Intel vs. AMD - Today's Generation Compared

Bender writes "The Tech Report compares 15 Core 2 and Athlon 64 processors from Intel and AMD — from sub-$200 to a cool grand, from slower dual cores to fast quad cores — in 32 & 64-bit apps in Windows Vista, including the new, multithreaded RTS game Supreme Commander. 'The release of Windows Vista and a round of price cuts by AMD prompted us to hatch a devious plan involving Vista, a new test suite full of multithreaded and 64-bit applications, fifteen different CPU configurations, and countless hours of lab testing. That plan has come to fruition in the form of a broad-based comparison of the latest processors from AMD and Intel... from the lowly Athlon 64 X2 4400+ and Core 2 Duo E6300 to the astounding Athlon 64 FX-74 and Core 2 Extreme QX6700.' Folding@Home in Linux, power use, and energy efficiency are tested, too."

Comments

Summary

[richdun]

14 pages of ads later...

Intel > AMD at high end, Intel >= AMD at low end, Core 2 > A64, Intel finally has a lead in both architecture design and process (65nm).

Re:Summary

[hal2814]

I disagree with their definition of "low end." Maybe low end as far as what they tested, but there are a lot of non-X2 Athlon 64s and Pentium/Celeron Ds being sold. At the true low end, AMD is still more than competitive. It's only when you near the most-horsepower-per-dollar peak that Intel really pulls away (and that's where they seem to start measuring here). It's worth noting that I have no dying love of AMD. I have two AMD processors and one Intel processor running in my current personal machines and plan to get a Core 2 as soon as the next significant price drop occurs.

Re:Summary

[itlurksbeneath]

Ads? I don't see no steenking ads... Adblock rulez!

Offtopic, yes, but I couldn't resist..

Re:Summary

[Barny]

In particular the x2 3600 and 3800 seem to be 2 of the best bang-for-buck chips out there.

Yes ANY of the CPUs tested will leave them for dead, but if your user is running WinXP, doing a bit of this (video transcoding) and a bit of that (watching streamed video) and even a bit of the other, they will do it and leave them thinking "damn this thing is fast" all for pocket change compared to these other chips.

Re:Summary

[ArcherB]

I don't see any adds either! All I see is:

Unable to connect

Firefox can't establish a connection to the server at techreport.com.

        * The site could be temporarily unavailable or too busy. Try again in a few
                    moments.

        * If you are unable to load any pages, check your computer's network
                    connection.

        * If your computer or network is protected by a firewall or proxy, make sure
                    that Firefox is permitted to access the Web.

Re:Summary

[LoudMusic]

14 pages of ads later...

Intel > AMD at high end, Intel >= AMD at low end, Core 2 > A64, Intel finally has a lead in both architecture design and process (65nm). There were ads on those pages? News to me. Firefox + Adblock. It will set you free.

As far as the data is concerned, it's good information but somehow unsurprising. Maybe Apple partnered with the right people after all.

Re:Summary

[Endo13]

Couldn't agree more. Intel's cheapest Core 2 Duo CPU is still $169 on Newegg. http://www.newegg.com/Product/Product.aspx?Item=N8 2E16819115013

And anything Intel has that's lower than that is still pwned by AMD CPUs that sell for half the price.

Just last night I finally bought some kit to get my system into the dual-core/DDR2 generation, and with AMD I was able to squeak in at a mere $300 at Newegg, including CPU, motherboard, RAM, and video card. The entrance fee with Intel (albeit with a significantly better CPU perhaps, but a 3600+ X2 will do what I need just fine)using a similarly equipped motherboard would have been nearly $394, or about 32% more. Easy choice for me.

Re:Summary

[Amiga Lover]

Intel > AMD at high end, Intel >= AMD at low end, Core 2 > A64, Intel finally has a lead in both architecture design and process (65nm).

I think what this review forgets is that AMD have the better design, and despite that they're running slower, the chips themselves are manufactured and designed right from the start to be more elegant internally.

The speed might be faster, but that's like comparing a slow BMW to a fast ratty corolla that some kid's bolted a turbo to. Technically sure the corolla is faster, but the BMW is still the better car, and that's what counts.

AMD is still king of the hill in my book

Re:Summary

[Anonymous Coward]

Thats really desperate fanboy talk there; 'Its worse by every metric, but under the hood its better designed, and so its better overall'. Ok...and thats ignoring the fact that AMD's designs ceased to be 'better' (care to explain what you mean?) about three or four generations ago.

I guess an Amiga user would be used to coming up with stuff like this though, carry on...

Re:Summary

[geekoid]

"Maybe low end as far as what they tested, but there are a lot of non-X2 Athlon 64s and Pentium/Celeron Ds being sold. At the true low end, AMD is still more than competitive."

do you relize that you're saying AMD is on it's way to absolences?

AMDs new slogan: "We're really fast on old CPUs!"

Re:Summary

[hal2814]

"do you relize that you're saying AMD is on it's way to absolences?"

Your point being? I've already stated that I have no great love for AMD. It wasn't karma whoring. It was truth.

Re:Summary

[High_Noonan]

14 pages of ads later... Only ad I saw was for their hosting provider. Either make the switch to Firefox or stop bitching about ads in IE.

Re:Summary

[Endo13]

I'm an AMD fan, but I have to strongly disagree with you. That may have been true when comparing Athlons with P4s, but the Core processors are a whole different ballgame.

It's incredibly ignorant to assert that AMD still has the better design when their CPUs have less processing power while consuming more electricity.

AMD may still be able to work with their current architecture and make it competitive with the Core 2 design, but only time will tell. It is worth noting that only now are they starting to convert to a 65nm fabrication, and they may well be able to significantly improve the work per watt ratio in doing this, however it's naive to assume that Intel isn't already working on similar improvements with an even smaller fabrication process or something similarly effective.

So though I tend to prefer AMD, in the end my spending dollars will go to whoever can give me the best performance at the price point I can afford. Currently, that is not yet Intel.

Re:Summary

[richdun]

I was in Safari, actually, so IE (a modern, supported version) isn't an option, and Firefox for Mac is less than good.

Stop apologizing for bad web design by blocking content.

Re:Summary

[Gilmoure]

I thought elegance died on the Titanic?

Re:Summary

[gbjbaanb]

but it was a review of multi-core CPUs. As far as the review/test is concerned, a AMD x2 4200+ is the low end chip.

at the true low-end, Intel is still the cheapest, as I can get a P3 1ghz off ebay for a fiver...

Re:Summary

[morcego]

In particular the x2 3600 and 3800 seem to be 2 of the best bang-for-buck chips out there.

Since I'll be buying some X2 3800+ computers soon, I'm really happy to hear that.

Re:Summary

[onkelonkel]

Beautiful Troll. OK, I'll bite. The undying AMD fanboy-ism is either really sad or really funny, I'm not sure which. It's just a processor. Sometimes AMD will be faster at a given price point and sometimes it will be Intel. For a while it was AMD, Right now it's Intel. Get over it.

Re:Summary

[whoop]

Just the other day, Newegg had the X2 4200 for $109. I was going to build a 3800 AM2 box, but then they dropped that, so I went with it. Check again if you're getting ready to build. ... I just checked, and damn, they've lowered it a few more dollars to $105.90. This CPU was just around $150 a few weeks ago before AMD cut prices, now Newegg has cut it a couple times more.

Re:Summary

[morcego]

Whoop, thank you for the tip, but I'm not in USA.

Also, I'm not building my own machines, but getting HP boxes.

Re:Summary

[drsmithy]

Intel > AMD at high end, Intel >= AMD at low end, Core 2 > A64, Intel finally has a lead in both architecture design and process (65nm).

Finally ? I think you mean, after AMD's brief lead, the status quo has been returned.

Re:Summary

[mikeburke]

Thanks! That's all I wanted. Good night. P.S Could you come work in my office?

Folding@home just released PS3 client as well...

