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
JGunther - Monday, August 9, 2004 - link
What the hell is the matter with you guys? I mean, I'm all for a good CPU wars, but comparing a desktop CPU to a server CPU? I mean, we're talking a $350 CPU vs. an $850 CPU. How, by any stretch of the imagination, is this a good comparison??Oh. my. god. Anandtech... I'm starting to wonder about this site. First you guys blast RAID 0 on the desktop (which Tweakers.Net just stated was COMPLETELY misleading in their own, more extensive battery of tests) and now this?
Man... this isn't even a question of AMD vs. Intel. The only people who would think that this is a decent comparison have to be Intel fanboys or something.
How did this get to print?
Viditor - Monday, August 9, 2004 - link
Soultrap - "The reason the benchmarks come out faster on the Intel part is because of it's higher clock."This is what I thought at first too...
The problem is that the benches being reproduced around the web for the A64 don't match up to Kris's...
menads - Monday, August 9, 2004 - link
#101 The mistakes in the results has nothing to do with the Intel clockspeed - it is the reviewer lack of knowledge how to use gcc (and its flags) and to understand what the benchmarks measure. And finally the poor judgement of selecting a badly positioned desktop CPU (even 754 pin 3400+ makes much more sense than the 3500+ used in the review let alone Opteron 150) versus the top end server CPU.manno - Monday, August 9, 2004 - link
"This is a sad day for all of us. Anandtech has now lost all credibility as an independent review site. First THG, now Anandtech, WTF is going on in the world? :("Get a life.
Soultrap - Monday, August 9, 2004 - link
The reason the benchmarks come out faster on the Intel part is because of it's higher clock. If you do the math you will see that the scores are proportional to the clock frequency on the benchmarks that the Intel part stomped the AMD part. This does not make the Intel part better then the AMD part! But, it does make it (much) faster a doing very simple tasks. The more complex the task is the more important all of the other features of the chip become. The "other" features is where the AMD chip realy shines, not in core frequency. (shorter pipeline, better instruction management, better memory access, ...)In any benchmark that truely uses the processor as a computer and not as simple calculator, like just about any of the gaming benchmarks out there you will find that in even comparisons (apples to apples) the Intel will be falling behind. When 64bit is mature you will realy see the weakness of Intel's 64 bit clunker.
By the way, just how much did you get paid by Intel to do this cornhole of a reveiw?
Never mind, I am sure they promised that you would disappear if you told on them.
Just kidding!
But really "Convictions are more dangerous enemies of truth then lies." - Friedrich Wilhelm Nietshe
And I beleive that when it comes to AMD vs Intel there are alot of convictions in the IT world.
bhtooefr - Monday, August 9, 2004 - link
#48, you're not going to see Doom 3 benchies because Doom 3 doesn't run on Linux, and even if it did, it'd be a regular x86 app, not x86-64/EM64T.chaosengine - Monday, August 9, 2004 - link
Yeah its really sorry to see this. I joined anandtech right now to post this damn message!!!So what next will we have? Since when sane people have started comparing Server chips against Desktop Chips???
I thought anandtech was much less biased than others but sadly not the case.
dougSF30 - Monday, August 9, 2004 - link
Problems with primegen benchmark:http://www.investorshub.com/boards/read_msg.asp?me...
"
What Anand's "primegen" was actually measuring:
*locking* and *unlocking* of the thread-safe version of putchar() was the bottleneck.
switching to unlocked putchar made the benchmark run twice as fast.
commenting out the putchar stuff entirely resulted in another factor of 2 faster.
So:
50% of time involves locking.
25% of time involved input/output
25% of time was actually doing arithmetic, calculating primes.
Gosh, I wonder if The Prescott New Instructions MONITOR and MWAIT have anything to do with the selection of this benchmark, and the performance of Nocona?
http://www.aceshardware.com/forum?read=115093892
"
snorre - Monday, August 9, 2004 - link
This is a sad day for all of us. Anandtech has now lost all credibility as an independent review site. First THG, now Anandtech, WTF is going on in the world? :(DrMrLordX - Monday, August 9, 2004 - link
In response to #93,I don't see 3500+ vs Opteron 150 being minor. It means the reviewer actually believes AMD's stupid PR scheme. Their PR scheme has always been wrong, and will likely always be wrong. The 3500+ is a glaring example of how screwed-up their PR scheme can get. Aside from dual-channel memory support, the 3500+ for socket 939 is the same cpu as the 3200+ Newcastle for socket 754. Previous Anandtech articles have highlighted this fact. Not only does KK's choice of cpus potray AMD as being in a position of weakness(somehow implying that AMD has nothing better to offer than the 3500+, that it's only purpose in the market is that its cheaper than Intel "just like always", etc), but it casts Intel in an unfavorable light by giving it a truly unworthy opponent. The 3500+ is a lousy processor for the price. The 3400+, which is clocked 200 mhz higher than the 3500+, is a superior CPU AND costs less. Dual-channel memory just doesn't do enough to justify any of the PR ratings AMD uses on its Socket 939 cpus.