AMD Athlon 64 FX-74 & Quad FX Platform Review

[mzs_biteme]

How is that FSB and massive cache coming along?...

Still dominating last time I checked....

 

[FlintMurdock]

This doesn't have to be an Intel vs. AMD thread. I always think of of it as AMD vs. AMD and the only boundaries being pushed are AMD's boundaries.

 

[Martyr]

This doesn't have to be an Intel vs. AMD thread. I always think of of it as AMD vs. AMD and the only boundaries being pushed are AMD's boundaries.

you forget one of the primary rules of the internet. have forum, will fight.

 

[FlintMurdock]

Well, we don't have to follow tradition..I mean we can have a lottery and pick a winner and stone some poor devil, but it's not going to change anything

 

[MsB]

Has anyone seen a fx70 review? It is a FX and should be unlocked right? If so, and it can make it to 2.8ghz -3.2 ghz than its a damn bargain and all the bandwidth will come in handy and will show much different result's in performance when running Vista pig OS and DX10 Games. Power hog yes but I'd like to how this baby could run a packed 50 man server while playin at highest resolutions with a 8800gtx compared to a kingsfield doing same at a lanparty.

 

[caboosemoose]

Here's the conclusion from another site, which sums up Quad FX's weaknesses and basic irrelevance quite nicely:

"AMD is going to have a very tough sell with Quad FX; although the CPUs are priced competitively, if the ASUS L1N64-SLI WS ends up just shy of the $400 mark it's a platform that is simply too expensive at no benefit to the end user. When only running one or two CPU intensive threads, Quad FX ends up being slower than an identically clocked dual core system, and when running more threads it's no faster than Intel's Core 2 Extreme QX6700. But it's more expensive than the alternatives and consumes as much power as both, combined.

There is the upgrade path argument, that eventually you will be able to put a total of eight cores in this Quad FX platform, but we can't help but wonder if the market for someone who wants a non-workstation 8-core setup for desktop use is a very small one. Although to AMD's credit we were able to create a scenario where even four cores won't cut it, making a case for the need for 8-core setups in the future. But the promise of eight cores in the future doesn't do a great job of justifying the Quad FX purchase today.

For those users who won't migrate to eight cores, once AMD's new micro-architecture debuts next year with native quad-core support, this expensive Quad FX platform will be notably slower than cheaper single socket systems. Quad FX is simply a very niche product, and in the era of power efficiency and performance per watt, AMD has released the proverbial SUV of high end desktops.

AMD hopes to sell more Quad FX processors than any FX processor in the past, which to us means that either AMD sees much more opportunity in this platform than we do, or that the previous FX processors simply didn't sell very well. Either way you slice it, there's only one AMD CPU we're really interested in and we won't get it until the middle of next year. Luckily for AMD, Intel doesn't appear to be doing anything huge between now and then either, so it looks like the CPU wars will cool down for a while after a heated few months.

Prepare to revisit this discussion in less than a year's time, and next time AMD will hopefully be much better prepared, armed with a new architecture and a cooler, smaller 65nm process. Until then, there's always Quad FX but you're better off with Kentsfield. "

 

[FrgMstr]

If you are going to rip folks content, at have the decency to link them please.

 

[osalcido]

umm yeah I'll bet that article is copyrighted too and you need to give them credit

 

[zachatfield]

One thing the review didn't make clear was the fate of consumers after AM2 - [H] says it'll be staying for another year at least in current machines, but what after that? Will we simply start seeing single Socket 1207 machines or will there be another "budget" socket?

I think what is going to happen here is this. AMD is using this utterly ridiculous socket, Socket 1207 to house its QuadFX processors because its the only one they saw fit to do it in such short notice; I smell marketing. The new "Barcelona" quad core from AMD is going to be their answer. The core is a native design meaning all four cores are on the same die as opposed to Intel's double Dual Core stack design. AM2 sockets will be filling with Barcelona's in the near future. I think AMD is fronting this "enthusiast" chipset and Quad FX marketing to bring attention to their company. Who the hell needs 20 usb ports, 12 sata ports for their 12 750GB hard drives, and their four GeForce 8800's; maybe BIll Gates. Damn!

