AMD 4X4 review

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

atom

Diamond Member
Oct 18, 1999
4,722
0
0
4x4 was actually looking ok till you got to the power consumption part. That really killed it IMO. If the power draw wasn't so insane it would be a decent workstation solution but as it stands now why the hell would anyone buy it.

After reading the article, I kinda wonder back about my old Abit BP6 system. Back then people really didn't look at power draw like they do now, I wonder how much power that system sucked up compared to single proc systems at that time.
 

beggerking

Golden Member
Jan 15, 2006
1,703
0
0
being a AMD fan ever since amd386, I have to admit this is a pretty pathetic attempt to regain market position. AMD should spend all its effert into developing the next gen CPU, not this....
 

tylerw13

Senior member
Aug 9, 2006
220
0
0
Originally posted by: beggerking
being a AMD fan ever since amd386, I have to admit this is a pretty pathetic attempt to regain market position. AMD should spend all its effert into developing the next gen CPU, not this....

thank you i would have to agree with you
 

beggerking

Golden Member
Jan 15, 2006
1,703
0
0
Originally posted by: GuitarDaddy
Wow! Deja vu

This is just as bad as when Intel came up with the Pentium D's to compete with X2's

exactly! I laughed at pentium D when it came out, thinking "how sad they had to glue 2 cpus together and call it a dualcore, high heat, power , etc etc ... "..

then this.. they don't even bother to glue the 2 cpus..
 

lopri

Elite Member
Jul 27, 2002
13,211
597
126
Why block Opterons via BIOS? Aren't Opterons even more expensive?
 
Oct 4, 2004
10,521
6
81
Originally posted by: GuitarDaddy
Wow! Deja vu

This is just as bad as when Intel came up with the Pentium D's to compete with X2's

Pentium D (Smithfield) came out a month or so before the X2s IIRC.

May 2005 vs June 2005
 

Furen

Golden Member
Oct 21, 2004
1,567
0
0
Originally posted by: beggerking
Originally posted by: GuitarDaddy
Wow! Deja vu

This is just as bad as when Intel came up with the Pentium D's to compete with X2's

exactly! I laughed at pentium D when it came out, thinking "how sad they had to glue 2 cpus together and call it a dualcore, high heat, power , etc etc ... "..

then this.. they don't even bother to glue the 2 cpus..

They would have glued the two CPUs if it was possible... It's physically impossible to fit two dual-core K8 dies under the same IHS because of their shape (they're pretty much square) and the fact that all the resistors and capacitors also go on the same side. Intel's CPUs are a bit longer and slimmer, so you can fit two of them in a square shape. They also have space on the "pin" side for capacitors and miscellaneous components.

Originally posted by: beggerking
being a AMD fan ever since amd386, I have to admit this is a pretty pathetic attempt to regain market position. AMD should spend all its effert into developing the next gen CPU, not this....

The sad thing is that it didn't have to be this way. AMD could have gone for more modest clock speeds (a 2.6GHz 65W part, if possible, or just 95W 2.8GHz parts), a more modest chipset (something that didn't NEED THREE FREAKING FANS to work) and we wouldn't have complained at all.

Two Opteron 2.4s (dual-core 95W) in a server (with all those server fans) idle at 200W, load at 265W. Going 50/100W over this would still have been acceptable, but 200W/300W more for idle or load is just sick. Hell, a Quad Opteron 8216 server (2.6GHz) with 16GB of RAM idles and loads at 520/657. 4x4 is obviously a badly engineered product, but this is only party AMD's fault. I mean, we can assume that the CPUs shouldn't use more than 200W at IDLE, so does that mean that the motherboard burns 200W just sitting there? I wonder just how much of that full-load power draw is also burned off by the motherboard...

Another thing to look at: An FX 62 system uses around 110W at idle and 210W loaded (from Anandtech's FX62 review)... you could almost load two whole FX-62 systems on the amount of power the Quad FX idles on.
 

peternelson

Junior Member
Mar 8, 2006
15
0
0

Few people buying a high end system like this would intend to run the 5 year old 32 bit windows XP as its operating system. XP32 doesn't support NUMA.

Newer OS like XP Pro x64, Server 2003 SP1 and Vista, not to mention Linux, DO support NUMA.

It is unsurprising that Anandtech found higher memory latency when using both cpus.

The bios setting "enable node interleave" is set for XP, whereas for NUMA aware OS like Vista it should be set to "disable".

Therefore I take performance benchmarks with scepticism until they are done on NUMA aware OS for which this is designed.


I'm pleased that Anandtech review answered positively my question "can I run this board with only one cpu". Even if AMD only sell them in pairs, I could build two systems and put one in each. Also enterprising IT resellers are likely to be willing to split packs for people wanting only one, or to upgrade. I can thus buy and use something now/Q1, and buy two of the lowerpower 65nm quadcores when they come out next year.

