Error
  • JUser::_load: Unable to load user with id: 67

Featured Articles

IHS teardown reveals Galaxy S5 BOM

IHS teardown reveals Galaxy S5 BOM

Research firm IHS got hold of Samsung’s new flagship smartphone and took it apart to the last bolt to figure out…

More...
Galaxy S5, HTC One M8 available selling well

Galaxy S5, HTC One M8 available selling well

Samsung’s Galaxy S5 has finally gone on sale and it can be yours for €699, which is quite a lot of…

More...
Intel lists Haswell refresh parts

Intel lists Haswell refresh parts

Intel has added a load of Haswell refresh parts to its official price list and there really aren’t any surprises to…

More...
Respawn confirms Titanfall DLC for May

Respawn confirms Titanfall DLC for May

During his appearance at PAX East panel and confirmed on Twitter, Titanfall developer Respawn confirmed that the first DLC pack for…

More...
KFA2 GTX 780 Ti Hall Of Fame reviewed

KFA2 GTX 780 Ti Hall Of Fame reviewed

KFA2 gained a lot of overclocking experience with the GTX 780 Hall of Fame (HOF), which we had a chance to…

More...
Frontpage Slideshow | Copyright © 2006-2010 orks, a business unit of Nuevvo Webware Ltd.
Tuesday, 22 November 2011 12:26

Bulldozer server benchmarks fail to impress

Written by

amd logon

Intel has little to fear but itself

 A few weeks ago AMD introduced its latest FX-series desktop processors, based on the new Bulldozer architecture. The new series offered underwhelming performance and failed to impress AMD enthusiasts, although they tend to offer pretty good value for money.

Now it is time for Bulldozer server parts and sadly it turns out they are not much better than their desktop siblings. Punters had expected the server parts to be a tad more competitive, thanks to their emphasis on hyperthreading. However, this does not seem to be the case.

Ars Technica compiled test results from several sources and went out of their way to compare the cost of new Opteron-based server systems and Intel Xeon systems. Although Opterons did manage to offer superior performance against comparable Xeons in a TPC-C scenario, they end up costing about 50 percent more, yet deliver an 18 percent improvement in performance.

In SPEC JBB2005 Java emulation Opterons also fared well against Xeon parts, but they appear to offer little to no improvement over previous generation Opterons. In virtualization and HPC tests conducted by Anandtech, the Opteron 6276 failed to beat Intel’s Xeon X5670 and pretty much continued the underwhelming streak.

The biggest issue is the lack of a clear performance advantage over existing Opteron 6100 parts, so the new architecture doesn’t appear to offer much incentive to upgrade from the previous generation. Worse, the new Opterons struggle to compete with Intel Westmere EP parts, despite the fact that they have been on sale for a year and a half.


More here.

 

E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it
blog comments powered by Disqus

Comments  

 
+4 #1 Vizeroy 2011-11-22 14:29
The results seem a bit strange, because if you use current level software (RHEL 6.2, Intel Compilers 2011R2 etc.), there definetely is a performance jump on the Bulldozer.

You refered to heise.de in another article, why didn't you have a look at their result in this case?

Compared to Anand and co. they really made an effort to find out the real performance.
(The whole test is in the c't magazine and not on the site, but it is definetely worth reading)

As cited before in a "XEON" story, the real SPEC results are quite impressive.
 
 
+3 #2 Vizeroy 2011-11-22 14:51
The price of the server allone is quite.. strange:
"The new Opteron 6200-based system? A cool $1 million and change".

The Opteron has roughly 10,000GB more storage capacity than the compared Xeon system.
The 90SSDs in the Opteron server alone add up to $1.12 million ($12,499.00 each).
The Xeon SSDs have the same capacity per drive, but there are just 81 and for some reason they cost just $5,999.00 each.

Therefore I'm not sure, if the Ars Technica comment on the price/performance comparison is justified.

The heise benchmark compares roughly equal systems. (1HE dual socket servers)
 
 
+1 #3 trajan2448 2011-11-22 16:59
Its going to be a very tough year for AMD.
 
 
+1 #4 trajan2448 2011-11-22 17:23
A four socket Xeon X7560 system (2.26 GHz, eight cores per processor, 16 threads per processor) system, with a total system cost of $880,000, achieved 1.8 million tpmC. Sure, the processors cost almost four thousand bucks apiece. But it doesn't matter, because the total system is cheaper and much faster than the new Bulldozers. It also uses slightly less power: 2.46 W/KtpmC for the Xeon versus 2.49 for the 6200.
 
 
+5 #5 maroon1 2011-11-22 18:36
Those results are not really surprising.

What is really surprising is AMD marketing team which keeps lying and lying.
 
 
-1 #6 Vizeroy 2011-11-22 20:57
Quoting trajan2448:
A four socket Xeon X7560 system (2.26 GHz, eight cores per processor, 16 threads per processor) system, with a total system cost of $880,000, achieved 1.8 million tpmC. Sure, the processors cost almost four thousand bucks apiece. But it doesn't matter, because the total system is cheaper and much faster than the new Bulldozers. It also uses slightly less power: 2.46 W/KtpmC for the Xeon versus 2.49 for the 6200.


Ok.. and how exactly is a 4 socket server with way less HD capacity in the same league as a 2 socket system?
 
 
+2 #7 Vizeroy 2011-11-22 21:00
Quoting maroon1:
Those results are not really surprising.

What is really surprising is AMD marketing team which keep lying and lying and manipulate people.


The strange thing is that everyone believes these benches without cheking the basis on which they are made.

Factor out the SSDs and the AMD system is much cheaper.
Just use the same SSDs as on the Xeon comparison system and the system cost will go down by about $600,000.

This kind of BOM manipulation is one of the cheapest ways to bend a test to your likings.
 
 
+1 #8 redisnidma 2011-11-22 21:28
Sometimes you have to wonder when in time did AMD became such a mediocre company.

I don't feel anymore pity for them. Too many f***d ups in such a small period of time.

Just hope they die fast.
 
 
+2 #9 Vizeroy 2011-11-22 21:42
Quoting redisnidma:
Sometimes you have to wonder when in time did AMD became such a mediocre company.

I don't feel anymore pity for them. Too many f***d ups in such a small period of time.

Just hope they die fast.



Sometimes I wonder how rare thinking has become....

Bulldozer was the wrong CPU for desktop. for server it works quite well (see real benchmarks).

APU and GPU production disaster wasn't the fault of AMD.

A dead AMD would mean Intel monopol and $1000.00 Pentium class CPUs for several years w/o improvements.
 
 
+2 #10 Kryojenix 2011-11-22 21:45
I guess the future will be in ARM.
 

To be able to post comments please log-in with Disqus

 

Facebook activity

Latest Commented Articles

Recent Comments