Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go »
No member browsing this thread
Thread Status: Active
Total posts in this thread: 286
Posts: 286   Pages: 29   [ Previous Page | 9 10 11 12 13 14 15 16 17 18 | Next Page ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 47172 times and has 285 replies Next Thread
pututu
Senior Cruncher
United States
Joined: Jan 3, 2016
Post Count: 241
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: THOR CHALLENGE [Y5-2020] by team CSF

Hi Thomas,

For the Thor Pure Challenge, only the credits obtained during the fifth week of the challenge and not the cumulative credits from the first to fifth week are counted, correct?

Edit. I was referring to this post:

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

----------------------------------------
[Edit 1 times, last edit by pututu at Sep 14, 2020 3:52:02 PM]
[Sep 14, 2020 3:25:00 PM]   Link   Report threatening or abusive post: please login first  Go to top 
[CSF] Thomas H.V. DUPONT
Master Cruncher
Réunion-France
Joined: Aug 25, 2013
Post Count: 1512
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: THOR CHALLENGE [Y5-2020] by team CSF

Thank you Thomas, however, I was looking for the top 50 after week 1.

Miklos, my link is the ranking of the 50 best teams at the end of the first week of challenge.
----------------------------------------
[Sep 14, 2020 4:03:16 PM]   Link   Report threatening or abusive post: please login first  Go to top 
[CSF] Thomas H.V. DUPONT
Master Cruncher
Réunion-France
Joined: Aug 25, 2013
Post Count: 1512
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: THOR CHALLENGE [Y5-2020] by team CSF

Have the week 2 invites gone out yet, I haven't recevied it yet.

Invitations will be sent out tomorrow, Yankton (a change from Monday to Tuesday for technical and logistical reasons). A post in this thread will be posted as soon as the invitations have been sent.
----------------------------------------
[Sep 14, 2020 4:06:06 PM]   Link   Report threatening or abusive post: please login first  Go to top 
[CSF] Thomas H.V. DUPONT
Master Cruncher
Réunion-France
Joined: Aug 25, 2013
Post Count: 1512
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: THOR CHALLENGE [Y5-2020] by team CSF

Hi pututu, only the WCG points obtained during the fifth week are counted.
But to be selected for this 5th week, the cumulative run time of the first three weeks will be counted.
----------------------------------------
[Sep 14, 2020 4:10:21 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Michael Goetz
Cruncher
United States
Joined: Dec 11, 2017
Post Count: 35
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: THOR CHALLENGE [Y5-2020] by team CSF

My team is participating in this challenge for the first time, and we're having a lot of fun. Thank you!

I am curious, however, about exactly how the scores are determined. I have a few questions:

1) Does the scoring count elapsed time, or CPU time? I believe it counts CPU time, correct?

2) Are tasks counted when they're returned, or when they're validated? If they're counted when they're returned, it might be easy to spoof the results and cheat by quickly returning fabricated results. On the other hand, if it's validated results, your scores are at the mercy of when your wingman decides to return his or her tasks.

3) For the first four weeks, which are based on time, how do we know that we're competing on a level playing field? It seems like a trivial exercise to recompile the BOINC client and have it inflate the elapsed and CPU times (plus a few other things to make it look "real" so it's not easily detectable. I won't mention what those are.) Essentially, it seems very easy to cheat. I know I'm not cheating. How do I know everyone else also isn't cheating?
[Sep 14, 2020 4:13:57 PM]   Link   Report threatening or abusive post: please login first  Go to top 
adriverhoef
Master Cruncher
The Netherlands
Joined: Apr 3, 2009
Post Count: 2070
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: THOR CHALLENGE [Y5-2020] by team CSF

You are probably expecting answers from Thomas, Michael; nevertheless I want to try to give some myself.
2) Are tasks counted when they're returned, or when they're validated? If they're counted when they're returned, it might be easy to spoof the results and cheat by quickly returning fabricated results.
While returning fabricated results would be possible, it seems more plausible that spoofed results will be marked as Invalid or even as an Error by the validation process. Even worse, returning spoofed results that wouldn't be spotted by the validator would not help the sciences at all, it would be counterproductive, it would spoil, rot and damage the whole advancement of the sciences. So, let's hope that spoofed results are spotted by the validator.

