• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

FormulaBOINC

Page 39 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Is it standard procedure that the server sends WUs with quorum = 1, but labels the returned task "inconclusive" (maybe after some simple partly validation on the server), replicates the WU, and validates after the replica returns?
 
Uhm, sounds weird to me Stefan.

Crap, Milkyway is HIGHLY optimized for double precision, which almost invalidates most modern cards, except AMD 79XX and R9-280X series cards, but even then, you aren't given enough tasks to last for very long without multiple instances of BOINC (for bunkering). I've not had much luck with multiple instances being stable on Win10, but...I'd best get started setting up my only 280X system.

And yeah! It messes with the SETI challenge (GPU).
 
I've started mw@h 2 days ago because I had a hunch about this weekend's sprint. It turns out having AMD card really such an advantage. I guess, I'll join the sprint this weekend while watching F1@spa.
 
As we start this sprint we are in 4th place in the Milkyway Marathon. Plenty of points ahead of 5th and lots behind 3rd. I don't foresee us gaining ground here nor losing any, so points will have to be gained through the sprint. May the AMD cards be with us. 😀

Milkyway@home............Marathon.........Start
1 Sicituradastra. .....................25 ........1,333,508,191
2 Overclock.net ......................18 ........1,129,268,894
3 Team Norway ......................15 ...........931,274,397
4 TeAm AnandTech ...............12 ...........420,715,302
5 US NAVY .............................10 ...........168,112,250
6 OcUK - Overclockers UK ......8 ............154,118,571
7 BOINC.SK ............................6 .............134,190,606
8 meisterkuehler.de Team .......4 .............130,855,435
9 AMD Users ............................2 .............117,915,563
10 SETI@Netherlands .............1 ...............73,517,655
 
I already have a good amount mw@h tasks to validate but I can't contact their server. It seems like their server is flooded or something.
 
Waaayyyy late with the stats, but here they be.
After the first hour of the sprint we were in 3rd, but have now secured second. 🙂


Milkyway@home............Sprint.........Day One
1 Team Norway ....................25 .........8,532,782
2 TeAm AnandTech ..............18 ........7,468,116
3 Crunching@EVGA .............15 .........5,701,164
4 OcUK - Overclockers UK ....12 ........4,864,458
5 Sicituradastra. .....................10 ........1,241,269
6 US NAVY .............................8 .........1,080,504
7 Overclock.net .......................6 ...........755,396
8 SETI@Netherlands ..............4 ...........711,252
9 AMD Users ...........................2 ..........707,584
10 The Scottish Boinc Team ...1 ...........441,579

Milkyway@home............Marathon.........Day One
1 Sicituradastra. .....................25 .........1,334,686,206
2 Overclock.net ......................18 .........1,129,880,280
3 Team Norway ......................15 ............939,181,079
4 TeAm AnandTech ................12 ...........427,493,493
5 US NAVY ..............................10 ..........169,118,274
6 OcUK - Overclockers UK .......8 ...........158,138,874
7 BOINC.SK ..............................6 ...........134,474,983
8 meisterkuehler.de Team ........4 ...........131,184,558
9 AMD Users ............................2 ............118,563,521
10 SETI@Netherlands ..............1 .............74,208,601
 
@bill1024, regarding your posts in the EVGA forum on app_config.xml:
Both <cpu_usage> and <gpu_usage> do not control the application's behavior directly. They only control how boinc-client predicts processor and coprocessor usage, and thus how many tasks boinc-client launches at the same time.

In total, boinc-client's processor usage prediction is controlled by
  • app_config.app.gpu_versions.cpu_usage (refers to CPU usage by GPU applications),
  • app_config.app_version.avg_ncpus (refers to CPU usage by CPU applications),
  • cc_config.options.ncpus (refers to the host; this value does not only influence boinc-client but also the project server),
  • and of course by the "Use at most __ % of the CPUs" setting,
  • furthermore by app_info parameters if you use the anonymous platform mechanism.
These dials in combination let you under- or over-subscribe processors, similar to how you use app_config.app.gpu_versions.gpu_usage to over-subscribe coprocessors.

Additionally, over-subscription of GPUs or/and CPUs can be achieved by running multiple client instances on the same host.

Of course most of the time we want to over-subscribe GPUs but not CPUs. There are certain situations in which it is useful to over-subscribe CPUs though.
 
Day two of the sprint and we are in a solid 2nd place. 🙂
Darm good with all the power that is going towards the Seti WOW event.😱
OcUK is making a move on 3rd and the Navy has moved up to 5th. Looks like we'll have to stay with it right to the end.