[Anonymous Coward]

Despite the FAH PS3 client has been out under 24h the PS3 client performance is overtaking all the CPU/GPU FAH clients combined!
http://fah-web.stanford.edu/cgi-bin/main.py?qtype= osstats
http://folding.stanford.edu/FAQ-PS3.html

Refreshing

[Visaris]

While the review is not perfect, it is a breath of fresh air compared to many of the tactics reviewers often use to skew the results in the favor of one company or the other (usually Intel). Tech Report presents benchmarks that each side wins. AMD takes a clear win in Cinebench and POV-Ray and some minor wins in a couple of other areas. It is good to see AMD get some accurate representation in a time when most are happy to claim that Conroe and the Core2 arch cannot be beaten. AMD's new architecture (new core enhancements as well as quad-core) will come out at the end of the second quarter this year, and if their claims of performance improvement on the per-core level is accurate, I think we may see another stage in the never ending game of leapfrog. Anyways, I'm pleased to see a mostly accurate review, even if I disagree with the commentary at times.

Re:Refreshing

[tomstdenis]

I didn't read the review [bah, ads] but the point isn't to be faster, it's to be better. Better often means things radically departed from simply faster.

For example, if AMD made a core, the same clock rate as their current mainstream core [say 2.4GHz], but take 40% less power at maximum, wouldn't that be better? It wouldn't render POV-Ray any faster but it would take less juice to do it. Aside from speed, size matters too. Smaller chips are cheaper to produce, reduces cost. Many design changes are for things customers don't even notice, like diagnostic support.

There is also the issue of the entire system. The CPUs are not the only power hungry thing in the typical box. The memory, south/north bridge, other controllers, storage, GFX, etc, all take power too. Especially north bridges [Intel camp in general] and GPUs take a lot of power.

So "leapfrog" doesn't always have to mean MIPS ratings. If the AMD core takes less power, but is marginally slower, that doesn't mean it's worse. As it stands now Intel is winning on both power and MIPS fronts. But they're competing against a 90nm process and are not winning by a landslide I might add. When AMD hits 65nm the energy scale may tip the other way again.

Tom

Re:Refreshing

[flaming-opus]

What gets to me is the way that most reviewers compare power usage by simply comparing the listed thermal envelopes. AMD lists the maximum power used, whereas intel lists the typical power used. Furthermore, for laptops and other machines where heat is your big concern, you do care a lot about the loaded maximum power used. However, for most desktops in which heat is not really an issue, you're more concerned about the cost of the electricity you burn. In that case it's almost more relevant to measure the idle power usage, since most desktops sit around doing nothing most of the time. Any good review should actually measure the system power between wall socket and PSU, otherwise it's not really infromative to the actual concerns of the user.

Re:Refreshing

[geekoid]

"(usually Intel)"
gah, talk about biased.

There skewed pretty evenly over all. gaame and 'geek' sites have a strong tendency to favor AMD. industry reviews have a tendency to favor Intel.

Re:Refreshing

[Anonymous Coward]

what year you stuck in? 2005? intel has been using maximum power used with the conroes.

Re:Refreshing

[0xABADC0DA]

Any good review should actually measure the system power between wall socket and PSU, otherwise it's not really informative to the actual concerns of the user. Well if you RTFA that's exactly what they did, but it's wrong. You don't measure the power consumption of a processor by comparing the total system draw unless the systems are otherwise the same (you can compare Intel to itself this way, not to AMD). The Core motherboard they tested is from Intel and has aluminum heatsinks, whereas the motherboard for AMD was by Asus and had heat pipes. Maybe they just put heat pipes on for look, but my bet is this MB adds quite a bit to the system power use.

The difference between most systems they tested was between 1 and 16 watts, so it's almost a certainty that the CPU was drowned out by the difference between motherboards. Judging by this benchmark it's a good bet that the AMD chips (with the exception of the quad ones) themselves draw significantly less power when idle than Core.

They could at least subtract out known power like drives, graphics card (use a low power one for doing the power tests), system fan, maybe processor fan, etc. Maybe turn the system on with no processor installed and use that as a baseline (if that doesn't destroy the system I don't know).

Re:Refreshing

[flaming-opus]

They did compare system power, thus the title "refreshing".
Correct, they did it wrong by not comparing identical systems. However, there is some merrit in this. I don't care if the power usage is from the processor, or from the northbridge needed to support said processor. Saying that a processor uses X amount of power makes for a good bullet point on marketing slides, but the actual user experience is with the whole platform, so I think it's appropriate to benchmark the power draw of the platform. Otherwise it's a purely accademic argument.

The only way in which the usage by the actual processor is interesting is when one is trying to estimate what a future product or related product will do. Since it's all speculation, though, it's not all that interesting of an exercise.

Re:Refreshing

[0xABADC0DA]

For instance I actually have a desktop system that draws 50 watts at idle, with two drives drives and a case fan (according to kill-a-watt...). If a different processor family takes 10 watts more you may call that academic, but I call it 20% more.

the actual user experience is with the whole platform, so I think it's appropriate to benchmark the power draw of the platform. That's exactly the point... virtually nobody is going to use their same configuration of 700W power supply, drive, memory, MB, CPU, video. So there is no user experience with these systems. It's almost entirely meaningless to give a power usage for them. And somebody building a low-power system still has no clue whether it's the Core processor that draws more, or the Intel motherboard that draws less, or the nvidia chipset that draws more, or how much the heatsink fan draws, etc.

Apparently you've never put together a low-power system, because you don't just try random combinations and see what the end result is... you look at individual components and find low-power combinations that work well together.

Re:Refreshing

[jonbryce]

In a desktop, I'm not that bothered about power consumption. In a laptop, I'm interested in battery life, so overall power usage is important.

Re:Refreshing

[Chandon Seldon]

Everything that you say is true, but remember one important thing: Comparing Athlon64 power consumption to Core 2 power consumption isn't an apples-to-apples comparison because of the memory controller. In order to get a useful comparison you need to compensate for that fact (the easiest way is to compare motherboard + processor consumption for both platforms).

Re:Refreshing

[Anonymous Coward]

Apples:Apples :: CPU:CPU
Apples:Apples :: MB:MB

If the memory controller is on the mobo then it's part of the mobo, and so that kind of CPU will tend to have mobos that draw more power than mobos for another kind of CPU. That will show up as a difference when you are comparing mobos. Here the component is the CPU, and that is what should be compared.

You don't compare "motherboard + processor consumption for both platforms" with two random motherboards... that's just stupid. You compare "min(motherboard)+min(processor)" for each platforms, or avg or max. I'm surprised that you /.ers are having such a hard time with this simple concept. The power consumption results that were reported are complete garbage. QED.

Re:Refreshing

[Chandon Seldon]

You don't compare "motherboard + processor consumption for both platforms" with two random motherboards... that's just stupid. You compare "min(motherboard)+min(processor)" for each platforms, or avg or max.

That's true, if the comparison is really going to be that sophisticated.

Here the component is the CPU, and that is what should be compared.

Only if you're working for the marketing department at Intel.

The fact of the matter is that the Athlon64 and Core 2 designs are different enough to make any direct comparison of power consumption misleading. A direct power consumption comparison is no more useful than directly comparing the TDP numbers that the two companies publish.

In order to publish an honest comparison of CPU power consumption, you would either need to compare MB + CPU platforms or publish the CPU power consumption numbers with a "The onboard memory controller on an Athlon64 consumes about X extra watts, on a Core 2 this power is instead consumed by the motherboard" disclaimer.

Correct link for second article

[Anonymous Coward]

Is this. Not this.

HTML is fine, but double check those URLs and HTML tags!

So...

[Southpaw018]

So I load this article up and a grand total of two comments have been posted. One says this test shows Intel beat AMD. The other says AMD beat Intel.

Beautiuful.

Re:So...

