4000 credit wus?

Message boards : Number crunching : 4000 credit wus?

To post messages, you must log in.

AuthorMessage
Profile Xaak

Send message
Joined: 20 Mar 06
Posts: 17
Credit: 3,701,702
RAC: 0
Message 47654 - Posted: 12 Oct 2007, 21:57:55 UTC

Sorry, but something isn't quite right here. Was perusing a few of the faster machines, and ran across this.

Several others on the same computer too:

https://boinc.bakerlab.org/rosetta/results.php?hostid=562651&offset=40

Is this a bug or has someone found a potential cheat?

XaaK



Join Us

XaaK



ID: 47654 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Jmarks
Avatar

Send message
Joined: 16 Jul 07
Posts: 132
Credit: 98,025
RAC: 0
Message 47655 - Posted: 12 Oct 2007, 22:11:25 UTC

If you look at the WU ids it say 'Rosetta Beta'
They report on seporate board, not here.
Jmarks
ID: 47655 · Rating: -2 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile dcdc

Send message
Joined: 3 Nov 05
Posts: 1829
Credit: 114,354,619
RAC: 52,117
Message 47656 - Posted: 12 Oct 2007, 22:19:15 UTC - in response to Message 47655.  
Last modified: 12 Oct 2007, 22:22:11 UTC

If you look at the WU ids it say 'Rosetta Beta'
They report on seporate board, not here.

it's a bit confusing (i posted saying I thought that label would cause problems when they first released 'beta' WUs!)

The 'beta' WUs aren't part of Ralph (the alpha test project) which is reported seperately. We're all running some of the beta-labelled WUs here - they're part of Rosetta and so those results will be counted here.

The error on those results is the phenomenal number of decoys being reported...

HTH
Danny
ID: 47656 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Nothing But Idle Time

Send message
Joined: 28 Sep 05
Posts: 209
Credit: 139,545
RAC: 0
Message 47657 - Posted: 12 Oct 2007, 22:21:25 UTC
Last modified: 12 Oct 2007, 22:22:20 UTC

How does one get more than 1000 decoys in just three hours? Tell me the secret; inquiring minds want to know.
ID: 47657 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile dcdc

Send message
Joined: 3 Nov 05
Posts: 1829
Credit: 114,354,619
RAC: 52,117
Message 47658 - Posted: 12 Oct 2007, 22:22:40 UTC
Last modified: 12 Oct 2007, 22:23:06 UTC

you have to have some corrupt files ;) or could this be a syncing issue between the cores?
ID: 47658 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Xaak

Send message
Joined: 20 Mar 06
Posts: 17
Credit: 3,701,702
RAC: 0
Message 47659 - Posted: 12 Oct 2007, 23:06:17 UTC

They're certainly being reported here, and they're calculated into the computer's RAC. And the computer's being granted well over 400 times what a similar processor is producing for the same runtime.
XaaK



ID: 47659 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
zombie67 [MM]
Avatar

Send message
Joined: 11 Feb 06
Posts: 316
Credit: 6,589,590
RAC: 2,906
Message 47663 - Posted: 13 Oct 2007, 0:25:45 UTC

Sweet! I want some of those too. =;^)
Reno, NV
Team: SETI.USA
ID: 47663 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Natronomonas

Send message
Joined: 11 Jan 06
Posts: 38
Credit: 536,978
RAC: 0
Message 47667 - Posted: 13 Oct 2007, 1:00:01 UTC - in response to Message 47663.  

The computer isn't claiming many credits, just being granted many... maybe because of the high decoy count, but it doesn't think they're worth that much...??
Crunching Rosetta as a member of the Whirlpool BOINC Teams
ID: 47667 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile dcdc

Send message
Joined: 3 Nov 05
Posts: 1829
Credit: 114,354,619
RAC: 52,117
Message 47668 - Posted: 13 Oct 2007, 1:05:05 UTC - in response to Message 47667.  

The computer isn't claiming many credits, just being granted many... maybe because of the high decoy count, but it doesn't think they're worth that much...??

yeah - the claim is based on benchmarks * time (both normal) while the granted is based on Nr decoys * average claim for that type of decoy.