Milkyway@home............Sprint.........Day One
1 Team Norway ....................25 .........8,532,782
2 TeAm AnandTech ..............18 ........7,468,116
3 Crunching@EVGA .............15 .........5,701,164
4 OcUK - Overclockers UK ....12 ........4,864,458
5 Sicituradastra. .....................10 ........1,241,269
6 US NAVY .............................8 .........1,080,504
7 Overclock.net .......................6 ...........755,396
8 SETI@Netherlands ..............4 ...........711,252
9 AMD Users ...........................2 ..........707,584
10 The Scottish Boinc Team ...1 ...........441,579

Milkyway@home............Sprint.........Day Two
1 Team Norway .......................25 .......14,961,364
2 TeAm AnandTech ................18 ........12,785,809
3 Crunching@EVGA ...............15 ........10,798,145
4 OcUK - Overclockers UK ......12 ........10,752,815
5 US NAVY ..............................10 ..........2,194,338
6 Sicituradastra. .........................8 ..........2,069,664
7 Overclock.net ..........................6 ..........1,618,719
8 SETI@Netherlands .................4 ..........1,268,348
9 AMD Users ..............................2 ..........1,236,516
10 meisterkuehler.de Team ........1 .............749,740

Milkyway@home............Marathon.........Day Two
1 Sicituradastra. ........................25 ..........1,335,472,679
2 Overclock.net .........................18 ..........1,130,739,491
3 Team Norway .........................15 .............945,592,673
4 TeAm AnandTech ...................12 .............432,739,416
5 US NAVY .................................10 ............170,159,666
6 OcUK - Overclockers UK ...........8 ............163,922,697
7 BOINC.SK ..................................6 ............134,656,080
8 meisterkuehler.de Team ............4 .............131,538,952
9 AMD Users .................................2 .............119,136,644
10 SETI@Netherlands ..................1 ................74,719,484
 
Han Solo made the Kessel Run in less than 12 parsecs, but let's make the Norwegian Milkyway Run in less than 20 (hours)! Engage Hyper-Drives!
 
The last minute dash payed big dividends with a first place finish.😀
OcUK managed a 3rd overtaking EVGA and gained a spot and two points in the Marathon. Congrats to them.🙂
Great race all.😎
Wednesday a new one gets announced.😱

Milkyway@home............Sprint.........Day One
1 Team Norway ....................25 .........8,532,782
2 TeAm AnandTech ..............18 ........7,468,116
3 Crunching@EVGA .............15 .........5,701,164
4 OcUK - Overclockers UK ....12 ........4,864,458
5 Sicituradastra. .....................10 ........1,241,269
6 US NAVY .............................8 .........1,080,504
7 Overclock.net .......................6 ...........755,396
8 SETI@Netherlands ..............4 ...........711,252
9 AMD Users ...........................2 ..........707,584
10 The Scottish Boinc Team ...1 ...........441,579

Milkyway@home............Sprint.........Day Two
1 Team Norway .......................25 .......14,961,364
2 TeAm AnandTech ................18 ........12,785,809
3 Crunching@EVGA ...............15 ........10,798,145
4 OcUK - Overclockers UK ......12 ........10,752,815
5 US NAVY ..............................10 ..........2,194,338
6 Sicituradastra. .........................8 ..........2,069,664
7 Overclock.net ..........................6 ..........1,618,719
8 SETI@Netherlands .................4 ..........1,268,348
9 AMD Users ..............................2 ..........1,236,516
10 meisterkuehler.de Team ........1 .............749,740

Milkyway@home............Sprint.........Final
1 TeAm AnandTech .................25 ..........23,648,066
2 Team Norway ........................18 ..........23,013,091
3 OcUK - Overclockers UK .......15 .........18,334,916
4 Crunching@EVGA .................12 .........15,499,032
5 Overclock.net .........................10 ...........7,476,569
6 US NAVY .................................8 ...........3,417,395
7 Sicituradastra. .........................6 ............3,219,135
8 SETI@Netherlands .................4 ............2,141,025
9 AMD Users ..............................2 ............1,861,620
10 meisterkuehler.de Team ........1 ............1,127,048

Milkyway@home............Marathon.........
1 Sicituradastra. .........................25 ..........1,336,649,589
2 Overclock.net ..........................18 ..........1,135,540,636
3 Team Norway ..........................15 .............953,687,572
4 TeAm AnandTech ...................12 .............442,990,452
5 OcUK - Overclockers UK .........10 .............171,719,608
6 US NAVY ..................................8 ..............171,464,787
7 BOINC.SK .................................6 ..............134,850,861
8 meisterkuehler.de Team ............4 ..............131,968,985
9 AMD Users .................................2 .............119,714,935
10 SETI@Netherlands ..................1 ................75,598,419
 
This sprint and the WCG sprint in April had in common that there was a team ranking very high which likely didn't even know that there was a sprint going on. If we are moved into league 1 next year, this will become the norm, I'm afraid.
 
StefanR5R said:
This sprint and the WCG sprint in April had in common that there was a team ranking very high which likely didn't even know that there was a sprint going on. If we are moved into league 1 next year, this will become the norm, I'm afraid.
Click to expand...
I'm fairly new in this DC stuff, but is this about certain coin team that we are talking about?
 
Gridcoin is of course always somewhere in the top ranks of each sprint. But of the more regular teams, for example SETI.Germany apparently didn't care for Formula Boinc throughout the last few years, but showed up high in a few sprints nevertheless.

But as Orange Kid noted in the opening post, the sprints were introduced just this year. Maybe more teams will become aware of them eventually. TennesseeTony has a talent to get the attention of otherwise sleepy dogs...
 
Last edited:
You must log in or register to reply here.

TRENDING THREADS

Back
Top