Linux and EM64T; Intel's 64-bit Suggestion
by Kristopher Kubicki on August 9, 2004 12:05 AM EST- Posted in
- Linux
Synthetic Benchmarks (continued)
TSCP
TSCP is a simple chess program, which you may read more about here. We compiled the program using our own Makefile, which you can download here. Once compiled, we ran the "bench" command inside the program. Using the -m64 flags provided no change in performance.
As you can see, there appears to be no advantage with HyperThreading for this application. This also appears to be the largest lead that the Intel processor takes over the AMD during the duration of our analysis.
Update:We have retested this part of the benchmark with the -O2 flag in the correct place for both machines. The score has changed to reflect this. br>
ubench
Finally, we have ubench, which stands as the definitive Unix synthetic benchmark. Feel free to learn more about the program here. We compiled the program using ./configure and make with no optimizations. The benchmark was run on a loop ten times to assure that we were getting a true average.
Here, we see HyperThreading working against the Xeon processor in a distinct fashion. According to the Ubench website, both of these machines with single processors outperform dual Xeon 2.4GHz machines, even though they are only running on one processor. The program runs several math-intensive floating point and integer operations over the course of three minutes.
Facebook
Twitter
RSS
275 Comments
View All Comments
Zebo - Thursday, August 12, 2004 - link
could acehardware have a worse forum tech? that's like 1995 BBS.JGunther - Thursday, August 12, 2004 - link
Don't mean to be a pest, but every moment that botched review is online is depressing. I'm looking forward to the new review as well. So where is this thing?tfranzese - Thursday, August 12, 2004 - link
About the Opteron vs. Xeon talk: Opteron scales better in SMP. Opteron is 8-way capable. And, I'd be willing to bet you Opteron will really stretch it's legs in 64-bit/64-bit once it's primetime, contrary to what I believe the Xeon will do - improve, but marginally.Aileur - Wednesday, August 11, 2004 - link
Tabajara - Wednesday, August 11, 2004 - link
I think that to call the reasoning of the author of this atticle as "nonsense" would actually be a compliment. Just take a look at the conclusion that someone thinking straight would get, using tha same info that is on the article:"In spite of the fact that this Xeon processor retails for $850 and the Athlon 3500+ retails for about $500 less, that this Xeon does not even exist in retail channels yet, that the AMD processor is clocked 1400MHz slower than the 3.6GHz Xeon and has 512 less kb of cache, IT STILL WINS MOST OF THE REAL WORLD TESTS!"
Another important factor: the price and performance difference of the mobos used for each processor probably gives the P4 an edge. To use a NVIDIA NForce3 250 Reference Board against a SuperMicro Tumwater X6DA8-G2 is just not fair.
Other caractheristic that makes this review resemble the ones done at the POS THG is that the synthetic benchmarcks seem to have been picked to benefit a CPU that has a higher clock and that excels at handling branching instructions (as chess based bechmarks, that have to calculate lots of possible moves). In other words, tests that show the best qualities of the P4.
Viditor - Wednesday, August 11, 2004 - link
KK - "yeah the review is done just pushing it live as soon as i can"You da man Kris! Now may I suggest you turn off the damn computer and go enjoy what's left of your vacation!!!!
Cheers,
Charles
snorre - Wednesday, August 11, 2004 - link
Kristopher: Read this:http://www.aceshardware.com/forum?read=115094123
You should also test Crafty.
KristopherKubicki - Wednesday, August 11, 2004 - link
Not really sure what the fuss is about the remote server? Its at Jason's place, SuperMicro gave it to him. You can email him about it if you like.Anyways, yeah the review is done just pushing it live as soon as i can. I think you will enjoy these benches much better.
-MySQL
-Postgress
-MentalRay
-Povray
-TSCP
-gzip
-mencoder
-lame
-JTR again but a different source - the AMD and Intel optimizations are highlighted as we compiled the code
-One synthetic benchmark
-Anything else i can think of in the next 20 min that is quick to test.
anything we compiled was done using -o2 and unroll-loops.
snorre - Wednesday, August 11, 2004 - link
Time's Up!snorre - Wednesday, August 11, 2004 - link
Correction, still about 20 minutes to go :PPrepare to be scrutinized, so this new review better be flawless ;-)