AMD 4X4 review

[Makaveli]

1st review is up guys over at Hardocp

http://enthusiast.hardocp.com/article.html?art=MTIzMywxLCxoZW50aHVzaWFzdA==

 

[Markfw]

Looks like the FX-74 holds its own OK by this review.

 

[Arkaign]

Pathetic is the only word I can summon.

Barely holds it's own in a few benchmarks, get owned in others.

Sits in a ridiculously priced mobo, while a C2Q can rock in a P965 at $130ish, and just wait till 650i! Don't get me started on price of the processor packs. Unless you bug the slug FX70 set, you're going to burn your wallet out on these underpowered procs.

Sucks down power like Michael Moore sucks down big macs ..

AMD has officially brought something out worthy of the massive suckage of their old K5, or perhaps as worthless as the Cyrix MediaGX.

 

[Accord99]

http://www.xbitlabs.com/articles/cpu/display/amd-quad-fx.html

 

[Stoneburner]

Performance wise pretty much expected.... amd's architecture is showing its strength.

 

[harpoon84]

AMD's Prescott - that's all I gotta say...

Btw, I like the Xbit review, they have the QX6700, Q6600, FX-74, FX-72 and FX-70 all benched. Top stuff!

 

[Accord99]

Originally posted by: Stoneburner
Performance wise pretty much expected.... amd's architecture is showing its strength.

And disadvantages. It's memory controller per socket gives it a lot of bandwidth but also increases latency. The FX-62 is able to beat the FX-74 in a lot of single-threaded or latency dependent benchmarks.

 

[MDme]

it seems that hardocp's and xbitlab's benchmarks don't show consistent results with one another. granted hardocp did not have a since am2 system to test with, but the slow scores of certain bechmarks which was blamed on NUMA did not seem to happen on the hardocp benchies.

oh well, 4x4 wasn't really gonna beat kentsfield, the real question will be answered when K8L/K10 launches. it will be further interesting how 8x4 will turn out.

 

[anandtechrocks]

Originally posted by: Accord99
http://www.xbitlabs.com/articles/cpu/display/amd-quad-fx.html

Originally posted by: XbitLabs

Our test platform used two Athlon 64 FX-74 processors with 3.0GHz nominal clock speed. Unfortunately, we couldn?t increase this frequency any further. If the clock generator frequency was set even 1MHz above the default 200MHz, the system wouldn?t boot Windows XP shutting down automatically. It was pretty strange also because nothing could be done to resolve this situation: no common tricks used to improve the stability of the overclocked system, such as increasing the processor voltage, bus voltage or chipset voltage or lowering the HyperTransport bus multiplier, would help. We could free some room for the clock generator frequency increase by reducing the processor clock frequency multiplier below the nominal value, but only until the CPU frequency hit 3.0GHz. Once the frequency got higher, the situation repeated: the system would simply shut down on us.

Everything we have just said suggests that AMD, Nvidia or ASUS introduced some kind of protection into their system that prevents the CPU frequency from being increased above the nominal. Of course, it is really hard to believe especially since AMD has never prevented overclockers from having their fun, but it is an undeniable fact. It is even harder to believe that the CPU is working at the utmost of its potential in the nominal mode already.

Ouch... not a single MHz...

 

[harpoon84]

Originally posted by: anandtechrocks

Originally posted by: Accord99
http://www.xbitlabs.com/articles/cpu/display/amd-quad-fx.html

Originally posted by: XbitLabs

Originally Posted by XbitLabs
Our test platform used two Athlon 64 FX-74 processors with 3.0GHz nominal clock speed. Unfortunately, we couldn?t increase this frequency any further. If the clock generator frequency was set even 1MHz above the default 200MHz, the system wouldn?t boot Windows XP shutting down automatically. It was pretty strange also because nothing could be done to resolve this situation: no common tricks used to improve the stability of the overclocked system, such as increasing the processor voltage, bus voltage or chipset voltage or lowering the HyperTransport bus multiplier, would help. We could free some room for the clock generator frequency increase by reducing the processor clock frequency multiplier below the nominal value, but only until the CPU frequency hit 3.0GHz. Once the frequency got higher, the situation repeated: the system would simply shut down on us.

Everything we have just said suggests that AMD, Nvidia or ASUS introduced some kind of protection into their system that prevents the CPU frequency from being increased above the nominal. Of course, it is really hard to believe especially since AMD has never prevented overclockers from having their fun, but it is an undeniable fact. It is even harder to believe that the CPU is working at the utmost of its potential in the nominal mode already.

