FormulaBOINC

[crashtech]

Do we want to be in League 1 next year? Because at this rate it might happen:

http://formula-boinc.org/index.py?lang=en&year=2017

 

[Orange Kid]

Ya, the overclockers will all be there, we gotta have someone to play wit.

 

[crashtech]

Well, we're only halfway in, and a lot can happen, but I did not expect to see us tied for 3rd in the Marathon at this point. Go TeAm!

 

[StefanR5R]

Marathon stats are a bit wonky. I think because DENIS has "Problems in the server" and has minus a bunch of our points. With the 25 for this Sprint we should be ahead of 3rd place Sicituradastra.
Does anyone remember what our Formula BOINC score was in DENIS?

Yesterday TeAm AnandTech was completely missing from the DENIS@Home marathon table, right? Now we are back in.

Still, the database restore at DENIS@Home must have been from an old backup indeed, or is not yet completed yet: According to Tony's last weekly stats, Fardringle, crashtech, and 10esseeTony should have some Recent Average Credit. But DENIS@Home's own team table shows virtually zero RAC, and crashtech is missing from the table.

 

[StefanR5R]

Do we want to be in League 1 next year? Because at this rate it might happen:

http://formula-boinc.org/index.py?lang=en&year=2017

Ya, the overclockers will all be there, we gotta have someone to play wit.

Medals for us will become rare then, but not absolutely impossible. Take this very LHC@home race for example. Current cross-league stats:

1    Gridcoin                 1,368,957    league 1
2    Czech National Team        535,705    league 1
3    TeAm AnandTech             398,369    league 2
4    SETI.USA                   370,587    league 1
5    SETI.Germany               323,644    league 1
6    Russia                     308,189    league 1
7    Overclock.net              261,154    league 2
8    [H]ard|OCP                 254,687    league 3
9    L'Alliance Francophone     215,830    league 1
10   Rechenkraft.net            184,629    league 1
11   BOINC@Poland               171,950    league 1
12   UK BOINC Team              164,329    league 1
13   OcUK - Overclockers UK     154,970    league 2
14   USA                        147,485    league 1
15   BOINC.Italy                 85,115    league 1
16   BOINCstats                  84,800    league 1
17   Team China                  68,574    league 1
18   Crunching@EVGA              62,844    league 2
19   BOINC Synergy               53,274    league 1
20   Planet 3DNow!               49,491    league 1

 

[StefanR5R]

Aaaand "Project is down".

Yesterday project admin Eric Mcintosh had a go at weeding out many of those short-running Sixtrack WUs. Maybe this involved the temporary downtime.

--------
Edit: There was a brief downtime on short notice earlier today:

There will be a (very) short interruption while I install a new sixtrack_validator. Should fix null/empty fort.10 and the nasty "outlier" problem. See SixTrack Application, sixtrack_validator for more news and details.

The sixtrack_validator was installed and the server processes re-started at 13:23:44. Eric.

LHC@home forum
:thumbsup:

So far, so good. HOWEVER, I think we are still missing one part of the puzzle.
Huge thanks to Juha who had a look at the Validator. I have now found at least one Work Unit with two Valid results, but one is an Outlier and the other is not!!!
As I understand this is NOT good. If you are the unlucky sidekick, the second result, you will NOT be treated as an Outlier, but worse you will then get a Real Time exceeded message and up to several hours of your valuable computer time are wasted. Fixing this does not appear to be trivial, to me at least.

Nonetheless we are processing a huge amount of work, most of it successfully. I am not complacent though, and this last piece of a huge puzzle may soon fall into place. Eric.

LHC@home forum

 

[Orange Kid]

Day two and still in first place

We have moved up to Third in the overall League 2 Marathon with the points in this race


LHC@home 1.0...........Sprint........Day One
1 TeAm AnandTech ................25 ...........339,564
2 Overclock.net .......................18 ...........213,657
3 OcUK - Overclockers UK .....15 ...........139,219
4 Crunching@EVGA ...............12 ............50,891
5 meisterkuehler.de Team .......10 ............35,272
6 France ....................................8 .............29,083
7 BOINC.SK ..............................6 .............24,234
8 Sicituradastra. ........................4 .............23,719
9 Ars Technica ..........................2 .............16,571
10 AMD Users ...........................1 .............14,180

LHC@home 1.0...........Sprint........Day Two
1 TeAm AnandTech .................25 ............544,326
2 Overclock.net ........................18 ............364,619
3 OcUK - Overclockers UK ......15 ............211,763
4 meisterkuehler.de Team .......12 ...............87,701
5 Crunching@EVGA ................10 ...............78,444
6 Sicituradastra. .........................8 ...............53,165
7 France .....................................6 ...............51,419
8 BOINC.SK ...............................4 ...............36,240
9 AMD Users ..............................2 ...............24,723
10 Ars Technica .........................1 ...............21,792


LHC@home.......Marathon
1 Sicituradastra........................ 25 ............39,949,649
2 Overclock.net ........................18 ............21,383,513
3 OcUK - Overclockers UK ......15 ............13,477,959
4 XtremeSystems .....................12 ............11,180,439
5 The Scottish Boinc Team .......10 .............9,745,556
6 TeAm AnandTech ....................8 .............8,051,218
7 meisterkuehler.de Team ..........6 ..............7,410,224
8 Team Norway ...........................4 .............7,012,958
9 AMD Users ...............................2 .............5,645,465
10 France ....................................1 ..............5,417,870

 

[StefanR5R]

Last night some of my machines got sixtrack tasks which run longer than a day.

