I had thought it was something wrong on the other equipment since I only ever recorded at live gigs with questionable setups (throwing warehouse parties leads to interesting audio setups). I had about 10 live shows recording get screwed up by this stupid bug. Like I said I will never build or own another Via based system.

Re:Who would buy intel?

[AsylumWraith]

And you'll be replacing that N2U400A within two years.

ECS used to make decent equipment. The K7S5A is exceptional for low-end applications, but it seems like everything they've put out since about the time of the K7SEM and K7SOM+ is utter, unreliable, crap.

Re:Who would buy intel?

[Anonymous Coward]

Wow, you sound like an ass. Ummm... Last time I checked if you're not a troll you're the typical whiny bitch sellout artist type... or doesn't or whatever. I don't care whether AMD or Intel is better, you're just annoying.

Re:mmm smell the toasty CPUs

[Pierre]

did you install it with a hammer?

HP's computer line is a joke everywhere.

[Behrooz]

HP's computer line is a joke everywhere. Well, everywhere outside Elbonia.

x700, ATI's answer to GF 6600, will launch Sept 21

[MojoStan]

with the new 6600 cards coming out, this is going to be a firm kick to the nether regions of ATI. There just isn't a card on the market that can hold a candle to it

Haven't you heard about ATI's Radeon x700 cards? If you haven't, they are based on the same core as the x800 cards and will be launched on September 21. Details here: Radeon X700 series to launch on 21st

Since the GeForce 6600's are not available in retail yet, I wouldn't declare the future's "mainstream" ($150-$200) winner until we see x700 benchmarks and both are on retail shelves. Until then, the current mainstream cards are GeForceFX 5700/5750 and ATI Radeon 9600/x600.

Intel is very dumb

[Anonymous Coward]

So far what I see is that AMD opteron provides 85% of the performance for 15% of the cost of the Itanium. Also the IBM Power5 simply wipes the floor with it on the higher end spec FP of 2700 vs 1700 and generally a better SMP architecture, higher throughput. much higher theoretical performance with a much small R&D investment.

They are fast but expensive, the compiler issue is a killer. They can not be made cheaply they are giant chips with large caches and tons of registers. Mainly the performance is not a result of any miracle of chip design but having the stack stored in registers.

Given the resources at intel in short time a reworked alpha could be developed with very high performance at a lower cost. Simply put intel has been out of the ISA game so long they really do not know what the hell they are doing

Re:Intel is very dumb

[Anonymous Coward]

Simply put intel has been out of the ISA game so long they really do not know what the hell they are doing

I'll agree with this. Intel has been making mistake after mistake lately. I should know; I work there. Management keeps saying we need to execute better, but the real problem is the stupid decisions that are made at the top. The executives think that telling people to work longer and harder, while there's an effective freeze on salaries, is somehow going to make things work out.

Itanic, dropping the ball on X86-64, overhyping LCOS and then realizing their schedule was unrealistic, and in the cellular arena they're trying to put out all kinds of products even though none of them are really catching on. What's worse, they have a hiring freeze, so while they're trying to do more and more projects, they're losing people to attrition and refusing to hire anyone new to replace them.

Personally, I think Intel is headed for a crash, buffered only by their dominance in CPUs (especially notebooks and corporate desktops) and chipsets. It's like Microsoft, where their only profitable business units are the OS and Office; everything is losing money. It's about the same at Intel, except that the CPUs and chipsets don't enjoy the kind of profit margin that MS's profitable units do, so they're not going to be able to afford to keep the other units afloat indefinitely.