[LBArrettAnderson]

The "Conclusion" page... (for those of you who don't want to go through the 10 pages of pretty graphs and charts).

The fact that Intel retains the overall performance crown comes as no surprise. As we said at the outset, AMD has no real answer to the Core 2 Extreme X6800 among its dual-core processors. Also, Intel's quad-core CPUs tend to scale better than AMD's Quad FX platform, especially for typical desktop-class applications. Our move to Windows Vista x64 has done little to alter this dynamic. At the same time, Core 2 processors tend to draw less power and to be more energy efficient--sometimes markedly so--than Athlon 64s. Right now, Intel has the magic combination of a superior processor microarchitecture and a more mature, fully realized 65nm manufacturing capability working together on its side.
This one-two punch has allowed Intel to maintain a performance edge at most price points, despite standing pat through AMD's aggressive pricing moves and new model introductions. AMD's current weaknesses manifest themselves most fully in its high-end models, like the Athlon 64 X2 6000+, which draws more power at peak than the Core 2 Extreme QX6700 yet is often outperformed by the less expensive Core 2 Duo E6600. The Athlon 64 looks more competitive in its lower-end incarnations like the X2 5000+ and 4400+, which match up better on both performance and power characteristics against the Core 2 Duo E6300 and E6400. These processors have the benefit of being available in 65nm form, and I'd say the minor performance penalty one pays in performance at 65nm (due to the slower L2 cache) is worth it for the reduced power draw.

This low-to-mid-range territory, incidentally, is where I'd be looking to buy. Many of our tests have shown the benefits of quad-core processors, but honestly, finding applications that will make good use of four cores is not easy--and the list of games that really use four cores is approximately zero. I'd probably grab a Core 2 Duo E6400 and overclock it until it started to glow, if I were putting together a system right now. I must admit, though, that I have an almost irrational fondness for the Core 2 Quad Q6600, probably because it's the most energy efficient processor in our Cinebench power test. The thing is by no means a great deal--two E6600s will set you back over $200 less than a single Q6600--but it's easy to imagine a near-silent multitasking monster built around one.

AMD would do well to expand its 65nm offerings into higher clock frequencies as soon as it can reasonably do so. That may take a while yet, given the limited overclocking headroom we've seen from early 65nm Athlon 64 X2s. Meanwhile, Intel isn't likely to sit still for much longer. Rumors of an April price cut abound, and in light of the Core 2's ample frequency headroom, higher speed grades are a definite possibility, as well. For AMD, its next-generation microarchitecture can't come a moment too soon.

Summary - too blanket

[Visaris]

Intel > AMD at high end, Intel >= AMD at low end, Core 2 > A64, Intel finally has a lead in both architecture design and process (65nm).

I would agree with that as a generalization, but I still think it is very important for people to consider the applications they use most often. TR's benches clearly show that someone working primarily with POV-Ray would get better performance for $599 with AMD than for $999 with Intel. I agree that Intel takes the overall win, but blanket statements like this really fail to catch the areas where some chips shine and others do not.

Re:Summary - too blanket

[trimbo]

TR's benches clearly show that someone working primarily with POV-Ray would get better performance for $599 with AMD than for $999

Ok, let's be realistic here. Does anyone use POV ray for anything other than processor benchmarks? I have yet to see one real production, student or otherwise, rendered with POVRay. Let's see benchmarks with PRMan and Mental Ray. Those are production renderers people are actually using.

Re:Summary - too blanket

[ari_j]

I have been involved with the use of povray as part of a DNA visualization suite in a commercial context. That said, I agree that there is a very low probability of anyone who would buy a CPU based on its povray performance and nothing else.

Re:Summary - too blanket

[insignificant1]

My comment addresses yours, but wanders, so apologies in advance.

I use POVRay for explaining engineering concepts to co-workers, in papers for external viewing, and for use by the marketing folks. Word on the street is that its renders find their way into publications such as Nature and Science.

You talk of "production," which sounds like "movie," but it isn't a chump app just because movies don't use it for their render engine. It's a free, not-too-restricted-source-code app that yields stunning results. (Side-note: PRMan is ~$1k, and I couldn't find a price on Mental Ray, but my attitude towards Autodesk is that of the indentured slave to his master. But I digress.)

It is the single most time-consuming task I do, doing a render, and about everything else runs "fast enough" for me on my three-year-old mid-grade P4 system. Sometimes I wish for greater performance with an Octave/Matlab script when I am playing around. But that's hard to benchmark & compare on systems and only rarely does it take the same order of magnitude of time that POVRay can take.

So in response to another, I am one (and likely not the only) person who would lean towards a system based upon its POVRay performance. I have just been overjoyed that this has started to be used for benchmarks. I personally find the frames-per-second on Doom4 benchmarks useless, but what it comes down to is one thing:

The more apps people benchmark (accurately), the more people benefit and can make informed decisions that address their specific situations.

So I would like to see the apps you mention benchmarked, too. CompUSA never let me install photoshop and run my personal tests on it back when Photoshop was important to me. And now the greatest computer selection is from online retailers; how will you compare the value of a computer for you? $3000 for a Core2Duo Extreme Quad SSE2 from Dell, $2800 for an AMD Dual Quad HyperTransport blah, blah, from HP, and a bare-bones system from somebody else? What is each worth to you? And performance, obviously, gets far more complicated when you move beyond the processor isolated in a system. Is it worth the extra $100 for me to get another 1GB of memory? Will I ever really know how that will effect my apps, or does it just come down to whether or not I am willing to hand over another $100 just in case it might help?

I guess that is it: What are people's expectations for spending their money? Those who look at benchmarks might just find the "best" and drop their cash (or credit) on that one. Those who don't look at benchmarks might hit a price point, and just grab the best-looking system or find something from a specific brand. People like myself who want to optimize on a personally-important criteria are mostly left to guess and always be uncomfortable with any choice they make. And this situation might only change when spending the $1000 (or whatever) is worth caring a great deal about.

Re:Summary - too blanket

[trimbo]

You talk of "production," which sounds like "movie," but it isn't a chump app just because movies don't use it for their render engine

Well, movies, commercials, games, TV series, student pieces... anything using 3D.

I never said it was a chump app, just that it's doubtful anyone reading is going to make a hardware decision specifically to run POVRay. While it might be used to render a the occasional 3D picture for academic purposes (i.e. no budget but needs high quality), I find it highly unlikely it's being used on several thousand, tens of thousands or million dollar farms rendering thousands of frames a day. On that scale, it would be really useful to see direct comparisons using the software people do use.

As far as price goes, considering that we had to go through 11 ad-laden pages to see their results, I'd hope the people who wrote TFA could buy a couple licenses to do tests with.

Btw, mental ray is not owned by Autodesk.

Re:Summary - too blanket

[Talinom]

but blanket statements like this really fail

Uh, this is Slashdot. We are all about making blanket statements and then ripping them apart.

Welcome aboard!

Nice summary

[Anonymous Coward]

What kind of 'editing' is this?

"The Tech Report compares 15 Core 2 and Athlon 64 processors from Intel and AMD from sub-$200 to a cool grand, from slower dual cores to fast quad cores in 32 & 64-bit apps in Windows Vista, including the new, multithreaded RTS game Supreme Commander. 'The release of Windows Vista and a round of price cuts by AMD prompted us to hatch a devious plan involving Vista, a new test suite full of multithreaded and 64-bit applications, fifteen different CPU configurations, and countless hours of lab testing. That plan has come to fruition in the form of a broad-based comparison of the latest processors from AMD and Intel, ranging from well under $200 to a cool grand, from two slow CPU cores to four fast ones, from the lowly Athlon 64 X2 4400+ and Core 2 Duo E6300 to the astounding Athlon 64 FX-74 and Core 2 Extreme QX6700.' Folding@Home in Linux, power use, and energy efficiency are tested, too."

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:repeating repeating summary summary?