Ouch...

LOL! So much for 4x4 being an 'enthusiast platform' - an overclocking lock? Please, say it ain't so!

Hopefully, it's just a BIOS issue that can be fixed in future, otherwise, what hope does 4x4 have against C2Q? Most Kentsfields are overclocking to around 3.5GHz, at that speed it would absolutely trounce a 3GHz FX-74.

 

[swtethan]

500watts full load!

 

[theprodigalrebel]

The Extremetech review had two FX-74s, 2 hard drives and a 7800GTX - they drew 595W with 4 instances of P95 & an instance of 3DMark.
Add 8800GTX SLI and let's see this baby pump some heat!

They gave it a whopping 5/10.

 

[DrMrLordX]

Those power consumption numbers are pathetic. Not only is Kentsfield overall cheaper, it beats the 4x4 rig on several tests and uses half the power (or less)? Give me a break.

We all know 4x4 would be pretty niche, but uh . . . who really wants a pig like that?

 

[AnandThenMan]

Flush goes AMD. Pretty sad really, let's all hope AMD pulls up their boot straps and gets back in the game. 4x4 is a power sucking pig. IF it was 40% faster (in AMD's dreams) then anything Intel had, the MAYBE I could excuse the power draw. As it stands, the system is stillborn.

 

[Lonyo]

I almost see an early FX thing all over again.
Anyone remember 940pin FX processors? They came, and then they went. And now anyone who bought them has nowhere to go but to buy a new system.

 

[Brunnis]

Crap. Competitive performance, yes, but who in their right mind would want something like this when Kentsfield is around? This is one stupid system and I wish AMD could have skipped it and been able to put "K8L" on the market this spring. It would have been so much better.

 

[DrMrLordX]

Somehow I doubt that 4x4 is delaying K8L's debut.

 

[theprodigalrebel]

One of the reviews mentioned something about Vista better harnessing the power of the NUMA architecture. Assuming that is true, and come January Quad FX starts approaching the price/performance territory of Core 2 Quad (let's toss aside performance-per-watt for now) and becomes a viable enthusiast platform:

Even if they resolve the BIOS bugs/locks and make overclocking on this possible - the system was already idling at 45 degrees celsius (damn, I'm getting reviews mixed up now - can't remember where I read this) with the stock HSFs.

Won't installing TWO enthusiast-grade HSFs be more than a little problematic given the size/layout constraints? And is there even any headroom in the architecture?

 

[DrMrLordX]

No clue if there's much OC headroom on those things. Likely not, at least not at 90nm anyway. 3.2 ghz might be doable but anything beyond that is sketchy unless you're got some very nice cooling.

Kentsfield, on the other hand, is good for 3.0-3.2 ghz.

 

[Accord99]

Hothardware got to 3.215GHz at 1.55 or 1.6v:

http://www.hothardware.com/viewarticle.aspx?page=4&articleid=911&cid=1

Tom's Hardware could only hit 3.125GHz:
http://64.233.179.104/translate_c?hl=en...fx74_quadcore_vierkern_cpu/page21.html

They also couldn't fit the Zalman CNPS9700 on it, only the 9500.
http://64.233.179.104/translate_c?hl=en..._fx74_quadcore_vierkern_cpu/page8.html

 

[Viditor]

Don't any of the reviews actually USE the 4 QSLI slots??

 

[rudder]

Originally posted by: Markfw900
Looks like the FX-74 holds its own OK by this review.

...at over TWICE the power consumption.

 

[shamgar03]

I've always been a AMD fanboy...but also a realist. I am not going to pay for an inferior product. Hopefully this whole 4x4 thing will pan out for AMD because I liked to see the competition they brought to the table. Unfortunately Intel is kicking their butts right now. I am still holding off upgrading my processor in hopes that AMD will come out with a product that wow's everyone

 

[Centurin]

Originally posted by: shamgar03
I've always been a AMD fanboy...but also a realist. I am not going to pay for an inferior product. Hopefully this whole 4x4 thing will pan out for AMD because I liked to see the competition they brought to the table. Unfortunately Intel is kicking their butts right now. I am still holding off upgrading my processor in hopes that AMD will come out with a product that wow's everyone

If it makes you feel any better, the Quad FX platform wow's me. Just not in a good way.

 

[LittleNemoNES]

Power Consumption is Insane.
There's no way in hell I'd consider this much less recommend it to some poor soul.
65nm > 90nm