My point here is that AMD is never going to be out of this race. Native quad core K8L's already tells us that they are thinking ahead and producing innovative products. Intel doesn't have that! Look at all of this discussion we're having over this. Look at how ridiculous an idea it is. The five fans on the mobo two dual core processors just to get four cores, and extreme power consumption that would make the EPA piss their pants. I bet you they don't make even tens of thousands of these boards or processors. They will continue to dominate the low end market and when "Barcelona" comes out it may just be a whole different ball game when all of you socket AM2ers are clickin over to newegg to get that huge upgrade you always wanted in the form of FOUR cores.
Merry Christmas and Chanukah peoples.

 

[emXry]

. . .

 

[Mega2]

I think AMD should start making power plants bulit in. like zero point energy module. 45nm of Perpetual motion growing energy farms in there cpus silicon.

why is ontario power so expensive? $280+/month

 

[EvilNando]

Hi Kyle.... i just wanna say that I cant imagine any1 willing to buy that 4x4 crap

performs slower its bulkier runs hotter its more expensive and in the near future a new (and hopefully better) solution will came that integrates 4 cores on a single chip

so yeah the FX-74 is a big POS , and please take note im a big AMD !!!!!!

and I dont need help when I need to go poo but if you want to help ok fine

 

[CopyThat]

Well, we don't have to follow tradition..I mean we can have a lottery and pick a winner and stone some poor devil, but it's not going to change anything

Please, Can I be stoned!!??

 

[jon67]

After RTFA I couldn't see any associations with Hummer H2 so much... another 4x4 classic may be more appropriate:

 

[sxotty]

Drabbit! DRABBIT! I've wanted to make my next box an AMD box, but I can't see where I'll get benefit from the AMD platform right now. The next box (to be pieced together over the next few months) is for gaming and general use. No encoding here. I've been building Intel boxes pretty much all my life and wanted to go with AMD this time. Oh, well, I'll have to look at them again in the next year to two years. I'm happy they've finally come out with something. I'm not so happy that what they have isn't the best value for me.

Since you were building Intel boxes when they were far inferior isn't it time to switch and build clearly inferior AMD boxes? After all that would make perfect sense from a symmetry standpoint.

 

[Taldren]

Kyle,
I respect your opinion, but looking at even speeds is one thing, but power consumption easily tilts the tables to the Intel side. Even if the AMD chip is cheaper, after factoring in the PSU cost and the future electric bill cost, Intel is the clear winner. I would ask myself, If someone were to ask me to build them a quad-core system, there is no doubt that I would choose Intel this time. Or am I missing something??

edit: I would not need a stop watch, just feeling the heat would be enough.

Um, yeah ... when you have to resort to talking about the power bill you have reached phanboy status.

It's 90nm, of course it is going to run hot and not be very overclocker friendly ... I consider it a proof of concept (a successful one) and will wait for the die shrink.

 

[wizzackr]

I can't wait when Barcelona comes out and I revive this thread for kicks. (...) Bitch and moan all you like about these 3 year old architecture cores (that are still neck and neck with Intel's latest) that suck up double the power.(...) I never seen so many people get into a tissy over a temporary processor solution that sucks up a little more juice compared to Intel's THIRD RESPONSE to K8 that came out only two months ago . I got the money to blow on this setup you think I give a rats ass about power consumption, especially since I only need it for 5 months until Barcelona.

While I am with on most parts, I still think it shouldn't be forgotten that you can buy a two socket/8 core setup from Intel for under 3800 as we speak.
So while Barcelona might kick some serious ass in the future, 4x4 is reviewed today - and if it is your thing to go massivley multi-core with the most power available ASAP, there seems little sense in shelling out that much cash NOW for a 4x4 platform /w two dual-core FXs, and then again some in 5 months when Barcelona is released.

Anyways, I'll definitely keep an eye open for 4x8 (HT3?) in summer '07.

 

[FlintMurdock]

After RTFA I couldn't see any associations with Hummer H2 so much... another 4x4 classic may be more appropriate:

LOL dude that is kinda cool. You're right though, its like lipstick on a corspe.

 

[67shelby]

Since I had to be reminded of performance issues on every other page due to the KillerNIC ad, I was just wondering if you tested this card on the 4x4 platform. Maybe it could have improved your gaming framerates by 41% which would been enough to defeat Kentsfield

 

[jstnomega]

Then again...AMD are not stoopid. Must somewhat pain the engineering staff to see such numbers/comparisons & commentary at this time.

So why'd they do it anyway, knowing full well ahead of time how it was gonna reflect on them all?