Re:Intel is very dumb

[ndykman]

Couple of things. First, the Power5 is a real threat, no doubt. And IBM was smart enough not to pull the plug on AIX (unlike HP, which seems to have shot HP/UX in the leg). And the Power5 does have more software running on it. I could very well be that the Power architecture does the Itanium in, but I hope Intel doesn't fall over and play dead just yet. As for the ISA, you can't blame Intel for that. HP did most of the work there. Most of it was based on a VLIW version of the PA-RISC architecture. And, from my looking at the market, 4 way Opterons are about the same as entry level Itaniums, so I'm a bit skeptical of the 15% of the cost argument.

athlon xp 3000

[Anonymous Coward]

runs at 46-53C under heavy video processing loads, in summer heat of 80F

old stolen story

[ruiner5000]

Nice, I like how the story is a week old, the facts are stolen, and even the graph has been ripped from another site.

Re:old stolen story

[Anonymous Coward]

huh? stolen from which site? the graph is in the press release that TFA links to?

Re:Intel is not faster than AMD

[GISGEOLOGYGEEK]

The best chip depends on the application.

I have a 2ghz P4 and a 2.088ghz Athlon XP (2600) both running Prime95 right now.

By your own reasoning, the Athlon has a huge advantage .... the slightly faster clock speed, and the huge ability to do more with one cycle than a P4 can.

nonetheless .. the P4 is chewing threw the Prime95 test much faster. About 0.086 seconds per iteration versus 0.121 seconds per iteration on the Athlon. About a 40% advantage for the wimpy piece of crap P4.

Both computers are running nothing else and are performing Prime95 tests of roughly equal difficulty.

Over the last 5 or 6 years there's pretty much always been such an advantage for the Pentiums on the hardcore FPU processing.

I am an AMD fanboy ... but unlike you, I dont sit there blindly screaming "I'm the best! I'm the best, you suck!" ... Instead, I use the best chip for the application at hand and get on with business.

Dell should look

[Anonymous Coward]

Sun destroyed their server sales last quarter with their Opteron servers. Sun was #1 in server sales I believe mainly because of the opteron. 32 bits looks pedestrian to IT purchasing compared to 64 bit goodness

I can go with this...

[Spacejock]

My last n desktops from oldest to newest:

Intel Pentium 90 (OC to 100Mhz after a year or so)
Intel Pentium 120 (OC to 133Mhz)
Intel Celeron 300A (OC to 450mhz of course)
Intel Celeron 633 (OC to something insane like 950 Mhz)
Intel Pentium III 1ghz (When the 633 blew up after a few short weeks)
Intel Pentium IV 2.26 Ghz - didn't seem any quicker than the PIII

For my latest PC I read up on the Northwood and Prescott CPUs. However, I live in Western Australia where summer daytime temps reach 40-45 deg celcius and nights don't get a whole lot cooler. My house has aircon, but last year everyone was told to switch it off because the power companies didn't see the massive switch to airconditioned homes coming a few years back, and their ageing equipment isn't up to the task. If you take a CPU running at 60+ degrees and 'cool' it with air at 35 degrees, it's not going to do much.

Having put together 4 Athlon XP machines for other people over the past 6 months or so, I've had my eye on their CPUs.

So... I now have an Athlon64 3400+ on a Gigabyte 'kitchen sink' MB with dual SATA 120 gig hard drives and a gig of ram. Bye bye, Intel.

Re:I can go with this...

[Anonymous Coward]

My last n desktops from oldest to newest:
C=64
486 DX2/66 (clone)
Athlon 550 MHz

Yes, that's the entire list, but I am considering getting a new computer within the next year or two. I like to make each new computer a noticable jump in technology.

Re:Rules of the game

[jcr]

last time I checked, the only people buying Macs for audio work were the ones using Protools systems

check again.

-jcr

Think McFly ...

[Anonymous Coward]

You can't sell less than 20% of the CPUs and more than 50% of the systems. Clearly AMD picked their data and foisted a PR piece on us. What really brough a chuckle to me is that Slashdot took the troll hook, line, sinker, pole and boat.

Reminds me of when Ford was getting their a$$ kicked by the Japanese car makers in the '80s and used to advertise the "best selling car designed and built in North America". Of course that died when the Japanese put their plants over here because costs were lower and worker quality was best.

