| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: Show us your Boinc Benchmarks. ( Comparison ) |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 119
|
|
| Author |
|
|
sk..
Master Cruncher http://s17.rimg.info/ccb5d62bd3e856cc0d1df9b0ee2f7f6a.gif Joined: Mar 22, 2007 Post Count: 2324 Status: Offline Project Badges: 
|
Vista Ultimate x64 SP1
Boinc 6.10.58 CPU: Phenom II 940 @ 3.0GHz K9A2 Plat board 4GB 800MHz DDR2 4 X GT240 @ 1600MHz Benchmark results Number of CPUs: 4 2450 floating point MIPS (Whetstone) per CPU 7612 integer MIPS (Dhrystone) per CPU |
||
|
|
Hypernova
Master Cruncher Audaces Fortuna Juvat ! Vaud - Switzerland Joined: Dec 16, 2008 Post Count: 1908 Status: Offline Project Badges: 
|
mikaok
----------------------------------------Processor: Intel(R) Core(TM)2 Quad CPU Q6600 G0 @ 3.00GHz OS: Microsoft Windows Vista x64, Service Pack 2 BOINC: 6.10.18 windows_x86_64 Number of CPUs: 4 2846 floating point MIPS (Whetstone) per CPU 8770 integer MIPS (Dhrystone) per CPU evilkats BOINC client version 6.10.58 for x86_64-pc-linux-gnu Processor: 4 GenuineIntel Intel(R) Core(TM)2 Quad CPU Q9550 @ 2.83GHz [Family 6 Model 23 Stepping 7] 2769 floating point MIPS (Whetstone) per CPU 14393 integer MIPS (Dhrystone) per CPU  I sense a problem here. Models and frequencies make the MIPS Dhrystone value of evilkats seem strange. It is even higher than my hyperprocessor.  evilkats any particularities on your rig, as your value seems stellar.  [Edit 1 times, last edit by Hypernova at Oct 7, 2010 2:31:20 PM] |
||
|
|
TimAndHedy
Senior Cruncher Joined: Jan 27, 2009 Post Count: 267 Status: Offline Project Badges:
|
I sense a problem here. Models and frequencies make the MIPS Dhrystone value of evilkats seem strange. Linux is much better at it. |
||
|
|
Hypernova
Master Cruncher Audaces Fortuna Juvat ! Vaud - Switzerland Joined: Dec 16, 2008 Post Count: 1908 Status: Offline Project Badges:
|
Linux is much better at it Fine but how. The code is compiled for the CPU. Once Linux has loaded it in RAM and it executes it is a matter of processor performance. I am talking only of the benchmarks. It is CPU and RAM. It is hardware that runs the benchmark not Linux. The only thing I see is that the executable code compiled under Linux is much more effective that the one compiled under Windows. But why. The code is generated for a given processors using the right instruction libraries available to carry the mathematical calculations. I really miss here a clear explanation. |
||
|
|
sk..
Master Cruncher http://s17.rimg.info/ccb5d62bd3e856cc0d1df9b0ee2f7f6a.gif Joined: Mar 22, 2007 Post Count: 2324 Status: Offline Project Badges:
|
Kubuntu x64 10.04
Boinc 6.10.58 CPU: Q6600 @ 2.4GHz Asus P5Q board 4GB 800MHz DDR2 GTX260-216 FOC (modest) Benchmark results Number of CPUs: 4 2166 floating point MIPS (Whetstone) per CPU 7891 integer MIPS (Dhrystone) per CPU |
||
|
|
anhhai
Veteran Cruncher Joined: Mar 22, 2005 Post Count: 839 Status: Offline Project Badges: 
|
this is what a few of my system looks like:
----------------------------------------Benchmark results: Number of CPUs: 16 2538 floating point MIPS (Whetstone) per CPU 6509 integer MIPS (Dhrystone) per CPU quantity vs quality I guess  |
||
|
|
evilkats
Senior Cruncher USA Joined: May 4, 2007 Post Count: 162 Status: Offline Project Badges: 
|
evilkats BOINC client version 6.10.58 for x86_64-pc-linux-gnu Processor: 4 GenuineIntel Intel(R) Core(TM)2 Quad CPU Q9550 @ 2.83GHz [Family 6 Model 23 Stepping 7] 2769 floating point MIPS (Whetstone) per CPU 14393 integer MIPS (Dhrystone) per CPU I sense a problem here. Models and frequencies make the MIPS Dhrystone value of evilkats seem strange. It is even higher than my hyperprocessor. evilkats any particularities on your rig, as your value seems stellar. It is strange becase the same banchmarks on the Windows XP are 2977 floating point MIPS (Whetstone) per CPU 5923 integer MIPS (Dhrystone) per CPU These came from the 2 identical HP desktops. One is running Windows XP, and on the other I installed Ubuntu 10.04 64-bit. What I do notice on Linux, is that my claimed points are almost always higher than what I receive.   |
||
|
|
sk..
Master Cruncher http://s17.rimg.info/ccb5d62bd3e856cc0d1df9b0ee2f7f6a.gif Joined: Mar 22, 2007 Post Count: 2324 Status: Offline Project Badges:
|
XP x86 SP3
Boinc 6.10.58 CPU: i7-920 @ 2.66GHz & No turbo 2 X GTX470 @ 715MHz Benchmark results Number of CPUs: 8 2364 floating point MIPS (Whetstone) per CPU 6350 integer MIPS (Dhrystone) per CPU |
||
|
|
TimAndHedy
Senior Cruncher Joined: Jan 27, 2009 Post Count: 267 Status: Offline Project Badges:
|
Linux is much better at it Fine but how. The code is compiled for the CPU. Once Linux has loaded it in RAM and it executes it is a matter of processor performance. I am talking only of the benchmarks. It is CPU and RAM. It is hardware that runs the benchmark not Linux. The only thing I see is that the executable code compiled under Linux is much more effective that the one compiled under Windows. But why. The code is generated for a given processors using the right instruction libraries available to carry the mathematical calculations. I really miss here a clear explanation. Code Libraries and compilers! No one writes an entire program. The program they write calls code. written by others. The efficiency of that other code makes a difference. http://en.wikipedia.org/wiki/Library_%28computing%29 People write abstract code to solve problems. The efficiency of the compiler determines the exact steps that are executed on a CPU. How efficient a compiler is makes a big difference in the speed of the created program. These are different in Windows and Linux. http://en.wikipedia.org/wiki/Compiler As to the specifics of what makes Windows so inefficient, this would take some careful research, which is made difficult without the source code. |
||
|
|
macusen
Advanced Cruncher Germany Joined: Mar 31, 2009 Post Count: 59 Status: Offline Project Badges: 
|
Well, I'm wondering that my results are not as bad as I thought...
----------------------------------------Although my Macbook is already 3 years old, the results don't seem to be SO bad in comparison... OS: Mac OS 10.6 32 Bit BOINC version: 6.10.17 CPU: Intel T7400 Core 2 Duo 2.16 GHz Number of CPUs: 2 2132 floating point MIPS (Whetstone) per CPU 6996 integer MIPS (Dhrystone) per CPU  |
||
|
|
|