• 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.

Formula BOINC Sprints 2021

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
well, its good I added more cores, as we LOST 3rd place by a few hundred points. And none of the new cores have finished a unit, but we are loosing to 2nd by the same 21k.

This should be interesting in a few hours.
 
Glad to hear you added more cores @Markfw ! It definitely looks like it will be a close one right down to the finish.

Current stats approximately 36 hours in-

3-20-21_stats.PNG


Name​
Sprint Points​
xii5ku​
614,159​
crashtech​
549,396​
10esseetony​
408,795​
Icecold​
382,595​
biodoc​
171,389​
Fardringle​
104,432​
The Great Cornholio​
76,189​
markfw​
56,465​
cellarnoise2​
10,891​
hca​
8,573​
Howdy​
7,558​
emoga​
3,847​
Lane42​
3,401​
Orange Kid​
3,002​
geecee​
2,060​
Kiska​
1,295​
Anitya​
124​
tony marklew​
86​
 
Crap, I had Winblows do an O.S. update (forgot to suspend) and took down one of my peon machines for several hours.

It appears that SiDock does not due incremental saves, so if your machine reboots, or crashes you lose all of the work?
 
Crap, I had Winblows do an O.S. update (forgot to suspend) and took down one of my peon machines for several hours.

It appears that SiDock does not due incremental saves, so if your machine reboots, or crashes you lose all of the work?
Unfortunately you are correct - SiDock does not do checkpointing (incremental saves) currently. It wasn't as big of an issue when WU's were shorter, but now that they can be 4+ hours I'm hoping they add it at some point soon.

I love a close contest! With P3D putting so much pressure on Xtreme, what they can do to help TBST might be limited!
What's really cool to me is how close it is across the board-

1st and 2nd are close

3rd and 4th are super close

5th, 6th and 7th are all very close

8th and 9th are close as well.

Of each of those groups, anybody could win.
 
Unfortunately you are correct - SiDock does not do checkpointing (incremental saves) currently. It wasn't as big of an issue when WU's were shorter, but now that they can be 4+ hours I'm hoping they add it at some point soon.


What's really cool to me is how close it is across the board-

1st and 2nd are close

3rd and 4th are super close

5th, 6th and 7th are all very close

8th and 9th are close as well.

Of each of those groups, anybody could win.
10th, 11th, and 12th are pretty close, too.
 
well, its good I added more cores, as we LOST 3rd place by a few hundred points. And none of the new cores have finished a unit, but we are loosing to 2nd by the same 21k.

This should be interesting in a few hours.
I'm looking forward to seeing you pass me very soon! I'm putting in all that I can, but your 'sprint' inside the sprint will definitely help our chances to finish in the top three. 🙂
 
Mr. Mark, you can relax!

I just fired up an old 2014 i7!!!! laptop! Not some of this new crapper-wear silicon! 2,400 MEGA-Hurts! 4 Cores, but more importantly 8 THREADS!

It is ready to run HARD as it has had 6 months of rest and you should have seen the dust and bugs that flew out of that thing with some real compressed air! Not your normal "can of blow". I fired up the real blower in the GARAGE for this effort! A full metered up to 90 PSI of pressure (U.S. standard)....

These old school laptop fans are loud though, but I am willing to sacrifice.

This post is brought to tEaMaNaNdTeCh, by the letters B and E and E and R! Not from some chilled fancy pint glass either. These are basement room temp 12 MEGA-OZ (sorry, regular aluminum 12 oz...) CANS!

I am late to this party as I don't check this forum but maybe twice a week. And this is my first "Sprint", if you did not detect 🙂

Go TeAm!
 
Well, in the last hour, not only did we lose 3rd place again, but we lost 13k against 2nd.
 
I am new to this, but these SiDock tasks seem to be taking longer the closer they get to being finished. I'm not joking.

I have Zen 1 threads at 3.7 ghz, that are now over 5.5 hours and still only 99%. Is this a problem? The progress meter in Boinc Manager is sill limping along, but this almost seems like a power virus app...
 
I have seen some tasks with very long run times. The % completed progresses normally, then slows down dramatically near the end. They do eventually complete (and reward lots of credits) as long as there isn't an error that interrupts the task. The do not have checkpoints, so any interruption of the processing will make the task fail.
 