In order to be able to point back at this thermal envelope fiasco when the nexgen K8L, or whatever it will be called, comes to market? How better to demonstrate the improvements they will have made by then & make everyone ooh & ahh? I mean, if AMD hadn't done this, how would we ever fully appreciate nexgen architecture & die shrink improvements? Compared to what?

Just a thought re, "Whatwuztheythinking!?!"

 

[LstBrunnenG]

Even then, this thing will only be adopted if the costs that go along with it - Motherboard, PSU, Memory, ETC - start going down. Rumors say $400 for the motherboard. That's a bit of a deal-breaker IMO.

 

[deeznuts]

Um, yeah ... when you have to resort to talking about the power bill you have reached phanboy status.

It's 90nm, of course it is going to run hot and not be very overclocker friendly ... I consider it a proof of concept (a successful one) and will wait for the die shrink.

Ok, so you call someone out for being a phanboy for mentioning power bill, which I might add AMD fans held over Intel fans, (of which I am neither, I've gone Athlon, northwood, xp, X2, Opty 165, now X6800 and QX6700; I'll defend Intel, I'll buy whatever I want, Intel doesn't pay my bills), for the duration of prescott. Who invented the term Press-Hot? And he didn't just mention the power bill, it's the combo of better performance, lower cost and power bill. How did you not see that?

So it's 90mm, of course it's going to run hot and not be overclocker friendly? Weren't various Athlon64 and X2's 90mm? I dont' remember them being called hot and not overclock friendly. Hell I bought a 3000+ and an X2 3800, and Opteron 165 because they were cooler and overclock friendly.

Now whether you consider this a successful proof of concept, that's up to you. Doesn't seem more to me then in essence a server board with an oc bios and the 4 PCIe slots. Don't think Intel cant grab one of their server platforms and perform the same type of surgery, and have 8 cores now instead of next year? They probably are thinking, "what the fuck for?" And I agree.

And I have an opinion. People are complaining about intel's practice of slapping two die on one package and calling it quadcore. I dont' agree with the criticism, but I don't deny it's validity. Performance differences between the two doesn't seem significant. But isn't this the ultimate Frankenstein stitching? Instead of two die on one package, it's two packages on one motherboard.

 

[Donnie27]

http://techreport.com/reviews/2006q4/quad-fx/index.x?pg=1

AMD's Quad FX platform
AMD decides to socket to 'em
by Scott Wasson — November 30, 2006

IN THE PC REALM, when you can't win by traditional means, there may be another reliable avenue available to you: move upmarket. This form of one-upsmanship has been masking technological shortcomings in increasing measure in recent years. Intel arguably started this trend in the CPU market when, on the eve of AMD's introduction of the Athlon 64, it uncorked the first Pentium Extreme Edition processor, basically a Xeon with scads of L3 cache pulled from the server market into service as a new flagship desktop part. At the prohibitive price of just one dollar short of a grand, the Extreme Edition wasn't intended to sell at high volumes. Its job was simply to defend the performance crown to the best of its prodigious ability. That's the beauty of the ultra-high-end product: a top product can rock the benchmarks yet only ship in a few hundreds or thousands of units.

Come on folks, see it for what it is? Unfortunately, both failed at defending the Crown LOL!

Doesn't matter if it is slightly slower, uses twice as much power, is expensive, gives off more heat, needs a larger Power supply, needs twice as much RAM, needs a Newer OS like Vista UE, and Higher electric bill, it's only money, right? The best thing about it is the Promise of something better in more like 8 months, not 5. It's AMD though so it's all good

 

[jvrobert]

http://techreport.com/reviews/2006q4/quad-fx/index.x?pg=1



Come on folks, see it for what it is? Unfortunately, both failed at defending the Crown LOL!

Doesn't matter if it is slightly slower, uses twice as much power, is expensive, gives off more heat, needs a larger Power supply, needs twice as much RAM, needs a Newer OS like Vista UE, and Higher electric bill, it's only money, right? The best thing about it is the Promise of something better in more like 8 months, not 5. It's AMD though so it's all good

Yeah, this thread is hilarious. People defending what is obviously a rushed and poorly thought out response to C2D and QXD.

Even if everything is bright and rosy and all works well and K8L _does_ come out in 6 months and also works with existing 4x4 MB's, why would you buy one now? Seems like a silly risk only a superAMDroid would take.

