FPU / Integer Performance

Questions and Answers : Unix/Linux : FPU / Integer Performance
Message board moderation

To post messages, you must log in.

AuthorMessage
Dotsch
Volunteer tester
Avatar

Send message
Joined: 9 Jun 99
Posts: 2422
Credit: 919,393
RAC: 0
Germany
Message 227482 - Posted: 7 Jan 2006, 12:37:09 UTC

Wich boinc version do you use on windows and linux ? - If you use different boinc versions, it could be that the benchmarks differs.
Can you please post the output of /proc/cpuinfo.
ID: 227482 · Report as offensive
Profile Crunch3r
Volunteer tester
Avatar

Send message
Joined: 15 Apr 99
Posts: 1546
Credit: 3,438,823
RAC: 0
Germany
Message 228573 - Posted: 9 Jan 2006, 22:48:48 UTC - in response to Message 228543.  
Last modified: 9 Jan 2006, 22:50:23 UTC

Wich boinc version do you use on windows and linux ? - If you use different boinc versions, it could be that the benchmarks differs.
Can you please post the output of /proc/cpuinfo.


cpuinfo: line 1: processor: command not found
cpuinfo: line 2: vendor_id: command not found
cpuinfo: line 3: cpu: command not found
cpuinfo: line 4: model: command not found
cpuinfo: line 5: syntax error near unexpected token `('
cpuinfo: line 5: `model name : Intel(R) Pentium(R) 4 CPU 1.60GHz'

---------------------------------

as for the linux boinc client, I am running 5.2.13. When I was running XP on it, it had 4.25 at the time the benchmarks were run.


This looks strange. You you try a "cat /proc/cpuinfo".
And if you're not happy with the default boinc benchmarks then you could try one of my versions.

Just click on the link in my sig.


Join BOINC United now!
ID: 228573 · Report as offensive
Profile Crunch3r
Volunteer tester
Avatar

Send message
Joined: 15 Apr 99
Posts: 1546
Credit: 3,438,823
RAC: 0
Germany
Message 228793 - Posted: 10 Jan 2006, 10:24:43 UTC - in response to Message 228606.  

Wich boinc version do you use on windows and linux ? - If you use different boinc versions, it could be that the benchmarks differs.
Can you please post the output of /proc/cpuinfo.


cpuinfo: line 1: processor: command not found
cpuinfo: line 2: vendor_id: command not found
cpuinfo: line 3: cpu: command not found
cpuinfo: line 4: model: command not found
cpuinfo: line 5: syntax error near unexpected token `('
cpuinfo: line 5: `model name : Intel(R) Pentium(R) 4 CPU 1.60GHz'

---------------------------------

as for the linux boinc client, I am running 5.2.13. When I was running XP on it, it had 4.25 at the time the benchmarks were run.


This looks strange. You you try a "cat /proc/cpuinfo".
And if you're not happy with the default boinc benchmarks then you could try one of my versions.

Just click on the link in my sig.


i'm a newb to linux, so i'm figuring this out as we go along. here is with the cat /proc/cpuinfo


processor : 0
vendor_id : GenuineIntel
cpu family : 15
model : 2
model name : Intel(R) Pentium(R) 4 CPU 1.60GHz
stepping : 4
cpu MHz : 1595.766
cache size : 512 KB
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 2
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm
bogomips : 3153.92

as for the benchmark results, i'm happy with whatever is normal. I'm just curious if my version of linux (Fedora Core 4) or boinc itself just has lousy processor support, or if windows has a tendency of overstating it's numbers.


This looks much better :)

As for the benchmarks, the windows compiler itself tends to optimize code a bit better with default settings. That's why there are optimized versions for linux to close the gap between windows and linux.

I suggest you to try this version --> http://www.bm-makler.de/setiathome/boinc/pc/boinc_5.2.14_SSE2.tar.bz2




Join BOINC United now!
ID: 228793 · Report as offensive

Questions and Answers : Unix/Linux : FPU / Integer Performance


 
©2025 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.