I should not joke around, but my puters are running CurieMarieDock on Bonic. Named "Eprot_v1_nb3di_####...." for what it is worth.

I'm out for the day!
 
26 hour stats-

View attachment 41497

Stats in Sega Genesis form(sorry for the super old logo it was the only one I could find)-

View attachment 41499

Name​
Sprint Points​
crashtech​
444,194​
xii5ku​
437,420​
10esseetony​
314,898​
Icecold​
281,509​
biodoc​
131,443​
Fardringle​
79,391​
The Great Cornholio​
56,976​
cellarnoise2​
6,456​
hca​
6,268​
emoga​
3,104​
geecee​
1,907​
Orange Kid​
853​
Howdy​
736​
Anitya​
124​

@cellarnoise and @Howdy are new additions to the stats since yesterday - thanks for joining in the sprint! It's a close race between us and TSBT for third place.

Here's a screenshot of league 2 also. It doesn't affect our standing, but the very close race between Crunching@evga and Boinc.Italy for position 2 will be interesting to watch. OCUK has a significant lead in first place.

View attachment 41498
Thanks Icecold!🙂

Good job over here as always gents.
 
well, we now have a 70k lead over 4th, and are back in 3rd, but I don't think we can catch 2nd, even though in the last hour we made up 8.6k, but we have lost like 50k in the last 3-4 hours.
 
Over at the SiDock message board...
hoarfrost said:
Now such deadline vitally for project, because if we increase deadline, project exhaust HDD space in current server. Short deadline -> short time for results hold. :) We plan to migrate to other server.
So the short reporting deadline, and certainly also the shark-tooth pattern of their work generation (with rather low max and low min numbers of tasks ready to send) are mainly due to limited space for files on the server.

It's good to see the SiDock admins being responsive and making rather good use of the resources at their disposal.
 
Today (shortly before 08:00 UTC), the reporting deadline of new tasks was increased to 48 h. This is perhaps a reaction to the new tasks taking about 4 times as long to compute than the previous batches.
 
I'm an hour early on the stats, but won't be in front of a computer when it hits exactly 48 hours. Hopefully there are no errors, I put it together in a hurry, but it should be right-

League 1 - looking good we are now well ahead of TSBT. Great work everybody!

3-21-21-league1.PNG
Name​
Sprint Points​
xii5ku​
1,065,645​
crashtech​
705,431​
10esseetony​
582,270​
Icecold​
555,597​
biodoc​
239,139​
markfw​
199,825​
Fardringle​
156,632​
The Great Cornholio​
109,080​
Lane42​
24,516​
emoga​
24,165​
Howdy​
20,324​
cellarnoise2​
19,277​
hca​
13,630​
Orange Kid​
9,876​
Kiska​
4,902​
geecee​
3,345​
tony marklew​
199​
Anitya​
124​


League 2 - Crunching@EVGA has a significant lead on positions 3 and 4 has closed OCUK's lead somewhat - this one will probably be close right until the end.

3-21-21-league2.PNG


Thanks Icecold!🙂

Good job over here as always gents.

Thanks yodap! With the strong sprint performance that puts you guys at third place overall for the year in league 2 - great job!
 
I have Zen 1 threads at 3.7 ghz, that are now over 5.5 hours and still only 99%. Is this a problem? The progress meter in Boinc Manager is sill limping along,
The newer tasks just take longer. But the initial estimation of the computational workload in a task, which the server sets in the task description when it creates the task, is still the same as for the older shorter tasks. Combined with the application not providing direct feedback on its progress to the BOINC client, this is confusing the progress bar in boincmgr. (AFAIU.)
 
Well, we are up to a 782k lead over 4th, but we have over 1.6m to catch second. (isn't going to happen)

I don't understand how 803 cores only gets me 364k in 2 days !
 
...I don't understand how 803 cores only gets me 364k in 2 days !

1.) You gradually added cores, correct?
2.) You have about another 170,000 in pending, waiting on a wingman
3.) I think your score is outstanding for a bit of a slow ramp up to 803. I suspect it will trend upward quickly and blast out past the Moon! 😀
 
Back
Top