Linux and EM64T; Intel's 64-bit Suggestion
by Kristopher Kubicki on August 9, 2004 12:05 AM EST- Posted in
- Linux
Since the excuse to not compare Athlon 64s to Intel Pentium based processors has always been "you can't compare apples to oranges," we found ourselves fairly entertained to come into the possession of a 3.6GHz EM64T Xeon processor. Intel's EM64T is Intel's true x86_64 initiative. This 3.6GHz Xeon processor is actually the exact same CPU in as the LGA775 Pentium 4F we will see in just a few weeks. We are offering a preview of an unreleased processor on 64-bit Linux systems. Now, we have Intel and AMD 64-bit x86 processors, 64-bit Linux operating systems and a few days to get some benchmarking done.
We are going to run the benchmarks for this review slightly different than we have in the past. We want to make our numbers easily replicable for those who have the necessary components, but we also want to show the fullest capabilities of the hardware that we have. Many of our previous benchmarks are not multithread (POV-Ray) or do not scale well. Unfortunately, this forces us to use a lot of synthetic benchmarks; but we feel the overall results are accurate and reflective of the hardware used.
The delicate bit for this review was using the SuSE 9.1 Pro (x86_64) installation rather than compiling it from scratch (à la Gentoo). This was done to preserve the ability to replicate our benchmarks easily. Fedora Core 2 refused to install on the IA32e machine because there was no recognized AMD CPU.
As there may have been a little confusion from the last review, the DDR PC-3500 only runs at 400MHz. The Infineon Registered RDIMMs used on the Xeon runs at slightly high latencies. All memory runs in dual channel configurations. We removed 1 CPU for the tests in this benchmark, but since HyperThreading was enabled, we used the SMP kernel. During the second half of the benchmarks, SMP was disabled and the tests were re-run under the single CPU generic kernel. These are both 64-bit CPUs, and so, all benchmarks are run on 64-bit OSes with 64-bit binaries wherever possible.
We are going to run the benchmarks for this review slightly different than we have in the past. We want to make our numbers easily replicable for those who have the necessary components, but we also want to show the fullest capabilities of the hardware that we have. Many of our previous benchmarks are not multithread (POV-Ray) or do not scale well. Unfortunately, this forces us to use a lot of synthetic benchmarks; but we feel the overall results are accurate and reflective of the hardware used.
The delicate bit for this review was using the SuSE 9.1 Pro (x86_64) installation rather than compiling it from scratch (à la Gentoo). This was done to preserve the ability to replicate our benchmarks easily. Fedora Core 2 refused to install on the IA32e machine because there was no recognized AMD CPU.
| Performance Test Configuration | |
| Processor(s): | Athlon 64 3500+ (130nm, 2.2GHz, 512KB L2 Cache) Intel Xeon 3.6GHz (90nm, 1MB L2 Cache) |
| RAM: | 2 x 512MB PC-3500 CL2 (400MHz) 2 x 512MB PC2-3200 CL3 (400MHz) Registered |
| Memory Timings: | Default |
| Hard Drives | Seagate 120GB 7200RPM IDE (8Mb buffer) |
| Operating System(s): | SuSE 9.1 Professional (64 bit) Linux 2.6.4-52-default Linux 2.6.4-52-smp |
| Compiler: | GCC 3.3.3 |
| Motherboards: | NVIDIA NForce3 250 Reference Board SuperMicro Tumwater X6DA8-G2 (Only 1 CPU) |
As there may have been a little confusion from the last review, the DDR PC-3500 only runs at 400MHz. The Infineon Registered RDIMMs used on the Xeon runs at slightly high latencies. All memory runs in dual channel configurations. We removed 1 CPU for the tests in this benchmark, but since HyperThreading was enabled, we used the SMP kernel. During the second half of the benchmarks, SMP was disabled and the tests were re-run under the single CPU generic kernel. These are both 64-bit CPUs, and so, all benchmarks are run on 64-bit OSes with 64-bit binaries wherever possible.