HTH
Danny
ID: 47668 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Xaak

Send message
Joined: 20 Mar 06
Posts: 17
Credit: 3,701,702
RAC: 0
Message 47670 - Posted: 13 Oct 2007, 3:40:30 UTC

Ok, this is getting rediculous. an 8500 credit wu? Why only his machine?

https://boinc.bakerlab.org/rosetta/result.php?resultid=111686926

He has a number of errored wus, so either the client is registering errored decoys as successfully completed, or he's hacking the result to show way too many decoys.

Can someone from the project have a look and comment please?
XaaK



ID: 47670 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Xaak

Send message
Joined: 20 Mar 06
Posts: 17
Credit: 3,701,702
RAC: 0
Message 47671 - Posted: 13 Oct 2007, 5:25:38 UTC

Found another one:
https://boinc.bakerlab.org/rosetta/results.php?hostid=567404&offset=20

Maybe the credit/decoy ratio on these are way off?
XaaK



ID: 47671 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile dcdc

Send message
Joined: 3 Nov 05
Posts: 1829
Credit: 114,354,619
RAC: 52,117
Message 47672 - Posted: 13 Oct 2007, 9:45:55 UTC

Xaak, I doubt anyone from the project will look at it until after the weekend. They can change anything retrospectively though...
ID: 47672 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Jmarks
Avatar

Send message
Joined: 16 Jul 07
Posts: 132
Credit: 98,025
RAC: 0
Message 47674 - Posted: 13 Oct 2007, 11:32:08 UTC - in response to Message 47672.  

Xaak, I doubt anyone from the project will look at it until after the weekend. They can change anything retrospectively though...


They are all have 2168 in them. If they are not ralph's then we should post them to the Problems with 5.80?

Jmarks
ID: 47674 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Jmarks
Avatar

Send message
Joined: 16 Jul 07
Posts: 132
Credit: 98,025
RAC: 0
Message 47677 - Posted: 13 Oct 2007, 11:49:43 UTC - in response to Message 47674.  

Xaak, I doubt anyone from the project will look at it until after the weekend. They can change anything retrospectively though...


They are all have 2168 in them. If they are not ralph's then we should post them to the Problems with 5.80?


I just did.
Jmarks
ID: 47677 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Conan
Avatar

Send message
Joined: 11 Oct 05
Posts: 150
Credit: 3,795,306
RAC: 2,327
Message 47679 - Posted: 13 Oct 2007, 13:18:11 UTC - in response to Message 47671.  

Found another one:
https://boinc.bakerlab.org/rosetta/results.php?hostid=567404&offset=20

Maybe the credit/decoy ratio on these are way off?


I don't believe the owner knows anything about this as all the WU's making the huge claims all have the same name, they all start with mcr1, for example

mcr1__BOINC_SYMM_FOLD_AND_DOCK_RELAX-mcr1_-short_mfr__2168_67176_0

On the other side of the coin the same computer is getting a lot of stuck WU's that are being terminated by the Watchdog that start with

STM0082_BOINC_MFR_ABRELAX_PICKED_

and they get just 20 credits for each of these WU's.
ID: 47679 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Xaak

Send message
Joined: 20 Mar 06
Posts: 17
Credit: 3,701,702
RAC: 0
Message 47681 - Posted: 13 Oct 2007, 14:45:46 UTC - in response to Message 47672.  

Xaak, I doubt anyone from the project will look at it until after the weekend. They can change anything retrospectively though...


I didn't mean this minute :-).

XaaK



ID: 47681 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Mod.Sense
Volunteer moderator

Send message
Joined: 22 Aug 06
Posts: 4018
Credit: 0
RAC: 0
Message 47683 - Posted: 13 Oct 2007, 14:57:13 UTC

I sent an EMail last night to David Kim and request a review of the situation.
Rosetta Moderator: Mod.Sense
ID: 47683 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Xaak

Send message
Joined: 20 Mar 06
Posts: 17
Credit: 3,701,702
RAC: 0
Message 47686 - Posted: 13 Oct 2007, 15:50:50 UTC - in response to Message 47683.  

I sent an EMail last night to David Kim and request a review of the situation.


Thanks!

XaaK



ID: 47686 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote

Message boards : Number crunching : 4000 credit wus?



©2024 University of Washington
https://www.bakerlab.org