[antifoidulus]

They were trying to optimize the summary's performance by using both cores, but ended up with a race condition where both wrote. But it is finally a chance to break out the "selfcontaineddupe" tag!

True, but some still skew.

[Visaris]

Sure, performance is not the only factor. Intel does get some points for having a great combination of outstanding performance and very good thermal characteristics. Core2 is a great architecture, and I don't think anyone is trying to say otherwise. However, many people take Intel's general win and skew this into the claim that Intel and Core2 are the best for everything, which clearly is not true. The tactics used in the past by many reviewers have been to run overclocked Intel chips against stock AMD chips. This isn't exactly "cheating," but it ensures that Intel will be in the top stops on the charts. Also, many reviewers simply choose to skip the benches AMD is strong at, like Cinebench and POV-Ray. I'm not here to claim any one chip beats the hell out of the other, I just wish a lot of the fanboyism and Intel's reviewer payments would go away so we could get more reviews like TechReport's, which show many of the strong points and weaknesses of both sides.

i'll take processors for a cool grand, alex.

[Anonymous Coward]

what kinda crap summery is that anyway? even an illiterate would catch onto that trash fast.

David v. Goliath?

[ThePhilips]

For AMD, its next-generation microarchitecture can't come a moment too soon.

Nice reading.

But of course conclusions are not that surprising. AMD is 10+ times smaller than Intel (judged by capitalization). Intel has many fabs - while AMD is constantly struggling expanding its production capacities.

Yet, AMD (with Athlon 64) had managed to pull quite a match against Intel. Kudos to AMD: without you Intel's CPUs for sure would have costed $2500 a piece.

Re:David v. Goliath?

[Anonymous Coward]

Yet, AMD (with Athlon 64) had managed to pull quite a match against Intel. Kudos to AMD: without you Intel's CPUs for sure would have costed $2500 a piece. That's not really fair to the rest of the sector. VIA makes x86 chips that are usably fast and dirt cheap (and tiny and low power), and Transmeta might still be around doing similar stuff if the competition had been them, VIA, and Intel chips at $2500. Likewise, for many applications you don't actually need x86 chips, and IBM's PPC processors like the G5 would have been a lot more common, particularly in servers etc, and perhaps ARM chips in laptops. Windows might even have come out for those architectures meaning the home desktop could move there too. Or IBM or Toshiba could have moved into making x86 chips.

In short, there is a lot more competition in the chip space than you think.

Re:David v. Goliath?

[geekoid]

Don't be stupid.

AMD is just one factor in prices. CPU's weren't 2500 dollars a piece before AMD.

It was the AMD competition that lead to the MHz war; which is why we are just now getting multiple cpus/cores into the desktop at the consumer level.

While competition is good, and I am glad Intel has some, it doesn't make everything perfect.

Re:David v. Goliath?

[Anonymous Coward]

wow. he may or may not be stupid, but you're an asshole.

Re:David v. Goliath?

[ozbird]

Kudos to AMD: without you Intel's CPUs for sure would have costed $2500 a piece.

... and a choice between either a 32-bit Pentium 4 or an Itanium. AMD's greatest contribution was bringing 64-bit CPUs to the x86 masses - without abandoning 32-bit compatibility.

Re:David v. Goliath?

[keeboo]

64-bit CPUs to the x86 masses - without abandoning 32-bit compatibility

I find this 64-bit frankenstein x86 regrettable.
It was a golden incentive for burying x86 and AMD destroyed that (does someone really think Intel only though on 64bit for x86 after AMD released that?).
Microsoft, obviously, loved the idea of staying with x86.

Re:David v. Goliath?

[drsmithy]

Microsoft, obviously, loved the idea of staying with x86.

Why ?

Hidden away on page 14

[Andy_R]

"...the list of games that really use four cores is approximately zero."

That's the most interesting part of the article for me. Apart from 3-D rendering and folding@home, they are really pushed to find any real-world reason for having 4 cores.

Maybe they should have waited for Adobe's CS3 when heavy Photoshop tasks should provide nice real-world benchmark, and perhaps Apple will finally give us that long-awaited an 8-core Macintosh to put up against high-end Vista machines.

Re:Hidden away on page 14

[mdarksbane]

It's so you get smooth performance when you're running a video game, a music player, a web browser, and that coding assignment you left up to fool your boss when you have to alt tab, all at once.

And games should become more multithreaded with the way consoles are shaping up. Just because they don't take advantage of it now (mostly because 90% of people still have single core systems, and programming in multiple threads is hard) doesn't mean that newer games won't.

Re:Hidden away on page 14

[ryanvm]

they are really pushed to find any real-world reason for having 4 cores.

On the desktop sure, but there is no shortage of even mid-size server scenarios where 4 or even 8 cores come in handy.

Re:Hidden away on page 14

[drinkypoo]

That's the most interesting part of the article for me. Apart from 3-D rendering and folding@home, they are really pushed to find any real-world reason for having 4 cores.

This argument only holds up if you only do one thing at a time. Even on my Athlon XP 2500+ (obviously, a single-core system) I would regularly burn a CD (or a DVD, but only at 2x max) while playing a game. It would work out but the game would sometimes stutter and the burn would sometimes underrun; the underrun protection would work, but it does slow down the burn.

With four cores, you can play your game AND burn a disc AND have some crap going in the background and not have to care unless you become I/O-bound.

Benchmarks do one thing at a time, so they're a shitty measurement of real-world performance for power users whose brains can cope with multitasking.

Re:Hidden away on page 14

[Anonymous Coward]

Virtualization.

For the first time in many many years, I am tempted to buy a computer *much* faster than a 1ghz PIII because I want to run several different operating systems on my computer at the same time. For that, I want as many cores per CPU as I can.

Re:Hidden away on page 14

[hal2814]

"With four cores, you can play your game AND burn a disc AND have some crap going in the background and not have to care unless you become I/O-bound."

That's s pretty big "unless" considering becoming I/O bound is such a common occurrence. "Until" is probably a much better word to use.

Re:Hidden away on page 14

[mgblst]

It is so that you can read email, and listen to music at the same time (at least, according to the PC World adds, the leading UK pc seller). I wonder how people used their computers before core duos?

Re:Hidden away on page 14

[drinkypoo]

That's s pretty big "unless" considering becoming I/O bound is such a common occurrence. "Until" is probably a much better word to use.

It's a problem that is easily remedied, however, by the use of RAID controllers, modern buses, and so on.

The computer I last got rid of has an onboard RAID controller on a 64 bit internal PCI bus...

Re:Hidden away on page 14

[Hoi Polloi]

I wonder if this is cheaper for licensing too. If Oracle (for example) charged by the CPU then having multicore CPUs would be a great way to save money, until Oracle changed licensing of course.

Re:Hidden away on page 14

[geekoid]

You can do the with SCSI and a celeron.
for example, My SCSI drive was too small. So I decided to put in a SATA drive. That was the only difference. I can no longer play WoW, and iTunes, and down load images from my camera at the same time without serious stuttering.

My point is, even 4 cores writing to a crappy hard drive architecture won't make much of a different for people using more then on program at a time.

Re:Hidden away on page 14

[AmigaBen]

"...the list of games that really use four cores is approximately zero."

"That's the most interesting part of the article for me... they are really pushed to find any real-world reason for having 4 cores."

Excuse me? Games=real world?

Sorry, must have missed the memo.

Re:Hidden away on page 14

[rjstanford]

Oracle does charge by the socket, its one of their competitive advantages. And paying a single CPU license fee for a nice quad Xeon is pretty damn sweet. It also means that you can put together a "2 CPU" cluster using Oracle Standard Edition (max 4 CPUs) with 8 cores of powerful goodness and the ability to scale up 100% with another two boxes without hitting licensing issues.

And yes, a cluster of 2 boxes is often the smallest deployment you can consider from a failover standpoint.