Aside from the power consumption, this is fine architecture with plenty I/O expansion capabilities. It is on the roadmap towards 8 interconnected sockets of quadcores which is the future AMD design.

Dissing QuadFX on the basis of 32 bit XP reviews is ridiculous.

 

Furen

Golden Member
Oct 21, 2004
1,567
0
0

Accord99

Platinum Member
Jul 2, 2001
2,259
172
106
Originally posted by: Furen
Originally posted by: ted
to know about 4x4 performance using windows vista, please follow this link
http://babelfish.altavista.com/babelfis...etta-amd-per-i-quattro-core_index.html

Wow, being NUMA aware really does make a difference. The same can be said about 64bits (the bottom of the megatasking Vista page). Doesn't make up for the massive power draw but still shows a much more positive picture on the performance side.
I think the gains have more to do with running in 64-bit mode than NUMA. Most of the gains come in 3D rendering which should show benefits to 64-bit optimization. The video rendering tests don't differ very much (other than DVD Shrink for the Quad FX). And the megatasking test shows no benefits with Vista.
 

IdaGno

Senior member
Sep 2, 2004
452
0
0
Originally posted by: Furen
Originally posted by: ted
to know about 4x4 performance using windows vista, please follow this link
http://babelfish.altavista.com/babelfis...etta-amd-per-i-quattro-core_index.html

Wow, being NUMA aware really does make a difference. The same can be said about 64bits (the bottom of the megatasking Vista page). Doesn't make up for the massive power draw but still shows a much more positive picture on the performance side.

That may be true in Vista, but what about in a real OS, such as Linux?

 

Viditor

Diamond Member
Oct 25, 1999
3,290
0
0
Originally posted by: IdaGno
Originally posted by: Furen
Originally posted by: ted
to know about 4x4 performance using windows vista, please follow this link
http://babelfish.altavista.com/babelfis...etta-amd-per-i-quattro-core_index.html

Wow, being NUMA aware really does make a difference. The same can be said about 64bits (the bottom of the megatasking Vista page). Doesn't make up for the massive power draw but still shows a much more positive picture on the performance side.

That may be true in Vista, but what about in a real OS, such as Linux?

Linux is also NUMA aware I believe...
 

Furen

Golden Member
Oct 21, 2004
1,567
0
0
Originally posted by: Accord99
Originally posted by: Furen
Originally posted by: ted
to know about 4x4 performance using windows vista, please follow this link
http://babelfish.altavista.com/babelfis...etta-amd-per-i-quattro-core_index.html

Wow, being NUMA aware really does make a difference. The same can be said about 64bits (the bottom of the megatasking Vista page). Doesn't make up for the massive power draw but still shows a much more positive picture on the performance side.
I think the gains have more to do with running in 64-bit mode than NUMA. Most of the gains come in 3D rendering which should show benefits to 64-bit optimization. The video rendering tests don't differ very much (other than DVD Shrink for the Quad FX). And the megatasking test shows no benefits with Vista.

Oh, you're probably right, to some extent. But, like you said, DVD shrink is quite an exception. TR's results are also quite close, and they are using a NUMA-aware 64-bit OS as well, though not Vista.
 

rchiu

Diamond Member
Jun 8, 2002
3,846
0
0
Man, my three main workstations have AMD processors in them, but I am about to lose all my respect for AMD with this launch. This new platform pretty much does what any tech savvy people like us can do with two Opterons years ago, with less flexibility and higher costs. The only reason I can see AMD doing this is because they cannot push their processor speed higher nor come up with a new processor, and the only way they can slap a FX74 or some big PR rating to compete with Conroe is to sell 2 cpu as one.

This is really pathetic.
 

DrMrLordX

Lifer
Apr 27, 2000
21,805
11,161
136
I don't know why that link to hwupgrade.com is suppose to show the FX-74 in a more positive light under Vista RC2. The QX6700 still wins most benches in that benchmark. It would be nice to see some other 64-bit OS environments tested, but . . .
 

beggerking

Golden Member
Jan 15, 2006
1,703
0
0
Originally posted by: Furen


The sad thing is that it didn't have to be this way. AMD could have gone for more modest clock speeds (a 2.6GHz 65W part, if possible, or just 95W 2.8GHz parts), a more modest chipset (something that didn't NEED THREE FREAKING FANS to work) and we wouldn't have complained at all.

Two Opteron 2.4s (dual-core 95W) in a server (with all those server fans) idle at 200W, load at 265W. Going 50/100W over this would still have been acceptable, but 200W/300W more for idle or load is just sick. Hell, a Quad Opteron 8216 server (2.6GHz) with 16GB of RAM idles and loads at 520/657. 4x4 is obviously a badly engineered product, but this is only party AMD's fault. I mean, we can assume that the CPUs shouldn't use more than 200W at IDLE, so does that mean that the motherboard burns 200W just sitting there? I wonder just how much of that full-load power draw is also burned off by the motherboard...

