mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > Software

Reply
 
Thread Tools
Old 2007-03-09, 23:54   #1
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

2·53·71 Posts
Default Core 2 Duo E6600 Benchmark please

I'm working on the formulas for CPU credit on the server. In the past I've used a Pentium-90 and a PII-400 to devise these formulas. It's time to update that! I'd like to use a mid-range Core 2 Duo as the baseline machine. Unfortunately, I don't own one.

So, if you have a non-overclocked Core 2 Duo E6600 running version 24.13 or 24.14, I'd be grateful for a benchmark. Exit prime95. Add the line FullBench=1 to prime.ini, run prime95, and do Options/Benchmark. Post the results here or email them to me. Thanks!!
Prime95 is online now   Reply With Quote
Old 2007-03-10, 02:52   #2
julian_emilio
 
Mar 2007

2 Posts
Default Core 2 Duo E6400 Benchmark (not E6400)

Here is a benchmark of Core 2 Duo E6400. You were looking for a benchmark of the E6600, but maybe this can help :

Intel(R) Core(TM)2 CPU 6400 @ 2.13GHz
CPU speed: 2133.85 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 32 KB
L2 cache size: 2048 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
Prime95 32-bit version 24.14, RdtscTiming=1
Best time for 4K FFT length: 0.061 ms.
Best time for 5K FFT length: 0.090 ms.
Best time for 6K FFT length: 0.107 ms.
Best time for 7K FFT length: 0.130 ms.
Best time for 8K FFT length: 0.134 ms.
Best time for 10K FFT length: 0.188 ms.
Best time for 12K FFT length: 0.229 ms.
Best time for 14K FFT length: 0.277 ms.
Best time for 16K FFT length: 0.289 ms.
Best time for 20K FFT length: 0.392 ms.
Best time for 24K FFT length: 0.477 ms.
Best time for 28K FFT length: 0.591 ms.
Best time for 32K FFT length: 0.615 ms.
Best time for 40K FFT length: 0.810 ms.
Best time for 48K FFT length: 0.985 ms.
Best time for 56K FFT length: 1.202 ms.
Best time for 64K FFT length: 1.286 ms.
Best time for 80K FFT length: 1.810 ms.
Best time for 96K FFT length: 2.201 ms.
Best time for 112K FFT length: 2.668 ms.
Best time for 128K FFT length: 2.819 ms.
Best time for 160K FFT length: 3.460 ms.
Best time for 192K FFT length: 4.285 ms.
Best time for 224K FFT length: 5.159 ms.
Best time for 256K FFT length: 5.813 ms.
Best time for 320K FFT length: 7.607 ms.
Best time for 384K FFT length: 9.339 ms.
Best time for 448K FFT length: 11.135 ms.
Best time for 512K FFT length: 12.346 ms.
Best time for 640K FFT length: 16.439 ms.
Best time for 768K FFT length: 20.186 ms.
Best time for 896K FFT length: 23.999 ms.
Best time for 1024K FFT length: 27.960 ms.
Best time for 1280K FFT length: 34.175 ms.
Best time for 1536K FFT length: 41.544 ms.
Best time for 1792K FFT length: 49.512 ms.
Best time for 2048K FFT length: 55.166 ms.
Best time for 2560K FFT length: 73.039 ms.
Best time for 3072K FFT length: 88.454 ms.
Best time for 3584K FFT length: 105.872 ms.
Best time for 4096K FFT length: 118.445 ms.
Best time for 5120K FFT length: 152.534 ms.
Best time for 6144K FFT length: 184.302 ms.
Best time for 7168K FFT length: 225.052 ms.
Best time for 8192K FFT length: 247.732 ms.
Best time for 10240K FFT length: 327.885 ms.
Best time for 12288K FFT length: 396.792 ms.
Best time for 14336K FFT length: 484.609 ms.
Best time for 16384K FFT length: 531.349 ms.
Best time for 20480K FFT length: 730.398 ms.
Best time for 24576K FFT length: 908.883 ms.
Best time for 28672K FFT length: 1084.860 ms.
Best time for 32768K FFT length: 1199.139 ms.
Best time for 58 bit trial factors: 5.245 ms.
Best time for 59 bit trial factors: 5.315 ms.
Best time for 60 bit trial factors: 5.309 ms.
Best time for 61 bit trial factors: 5.254 ms.
Best time for 62 bit trial factors: 8.375 ms.
Best time for 63 bit trial factors: 8.431 ms.
Best time for 64 bit trial factors: 8.066 ms.
Best time for 65 bit trial factors: 7.987 ms.
Best time for 66 bit trial factors: 8.026 ms.
Best time for 67 bit trial factors: 7.930 ms.
julian_emilio is offline   Reply With Quote
Old 2007-03-10, 03:14   #3
Xyzzy
 
Xyzzy's Avatar
 
"Mike"
Aug 2002

202016 Posts
Default

We like P90 years.