You could buy an E6600 and get comparable performance in most tasks and _better_ performance in others RIGHT NOW. Then, in RosyWorld if all works out well you could rebuild a new"4x4" machine next year and probably save money versus buying one of these overpriced junkheaps now.

What's most amusing is the shameless difference in tone between this review and the C2D review. C2D has utterly changed the performance PC landscape in the last 4 months, and according to the illustrious reviewer it was basically ho-hum and offered little "real-world" advantage.

Now, we have this monstrous piece of 4x4 crap (and it is, no matter what its potential future could possibly be in X months if so and so happens and everything works out perfectly) and it basically gets an "it's OK" review.

Now, if you read both of these reviews you'd probably think that both pieced of technology are similar in importance. Take this in conjunction with established fact, and I think it's fairly easy to draw some conclusions that a lot of people on these boards have been bringing up for some time about who pays the bills and how things work around here.

 

[chrisf6969]

Yeah, this thread is hilarious. People defending what is obviously a rushed and poorly thought out response to C2D and QXD.

<snip>

What's most amusing is the shameless difference in tone between this review and the C2D review. C2D has utterly changed the performance PC landscape in the last 4 months, and according to the illustrious reviewer it was basically ho-hum and offered little "real-world" advantage.

Now, we have this monstrous piece of 4x4 crap (and it is, no matter what its potential future could possibly be in X months if so and so happens and everything works out perfectly) and it basically gets an "it's OK" review.

Yeah, I was thinking the same thing!! Look at how many people have jumped ship from AMD to Intel for Core 2 b/c it simply rocks at everything, yet the [H] gave it a fairly lackluster review. Saying basically "Ah its about the same as an X2 in realworld, b/c videocards bottleneck everything!" (paraphrased obviously)

But then giving AMD kudos for the 4x4 monster "Hummer H2".

 

[Cannydog]

I want one!! It looks like it should make an Extraordinary Space Heater.

 

[PunjabiPlaya]

Ahmdahl's law comes to mind...
Anyways, here's some more fuel for the fire that is the insanity of the enthusiast market:
http://babelfish.altavista.com/babe...-la-ricetta-amd-per-i-quattro-core_index.html

Nice to see some Vista/64 bit benchies rather than whining, cussing, and waving e-peens and Hummers around.

/straightjacket

oh and - NUMA NUMA NUMA
Anyone else think AMD should be playing that song?

edit: Space Heater FTW, it's freaking cold outside

 

[theelectic]

Ahmdahl's law comes to mind...
Anyways, here's some more fuel for the fire that is the insanity of the enthusiast market:
http://babelfish.altavista.com/babe...-la-ricetta-amd-per-i-quattro-core_index.html

Nice to see some Vista/64 bit benchies rather than whining, cussing, and waving e-peens and Hummers around.

My italian is rusty, but it looks to me the FX-74 only won two benches and was fairly close in the rest.

 

[BigMacAttack]

I want one!! It looks like it should make an Extraordinary Space Heater.

Save your money and got to Wal-Mart for a cheaper one!

 

[Donnie27]

Yeah, this thread is hilarious. People defending what is obviously a rushed and poorly thought out response to C2D and QXD.

Even if everything is bright and rosy and all works well and K8L _does_ come out in 6 months and also works with existing 4x4 MB's, why would you buy one now? Seems like a silly risk only a superAMDroid would take.

You could buy an E6600 and get comparable performance in most tasks and _better_ performance in others RIGHT NOW. Then, in RosyWorld if all works out well you could rebuild a new"4x4" machine next year and probably save money versus buying one of these overpriced junkheaps now.

What's most amusing is the shameless difference in tone between this review and the C2D review. C2D has utterly changed the performance PC landscape in the last 4 months, and according to the illustrious reviewer it was basically ho-hum and offered little "real-world" advantage.

Now, we have this monstrous piece of 4x4 crap (and it is, no matter what its potential future could possibly be in X months if so and so happens and everything works out perfectly) and it basically gets an "it's OK" review.

Now, if you read both of these reviews you'd probably think that both pieced of technology are similar in importance. Take this in conjunction with established fact, and I think it's fairly easy to draw some conclusions that a lot of people on these boards have been bringing up for some time about who pays the bills and how things work around here.

Yup, you got just what I was saying It is crap and mzs biteme was correct as well. How do you just talk about two times 95 = 190 being 50 more watts when you have all of the associated circuitry around the proc nicely brushed aside?