Facebook
Twitter
RSS
275 Comments
View All Comments
T8000 - Monday, August 9, 2004 - link
I think it is justified to put the 3500+ in its place here. Either it is overrated or its 64 bit support is more of a marketing statement.For those wanting an 3800+ in the review, just add a little under 10% to the 3500+, since the 3800+ has about 10% more clockspeed and it usually scales close to that. Not that it would change that much, it would only make things worse for AMD fanboys, seeing the 3800+ unable to hold its own against a real 3.6 Ghz CPU.
Besides, you are welcome to try these tests at home with your FX53 or Opteron x50 and submit some scores. Don't have one? Don't worry, AMD hardly sells them anyway, especially to home users, since most users that can afford them do not buy AMD.
matman326 - Monday, August 9, 2004 - link
Very disappointing.... thats all i'm gonna say.wildguy2k - Monday, August 9, 2004 - link
"Even a intel fanyboy has to laugh at how off sided this failure of a "review" was.Come on anandtech, if we wanted to read stuff like this we go to tom's"
Exactly. Also, to all those who say that 200MHz & 512KB of cache don't really make much difference, there's an article on this same site that may point out the very difference they provide. Now, I know it's not utilizing the 64bit extensions, but this image
(http://images.anandtech.com/graphs/athlon%2064%203... DOES show a 10% difference between the FX-53 & the 3500+ while compiling...
rocketbuddha - Monday, August 9, 2004 - link
AlexWade - Monday, August 9, 2004 - link
Okay, deep breath ...Obviously, the Prescott does some nice math work. I'll keep that in mind. But, most of us, and I'll tend to believe most servers, don't crunch numbers all day. I wanted to see more benchmarks on stuff that is more likely to get done, not finding prime numbers. More encoding, more games, more SQL, more compiling, only one or two math benchmarks.
It is NOT a fair comparision of A64's weakness vs. Prescott's new strength on 60%+ benchmarks.
It is a fair comparision of CPU's. Although, not the best. The Opteron and Athlon64 come from the same mold. Variations aren't going to be that minor.
Please, next benchmark, make it more well-rounded. I could give a flip about Super Pi .
classy - Monday, August 9, 2004 - link
I see no basis at all for this article. If you only had benchmarks for the 3500+ you should have even written this article. In all the years on Anandtech, I don't ever recall an article as uninformative as this one. Its nice to see the Athlon win a couple of benchmarks, but this is a very needless comparison.Marlin1975 - Monday, August 9, 2004 - link
Even a intel fanyboy has to laugh at how off sided this failure of a "review" was.Come on anandtech, if we wanted to read stuff like this we go to tom's
Pollock - Monday, August 9, 2004 - link
I just want to throw my comment in here that I agree with most of the other people here. The conclusion is what I find most ridiculous."Without a doubt, the 3.6GHz Xeon trounces over the Athlon 64 in math-intensive benchmarks."
Like many other people said, I find that statement very unfair, again considering it wasn't against a similar chip.
bhtooefr - Monday, August 9, 2004 - link
OK, I posted that last comment in reply to one on the previous page, and didn't realize that some of the benchmarks were 32-bit ones, either by accident, or to make the AMD smearing more obvious...bhtooefr - Monday, August 9, 2004 - link
All of the benchmarks were 64-bit. They couldn't have thrown in a regular Xeon 3.6, because it wouldn't be able to run the OS or the benchmarking apps.They should be testing chips against others in their price range and PR rating range. So, here's what they should have tested (I noticed further down why they used the 3500+ ($346):
Pentium 4 560 EM64T (3.6GHz, $637)
Pentium 4 550 EM64T (3.4GHz, $417)
Pentium 4 540 EM64T (3.2GHz, $278)
I obtained these P4 prices from Intel's price list. While these prices are for the NON-EM64T chips, I read in a press release that Intel isn't charging any more for EM64T.
If they did it right, with the Xeon DP 3.6, here are the CPUs:
Xeon DP 3.6 EM64T ($851)
Opteron 250 ($851 - looks like it's aligned EXACTLY against the DP 3.6)