| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: Is my computer a malicious user?! |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 18
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Am I getting the short end of the stick or does my computer tries to cheat?!
 "This process eliminates the ability for malicious users to artificially claim higher points for their work." http://www.worldcommunitygrid.org/help/viewTopic.do?shortName=points#177 I'm working on all the projects not one specifically and it's the same for all of them: my computer claim about +/- 15% more than it gets 9 times out of 10 (I think it would be even closer to 20% if I took the time to look at all the results). I understand there is an average being done on all claims associated with one work unit and that's why I get what I get but why does my computer always claim so much more than the other computers, even when it has similar CPU time? I've put my claim first for comparison purposes. CPU Time | Claimed / Granted Example 1 9.81 | 67.5 / 62.5 (+8% claimed) 7.37 | 56.0 / 62.5 6.55 | 64.1 / 62.5 Example 2 17.60 | 83.0 / 72.2 (+15 claimed) 14.75 | 72.9 / 72.2 7.74 | 60.5 / 72.2 Example 3 1.93 | 19.3 / 15.7 (+23% claimed) 2.26 | 13.8 / 15.7 2.16 | 13.9 / 15.7 Computer is a laptop: T2500@2.00Ghz w/ 2GB of Ram BOINC CPU benchmark (three times; first time there were a lot of programs opened): 2 CPUs 1609 Whetstone per CPU 3661 Dhrystone per CPU 1795 Whetstone per CPU 3538 Dhrystone per CPU 1792 Whetstone per CPU 3559 Dhrystone per CPU Thanks for helping me understand why this is happening. * And also, I guess there is no point fixing anything as any lower claim by my computer will only result in a lower average for everyone, but I still don't understand why my computer claims more than the rest. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Looks to be well within the normal variation. Don't give it another thought.
If you have a very efficient cache or fast memory, the benchmark tends to come out a bit high. There are all sorts of other variables, too. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Not sure about the normal variation, maybe the results I gave as example were not very impressive. Here are some more:
CLAIMED / GRANTED: 124.9 / 77.7 (+59% claimed) 163.7 / 66.7 (+145% claimed) 163.2 / 76.0 (+115% claimed) 161.1 / 63.9 (+152% claimed) 109.6 / 65.2 (+68% claimed) 95.0 / 53.4 (+78% claimed) 95.8 / 58.8 (+63% claimed) 99.6 / 61.5 (+62% claimed) 130.2 / 87.8 (+48% claimed) 124.9 / 81.7 (+53% claimed) 94.8 / 60.8 (+56% claimed) 101.8 / 67.7 (+50% claimed) 97.1 / 61.2 (+59% claimed) 121.8 / 82.8 (+47% claimed) Most of these results were returned in the last week and most of them were from the Human Proteome Folding 2. As it is, I have decided to stop participating in the Human Proteome Folding 2 as it seems this project does not work very well with my machine. Although the claim/granted difference is astronomic in the case of the HPF2 project, there is still a problem with the other projects as well. I think all this project does is showing that this is not within the normal variation and there is a problem with how my machine request its claims based on I don't know what. Wish I knew, if not for fixing it at least just to understand. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Time for some revelations:
----------------------------------------1.OS? 2.BOINC version? 3.Stock, home compile or other 3rd party compile like Crunch3r? 4. Any other project outside of WCG you participate in? 5. If so, similar percentages deviation? I can tell you straight up, there is not a single WCG job that, apart for rare exemptions, will grant claimed credits that we see here. They bare no relation to the benchmarks and typical job durations at WCG. My C2D in the ballpark is very close to what you have here and rarely claims over 80/85 on a longer FAAH job.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
retsof
Former Community Advisor USA Joined: Jul 31, 2005 Post Count: 6824 Status: Offline Project Badges: 
|
This method was just added recently. It could be swinging wildly if a new user shows up. It should settle down as more workunits are submitted.
----------------------------------------I brought up a question of a computer that had been upgraded and/or overclocked, but the techs weren't too worried about it.
SUPPORT ADVISOR
Work+GPU i7 8700 12threads School i7 4770 8threads Default+GPU Ryzen 7 3700X 16threads Ryzen 7 3800X 16 threads Ryzen 9 3900X 24threads Home i7 3540M 4threads50% |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
You said it retsof..... an existing member, since oct-2006 should not see it and i certainly don't see it. As knreed explained, nothing should be effected but for a few percentage points of the members at the outlier ends.
----------------------------------------Very interested to get the 5 point info. edit: "do" should have read "don't" 
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Apr 28, 2007 6:15:06 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Time for some revelations:
1.OS? WINDOWS XP PRO SP2 2.BOINC version? 5.4.11 3.Stock, home compile or other 3rd party compile like Crunch3r? Hum, no idea... I guess not 4. Any other project outside of WCG you participate in? NO 5. If so, similar percentages deviation? N/A I can upggrade BOINC and I can also participate in another project if that can helps |
||
|
|
retsof
Former Community Advisor USA Joined: Jul 31, 2005 Post Count: 6824 Status: Offline Project Badges:
|
BOINC 5.4.11 is a decent version, but is getting rather old. I have been running 5.8.15 for awhile, so it is quite stable. They have done a lot with the statistics.
----------------------------------------5.8.16 is the current production release, which has some more checkpoint info in the messages. 5.8.17 is experimental, and has had some problems. Stay away from that one. 5.6 and above have file compression. Since WCG does that now, that should save time uploading and downloading. There could be some differences that require a newer BOINC, like the simplified interface, etc. etc. Upgrading BOINC couldn't hurt, and is fairly painless. After downloading the new version, you aren't supposed to have to stop your workunits any more, but I always do and exit before upgrading. It's always worked.
SUPPORT ADVISOR
----------------------------------------Work+GPU i7 8700 12threads School i7 4770 8threads Default+GPU Ryzen 7 3700X 16threads Ryzen 7 3800X 16 threads Ryzen 9 3900X 24threads Home i7 3540M 4threads50% [Edit 2 times, last edit by retsof at Apr 28, 2007 3:07:42 AM] |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Hi UofC,
----------------------------------------Odd indeed these wild claims, particular the 160+. The new rules of computing the BOINC 'claimed' credit should normalize after so many results on 1 machine as the servers now keep a history and compare that in addition. One of the progresses of the latest system is, that no more 50% penalty is given, just the 'fair' credit. Stock means: Downloaded from WCG or the official page: http://boinc.berkeley.edu/download_all.php Home Compile means: BOINC is open source and some people compile the executables themselves, with sometimes wrong settings. 3rd party compile means: The mentioned inflates the benchmark for the Integer part for instance by factor 3, making a 3.3ghz machine calculate a claim as were is a 10ghz machine (which don't exist of course). Not appreciated for use at most projects!!! Didactylos had an experience of a damaged L2 cache on a CPU turning his machine into an overclaimer, but as said your T2500 2ghz does a consistent benchmark very much in line with it's abilities. Wait it out a few days more and advice if still seeing the exessive deviations. cheers
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Apr 28, 2007 6:32:02 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Then I would have stock. I downloaded the new version; will let it work for a week and see.
Thanks. |
||
|
|
|