These are no-name PCs sold through the retail channel only. I'm not so sure I'd want to be known as the CPU most prefered by guys who sell off the back of a truck.

--------
PR ... for when the truth just won't do.

Hey: Good for AMD!! screw intel overpriced crap!!

[Anonymous Coward]

Its about time that a decent company came out on top (I remember 15 years ago when intel was crushing good competition and a 15mhz 386 chip (no motherboard) cost $1200 US and you were not allowed to buy a computer without having microsoft installed on it (even though there were competitors availible), because the belief was that you were going to pirate an OS from MS, and companies selling systems (where I worked) were locked into contracts with intel and MS and if you even said AMD out loud, intel could cancel your contract.. I use an AMD machine now, and I refuse to use overpriced, under-engineered orverated intel machines)....same with software, the sooner I get rid of microsoft and go to linux, the better. I can't wait to get a 64bit AMD machine!

They never let you gohhhh....

[davidsyes]

Maybe AMD will sing this 70's jingle to taunt Intel?:

"OUR chips match YOUR chips --they RACE you they -- MOLD you-- they NEHvur let 'chu gohhhh."

* (or, SCOLD, or SCALD, or PROBE)

(A play on Leggs' "OUR Leggs fit YOUR legs -- they HELP you, they HOLD* you! They NEHHHvur let you go....")

DAMN! TOO much TV in the '70's!!!!

Re:Intel is not faster than AMD

[Anonymous Coward]

The P4 is using SSE2, the Athlon XP (an older family of processor) is using SSE1.

Try that Prime95 test with a 2GHz Athlon 64 (all K8's have SSE2) and we'll see apples to apples.

I think you can even force Prime95 to use SSE1 on the P4, which would give you an objective benchmark.

Any application that doesn't use SSE2 is usually SOL on a P4 when run against an A-XP, from the benchmarks I've seen.

More interesting than IDF ?

[Anonymous Coward]

So this totally random sales statistic is more interesting than a story about the Intel Developers Forum ?

How about the showing of Montecito , a chip with 1.7 Billion transistors . "News for nerds .. Stuff that matters " , yeah right .

Re:More interesting than IDF ?

[Anonymous Coward]

so where is your story submission?

Blah

[Anonymous Coward]

blah blah blah blah blah intell blah blah blah blah amd is the

If this keeps up

[Anonymous Coward]

If this keeps up we'll soon have to switch to intel to ensure we ssupport the underdog

Unlikely

[S.I.O.]

It's hard to hear anything inside Intel's ass.

Intel is partially to blame for this

[MemoryDragon]

For plain GHz monsters, AMD simply is the better price/performance deal. Now lets look at the situation from a different angle. Intel has a kick ass processor in their line. Yes the Pentium-M faster than most except the high end P4s only sold in servers and laptop computers. Outside of the US there is a huge market for machines which save energy (well in the US nobody thinks about energy, except for a god given right to be consumed probably) But the market currently is dominated by the rather measly VIA CPUs which have a huge following over here in Europe (and probably Asia) Well AMD currently reacts to the trend with their own line of new fast energy savers (which we will see probably in desktop boards soon, but definitely not from via :-) ) Via currently sells boatloads of their C3 stuff, and Transmeta probably would also if their stuff was available. So where is Intel in that game. Basically nowhere, Intel itself says this is a notebook processor only. Some third companies already produce industrial boards because the advantages of the PM over other intel designs are huge, blazingly fast, with a rather low power consumption. But those boards cost a fortune. But the end user market is left to VIA. What happens here is basically the same thing Intel did in the 64 bit market, which basically was handed over to AMD. And if AMD can get their act together and have several companies producing boards in the ATX format using their new low powere cores, they basically will win the slowly but rapidly emerging home server market, which currently is a hobbyists market, but in a few years will become the mass market.