Why they acted surprised Q6700 did so well against $X$?

http://www.hardforum.com/showthread.php?p=1030139907#post1030139907

Laid the smackdown on what? 4x4 hasnt been released yet, and as far as I know nobody has seen benches yet. So what did the Woodcrest lay the smackdown on, and how does this have anything to do with 4x4? Who has the advantage in video subsystems?
I'm only asking becouse your not being quite clear. Please help me understand.?

Two Socket Dual Core Opteron vs. Two Socket Dual Core Woody have been done already. Do you think AMD made improvements architectural wise to FX-7x? The DP and Two FSBs do work well together at 1333MHz.

Hell, I underestimated how large the PSU would need to be.

 

[speedyman]

Dear Mr. Bennett,

Well, this thread is certainly getting hot.

I laughed as I read these posts yesterday, and dismissed them as the rants of a small group of computer geeks. But the posts keep coming... obviously this 4x4 review has touched a nerve. Or has it? I no longer believe it's the review that is at issue here. Please read on...

I saw this mention of the [H] review here,

which links to the blog here.

Relevant quote: "Ironically, three of the four benchmark sites I link to give such contradictory glowing conclusions for the Quad FX in spite of their own data showing AMD being slaughter that Baghdad Bob would be proud. Here are the four reviews of which only TomsHardware had a realistic conclusion that matched their actual data."

One of the four sites listed was [H] Enthusiast.

Forget about the actual review. Now your reputation seems to be at risk here. But I can help you.

Mr. Bennett, regardless of your opinions regarding the AMD 4x4, it seems that you have become the focus of a growing controversy. Is all this attention good for your web site? Maybe not, if it leaves the readers with a nagging sense of doubt about the veracity of the reporting they find here, or worse, a feeling of being betrayed by a trusted news source. And now others in your profession see an opening to publicly ridicule. The reference to Baghdad Bob, one of the most notorious liars in the history of warfare, is especially cruel. And then the final cut: Tom's Hardware is the only source to be trusted for this type of news. Ouch.

What to do when even self-admitted geeks are snickering at you, and the snarling, snapping packs of opportunistic dogs called your "peers" are circling, looking for a quick meal? What to do, indeed!

Things are getting ugly fast. And you are like the guy at a local bar who, after feeling slighted, calls the offender out for an exchange of ideas in the back alley. Only when you turn around, he has brought five friends with him to join in the discussion. You are not going to change any of their opinions at this point.

It's time for self-preservation.

If you were to retain a public relations firm to guide you through this crisis of public confidence, they may well advise you:

(1) Immediately admit you were wrong, even if you don't mean it. You are being perceived as wrong and bullheaded. Continued obstinance on your part will only further degrade your reputation among the readers. Let them think they have changed your mind. Let them rejoice. You can stick it to them later, maybe with some sort of fee-based web access system. Or tell them to buy the most expensive Intel CPU out there. It's not your money. Pandering works for the majority.

(2) Promise to update the review with any new insight you may now have. This will give the appearance of fairness, while also giving you an opportunity to publicly defend your conclusions in perhaps a more subtle way.

(3) Promise to do whatever you can to ensure this doesn't happen again. Give the impression that you have learned from this experience; the majority will always be inclined to forgive. P.T. Barnum was right.

Also, (4) Cease directing any childish taunts at your readers. These are your customers, not your enemies, even though that seems in doubt sometimes.

In order to regain the trust of your readers, many of whom feel passionate enough about this to post comments here, you must now display a professional composure during an embarrassing period. At the risk of sounding Machiavellian, the above advice seems to work for just about everybody, including sociopathic career politicians who are genetically incapable of feeling any true remorse.

Give it a try, and the same people who laugh at you now will hoist you upon their shoulders and cheer you.

Just be sure to check for your wallet when they set you back down.

Regards,
speedyman

 

[BoogerBomb]

Its surprsing that so many intel zealots are bragging that a cpu that is 4 generations ahead of the amd tested here was just BARELY better. What was tested here was a platforum against a cpu. I saw a margin of a few points between most test with the exception of gaming. That is total and utter domination? I will say however that i would like to see ths same platform tested with dual quad core amds pitted against dual qud core intels. Does intel have plans for dual quad core? Then we can give the true test of intels architecture as it wil be new tech against new tech instead of ancient tech against new tech.