Re:Hidden away on page 14

[jim3e8]

`make -j4` is a compelling real-world reason to have 4 cores. SMP is not limited to multithreading.

Re:Hidden away on page 14

[mandolin]

"...the list of games that really use four cores is approximately zero."

Do you play chess? Many chess engines will use as many cores as they can get (leaving out the fact that it could probably still beat you with just one).

Re:Hidden away on page 14

[DamnStupidElf]

That's the most interesting part of the article for me. Apart from 3-D rendering and folding@home, they are really pushed to find any real-world reason for having 4 cores.

Just wait for real time raytracing. From what I've seen of OpenRT, it only takes about 8 cores to get photorealistic real time raytracing for complex models entirely with software. Combining that with the advanced shaders on the GPU should make for some very visually stunning games with perfect reflections, refractions, shadows, and surface properties.

Re:Hidden away on page 14

[suggsjc]

And yes, a cluster of 2 boxes is often the smallest deployment you can consider from a failover standpoint.

Ok, you've really got my stumped here. I've been thinking about this for a while and I can't imagine a deployment smaller than 2 that would be able to survive a failover.

Re:Hidden away on page 14

[Icculus]

Sure you can do it with SCSI hardware, but for how much extra you spend on controllers and disks you might as well buy an entire separate machine to burn CDs.

Re:Hidden away on page 14

[Emetophobe]

With four cores, you can play your game AND burn a disc AND have some crap going in the background and not have to care unless you become I/O-bound.

You mean with two cores. I play games and backup dvds in the background all the time while playing mp3s at the same time on my lowly socket 939 Athlon64 X2 3800+. I can play ANY game at 1680x1050 with 4xAA and 8xAF, max quality settings and still burn a dvd in the background with no noticeable drop in frame rate. Dual cores have been able to do this for a while now, not just with quad cores like you mentioned.

Re:Hidden away on page 14

[drinkypoo]

Dual cores have been able to do this for a while now, not just with quad cores like you mentioned.

you missed the "AND have some crap going in the background" part. I am sitting at a dual core system right now (Core Duo T2600) and I am well aware of what it is and is not capable of doing.

Your antialiasing and filtering settings are completely and totally irrelevant as they are handled solely by the GPU.

Regardless, games ARE heading towards being more multithreaded. The Xbox 360 is going to provide a gigantic boost in this direction, as a SMP XNA platform. You HAVE to go multithreaded to get good performance out of the Xbox 360, and for the most part porting to Windows Vista should involve recompiling (as well as recoding any PPC-specific optimizations in x86 asm.)

This is going to filter down to several game engines used on the PC, of course, because they're also used on the 360.

And of course, the PS3 is REALLY going to get developers thinking in terms of parallelization, but it won't be much good as training for other systems. It will only put people in the right mindset.

Re:Hidden away on page 14

[Minwee]

Absolutely. If they were looking for real world applications, they should have started with porn.

Re:Hidden away on page 14

[Idbar]

"they are really pushed to find any real-world reason for having 4 cores."

Let's see, for an average PC user (Vista?):
1. AOL/AIM (That's worse than a virus, but used by many americans, that takes one core)
2. Music of course: Windows Media Player, another core gone (no further comments)
3. Browser: Try opening pages with heavy flash advertisement, like several newspapers these days.
4. Put here the real reason you are supposedly using the PC for, added to extensions of your desktop.

Re:Hidden away on page 14

[drsmithy]

This argument only holds up if you only do one thing at a time. Even on my Athlon XP 2500+ (obviously, a single-core system) I would regularly burn a CD (or a DVD, but only at 2x max) while playing a game. It would work out but the game would sometimes stutter and the burn would sometimes underrun; the underrun protection would work, but it does slow down the burn.

Something is wrong with your PC. Ca. mid-1997 I was burning CDs (at a blistering 4x) and playing Quake at the same time, on a 133Mhz Pentium with 40MB RAM running NT4.

Re:Hidden away on page 14

[pipingguy]

With four cores, you can play your game AND burn a disc AND have some crap going in the background and not have to care unless you become I/O-bound.

You mean like RAM management?

Re:Hidden away on page 14

[rjstanford]

Badly stated, perhaps - you can get away with one box if your failover requirements don't exist, which is more common than you might think.

Evil flash banners

[compwizrd]

I clearly need a dual or quad core system to handle the flash skyscraper banners on the sides.

Re:Evil flash banners

[qube99]

Nice and off topic but....what ads? Anyone who sees ads on that site is getting duped into thinking they need quad cores to get through the inter-tubes-web.

Re:Evil flash banners

[Anonymous Coward]

Sorry for going OT, but really, net-educated people have little or no reason to complain about ads in a world where Firefox and the Adblock extension is free.

I noticed absolutely zero ads on this site, simply because I had Adblock configured to kill all addresses [ad.*] and [ads.*].

Adblock is a sanity-saver, highly recommended.

https://addons.mozilla.org/firefox/10/

Makes no sense

[Anonymous Coward]

Why didn't they test performance with any FOSS games running on Linux? Some of those are pretty processor intensive. And popular, too.

Repeating Summary

[Anonymous Coward]

I wonder if they tested CPU's "ranging from well under $200 to a cool grand"...

DuggMirror????

[ThirdPrize]

Whoops, wrong site.

Bogus test benchmarks..

[FirstOne]

Most of these benchmarks are targeted towards unified caches.. (Intel)

Meanwhile real world apps favor separate caches per core.
(Where one user app isn't flushing cache entries of another app executing on different core.)

If they wanted to make it fair..
  They should execute n-copies of each benchmark compiled separately using different module names. (no unified cache sharing.)

Next item.. Graphics & games. What are they really measuring?
    The ability of some device driver writer to take advantage of some esoteric CPU optimization?

Last item they disabled Cool and Quiet on over clocked AMD configuration s it should have never been published.. I.E. They're simulating certain AMD configurations and aren't testing the real thing..

Re:Bogus test benchmarks..

[TheSunborn]

Hu? What kind of workload do you have, that have multiple tasks consuming much cpu time?

Re:Bogus test benchmarks..

[Anonymous Coward]

Poor AMD fanboi - did the article make you wet your pants?

Re:Bogus test benchmarks..

[Emetophobe]

Last item they disabled Cool and Quiet on over clocked AMD configuration s it should have never been published..

Pretty much any overclocker will tell you to disable Cool'n'Quiet if you plan on overclocking, mainly because C'n'Q can cause weird behaviours while attempting to overclock.

Cool'n'quiet down clocks your processor, the opposite of what you are trying to accomplish when you overclock. They don't really play very well together.

Re:Bogus test benchmarks..

[Anonymous Coward]

What's also bogus, is the power consumption test, testing only one specific load, instead e.g. a server profile (say a load curve from 40% to 90% over the day) and a desktop profile (say 85% idle).

Cheap testing, cheap results.

Manufacturing Process

[MobyDisk]

AMD is producing chips using a 90nm process and moving to 65nm, while Intel is moving from 65nm to 45nm. It is very difficult to compete in design when you are working with something 4 times less dense. AMD has always been behind in this area (except when they were using IBM fabs, and they had copper interconnects before Intel).

Simultaneously with this story, I see an announcement that Intel has announced another 45nm processor for ultra low power consumption.

Re:Manufacturing Process

[evilviper]

It is very difficult to compete in design when you are working with something 4 times less dense.

And yet, AMD is just barely behind Intel, in both performance, and power consumption. The only exception is the very high-end, where Intel has a quad core, and AMD needs two dual-core SMP CPUs.

In the article it's specifically mentioned that you can find certain AMD CPUs priced much cheaper than Intel counterparts. But the part that interests me much more is that AMD's low-end CPUs aren't crippled, where Intel removes SpeedStep on it's Core-based Celerons, driving power consumption way up.

And even more than that, Intel's Turions appear to own the low power consumption market, with better performance than anything Intel has to offer with 1/3rd lower power/heat requirements. That's what I'm look at getting right now, for a nearly-silent DVR system, and it's dirt-cheap.