Re:Intel is partially to blame for this

[evilviper]

Even stock AMD processors can beat the best VIA processors, if you use a motherboard that fvcool works well with.

The one I've had best success with is a VIA/SIS KT133... It only supports processors up to about 1.4GHz, but with fvcool running, that 1.4GHz processor plus mobo chipset, plus RAM, plus hard drive, plus power supply, will only use 40watts.

The biggest problem with AMD is that they haven't wipped their motherboard makers into like, like Intel has.

Re:Who would buy intel?

[Lord Kano]

My last board was a K7S5A, that board and the Athlon XP 1800+ that I got with it are now in my girlfriend's machine. With the exception of losing the bios settings every few weeks, that board is fantastic. Cheap and stable, even if not all that fast compared to its contemporaries. But I challenge anyone to find a better board new for $40.

LK

Intel is trying to beat their worst competitor...

[Kjella]

...which is Intel. A few percent here and there is nothing compared to making all the current computer owners out there buy a new CPU. Primarily by making people feel that their old CPU is missing lots of buzzword features.

Kjella

this is good news

[maxpublic]

Unless you're some sort of fanatically loyal Intel customer the better that AMD does the more likely it is that Intel will produce improved products to compete. Unlike MS, Intel isn't in a position to force its competitors into bankruptcy through illegal monopoly practices, and Intel doesn't have what it takes to buy AMD out.

I only wish there were a half-dozen major players in the chip market, rather than just two. Don't get me wrong, I've chosen AMD over Intel for years now and have been quite happy with the results (despite the bizarre brand-loyalty FUD that some Intel customers start screeching the moment they hear the initials "AMD"). But six different companies have a better chance of coming up with innovative products than just two, at least most of the time.

So as a customer I wish both Intel and AMD heaps of success, if only so both will remain in the market competing with one another. It's better for me, it's better for you - it's better for everyone except a few at the top of either corporation who long for an MS-style monopoly.

Max

Re:Who would buy intel?

[ecc0]

That was very much beside the point, though. I considered mentioning that, but this is really an AMD/Intel discussion, and if you're going to do audio work on an IBM compatible, you will be using Intel.

Re:Who would buy intel? Who would use onboard...

[ecc0]

Listen -- I speak from experience, I KNOW what I am talking about. I use M-Audio soundcards, and there is a WELL-DOCUMENTED BUG in several VIA chipsets that makes PROFESSIONAL -- not AC'97 gaming -- use more or less impossible with these chipsets. As other posters have mentioned, there are some patchese, but in my experience those help slightly at best and make the problem worse at worst. KINDLY know what you are talking about the next time you are going to reply.

Re:Who would buy intel?

[ecc0]

THANK you. I'm glad to have someone else who actually knows exactly what problems I am talking about confirm it. So apparently, some nVidia chipsets now work better than VIA crap. I don't think I deserve -1, Troll for not knowing that -- alright, some newer chipsets might work better, but I'm never buying AMD for serious use again after experiencing what I have, and you need only search Google for a couple of minutes to see that I'm not alone.

Although, what do I expect posting a pro-Intel post on Slashdot... It's not like I wouldn't want to see the cheaper, less evil stuff be better, but from my experience, it just isn't. For advanced PCI stuff like pro audio cards, it's worth spending the extra bucks to get the reference implementation and be ASSURED things will work.

Re:Who would buy intel? Who would use onboard...

[ecc0]

VIA's Intel boards might be cheap and crappy as well, but they do not have the well-documented bug that several of VIA's AMD chipsets have that makes high-bandwidth PCI usage a fucking hassle at best and impossible at worst. Again, I DO NOT USE A SHIT SOUNDCARD. I said I speak from experience, so kindly believe me... Even if this is Slashdot, infamous for people talking out of their asses.

Re:Who would buy intel? Who would use onboard...

[evilviper]

The problems were enough to put me off VIA permenantly. I now use SiS chipsets for my AMD systems

HAH!