3) […about recompiling the software…] Essentially, it seems very easy to cheat. I know I'm not cheating. How do I know everyone else also isn't cheating?
I guess you will never know. Why? a) Let's say someone is cheating. Let's say the validator doesn't detect this. Then my question would be: How would one of the techs know someone is cheating? And if the techs don't know, how would we — crunching members — know someone else is cheating? I presume there is only one way: when that certain someone confesses he is cheating …; b) Let's say someone is cheating (again). Let's also say the validator detects this. How are we to find out that this happened? I mean, how can we — crunching members — tell that the validator did detect something suspicious, something unusual, among the millions of results returned daily? Does a result get marked Error or Invalid or …? And if we could tell, wouldn't it be a coincidence that someone sees a spoofed result among the millions of results returned daily? c) Let's say the validator detected a spoofed result. Let's say the techs found out about this, too. Let's say we — crunching members — don't know about this. Do you think the techs or other WCG staff will tell this to us? Yes, there is a possibility that they will, but there is no guarantee.

So, let's hope nobody is spoofing results intentionally, except when explicitly told to do so by the techs (e.g. when testing an application). I don't know if this is impossible, so let's say I find it questionable. In any case I've tried to provide some answers to the best of my knowledge.
[Sep 14, 2020 11:11:38 PM]   Link   Report threatening or abusive post: please login first  Go to top 
[CSF] Thomas H.V. DUPONT
Master Cruncher
Réunion-France
Joined: Aug 25, 2013
Post Count: 1512
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: THOR CHALLENGE [Y5-2020] by team CSF

Hi Michael, glad to see you here.
And happy to see that you and your team, Antarctic Crunchers, are having a lot of fun for your first participation in the THOR Challenge and THOR Challenge Pure.
CPU time for the scoring count elapsed time.
For the rest, we count on the fair play of the participants.
Obviously, there can be cheating but we would be quite pretentious to say that we can control it, even identify it.
One thing is regrettable: that we cannot do, as on the PrimeGrid project of which you have long been one of the very high quality Admins, to validate the results as soon as they return without waiting for a double validation and this during the challenges only.

And thanks so much, adriverhoef, for answering Michael's questions so brilliantly.
----------------------------------------
[Sep 15, 2020 5:50:47 AM]   Link   Report threatening or abusive post: please login first  Go to top 
BobMcG
Senior Cruncher
USA
Joined: Nov 14, 2005
Post Count: 170
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: THOR CHALLENGE [Y5-2020] by team CSF

Please tell me how to cheat. Not that I would ever do such a thing;)
But seriously, why would someone cheat? What would they get out of it besides bragging rights?
----------------------------------------

----------------------------------------
[Edit 1 times, last edit by BobMcG at Sep 15, 2020 6:52:41 AM]
[Sep 15, 2020 6:47:29 AM]   Link   Report threatening or abusive post: please login first  Go to top 
[P3D] Crashtest
Cruncher
Joined: Oct 9, 2009
Post Count: 1
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: THOR CHALLENGE [Y5-2020] by team CSF

Well some teams have large systems or lots of systems or even both - some dont!

Example:
Team 72 "The Gregory Team" created a total runtime of 126d 21h 36m 38sec

My AMD EPYC 7V12 can do this in 24h !


I have (if needed):
1x Ryzen5 1600
5x Ryzen7 1700
1x Ryzen7 2700
1x Ryzen5 3600
1x EPYC 7V12
12x Opteron 3280

in total 344 Threads -> 344d Runtime in 24h

Planet3Dnow has many systems incl. more than 10 AMD EPYC 7V12
[Sep 15, 2020 9:04:01 AM]   Link   Report threatening or abusive post: please login first  Go to top 
[CSF] Thomas H.V. DUPONT
Master Cruncher
Réunion-France
Joined: Aug 25, 2013
Post Count: 1512
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: THOR CHALLENGE [Y5-2020] by team CSF



Captains, the invitations have just been sent!
I invite you to check the Challenge Control Panel of your team.
We look forward to seeing you registered for this second week of competition.
50 teams in the run and only the 25 best will be selected!




----------------------------------------
[Sep 15, 2020 11:58:55 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 286   Pages: 29   [ Previous Page | 9 10 11 12 13 14 15 16 17 18 | Next Page ]
[ Jump to Last Post ]
Post new Thread