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
fifi - Monday, August 9, 2004 - link
oh blasted, can't edit!that post above was addressed to manno.
tfranzese - Monday, August 9, 2004 - link
Edit buttons would be great, but anyway... I meant to say objective in place of subjective, but for anyone looking for someone's opinion there is also a link in there to a subjective view on Intel's implementation of AMD64.fifi - Monday, August 9, 2004 - link
why is it that it MUST be AMD-fans versus Intel-fans? can we not complain when we see shoddy work done?It's shoddy work when there's no comparison between 32-bit and 64-bit on the P4.
It's shoddy work when the benchmarks posted are simply wrong (look to his earlier review on A3500+).
No, it's not AT being bought by Intel. It's just plain shoddy work.
I would have preferred if Kristopher took his time to run the benchmarks properly, and checked all the numbers are correct and ran all the control tests and present a complete picture, rather than just trying to be the first out of the door with a review on the EM64T.
it's just plain shoddy work.
So keep on the sarcasm and hopefully it will improve your mental abilities which are clearly being impaired by the background EM waves, you SHOULD have bought those aluminium hats like I told you to...
tfranzese - Monday, August 9, 2004 - link
#55, 60, aka fanboy. Anandtech was no where the first to post Nocona benches. Not even the second, and probably not even the third.Here's some further Nocona reading which is far more informative and subjective for those interested that I've collected here: http://www.overclockersonline.com/index.php?page=w...
As for this review I can only say my respect for this site has been lowered a notch. Congrats Kris and Anand!
Viditor - Monday, August 9, 2004 - link
About what you'd expect between these 2 chips...1. 512k cache vs 1 meg cache
2. 3.6 GHs vs 2.2 GHz
3. single CPU (Opteron and Athlon64 perform much better as you add CPUs because of Hypertransport)
I really feel that this review should have been held until some comparable chips could have been tested as it appears quite biased in it's present form...
Questions though...
1. Was the setting for the memory timing on the Athlon 64 set to 1T or 2T?
2. Since both CPUs are 64bit, why use only 1 Gig of Ram? (4 Gig would have been a better demonstration...) The reason this is interesting is that there is some confusion as to Xeon's ability in handling larger amounts of memory well.
manno - Monday, August 9, 2004 - link
Oh cry me a river you poor poor souls."Alot of the benchmarks are incorrect due to setup errors. It's not just the scores that we're bitching about, but the SLOPPINESS of the review.."
Yeah it's got to be that, and not the fact that it doesn't paint AMD as the savior of the free world. Of course!! why didn't I see that the first time a lame fanboy posted?
God I feel so stupid!
You know what you're right I must of been crazy to think that Anand isn't in a secret Cabal with Intel to paint The A64 in the worst possible light. Thank you... no I truly mean it from the bottom of my heart thank you for showing me the error of my ways, and keeping evil Intel, and the even more devious Anantech.com out of my life. Well I'm going to head out now and buy myself one of those shiny aluminum hats to keep them from taking over my brain waves. Take care and have fun flaming me.
-manno
Xspringe - Monday, August 9, 2004 - link
It'd be nice to see more balanced benchmark set in the next similar review with less synthetic benchmark and more benchmarks relative to real world usage. These current benchmarks aren't very useful and incomplete as has been stated by a few people before me.hirschma - Monday, August 9, 2004 - link
I'd really love to know the sizes of the executables where the Xeon won by a significant amount. I'd bet that everything just fits in the cache for the Xeon, but not for the Athlon.If that is, in fact, the case, then Anandtech's conclusion should be: sometimes 1 meg of cache beats 512k of cache.
Sloppy, sloppy, sloppy.
WizzBall - Monday, August 9, 2004 - link
Uh you're right manno, how awful of us to not behave and thank KK for his effort... So here it is, thank you for wasting my time Kristopher!And btw, what's this bs about the 'service'... if WE wouldn't come here to read on a daily basis they would have no job, get it manno? Now go back and bow again to KK, thank you very much.
*We'll send kudos when we see something done right although it's true you can't please everybody but this time a majority of the people expressing their opinion think this is crappy work. It's weird how all the fanboys gather here on AT, huh?
Carfax - Monday, August 9, 2004 - link
Manno, have you read ANY of the replies? Alot of the benchmarks are incorrect due to setup errors. It's not just the scores that we're bitching about, but the SLOPPINESS of the review..