What is helpin intel right is pure manufaturing capabilities. They can do 65 which amd cannot and when amd can intel will be doing 45. Amd needs to just start building the shit out of fabs or make the current one more upgradable friendly. A die shrink or die size difference usually favors the smaller one as is the case here.

QuadFX wasnt meant for the older tech and amd shouldnt have brought it out until the cpu for which is was made was able to be tested.

 

[InorganicMatter]

Something to consider:

Going by the [H]'s number on power draw, a full 4x4 system draws on average 250W more than a full Kentsfield (assuming 50% idle and 50% load on average). 0.25kW times 24 hours times 365 days times 15 cents a kW (the current rate in my area), you come out to $320. A 4x4 system costs $320 per year more than an Intel system does per year in power.

Suddenly, power usage isn't a "phanboy" arguing point any more, is it?

 

[mzs_biteme]

In one of the links form speedyman...:

WTF! Are people at ZDNet blogs actually reading reviews??? Looks like ZD is overrun by some hard-core AMD-r-o-i-d-s...

 

[deeznuts]

Its surprsing that so many intel zealots are bragging that a cpu that is 4 generations ahead of the amd tested here was just BARELY better.

Lol ahahahahahah. 4 generations? Ok, please, tell me the 3 generations that preceeded Conroe after the Athlon64 generation.

Netburst was introduced in 2000. Athlon 64 in 2003. Were you defending Intel against the Athlon 64 saying Intel's architecture was 3 years old?

Of course surely you don't mean you are counting Presler, Smithfield, prescott, etc. Because then you have Orleans, San Diego, Venice Winchester etc. So please explain your 4 generations comment.

What was tested here was a platforum against a cpu.

Yeah and that platform lost. You could throw a Kentsfield in any platform and it would have still won.

I saw a margin of a few points between most test with the exception of gaming. That is total and utter domination?

yes when one costs way less, uses a ton less power, is only one CPU compared to TWO cpus (people keep missing this point) is available now (paper launch anyone?) ties it in most tests while beating it handily in 1/3 tests here (even more elsewhere) oh yeah and runs 300 mhz less. that's total and utter domination buddy.

I will say however that i would like to see ths same platform tested with dual quad core amds pitted against dual qud core intels.

Yeah well first AMD doesn't have quad cores just yet, and Intel doesn't have dual core desktop boards. It's a valid wish though.

Does intel have plans for dual quad core?

They probably think the same way I do ... what the fuck for? Usually computers that need 8 cores DO NOT need 4 PCIe. Usually.

Then we can give the true test of intels architecture as it wil be new tech against new tech instead of ancient tech against new tech.

Again, throw out all Athlon 64 comparisons to Pentiums then. C'mon throw them out!

As a matter of fact, with architectures so staggered between AMD and Intel, comparisons should never be made again! No point in comparing what's available today (I say that loosely, 4x4 not available until January?) since they are never concurrent generations! Shut down all the site, you are not relevant.

QuadFX wasnt meant for the older tech and amd shouldnt have brought it out until the cpu for which is was made was able to be tested.

Hey I wholeheartedly agree with you there that AMD shouldn't have brought it out. But they did, so it is.

 

[deeznuts]

In one of the links form speedyman...:

WTF! Are people at ZDNet blogs actually reading reviews??? Looks like ZD is overrun by some hard-core AMD-r-o-i-d-s...

That is your typical demographic bro, the masses are confused!

Unless they don't track voters, so they can push it a few hundred times like what my ex-gf was complaining about cuz some kid voted 300 times for Ruben whatever his name was over Clay Akin.

 

[BoogerBomb]

I like how power consumption is now for some reason the biggest consideration when buying a cpu. Back in the day not many people based their decision on power consumption and that aspect was rarely reviewed.

The folowing is from: http://www.itjungle.com/breaking/bn113006-story02.html

With the top-end Athlon 64 FX-74, 4 GB of memory, all 12 disk drives, and four video cards, McNaughton says that such a machine could keep all of its parts whirring using an 850 watt power supply and still have some room left over. In tests at AMD's labs--which did not stress all the video cards and drives to their maximum--the highest power that the Quad FX platform drew was around 740 watts.

So who is more correct? Kyle measurements with one card un maxxed or AMD's estimate with 4 cards running?

