| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: The Clean Energy Project - Phase 2 Forum
Thread: SuperComputer Or ? |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 8
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Is this some sort of record for the same task, mine is the 8hrs+ on a Q6600 it has done them a bit faster sometimes but thing is quick.
----------------------------------------E202520_ 184_ C.27.C19H9N5OS2.00223625.4.set1d06_ 1-- 640 Valid 6/26/11 21:29:38 6/27/11 07:47:01 ( 8.30 ) 202.0 / 201.4 E202520_ 184_ C.27.C19H9N5OS2.00223625.4.set1d06_ 0-- 640 Valid 6/26/11 21:28:06 6/27/11 00:50:02 ( 2.99 ) 200.8 / 201.4  [Edit 1 times, last edit by Former Member at Jun 28, 2011 5:51:06 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Not entirely sure, but doubt. CEP2 tasks on occasion continue to loose the proper CPU seconds on restore point of a job. Open the wingman log and see the wallclock time of start and end to get a more correct indication [ consider efficiency isnot 100%]. Had a wingman recently who had 976 seconds until the 15th job and ended with 2.23 hours after completion of the 16th, though on wallclock it took 8 hours.
--//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi SekeRob.
This is from there result info, looks close to me don't know for sure. <core_client_version>6.10.59</core_client_version> <![CDATA[ <stderr_txt> INFO: No state to restore. Start from the beginning. [06:34:35] Number of jobs = 16 [06:34:35] Starting job 0,CPU time has been restored to 0.000000. [09:00:28] Starting job 15,CPU time has been restored to 8289.880000. [09:00:28] Starting new Job [09:00:28] Qink name = fldman [09:00:29] Qink name = gesman [09:00:29] Qink name = scfman [09:37:02] Qink name = anlman [09:43:40] End of Job [09:43:41] Finished Job #15 09:43:44 (3177): called boinc_finish |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Yes, the log confirms that your Q6600 is no where near the performance of that wingman.
Subdued [as of now] Happy Crunching ;P --//-- |
||
|
|
mikaok
Senior Cruncher Finland Joined: Aug 8, 2006 Post Count: 489 Status: Offline Project Badges: 
|
Yes, the log confirms that your Q6600 is no where near the performance of that wingman. If we compare Q6600 and 2600K with points they contribute in a day, then SB might be around that much quicker. My Q6600 (@3.0GHz /w linux 64bit and HCC tasks) gets around 21k ppd. Here is 2600K @4.1GHz reported to be making 38k ppd. Hardware just gets old quickly.
to infinity and beyond
---------------------------------------- [Edit 1 times, last edit by mikaok at Jun 28, 2011 12:04:17 PM] |
||
|
|
Jack007
Master Cruncher CANADA Joined: Feb 25, 2005 Post Count: 1604 Status: Offline Project Badges: 
|
I bought a Q6600 5 years ago, it IS old :)
----------------------------------------but still has crunching potential  |
||
|
|
Rickjb
Veteran Cruncher Australia Joined: Sep 17, 2006 Post Count: 666 Status: Offline Project Badges: 
|
@Mikaok: "Here is 2600K @4.1GHz reported to be making 38k ppd."
The 2600K would be using Intel's Hyperthreading (HT). The WCG/BOINC points award system does not deal with HT properly, and awards many more points than is deserved by the true WU throughput of these machines. I'm not saying that the 2600K is no faster than a Q6600, it's just that you need to compare throughputs of WUs per day, with the same O/S and project mix. I don't have a 2600K (yet), but here are a few pointers to ccomparing throughputs: - The 45nm high-end Intel Yorkfields, eg Q9550 & Q9650, are about 5% faster, clock-for-clock, than your Q6600. (I'm running 2 of these). - Movieman's i7-920, with HT enabled, was about 30% faster, c-4-c, than my Q9650, on the old HCC under Windows. - Sandy Bridge CPUs are considered to be about 5% faster. c-4-c, than the i7-9xx series. - Since you are overclocking your Q6600, it's relevant to mention that the later CPUs can be run faster. 3.0GHz is a low-end overclock for a Q6600, and the equivalent for SB would be about 4.0GHz. And the SB at this speed would consume less power. 21kppd is very high for a Q6600, especially @3.0GHz, and must be due to the efficiency of the current newer HCC software running under Linux-x64. 15-16kppd was considered the normal range for a Q6600 @3.6GHz under Windows. 2600K @ 4.1GHZ + HT + HCC + Linux-x64 might give interesting numbers. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I bought a Q6600 5 years ago, it IS old :) but still has crunching potential The Q6600 will be a workhorse for years to come. Mine is as old and it would get retired quickly if there was a system around doing half the watt use and equal or more WU crunch power... mine is paid for with 2.5 years of electricity consumption i.e. including operating cost I've now bought 3 Q6600's.  --//-- |
||
|
|
|