I see an announcement that Intel has announced another 45nm processor for ultra low power consumption.

Except it's more rumor than announcement, and there's no info to be had on performance, power consumption, etc.

Re:Manufacturing Process

[MobyDisk]

And even more than that, Intel's Turions appear... You meant AMD's Turions. I have one in my HTPC and I love it. 98% silent, and it is the first system I've built that sits in an entertainment center without overheating.

Re:Manufacturing Process

[evilviper]

You meant AMD's Turions.

Whoops! Yes I did.

Re:Manufacturing Process

[zaibazu]

AMD used Silicon on Insulator in its fabs. You can't just compare the nanometers like you can't just compare the clockspeeds of processors. Remember when Intel introduced the desastrous Pentium4 Prescott ? It was IIRC their first mainstream 90nm processor, but performed worse on pretty much every front than the older Pentium 4 Northwood.

Um, problem

[sudden.zero]

I went to read the article and after page 2 I am getting a 404 error.

Re:Um, problem

[MrNaz]

The Slashdot department apologizes for any inconvenience.

Re:On a side note

[Zo0ok]

Sleeping with Apple?

Verdict: Core 2 Duo E6400

[digitaldc]

Summary: "I'd probably grab a Core 2 Duo E6400 and overclock it"

Save your money and buy the cheaper Core 2 duo. Then you can find out the Core 8 Octo will be released in a few weeks for about the same price.
Oblivion looks amazing IMHO.

Re:Verdict: Core 2 Duo E6400

[steve-san]

You got modded "Funny", but your comment holds some useful truth for anyone who might be in the market for an upgrade.
The C2D line gives you a chance to put together a fast machine on a reasonable budget with considerable room to grow.

For upgrade cycles, I can beat some of the folks who already replied -- my recent January purchase brought me into the 21st Century from a P3-800 (O/C'd to 1.1GHz, of course).

I went with the Core 2 Duo 4300, a 1.8GHz part.
- It's relatively cheap, at less than $170.
- It comes set at 9x200FSB, which means it'll overclock to the 300FSB world without super-fancy 1000+MHz RAM.
- It crosses the 3GHz mark on it's own stock cooler. With 3rd-party air cooling, I hit ~3.4Ghz by setting 9x375 (using DDR2-800 RAM).
- Almost any of the boards you put it in will be able to take a quad-core CPU; some of the better ones will hit much higher FSB's for later CPU's.

You can't ever future-proof a computer, but it's nice to be able to stretch out your investment a bit. If I can drop in a quad/octo/whatever in a few years, great!

POV-Ray

[muridae]

Single core, chess2 only takes me 347.58 seconds to render! I know that 'beta' makes it seem like a newer and better test, but look at the 'known issues' of the beta. They could have just stuck with 3.6 and one of the multi-thread patches.

Actually, I'm just upset that they didn't list what kinds of fps they got with POV-Ray's real time raytrace demos.

mac threading SUCKS

[Anonymous Coward]

"ong-awaited an 8-core Macintosh"

Why, so you can wait around longer for stuff to happen? Unless Apple does a major overhaul of their kernel (ie: get rid of that primitive microkernel), Windows (and Linux even more so) will beat OS X hands down in any multithreaded application benchmark. Sad for the zealots, but true nonetheless.

Re:mac threading SUCKS

[Anonymous Coward]

"Unless Apple does a major overhaul of their kernel (ie: get rid of that primitive microkernel), Windows (and Linux even more so) will beat OS X hands down in any multithreaded application benchmark."

Mac OS X doesn't use a microkernel.

Use of Mach doesn't automatically make it a microkernel. Think of Mach as a sort of operating system library, a chunk of code which provides a subset of the services usually associated with an OS kernel. You could flesh it out by adding the missing services in the form of 'server' processes in isolated address spaces using message passing to communicate -- the classic definition of a microkernel design. However, there's nothing preventing you from adding those services in the same address space as Mach with no message passing, just function calls. That's how MacOS X is structured.

As for your assertion that Windows and Linux must automatically win any multithreaded benchmark -- prove it. I wouldn't be surprised if Linux won most (it's by far the leanest and most efficient of the three kernels), but in the past there have been many benchmarks which purported to show how bad MacOS X was which were actually testing other things. For example, one benchmarker lambasted OS X for bad thread performance because the guy recompiled his Linux code and it went slow. Turned out that his code was highly sensitive to the choice of malloc() algorithm and pool size, and MacOS X had different default choices for both. (He had never paid attention to allocators on Linux, Linux just happened to have defaults which worked well for his code.) When others pointed this out and tested with the same malloc() and pool size on OS X, the performance gap disappeared.

The real performance problems in OS X aren't related to threading at all. Its threading support is fine. The major problem is that Mach's VM system is both extremely sophisticated and slow, and it's questionable whether the sophistication is a sufficient justification for the performance impact. Slow VM impacts things like context switching between processes, but has few implications for threads inside a single process.

x2 4400 low end???

[brennanw]

I don't even have a dual core chip. I guess that makes my computer non-existent... ...

Egads. I've been looking forward to getting a single-core 3800 -- that would be an upgrade for me.

Re:x2 4400 low end???

[corychristison]

I don't even have a dual core chip. I guess that makes my computer non-existent... ...

Egads. I've been looking forward to getting a single-core 3800 -- that would be an upgrade for me.

I hear, ya. I'm still running an AMD Athlon XP 2500+ (Barton Core). No overclocking, 1GB of cheap generic DDR Memory.
I can't play any games [ASUS TNT Geforce 128MB video card (I think?)], but I can still run WinXP with Photoshop CS2 (that's all I use it for) in Vmware on top of my Gentoo install like a dream.
To anyone looking to upgrade, wait until AMD launches its next line of chips! I am waiting for that day... *dreamy eyes*

Re:x2 4400 low end???

[MobyDisk]

I'm in the same boat. I am using a single core 3400+ because the new chips aren't really any faster - just more cores. So they won't really buy me much performance, especially considering the fact that I have to swap-out every component in my system again. When they have 4Ghz multi-core chips, and everyday apps run multi-core, then the upgrade will be worth it. For now, for most apps and games, a 5000+ and 3500+ are about the same.

Re:x2 4400 low end???

[Overzeetop]

Even worse...I "upgraded" to a 3.4 P4 in my soon-to-be HT box because it was as far as the MB could go. If I want a faster chip, it means I need a new MB. If I get a new MB, I'll have to change the main video card (AGP to PCI-E) and upgrade the memory, and - just to put a little salt on that potential wound - my MB is an OEM, which means if I swap it out I'll have to buy a new copy of Windows. Makes you wonder why they socket the chips at all instead of soldering them on.

Re:x2 4400 low end???

[Deagol]

No kidding!

Last spring I fried my MB, so I grabbed the Asus A8V w/ an Athlon 64 3200+ (2GHz, though I have it clocked to 2.2GHz). I was thinking about putting in a 2nd GB of memory and topping out the processor this spring (tax return time!), which for the A8V is (I think) the X2 3800. Sheesh. Looks like I can't get out of "low end" now w/o upgrading the MB, too.

What happened to the days of 3-year-old hardware being "low end"? :)

Re:x2 4400 low end???

[MobyDisk]

...just to put a little salt on that potential wound - my MB is an OEM, which means if I swap it out I'll have to buy a new copy of Windows. Depending on who the OEM is, there are ways around that. Personally, that would piss me off enough that I would go pretty far to make it work, on principle.

Re:x2 4400 low end???

[Overzeetop]

Dell, of course. The beauty is that any disc works on any dell bios. The trap is that you can never leave (cue evil laugh).

I don't have the time or energy to defeat it, and if I were really worked up I would probably just buy a new dell machine and upfit it - it would probably end up about the same price as a piecemeal. I have a business account and get good tech support, so for me there's usually a deal that makes it worth my while. I've got so many spare (proprietary) parts lying around the office I never really care about needing one.