 

[SLee]

I like how power consumption is now for some reason the biggest consideration when buying a cpu. Back in the day not many people based their decision on power consumption and that aspect was rarely reviewed.

Of course back in those days, the typical computer wasn't capable of doubling the average power consumption of a home.

With the top-end Athlon 64 FX-74, 4 GB of memory, all 12 disk drives, and four video cards, McNaughton says that such a machine could keep all of its parts whirring using an 850 watt power supply and still have some room left over. In tests at AMD's labs--which did not stress all the video cards and drives to their maximum--the highest power that the Quad FX platform drew was around 740 watts.

So who is more correct? Kyle measurements with one card un maxxed or AMD's estimate with 4 cards running?

Too bad they didn't provide more details.

What kind of video cards? could be 4X 7300GS which use less power combined than a single high-end video card.
What kind of CPU test?

 

[Taldren]

Ok, so you call someone out for being a phanboy for mentioning power bill, which I might add AMD fans held over Intel fans, (of which I am neither, I've gone Athlon, northwood, xp, X2, Opty 165, now X6800 and QX6700; I'll defend Intel, I'll buy whatever I want, Intel doesn't pay my bills), for the duration of prescott. Who invented the term Press-Hot? And he didn't just mention the power bill, it's the combo of better performance, lower cost and power bill. How did you not see that?

When you can name every PC you ever had and who coined what term ... never mind ... anyway ... didn't mean to start a geek war here. Just pointing out the absurdity of talking about power consumption when on the subject of enthusiast hardware. That is like talking about fuel efficiency of dragsters.

So it's 90mm, of course it's going to run hot and not be overclocker friendly? Weren't various Athlon64 and X2's 90mm? I dont' remember them being called hot and not overclock friendly. Hell I bought a 3000+ and an X2 3800, and Opteron 165 because they were cooler and overclock friendly.

Um yeah, all CPU's cores were at one time very overclockable at a certain die size. Then a time comes when you can't really push them much further due to high voltage/heat requirements.

Now whether you consider this a successful proof of concept, that's up to you. Doesn't seem more to me then in essence a server board with an oc bios and the 4 PCIe slots. Don't think Intel cant grab one of their server platforms and perform the same type of surgery, and have 8 cores now instead of next year? They probably are thinking, "what the fuck for?" And I agree.

Proofs of concept don't really require polish ... They are PROOFS OF CONCEPT. As for Intel doing the same thing ... don't think it will work as well without something like AMDs Hypertransport technology.

I don't hold any loyalty to any one company ... because ... they certainly don't hold any loyalty to me. However, I have to agree with Kyle that I have to see a noticeable difference in performance ... noticeable by any of the 5 senses god gave me (although I don't think I will taste a CPU any time soon) ... before I will fork over any amount of money to upgrade needlessly.

As for the argument "Who is better" ... I accept that Intel is currently 'better' for 32-bit apps while AMD is currently the clear winner for 64-bit apps. However the difference in 32-bit performance isn't going to make me throw my current 64X2 machine away for something I can't see, hear, or feel.

What I am more interested in is if Vista 64-bit will make in impact on this ongoing feud between these chip makers.

 

[Donnie27]

When you can name every PC you ever had and who coined what term ... never mind ... anyway ... didn't mean to start a geek war here. Just pointing out the absurdity of talking about power consumption when on the subject of enthusiast hardware. That is like talking about fuel efficiency of dragsters.


Um yeah, all CPU's cores were at one time very overclockable at a certain die size. Then a time comes when you can't really push them much further due to high voltage/heat requirements.


Proofs of concept don't really require polish ... They are PROOFS OF CONCEPT. As for Intel doing the same thing ... don't think it will work as well without something like AMDs Hypertransport technology.

I don't hold any loyalty to any one company ... because ... they certainly don't hold any loyalty to me. However, I have to agree with Kyle that I have to see a noticeable difference in performance ... noticeable by any of the 5 senses god gave me (although I don't think I will taste a CPU any time soon) ... before I will fork over any amount of money to upgrade needlessly.

As for the argument "Who is better" ... I accept that Intel is currently 'better' for 32-bit apps while AMD is currently the clear winner for 64-bit apps. However the difference in 32-bit performance isn't going to make me throw my current 64X2 machine away for something I can't see, hear, or feel.

What I am more interested in is if Vista 64-bit will make in impact on this ongoing feud between these chip makers.

Where'd you read AMD runs 64bit apps faster?