That's damn funny!

You were put-off by VIA, so now you only use chipsets by a company they own! Wonderful!

The only alternative to VIA is NVidia's NForce boards.

I've always bought AMD....

[Anonymous Coward]

I've always bought AMD simply due to the fact that less price gets you equivalent or better performance.

Have just upgraded my box to an Athlon 64 (w/motherboard and ram) for just £250 ($400)! Am well pleased with the results....

Circle

[Remlik]

Ok so Intel has, until recently, held the market share of processors. Because of this they have been able to keep processor prices high, knowing that people will pay. Hence they have a small monopoly, and are kicking consumers arounds. INTEL BAD!

Enter AMD. Everyone likes to root for the underdog! GO TEAM! Now AMD has made some strides into and over Intel's market share. Soon they may even topple Intel. Where does that leave the consumer? We still have a small monopoly controlled by one company. You can't tell me that given enough time AMD won't become the next Intel with high prices in a locked market. /.ers do this constantly, yea small guy, boo big guy. Then the tables turn and we are back to cheering the new small guy. Pick a side and stick with it please.

Re:HT -- MultiCore

[EnglishTim]

I've read they will have a 940 pin part (Opteron) and a 939 pin part (Athlon64).

I wonder what that extra pin is for? Perhaps it sends the chip an 'I am a server' signal?

99% total

[MacGod]

If AMD got 54% and Intel 45%, that only leaves 1% unaccounted for. I find that a little difficult to believe. Do you mean to tell me that IBM/Motorola (PowerPC) chips, VIA chips, and Transmeta chips together only added up to 1% of the market?

Seems a little fishy to me

Re:Intel is not faster than AMD

[Anonymous Coward]

WOW!
How on earth does it achieve that? Does it perhaps select between an x87 code path written in C and a hand-tuned asm SSE2-path depending on features?

Seriously. That's the only way I could explain the difference. Per clock, AMD has roughly twice as fast x87-unit and while its SSE doesn't achieve similar speed difference, it isn't supposed to be slower than P4's either.

Intel CEO a liability

[Anonymous Coward]

While I think that AMD might have a slight edge for the moment, for my use, programming, both products did the job adequately. The clincher was remembering Craig Barrett's statement pointing to the U.S. education system as a cause for offshoring. We should work to improve education in the U.S., but no teacher held a brownie to Mr. Barrett's head and said "the children are stupid, Craig, make your chips somewhere else..."

Mr. Barrett probably makes proportionately more money than most CEO's, or so recent studies woud promulgate. I'm confident that he would tell you differently regardless of the truth however.

Thanks for the choice, I enjoyed it.

Re:Who would buy intel?

[Fareq]

Yay for slashdot!

Once again, pro Intel? You must be a troll, nobody actually likes Intel... Pro AMD? Ahh, there ya go, that's nice and informative...

why am I not surprised......

I happen to agree with the parent. There's nothing quite like an Intel CPU running on an Intel Chipset. I don't usually run Intel Motherboards, being an MSI and Gigabyte fan, but I always pick out an Intel chipset... They just work more often...

Don't like it? BFD. You can buy whatever you want.

Re:Who would buy intel?

[Fareq]

Bullshit.

Itaniums are currently very expensive. Mostly because they are designed for servers. Also partly because the fixed-cost of designing the damn thing has to be recovered with fewer units, because its a server thing, and only really for servers that need 64-bit.

The reason that Intel is going to make "AMD compatible" 64-bit processors is because the AMD version extended x86, so that you don't have to emulate "old 32-bit apps" Since 99.9999999999999999999999999999% of all apps normal people run are of the "old 32-bit" variety, Intel hasn't been able to get eople to switch to a new architecture because it would run their apps slower. At least, until they had 64-bit apps to run.

Intel recognized that x86 is a painful architecture and tried to make a new one (not sure how much beter IA64 is, haven't digested the specs). AMD thought they should bet on a proven winner.

