| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Help Cure Muscular Dystrophy - Phase 2 Forum
Thread: Strange "Points Awarded" |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 13
|
|
| Author |
|
|
joeperry39@gmail.com
Advanced Cruncher USA Joined: Nov 22, 2006 Post Count: 140 Status: Offline Project Badges: 
|
This one of the strangest things I have seen on these projects. Take a look at the 3 bold type fields -- they are, in order, elapsed time of run, claimed points and granted points. I have never seen such a wild discrepancy in the numbers.
----------------------------------------Workunit Status Project Name: Help Cure Muscular Dystrophy - Phase 2 Created: 6/17/10 Name: CMD2_0610-EZRIA.clustersOccur-1XM9_A.clustersOccur_85 Minimum Quorum: 2 Replication: 2 CMD2_0610-EZRIA.clustersOccur-1XM9_ A.clustersOccur_ 85_1 614 Valid 6/25/10 13:11:47 6/26/10 19:26:16 6.00 68.0/37.4 CMD2_0610-EZRIA.clustersOccur-1XM9_ A.clustersOccur_ 85_0 614 Valid 6/25/10 13:09:03 6/27/10 21:55:11 4.69 83.6/462.0 BTW, I'm not complaining -- I'm the one that got the highest points but I still can't figure out what would cause such a huge difference. Any thoughts or ideas how this would happen?   "Everything in moderation, including moderation" -- Mark Twain [Edit 1 times, last edit by osugrad at Jun 27, 2010 10:50:36 PM] |
|||
|
||||
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Sorry, but we've seen even greater aberrations. It's driven by wingman A with Client/OS/benchmark offness, meeting wingman B with client version/OS/benchmark offness and mostly 1 doing [small] part and the other doing whole. After, for some the over award of credit persists for about 10-15 more tasks before they normalize again.
----------------------------------------WCG is not going to do anything... just take it and crunch crunch crunch... until the [in my eyes] fata morganan new scheme under development at Berkeley hits ground right before our eyes and starts awarding credit based on actual work and resources used... [To me it's preposterous that a 8MB RAM using science gives same credit as a 2 GB reserving science, that looses time due for instance HD exchanges]. So sit back, relax and wait... it's going to be a long hot summer... most all Meteorological services around the world are saying so.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Of course, never talk about the horned man as then he strikes out:
----------------------------------------CMD2_ 0609-EZRIA.clustersOccur-3BYH_ A.clustersOccur_ 135_ 399834_ 400515_ 400455_ 400515_ 1-- 614 Valid 26-6-10 19:33:23 28-6-10 06:04:29 6.09 59.7 / 83.4 CMD2_ 0609-EZRIA.clustersOccur-3BYH_ A.clustersOccur_ 135_ 399834_ 400515_ 400455_ 400515_ 0-- 614 Valid 26-6-10 19:32:36 28-6-10 03:08:47 6.74 210.8 / 164.1 Think both will still be quite happy... 24.34 credit per hour for the fast cruncher.
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Jun 28, 2010 8:06:53 AM] |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
PS: My slow got 13.69 p/h where it always benchmarks at around 10... would not be surprised the fast one was an hyperthreaded I7 on 64 bit OS... log only reveals it's a 6.10.18 client.
----------------------------------------
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Well my usually fast HCMD2 executing quad was now deemed a slug...
----------------------------------------CMD2_ 0611-EZRIA.clustersOccur-1YDI_ A.clustersOccur_ 84_ 1-- 614 Valid 6/27/10 08:52:06 6/27/10 19:28:53 8.20 143.2 / 312.8 CMD2_ 0611-EZRIA.clustersOccur-1YDI_ A.clustersOccur_ 84_ 0-- 614 Valid 6/27/10 08:51:38 6/28/10 11:40:11 6.00 132.7 / 86.0 < My Crawler To consider that the 6 hour cut-off could have been just barely shy of 60%. Regrettably the result logs to not show the positions computed to even attempt a fair reconstruction. Time and credit claims just does not work. The award infers only 27% of the faster device was processed. Well there are no 4x faster devices yet... plain and simple... but the HW boys are appealed to, to not take this off-topic.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Dataman
Ace Cruncher Joined: Nov 16, 2004 Post Count: 4865 Status: Offline Project Badges: 
|
PS: My slow got 13.69 p/h where it always benchmarks at around 10... would not be surprised the fast one was an hyperthreaded I7 on 64 bit OS... log only reveals it's a 6.10.18 client. It makes no sense to me Sek. My I7's with Win7 64-bit are always the low ones. I quit crunching it at 3 years and moved on to other projects. I agree with you that WCG is not going to do anything about it.   |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
low or slow? I meant to say that the one doing the whole task at 6.74, yet meating the grim credit slasher may be a I7-HT... higher claim, lower award. If I understand correctly you're saying that your getting less than the 6.09 hours / cut-off wingman?
----------------------------------------
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Dataman
Ace Cruncher Joined: Nov 16, 2004 Post Count: 4865 Status: Offline Project Badges:
|
Yes. Low. Not implyin it is often as it is rare but just that I seem to be the low wingman and I have some rather good machines. No example as I am not crunching this project and have rolled off stat's.
----------------------------------------When I was in college I should have taken a course in BOINC/WCG Math.   |
||
|
|
KerSamson
Master Cruncher Switzerland Joined: Jan 29, 2007 Post Count: 1684 Status: Offline Project Badges: 
|
Hi,
----------------------------------------I cannot explain this "Fata Morgana" but I can perhaps add some observations.
This host is reliable and it came back to "normal" claim afterward. Cheers, Yves |
||
|
|
Hypernova
Master Cruncher Audaces Fortuna Juvat ! Vaud - Switzerland Joined: Dec 16, 2008 Post Count: 1908 Status: Offline Project Badges: 
|
"Fata Morgana" in the arthurian legends had a love affair with king Arthur.
----------------------------------------It is also an expression for optical illusions in italian. "Mirage" in french. Anyway she is a magician and as you know, only magical powers can change things in our cruncher's world. Or maybe everything is an illusion   [Edit 2 times, last edit by Hypernova at Jul 1, 2010 2:02:37 PM] |
||
|
|
|