Linux and EM64T; Intel's 64-bit Suggestion
by Kristopher Kubicki on August 9, 2004 12:05 AM EST- Posted in
- Linux
Conclusions
Although the Athlon 64 3500+ and the Xeon 3.6GHz EM64T processors were not necessarily designed to compete against each other, we found that comparing the two CPUs was more appropriate than anticipated, particularly in the light of Intel's newest move to bring EM64T to the Pentium 4 line. Once we obtain a sample of the Pentium 4 3.6F, we expect our benchmarks to produce very similar results to the 3.6 Xeon tested for this review.Without a doubt, the 3.6GHz Xeon trounces over the Athlon 64 3500+ in math-intensive synthetic benchmarks. Again, not that it is really a comparison between the two chips yet anyway, but perhaps something of a marker of things to come. However, real world benchmarks, with the exception of John the Ripper is where AMD came ahead instead. Even though John uses several different optimizations to generate hashes, in every case, the Athlon chip found itself at least 40% behind. Much of this is likely attributed to the additional math tweaking in the Prescott family core, and the lack of optimizations at compile time.
That's not to say that the Xeon CPU necessarily deserves excessive praise just yet. At time of publication, our Xeon processor retails for $850 and the Athlon 3500+ retails for about $500 less. The 3.6F processor the Xeon represents does not even exist in retail channels yet. Also, keep in mind that the AMD processor is clocked 1400MHz slower than the 3.6GHz Xeon. With only a few exceptions, synthetically the 3.6GHz Xeon outperformed our Athlon 64 3500+, whether or not the cost and thermal issues between these two processors are justifiable.
We will benchmark some SMP 3.6GHz Xeons against a pair of Opterons in the near future, so check back regularly for new benchmarks!
Update: We have addressed the issue with the -02 compile options in TSCP, the miscopy from previous benchmarks of the MySQL benchmark, and various other issues here and there in the testing of this processor. Expect a follow up article as soon as possible with an Opteron.
Facebook
Twitter
RSS
275 Comments
View All Comments
tfranzese - Tuesday, August 10, 2004 - link
Which tests were re-ran? I guess I should ask that first.KristopherKubicki - Tuesday, August 10, 2004 - link
#203 can you be more specific? What is wrong with my conclusion.Kristopher
tfranzese - Tuesday, August 10, 2004 - link
What more do we want? Fix the rest of the article, including the conclusions you drew.KristopherKubicki - Tuesday, August 10, 2004 - link
#200:Jasons article is seperate from mine. His is running on windows anyway with two processors.
"f-off people" is strictly not my attitude. I have made changes to the article that were suggested; i fixed the broken makefile, i even did another article on my vacation with an opteron 150 to be posted within the next 24 hours.
What more do you want? Really?
Kristopher
MikeEFix - Tuesday, August 10, 2004 - link
There is nothing wrong with an engineers’ pov. We tend to like symmetry and base results on facts. Unfortunately there is little symmetry and too many variables for accurate results.Experiments should be kept in the lab
allnighter - Tuesday, August 10, 2004 - link
A few days ago when Nocona was announced I shouted I wanted some benchmarks.When Jason Clark mentioned AT is working on some benchmarks I was very excited. Now I am am not. This article is... well in attempt to be polite... of a very questionable quality or simply crap.As many people posted their observations, whith most of which I agree, this is not quality we're used to from AT. And all the AMD and Intel fanboyish escapades in here are just making all this worse.
I feel like something should be done. I have no idea what but just one 'f-off people' type of comment from the author is just not doing it for me. This is bad.
Pumpkinierre - Tuesday, August 10, 2004 - link
Good on ya Kristopher, you got a bit of scientist in you unlike some of these machine like engineer types. Some of us know where you are coming from (and it aint Intel). A little knowledge can be a dangerous thing but its still better than no knowledge at all. Looking forward to your later articles on the subject.MikeEFix - Tuesday, August 10, 2004 - link
Well the nvidia 250 reference board isn't exactly a Tyan Thunder K8W(S) either.The Xeon is using platform $600.00 server mainboard while the desktop variant, a64, is using the generic desktop solution.
Macro2 - Tuesday, August 10, 2004 - link
This is the kind of article i'd expect from Tom's not Anandtech. I guess Anand is out with the lady.srg - Tuesday, August 10, 2004 - link
Well with these refinements an updates to the benchmarks, not forgetting setting them up right, the gap is closing to what I would have though it would have been, although the Jon-The-Ripper benchmark does seem odd (explained better my an earlier poste, if he's right then the 3500+ is basically running 64-bit'ised K6 code).I think what Kris has learnt here will be valuable for later reviews (and no, the simple fact that Intel beat AMD won't bring a torrent of flames like this one).
srg