It is entirely forseeable, AMD won that battle. That's why Intel will make x86-64 chips. NOT because they can't make others.

That's not neccessarily Perl

[mr_mischief]

Your perl might choke on it because it's not Perl. The vi editor would have little problem understanding that, so long as it's in command mode and you typed a colon first.

Yes, Perl is good with regular expressions and has a highly extended regex syntax compared to some other tools. I love Perl and do almost all my coding in it. It is not, however, the only place regexes are used.

buying generic Dells with Intel inside

[Some_Llama]

My neighboor spent 900 on a brand new dell dimension 2400, when i showed him my ssytem (AMD inside) that costed 750 dollars and blew his system away in terms of performance and upgradability (the dimension doesnt even have an agp slot) he immediately asked me to sell his Dell on ebay and help him build a new computer using AMD...

i think most people just don't know enough about the nuts and bolts of computer hardware to make an informed decision with their dollars.. they see dell or maybe gateway/compaq and buy the best they can afford...

Too bad their stock isn't doing well

[Anonymous Coward]

a real bummer

Re:Who would buy intel? Who would use onboard...

[Anonymous Coward]

Creative Labs is not professional calibre cards.

I've had no issues with my Event Darla in and AMD box.

Re:mmm smell the toasty CPUs

[insomniakxz]

So you expect to achieve different results by using an AMD processor, instead?

Re:Intel is not faster than AMD

[GISGEOLOGYGEEK]

Well DUHHHHH

damn you anonymous coward fanboy fools!

My athlon 2600 is newer than the P4 that I have it up against, yet the older, slower P4 kicks its ass on the Prime95 test. Did I say I was testing a brand new processor? No.

It's in your face but you can't deal with even one small corner of the processor world being absolutely ruled by the P4.

I am curious how an Athlon 64 would do ... not because of the sse2 stuff, but to see if it can compete close to the 8 instructions per clock cycle FPU of the latest P4's ... You do realize the latest P4's are not the same ones that were around when my XP was new don't you?

Re:Intel is not faster than AMD

[GISGEOLOGYGEEK]

wooot, here's a link

http://www.mersenneforum.org/archive/index.php/t -1 947.html

YOU were SOO WRONG! hahahahaha

Check the benchmark number in the second post on that thread, for the '1536K FFT' test of 95.939ms (0.095 seconds per iteration) for an Athlon 64 3200, that has been OVERCLOCKED by 15%!

That '1536 FFT' notation puts the benchmark in the same category as the numbers I'm testing on my Athlon XP and P4 that I mentioned in my first post.

Yet the OLD 2ghx P4 is still faster by more than 10%!

For Prime95, A 2ghz stock P4 kicks ass over a shiny new Athlon 64 overclocked running at 2362mhz!!!!!

Damn! Maybe I'm not an amd fanboy anymore!

Re:Intel is not faster than AMD

[GISGEOLOGYGEEK]

wooot, here's a link

http://www.mersenneforum.org/archive/index.php/t -1 947.html

YOU were SOO WRONG! hahahahaha

Check the benchmark number in the second post on that thread, for the '1536K FFT' test of 95.939ms (0.095 seconds per iteration) for an Athlon 64 3200 with SSE2, that has been OVERCLOCKED by 15%!

That '1536 FFT' notation puts the benchmark in the same category as the numbers I'm testing on my Athlon XP and P4 that I mentioned in my first post.

Yet the OLD 2ghx P4 is still faster by more than 10%!

For Prime95, A 2ghz stock P4 kicks ass over a shiny new Athlon 64 overclocked running at 2362mhz!!!!!

Damn! Maybe I'm not an amd fanboy anymore!

Check out the next post in that thread link, that describes your other error, that the A64 FPU has the same throughput per clock cycle as the P4 ... but with a far slower clockspeed.

Heck, the non-overclocked test is nearly as slow as my Athlon XP!