Re:x2 4400 low end???

[steevc]

My 1200 Duron is still working fine, but I guess I would notice some difference if I went for even a low end dual core. It would not be worth upgrading my box, so it means a whole new PC. I know they are cheap these days, but I hate waste. It just seems ridiculous that people are buying dual core PCs just for everyday surfing/email/letters to granny. They will only really stress the PC if they play the latest games. I don't play many games, but have been getting addicted to Frozen Bubble lately.

Re:x2 4400 low end???

[Emetophobe]

I also found it kind of funny that they consider the X2 4400 low end, considering I have a X2 3800....Also, you can now get X2 3600's which are even lower end dual cores...

PS. I love my X2 3800+, it doesn't feel very "low end" to me.

Re:x2 4400 low end???

[adfour]

I wondered about that too. That 4400 would be the fastest box in my house. Of course I mostly just use openoffice, firefox, cedega and a few games (and cinelerra & kino for video, and audacity) on Ubuntu dapper desktops. Once in a while I compile something. Everything I do at home runs just fine on much lesser systems than those described. I'm happy to keep it that way, and happy not to get sucked into paying $$ to have the "most powerful" desktop available for ....what..the next 10 minutes? ('Til some other "must have" upgrade comes out). Something not mentioned: Wasn't INTEL doing some fairly nefarious DRM stuff recently? Planning to? Have they stopped, or do we just care about DRM from OEMs like Apple, or software vendors like MS?

Another data point about Xeons...

[rmdyer]

We've found after extensive testing that the Xeon line, the 5000 series, and specifically the 5150@2.66, are several percentage points slower than the Core 2 line of processors. The Core 2 Duo 2.66 is faster than the 5150 2.66 processor. So buying the Xeon processors apparently only gets you SMP capability for the higher price(?)

Re:Another data point about Xeons...

[rmdyer]

Note(s) to my previous post...

1. Most of the tests were conducted using 32 BIT Windows XP Pro SP2, some were using Red Hat Linux (also 32 BIT).

2. The machines tested were the Dell Precision Workstation line vs. the business Optiplex series.

3. Some of the software...

          ProEngineer Wildfire 2 and 3.
          Matlab
          Quake 4 Benchmarks.
          Spec tests for ProEngineer.

The Xeons may in fact run 64 BIT OSs and software better, but we have no data to show that.

Re:Another data point about Xeons...

[Anonymous Coward]

It's the memory system. Intel's chipsets for 51xx CPUs require use of FBDIMM memory rather than conventional DDR2. The FBDIMM channels offer much more bandwidth than DDR2 DIMM channels could in the same pin count, which needed to support 2 CPU sockets, but the flip side is that FBDIMM latency is higher than plain DDR2.

In some applications FBDIMM should actually be faster than plain DDR2 -- specifically, server applications which heavily load the memory channel with lots of transactions. It's much better at handling such heavy loads. However, for many other applications FBDIMM is a net performance penalty, as you've found.

BTW, 51xx CPUs paired with one of the variants of the 5000 chipset are quite sensitive to memory configuration, and systems aren't always delivered with the best performance setup. The 5000 chipset has four FBDIMM channels (*), and performs best when there are at least two 'ranks' per channel. 512MB FBDIMMs are typically single rank, 1GB or larger are typically dual rank. So a good baseline configuration is four 1GB FBDIMMs, one installed in each channel. You also want to try to stay at 1 DIMM per channel for absolute best performance; FBDIMM channels support daisy chaining a very large number of DIMMs but each DIMM in the chain causes access latency to increase. So if you need 8GB, try to use 4x2GB instead of 8x1GB.

(*) Most 5000 chipset systems are configured to bond the FBDIMM channels in pairs, so you end up with an effective 2 channels. You then install pairs of identical FBDIMMs to fill channels. All the same things apply re: best performance, you're just constrained to fewer possible configurations.

Be interesting to know

[hrieke]

What effect the L2 cache on the Intel chips have on the numbers.

Did you say AMD has the better real low end chips?

[Anonymous Coward]

Newsflash, AMD and Intel could care less about the chips that are years old selling for sub $100.

It is just not a relevant issue. Next you will want to talk about what a POS the P90 is and how Intel is the bestest in the world because no one can compete with the P2 166.

Also to the "BUT AMD was better at......" Yeah...... that's why he put the data in. As stated at the start this was a product line comparison and he stated the obvious overall winner.

Don't get your panties in a bunch AMD fanboys!

In 6mo I would expect us to be back to AMD having the best top line chip and another 6mo they will have the best line in an identical test (/sarc BUT INTEL WAS BETTER AT.....). Rinse/repeat.

Re:Folding@home just released PS3 client as well..

[KDR_11k]

Yep, "supercomputing" is what the Cell excells at. It may not be the best processor for all purposes but for that one purpose it's one of the best there is.

OT

[Anonymous Coward]

Your sig is Pi in roman numerals all bound together? Man that's tough to read...

Re:OT

[TeknoHog]

Your sig is Pi in roman numerals all bound together? Man that's tough to read...

Score: +V, informative!

In other news, Dr. Watson commented on the lack of fecal matter.

I culdn't agree more

[geekoid]

AMD is for people who prefer cheap over performance. You know, people who will be buying another AMD chip before they would have bought another intel chip because the AMD chip will fall behind the perfornace curve faster.

Re:I culdn't agree more

[Wdomburg]

Or for people who's performance needs are met by any currently marketed processor from either company. For typical desktop applications - internet applications, office suites, finance applications, basic photo-editing - even midrange processors are a waste.

The only reason I recently retired my primary desktop at home (an Athlon 800 built in 2000) is because a free Athlon 64 fell into my hands.

Re:I culdn't agree more

[botik32]

I do not understand this fixation with CPU speed. Isn't the bottleneck in memory speed and throughput nowadays? Why pay through the nose when the CPU is more than adequate for 99% of the users out there?

Bah, I am writing this on a 3-yr old machine and it is running fine, so I reasonably expect that a new low-end CPU would be just great for a new machine.

Re:Folding@home just released PS3 client as well..

[Slashcrap]

Despite the FAH PS3 client has been out under 24h the PS3 client performance is overtaking all the CPU/GPU FAH clients combined!

It's probably because the PC owners are wasting CPU time playing games.

Re:On a side note

[Targon]

Intel tends to push the full "Intel processor+Intel chipset" which also favors using Intel graphics. Pretty much every AMD based machine uses either an AMD/ATI chipset, or a NVIDIA chipset with the appropriate graphics. For Vista, the "experience" favors having a decent GPU, so AMD is the better platform there on the low end.

Every advertisement out there is about the low end of the dual-core, with a few mentions of the higher end products. So, you are looking at Athlon 64 X2 machines, or Core 2 Duo machines. Benchmark a low end Core 2, and you include the Intel video. It is NOT a good demonstration of Vista with a low end Core 2 with Intel graphics based machine.

good thing it's official?

[NRISecretAgent]

If you didn't know the results of this testing BEFORE now, raise your hand... That's what I thought, wonderful article.

Re:Folding@home just released PS3 client as well..

[schnipschnap]

Yeah, but much more TFLOPS/PU on the GPU row than on the PS3 row. But I think the Pony Slaystation 3 (sorry ponies) could theoretically get some more out of its hardware if it'd be using the GPU as well on the FAH client (correct me if it actually does). The fact that the client has been out for a short time only probably wouldn't make the measured performance better.

Re:Bogus test benchmarks.. Bogus analysis

[MerlynEmrys67]

Meanwhile real world apps favor separate caches per core. (Where one user app isn't flushing cache entries of another app executing on different core.)