Xyzzy is offline   Reply With Quote
Old 2007-03-10, 03:50   #4
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

2·53·71 Posts
Default

Quote:
Originally Posted by Xyzzy View Post
We like P90 years.
I just knew this request would start this discussion!

Examining v24.14 640K to 2048K FFT benchmarks from a P100, PII-400, P4-2000 and a C2D-2400 we get the following architectural multipliers (after factoring out the MHz difference):

One core of a C2D = 1.68 P4.
A P4 = 3.44 PIIs
A PII = 1.12 Pentium

Using this we can convert from one Intel-based measure to another. Even if the server internally maintains C2D CPU seconds, we can display reports in any measure we want.

Since almost no one here is doing LL tests using a P-90 now, I'm sure we can agree that granting CPU credit based on a P4 or C2D would be fairer. If I continued to use the timings from my PII-400, its small L2 cache would at some point inflate the CPU credit given to larger FFT sizes.
Prime95 is online now   Reply With Quote
Old 2007-03-10, 07:36   #5
Cruelty
 
Cruelty's Avatar
 
May 2005

65816 Posts
Default

I have already done that some time ago. http://www.mersenneforum.org/showpos...&postcount=248

Last fiddled with by Cruelty on 2007-03-10 at 07:38
Cruelty is offline   Reply With Quote
Old 2007-03-10, 11:11   #6
davieddy
 
davieddy's Avatar
 
"Lucan"
Dec 2006
England

2×3×13×83 Posts
Default

Quote:
Originally Posted by Prime95 View Post
Since almost no one here is doing LL tests using a P-90 now, I'm sure we can agree that granting CPU credit based on a P4 or C2D would be fairer. If I continued to use the timings from my PII-400, its small L2 cache would at some point inflate the CPU credit given to larger FFT sizes.
And using updated FFT ranges on the status page would
also be more representative
davieddy is offline   Reply With Quote
Old 2007-03-10, 11:20   #7
ET_
Banned
 
ET_'s Avatar
 
"Luigi"
Aug 2002
Team Italia

113178 Posts
Default

Quote:
Originally Posted by Prime95 View Post
I'm working on the formulas for CPU credit on the server.
So someone IS still working on Primenet V5.0

Luigi
ET_ is offline   Reply With Quote
Old 2007-03-10, 13:14   #8
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

11101011001102 Posts
Default

Thanks Cruelty -- just what I was looking for.
Prime95 is online now   Reply With Quote
Old 2007-03-10, 14:35   #9
Cruelty
 
Cruelty's Avatar
 
May 2005

23×7×29 Posts
Default

Glad I could help
Cruelty is offline   Reply With Quote
Old 2007-03-10, 21:28   #10
S485122
 
S485122's Avatar
 
Sep 2006
Brussels, Belgium

2·3·281 Posts
Default

Basing a new norm on the best times of a benchmark seems not fair to me. I benchmarked one machine quite extensively and came to the conclusion that results have a big standard deviation. I could quote someone on this :
Quote:
Originally Posted by Prime95 View Post
The benchmark is a poor tool to use. Since it reports the BEST iteration time.

You really need to run 4 LL tests which will report the more important average iteration time.
;-)
Jacob
S485122 is offline   Reply With Quote
Old 2007-03-10, 22:43   #11
tha
 
tha's Avatar
 
Dec 2002

2×11×37 Posts
Default

Quote:
Originally Posted by S485122 View Post
I benchmarked one machine quite extensively and came to the conclusion that results have a big standard deviation.
Here are the timings of my machine:


Compare your results to other computers at http://www.mersenne.org/bench.htm
Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz
CPU speed: 2400.31 MHz, 2 cores
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 32 KB
L2 cache size: 4096 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
Prime95 32-bit version 25.1, RdtscTiming=1


Compare your results to other computers at http://www.mersenne.org/bench.htm
That web page also contains instructions on how your results can be included.

Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz
CPU speed: 2399.92 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 32 KB
L2 cache size: unknown
L1 cache line size: 64 bytes
L2 cache line size: unknown
Prime95 32-bit version 24.14, RdtscTiming=1
Best time for 4K FFT length: 0.054 ms.
Best time for 5K FFT length: 0.080 ms.
Best time for 6K FFT length: 0.095 ms.
Best time for 7K FFT length: 0.115 ms.
Best time for 8K FFT length: 0.119 ms.
Best time for 10K FFT length: 0.168 ms.
Best time for 12K FFT length: 0.202 ms.
Best time for 14K FFT length: 0.246 ms.
Best time for 16K FFT length: 0.257 ms.
Best time for 20K FFT length: 0.349 ms.
Best time for 24K FFT length: 0.424 ms.
Best time for 28K FFT length: 0.525 ms.
Best time for 32K FFT length: 0.547 ms.
Best time for 40K FFT length: 0.720 ms.
Best time for 48K FFT length: 0.872 ms.
Best time for 56K FFT length: 1.068 ms.
Best time for 64K FFT length: 1.140 ms.
Best time for 80K FFT length: 1.608 ms.
Best time for 96K FFT length: 1.955 ms.
Best time for 112K FFT length: 2.369 ms.
Best time for 128K FFT length: 2.500 ms.
Best time for 160K FFT length: 3.063 ms.
Best time for 192K FFT length: 3.783 ms.
Best time for 224K FFT length: 4.498 ms.
Best time for 256K FFT length: 4.991 ms.
Best time for 320K FFT length: 6.381 ms.
Best time for 384K FFT length: 7.851 ms.
Best time for 448K FFT length: 9.439 ms.
Best time for 512K FFT length: 10.548 ms.
Best time for 640K FFT length: 14.238 ms.
Best time for 768K FFT length: 17.527 ms.
Best time for 896K FFT length: 20.949 ms.
Best time for 1024K FFT length: 23.175 ms.
Best time for 1280K FFT length: 29.503 ms.
Best time for 1536K FFT length: 35.997 ms.
Best time for 1792K FFT length: 42.803 ms.
Best time for 2048K FFT length: 47.554 ms.
Best time for 2560K FFT length: 62.782 ms.
Best time for 3072K FFT length: 76.475 ms.
Best time for 3584K FFT length: 92.202 ms.
Best time for 4096K FFT length: 103.410 ms.
Best time for 5120K FFT length: 140.487 ms.
Best time for 6144K FFT length: 170.367 ms.
Best time for 7168K FFT length: 206.901 ms.
Best time for 8192K FFT length: 226.137 ms.
Best time for 10240K FFT length: 302.304 ms.
Best time for 12288K FFT length: 365.883 ms.
Best time for 14336K FFT length: 442.444 ms.
Best time for 16384K FFT length: 485.851 ms.
Best time for 20480K FFT length: 657.477 ms.
Best time for 24576K FFT length: 793.154 ms.
Best time for 28672K FFT length: 945.729 ms.
Best time for 32768K FFT length: 1032.329 ms.
Best time for 58 bit trial factors: 4.724 ms.
Best time for 59 bit trial factors: 4.660 ms.
Best time for 60 bit trial factors: 4.621 ms.
Best time for 61 bit trial factors: 4.659 ms.
Best time for 62 bit trial factors: 7.436 ms.
Best time for 63 bit trial factors: 7.434 ms.
Best time for 64 bit trial factors: 7.121 ms.
Best time for 65 bit trial factors: 7.081 ms.
Best time for 66 bit trial factors: 7.082 ms.
Best time for 67 bit trial factors: 7.045 ms.

Last fiddled with by tha on 2007-03-10 at 23:03
tha is offline   Reply With Quote
Reply



Similar Threads
Thread Thread Starter Forum Replies Last Post
Core i5 2500K vs Core i7 2600K (Linear algebra phase) em99010pepe Hardware 0 2011-11-11 15:18
exclude single core from quad core cpu for gimps jippie Information & Answers 7 2009-12-14 22:04
Optimising work for Intel Core 2 Duo or Quad Core S485122 Software 0 2007-05-13 09:15
Intel e6600 Dual Core Problem - How to use both cores with Prime95? Shoallakeboy Hardware 2 2006-11-06 17:55
4 cores in 1,Conroe E6600 8Mb Cache,Kensfield result@MSI 975X IronBits Hardware 4 2006-09-17 15:34

All times are UTC. The time now is 00:03.


Sat Jul 17 00:03:28 UTC 2021 up 49 days, 21:50, 1 user, load averages: 1.22, 1.48, 1.47

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.