PrimeGrid
Please visit donation page to help the project cover running costs for this month

Toggle Menu

Join PrimeGrid

Returning Participants

Community

Leader Boards

Results

Other

drummers-lowrise
11) Message boards : Number crunching : Pi Approximation Day Challenge (Message 156531)
Posted 60 days ago by Profile Michael GoetzProject donor
You might let us know if users doing only the check versions of PPS would help you process the results faster and reduce your disk space shortage.



There's no practical way for users to select just DC tasks, and it wouldn't matter if there was because they're so small.

Between the server prioritizing sending out the DC tasks, and them being so small, they get done very quickly.
12) Message boards : Number crunching : Pi Approximation Day Challenge (Message 156526)
Posted 60 days ago by Profile Michael GoetzProject donor
Uploads of results (all results, not just PPS) have been disabled.

The next couple of days may be a little rough.
13) Message boards : Number crunching : Pi Approximation Day Challenge (Message 156525)
Posted 60 days ago by Profile Michael GoetzProject donor
Did they end the challenge early then?

It seems we made it to the end. The server may still die, but at least the challenge is over.
14) Message boards : Number crunching : Pi Approximation Day Challenge (Message 156506)
Posted 61 days ago by Profile Michael GoetzProject donor
It's not related to the database, and aborting unstarted tasks would help, yes.
15) Message boards : Number crunching : Pi Approximation Day Challenge (Message 156500)
Posted 61 days ago by Profile Michael GoetzProject donor
Most unfortunate if it comes to pass! I'm guessing that "shutting down the server" means all subprojects will be impacted?

Regardless, it's my turn for a local heatwave to hit, so I've already NNT'd some of my systems early so as not to cook the house more than mother nature is.


Yes, all subprojects.

Many of us are already setting our computers to No-New-Work, or switching to other projects. Every bit helps.
16) Message boards : News : Challenge may end early (Message 156497)
Posted 61 days ago by Profile Michael GoetzProject donor
Due to unanticipated problems with the "Pi" challenge, we may, without notice, end the challenge early.

Details can be found in this forum thread.

We apologize for any inconvenience this may cause and thank you for your cooperation.
17) Message boards : Number crunching : Pi Approximation Day Challenge (Message 156496)
Posted 61 days ago by Profile Michael GoetzProject donor
**VERY** IMPORTANT ANNOUNCEMENT

The administrators may need to end this challenge early, without notice. There's been unanticipated problems with the server, and it may not make it to 22:00. We therefore may end the challenge abruptly and without warning by shutting down the server until we can process the backlog.

If you're bunkering tasks until the last minute, you may not be able to return them during the challenge.

We apologize for any inconvenience this may cause.
18) Message boards : Number crunching : Pi Approximation Day Challenge (Message 156478)
Posted 61 days ago by Profile Michael GoetzProject donor
I'll also go with 5. I think 10 and above is unrealistic.


10 were found on the first day of the challenge.

There undoubtedly have been as many or even more on the second day, but the server is having some problems and is way, way behind in processing the results. It will be a while before the primes are disclosed.
19) Message boards : Proth Prime Search : PPSE (Proth Prime Search Extended) too small for Top 5000 now (Message 156329)
Posted 71 days ago by Profile Michael GoetzProject donor
Considering:

- PPS will close when it reaches 1M digits


PPS isn't closing. When it reaches 1M digits it will leapfrog ahead of PPS-MEGA and test small Ks at moderately high Ns.
20) Message boards : Number crunching : Way too much work (Message 156214)
Posted 79 days ago by Profile Michael GoetzProject donor
Is a waiting task (for 3 or so minutes) really that bad? For a large portion of my "fast" gpu tasks the time difference between the tasks sent between my computer and my wingman is greater than the 3 minute idle time. Sometimes in my favor sometimes not. So for me having a task sit idle is not really a problem.


Sometimes. If you're running DYFL it doesn't matter at all. But if you're running GFN-15 on a fast GPU, that 3 minute hit is HUGE. The average run time is under three minutes, and fast GPUs can do them a lot faster than that. Adding 3 minutes significantly hinders your chances of being first if the other guy doesn't have that delay.


Next 10 posts
[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2022 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 0.52, 0.47, 0.62
Generated 24 Sep 2022 | 19:37:07 UTC