Real world apps may not favor an integrated cache - but real world workloads do. Why do I want to give my simple little 500K application a 2MB cache to run in. Why not give it 500K and let the other 3.5MB go to the larger application that is also running and needing cache space. That said - yes I agree that in a case of an application needing 4 MB cache, and a second application needing 500K - the 500K application will run much better with a dedicated cache. However, I defy you to show me a system where a dedicated cache will perform better than a unified cache for a simple workload like this.

Now, that said you need a good caching strategy - there is a LOT of effort working on caching the correct memory to have it available quickly to the processor. I don't know where you get that one active process will flush the cache at the expense of another application that also needs a heavily used memory location... YMMV

Re:Did you say AMD has the better real low end chi

[Endo13]

Newsflash, even today the average new PC bought by a home user still does not have a CPU as good as any of the ones used in the benchmarks.

Whether AMD and Intel care about sub $100 CPUs I don't know. But considering that by far the majority of inexpensive new PCs being sold today have precisely this range of CPUs in them, I'd say it's very relevant.

Bad timing

[Sterling Christensen]

Not a good idea to do a price/performance comparison when prices and lineup are about to change.

Intel will be releasing a few new CPUs and cutting prices on April 22. The E6320 and E6420 for example, identical to their 6x00 counterparts except with 4mb of L2 cache. They'll go for $163 and $183 respectively.

Benchmarks for next month's processors with price list:
http://www.xbitlabs.com/articles/cpu/display/core2 duo-e6420.html

A 20-30% price cut is expected from AMD on April 9.

Even now the prices Techreport lists are outdated! The Athlon X2 4600+ dropped to $122 a week ago - faster and cheaper than the $170 4400+ techreport tested (which is actually more like $159).

Re:Bad timing

[penp]

Maybe I'm nitpicking, but why are you comparing the price of a processor made at 90nm (Windsor - AMD gives codenames to all of the cores of each cpu they produce) with one at 65nm (Brisbane)? There is more to a processor than just clocked speed. It's a lot harder to make a higher density chip, so a 65nm chip clocked at 4400+ is going to cost more than a 90nm one at 4600+.

As far as I can tell there isn't even a Brisbane 4600+, but for a drop in "200" mhz, you'll be consuming a less power to run the Brisbane 4400+. Less power means less heat, which means longer life at those clocked speeds.

I just wanted to point this out, as a lot of people seem to have the misconception that the only part of a processor that matters is the clock speed. There is a lot more that goes into it. The same concept was around back in the day when the Barton core processors were released. While clocked at the same speeds of the Thoroughbred core, the Barton core has more cache and a higher front-side bus speed, giving it a significant performance increase.

Re:Bad timing

[Sterling Christensen]

Maybe I'm nitpicking, but why are you comparing the price of a processor made at 90nm (Windsor - AMD gives codenames to all of the cores of each cpu they produce) with one at 65nm (Brisbane)? There is more to a processor than just clocked speed. It's a lot harder to make a higher density chip, so a 65nm chip clocked at 4400+ is going to cost more than a 90nm one at 4600+. Benchmarks show that aside from situation where Brisbane's slightly higher cache latency causes a negligible difference, they perform identically. 4400+ Brisbane and 4600+ Windsor have the same amount of L1/L2 cache, same almost everything because Brisbane is simply a 65nm shrink of Windsor. So performance wise it's still apples to apples - higher rating really does mean faster, at least in this case anyway.

As far as I can tell there isn't even a Brisbane 4600+, but for a drop in "200" mhz, you'll be consuming a less power to run the Brisbane 4400+. Less power means less heat, which means longer life at those clocked speeds. That's why I linked to an EE version of the 4600+. Same 65 watt thermal design power.

Yeah or you could just use an OS with a scheduler

[Dion]

Way back in the day when CDRs were expensive, way before burnproof I used my 166MHz Pentium to burn a CD while playing Quake at high enough resolution to make the audio stutter.

Without any getting a coaster.

I'll let you guess if I was running Linux or windows:)

CPUs and Supreme Commander

[Anonymous Coward]

I just bought SupCom because I was a huge Total Annihilation fan back in the day. As my current system is an AMD 2100+ with 768MB of RAM, SupCom does not have to do much to bring it to it's knees.
So, it's new system time for me which makes this review very timely. And I was looking at dual cores as my CPU. This review seems to confirm that, but the power consumption benchmarks make it appear that the E6400 is a better value than the E6300.
This also points out to me that building a decent system for under $1K will be *difficult* if I have to include the E6400 ($229 is lowest pricewatch price for the E6400 right now).
*sigh* What's a cheapass gamer like me to do? I'm also looking forward to Unreal Tournament 2007. Maybe I'm just screwed right now.

Real 64-bit performance

[ceeam]

I have read somewhere that when moving from 32-bit to 64-bit arch AMD CPUs gained about 15-20% whereas Core2's gained 0-5% (apparently Intel skips some hardware "optimizations" in x64 code). I wonder if you use a 64-bit machine for a (Unix of course) server is Opteron still a better deal that Core2/Xeon? I don't trust Microsoft's 64bit implementation to be seriously good of course. Any x64 server app benchmarks?

They bothered with this why?

[Gr8Apes]

Surprise surprise. a brand spanking new architecture on a smaller process beats 4 year old tech on a 3 year old process. I want to see Barcelona tested against Core 2.

Core 2 had damn well better be a large step up from A64. It's too bad that it's really limited to 2 physical CPUs at the moment. AMD still pwns the medium and large scale server market.

Yes, I own an AMD X2 3800+ that I recently bought and put together for a whole $200 out of pocket. C2 would have forced me to spend over $600 at that time. (I had RAM for the X2, C2 would have also required a RAM purchase) The X2 handles what I need just fine, and I can't argue with the results.

I also own a C2 laptop. It's great and I use it more than my desktop.

What would be the best processor?

[subitophoto]

To do the following task? Resize image in batch, rotate them...(thousand of them). Is a processor company better than the other for these specific task? Claude

Re:What would be the best processor?

[Chandon Seldon]

Unless the images are really massive, a couple thousand pictures wouldn't take that long to crunch through on any modern processor.

If they *are* really massive, you're going to be bottlenecked on memory bandwidth which is somewhat better on AMD processors.

Windows Media Video codec is 4-way threaded

[benwaggoner]

The current Windows Media Video 9/ 9 Advanced Profile codec is 4-way threaded.

To get the current version, you need to install any of:

Windows Media Player 11
Windows Media Format SDK 11
Windows Vista

your sig

[Anonymous Coward]

ACs are modded -6. I don't read you, I don't mod you, I don't see you. Don't like it? Don't be a coward.

Go fuck yourself you piece of shit.

You could tell Intel was gaining ascendency...

[Anonymous Coward]

when Dell started using AMD and Apple switched to Intel.

chip prices before AMD

[pbhj]

Sorry, had to check this one out.

AMD established in 1969.

Intel's first microprocessor chip (4004, see http://www.intel.com/museum/online/hist_micro/hof/ index.htm) was released in 1971 and cost less than $100. They established in 1968 however.

There are some suggestions that the R8000 was the most expensive microprocessor: "As the price list indicates a Power Indigo with a R8000 CPU and Extreme graphics was priced at $61,500 with 128mb ram and a 2gb system disk (1993-94 pricing). " (from http://www.blackcube.org/sgi.html ). In '92 an R4000 based system was around $40k. So, it seems reasonable that the R8000 chip cost (retail) about $20k more than the R4000.

Granted this wasn't "before AMD" but was probably before the competition between AMD and Intel made such a difference to the home PC market. Nevertheless I get geek-cred points here surely!!!???!!11

Re:Yeah or you could just use an OS with a schedul

[leenks]

Way back in the day (1996-7) when CDRs were still expensive, before burnproof I used my 200MHz Pentium to burn a CD while playing Quake at a sensible resolution, but still to the point of using up all the CPU (well, you are doing that essentially unless it hit the FPS cap easily anyway).

Without getting a coaster.

I used Windows (NT).