Edit:
Sadly, there is now a considerable percentage of completed sixtrack tasks being held up with "validation inconclusive". I guess the new validator which went online yesterday plays a role.

 

[Orange Kid]

Looking at my tasks, I haven't had a validated task since yesterday sometime. everything else that has been returned since is either waiting for validation or inconclusive. This project has more problems than any I've run.

 

[crashtech]

Interestingly, I am seeing validations continuing into today for the two boxes that are crunching more variety, namely LHCb Simulation and Theory Simulation. But my last Sixtrack validation occurred on: 24 Jun 2017, 23:25:52 UTC.

I hope the new validator is not broken!?

 

[TennesseeTony]

Well, this Sprint is over, and somehow we won again! Congrats all!

Sure wish I had been able to get some tasks after the first day, to help more.

We were third across all three leagues, which is even more impressive!

 

[Orange Kid]

Another first and 25 more points.
This has move the TeAm to third in the overall League Two standings and given breathing room in the Marathon.
Congrats to all.
Another sprint next week already. Lets hope it is a better behaved project

LHC@home 1.0...........Sprint........Day One
1 TeAm AnandTech ................25 ...........339,564
2 Overclock.net .......................18 ...........213,657
3 OcUK - Overclockers UK .....15 ...........139,219
4 Crunching@EVGA ...............12 ............50,891
5 meisterkuehler.de Team .......10 ............35,272
6 France ....................................8 .............29,083
7 BOINC.SK ..............................6 .............24,234
8 Sicituradastra. ........................4 .............23,719
9 Ars Technica ..........................2 .............16,571
10 AMD Users ...........................1 .............14,180

LHC@home 1.0...........Sprint........Day Two
1 TeAm AnandTech .................25 ............544,326
2 Overclock.net ........................18 ............364,619
3 OcUK - Overclockers UK ......15 ............211,763
4 meisterkuehler.de Team .......12 ...............87,701
5 Crunching@EVGA ................10 ...............78,444
6 Sicituradastra. .........................8 ...............53,165
7 France .....................................6 ...............51,419
8 BOINC.SK ...............................4 ...............36,240
9 AMD Users ..............................2 ...............24,723
10 Ars Technica .........................1 ...............21,792

LHC@home 1.0...........Sprint........Final
1 TeAm AnandTech ..................25 ..............586,752
2 Overclock.net .........................18 ..............401,302
3 OcUK - Overclockers UK ........15 ..............235,571
4 meisterkuehler.de Team ..........12 .............109,222
5 Crunching@EVGA ...................10 ...............95,780
6 Sicituradastra. ............................8 ...............62,611
7 France ........................................6 ...............58,734
8 BOINC.SK ..................................4 ...............38,066
9 AMD Users .................................2 ...............31,713
10 Portugal@Home .......................1 ...............22,552


LHC@home..........Marathon
1 Sicituradastra. ..........................25 ...............39,973,982
2 Overclock.net ...........................18 ...............21,507,024
3 OcUK - Overclockers UK ..........15 ..............13,548,274
4 XtremeSystems .........................12 ..............11,185,121
5 The Scottish Boinc Team ..........10 ................9,746,223
6 TeAm AnandTech .......................8 .................8,197,672
7 meisterkuehler.de Team ..............6 ................7,465,106
8 Team Norway ..............................4 ................7,019,131
9 AMD Users ..................................2 ................5,655,374
10 France .......................................1 .................5,436,679

 

[StefanR5R]

Not sure what makes me more happy right now: That we achieved Gold again, or that we are done now with LHC@home races for the year.

 

[GLeeM]

This has move the TeAm to third in the overall League Two standings and given breathing room in the Marathon.

Not sure if you noticed, we took first place (up from sixth!) and gained 13 Marathon points in Climate Prediction Thanks Tony!

 

[Ken g6]

And now they add more Atlas work.

 

[crashtech]

Looking at the Marathon stats, I was a bit shocked to realize that we don't even have 1 point yet in PrimeGrid!?

http://formula-boinc.org/projets.py?lang=en&prj=32&year=2017

 

[StefanR5R]

Loads of SixTrack credits are incoming for the LHC@home marathon (for all teams which participated in the sprint, notably). A few hours ago, the LHC@home admins reverted the SixTrack validator and moved all WUs with status "validation inconclusive" to "validation pending", and the validator is getting busy now reevaluating all those WUs.

 

[crashtech]

Hmm, two hours until project announcement!

Guys, is there a trick to getting work from BURP? I've been trying for days to see if I can get a WU.

 

[GLeeM]

Hmm, two hours until project announcement!

Are you sure? That would mean 15 minutes from now?
I always have to ask when, so ... but I figure announcement in two hours and 15 minutes from now

 

[crashtech]

I'm chronologically challenged. You are right...

 

[Orange Kid]

...........and the winner is

Collatz Conjecture

 

[Orange Kid]

and a tuff one to bunker as it is 200 tasks per.

EDIT: that's 100 per

10 minutes for my 1070 per wu
15 minutes for my 970 per wu

 

[TennesseeTony]

Uh oh. Some "Collatz SuperStars" on the other teams. So long as the tasks get steadily distributed, this may be an interesting battle for us.

 

[StefanR5R]

But CPU tasks look very easy to bunker.

 

[crashtech]

Is CPU worth running?

And is there a reason my G4560 on an Asrock E3V5WS board with a Sapphire R9 290 will not run Collatz? It's not overclocked. All I get is all "Computation Errors."