Another thing to look at: An FX 62 system uses around 110W at idle and 210W loaded (from Anandtech's FX62 review)... you could almost load two whole FX-62 systems on the amount of power the Quad FX idles on.

exactly.. the power requirement and heat is just plain sick.. what is wrong w. their engineers?.. where were those who thought up the original athlon 64s ?
 

coldpower27

Golden Member
Jul 18, 2004
1,677
0
76
Originally posted by: beggerking
Originally posted by: Furen


The sad thing is that it didn't have to be this way. AMD could have gone for more modest clock speeds (a 2.6GHz 65W part, if possible, or just 95W 2.8GHz parts), a more modest chipset (something that didn't NEED THREE FREAKING FANS to work) and we wouldn't have complained at all.

Two Opteron 2.4s (dual-core 95W) in a server (with all those server fans) idle at 200W, load at 265W. Going 50/100W over this would still have been acceptable, but 200W/300W more for idle or load is just sick. Hell, a Quad Opteron 8216 server (2.6GHz) with 16GB of RAM idles and loads at 520/657. 4x4 is obviously a badly engineered product, but this is only party AMD's fault. I mean, we can assume that the CPUs shouldn't use more than 200W at IDLE, so does that mean that the motherboard burns 200W just sitting there? I wonder just how much of that full-load power draw is also burned off by the motherboard...

Another thing to look at: An FX 62 system uses around 110W at idle and 210W loaded (from Anandtech's FX62 review)... you could almost load two whole FX-62 systems on the amount of power the Quad FX idles on.

exactly.. the power requirement and heat is just plain sick.. what is wrong w. their engineers?.. where were those who thought up the original athlon 64s ?

Considering they need to up the voltage to 1.475V AMD's seems to be still getting the same levels of TDP as before on their parts, and can't really reach 3.0GHZ reliably on voltages similar to the rest of the lineup.

AMD could have released a 95W 2.6GHZ part or a 68W 2.4GHZ, both would have lost to the Kentsfield, in terms of performance but the power consumption at load would have been much more what's the word reasonable.

And anyway, now that Anandtech has figured out how to work Cool'n'Quiet on the 4x4 platform, the idle temperatures are similar to what Kentsfield gets when idling. Though to be fair, AMD's do idle at 1.0GHZ while the Intel's idle at 1.6GHZ.
 

coldpower27

Golden Member
Jul 18, 2004
1,677
0
76
Originally posted by: Mr Vain
AMD coming back with Real Quad core CPU?s on this and AM2 platform!

2 quad cores on this platform should be good.

Non believers will eat humble pie.

http://www.tgdaily.com/2006/12/01/amd_aims_to_reclaim_cpu_crown/



That doesn't matter currently as were interested in performance I can use today, not future performance that isn't here where I can't use it yet.

Yes you can plop down 2 Quad Core processors down the road, but that is something else entirely as by that time Intel will likely have some other alternatives as well.

There is no humble pie to eat, 8 Cores on the desktop even by Mid 2007 is even a more limited niche then a single quad core is now.

"Real" Quad Cores CPU's, nice wording lol, you mean AMD's native Quad Cores, there is nothing real about it, Intel current Quad Core CPU's are real, the only definition required for having a Quad Core CPU is that all 4 Cores are on the same socket. That is all. So currently the case is Intel has shipping Quad Core processors AMD doesn't period.

AMD is using a Quad Core system right now to compete with Intel Quad Core processor, suffice it to say, the solution requires alot more energy to operate. Since you have 2 Dual Cores on 1 motherboard with each Dual Core on it's own Socket.

Currently the solution is not very enticing, it's is likely to become more enticing down the road, but for the moment we don't want it.
 

Mr Vain

Senior member
May 15, 2006
708
1
81
Good things come to those that wait.

I'm waiting; Intel has nothing new till end of 2007.

I will not buy a quad core for sometime, not many apps to take advantage of it for me and the prices are way too high.
When I?m ready for Quad cores, then I?ll make an informed decision not based on the here and now but rather on performance, future roadmaps, upgrade paths etc.

I hope the 4x4 platform is more streamlined/efficient by the time AMD introduces the Agena cores as it promises a more future proof path IMO.

http://img65.imageshack.us/my.php?image=amdbarcelonaquadcoremm4.jpg

AMD chief sales & marketing officer Henri Richard compares AMD's "Barcelona" Opteron quad-core (left) to Intel's dual-die Clovertown (Xeon 5300) quad-core CPU.









 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |