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 : 2019 Tour de Primes (Message 127205)
Posted 2 days ago by Profile Michael GoetzProject donor
Odd,
https://www.primegrid.com/workunit.php?wuid=600136042


Middle task has:
Exit status 194 (0x000000C2) EXIT_ABORTED_BY_CLIENT

Not sure how that ended up being "completed and validated"


We try to fix "broken" tasks that actually returned the correct result. The goal is to award credit whenever possible if the work was done. We don't want you to be penalized when some glitch causes the BOINC client to report a failure even when it completed the task correctly.
12) Message boards : Number crunching : 2019 Mega Primes (Message 127204)
Posted 2 days ago by Profile Michael GoetzProject donor
#307 is a GFN-17-MEGA.
13) Message boards : Number crunching : Grace Period list for the apps. (Message 127198)
Posted 2 days ago by Profile Michael GoetzProject donor
Should I add <fraction_done_exact/> to the PrimeGrid app_config.xml to enhance estimates?


YES!!!

We have that forced on at the server, but unfortunately when you use app_config it turns off fraction_done_exact.

The estimates are much, much more accurate when fraction_done_exact is enabled.

EDIT: I don't know if changing that setting will affect the tasks currently in progress. You can always divide the current elapsed time by the progress percentage to get an accurate estimate of the total run time. (Don't forget to divide the percentage by 100!)
14) Message boards : Number crunching : 2019 Mega Primes (Message 127197)
Posted 2 days ago by Profile Michael GoetzProject donor
#306 is a GFN-17-MEGA.

I seem to have missed one...

#305 is a PPS-MEGA.
15) Message boards : Generalized Fermat Prime Search : GFN prime discoveries II (Message 127192)
Posted 2 days ago by Profile Michael GoetzProject donor
We've found an n=17-Mega (131072) prime and eight n=16 (65536) primes!


On 19 Feb 2019 6:52:46, the 1,017,169 digit prime 57594734^131072+1 was found by dh1saj using the OCL2 (combined) app.

The double checker of this task was FishFry on 19 Feb 2019 7:10:46 using the OCL2 (combined) app.


On 19 Feb 2019 6:43:12, the 507,138 digit prime 54739648^65536+1 was found by [AF>EDLS]GuL using the OCL2 (combined) app.

The double checker of this task was vinn@[CNT] on 19 Feb 2019 11:59:21 using the OCL2 (combined) app.


On 19 Feb 2019 13:01:18, the 507,216 digit prime 54889620^65536+1 was found by Tornlogic using the OCL2 (combined) app.

The double checker of this task was Jordan Romaidis on 19 Feb 2019 13:09:42 using the OCL2 (combined) app.


On 19 Feb 2019 15:08:42, the 507,241 digit prime 54937732^65536+1 was found by Olgar using the OCL2 (combined) app.

The double checker of this task was =Lupus= on 21 Feb 2019 1:52:04 using the OCL2 (combined) app.


On 19 Feb 2019 15:35:00, the 507,247 digit prime 54949840^65536+1 was found by orvar* using the OCL2 (combined) app.

The double checker of this task was Woodles on 19 Feb 2019 15:36:46 using the OCL2 (combined) app.


On 19 Feb 2019 20:48:30, the 507,310 digit prime 55072040^65536+1 was found by DeleteNull using the OCL2 (combined) app.

The double checker of this task was wizanhi on 19 Feb 2019 20:50:23 using the OCL2 (combined) app.


On 19 Feb 2019 22:38:04, the 507,330 digit prime 55110930^65536+1 was found by Honza using the OCL2 (combined) app.

The double checker of this task was sentient_life on 19 Feb 2019 22:43:16 using the OCL2 (combined) app.


On 20 Feb 2019 8:43:47, the 507,424 digit prime 55292592^65536+1 was found by bcavnaugh using the OCL2 (combined) app.

The double checker of this task was mikey on 21 Feb 2019 2:48:47 using the OCL2 (combined) app.


On 20 Feb 2019 9:59:27, the 507,437 digit prime 55317972^65536+1 was found by Olgar using the OCL2 (combined) app.

The double checker of this task was Jewfro on 20 Feb 2019 10:00:32 using the OCL2 (combined) app.


Congratulations!
16) Message boards : Number crunching : 2019 Tour de Primes (Message 127182)
Posted 2 days ago by Profile Michael GoetzProject donor
The results are in for the 2019 Tour de Primes Mountain Stage.

Congratulations to [AF>EDLS]GuL for winning this year's polka dot jersey! It took two primes with a tie-breaking prime score of 256.72977589216987 to win this year.

Congratulations!
17) Message boards : Sophie Germain Prime Search : Sophie Germain Prime Search (Message 127178)
Posted 2 days ago by Profile Michael GoetzProject donor
Will the Sophie Germain primes be too small forever to reach the Top 5000 list or would they catch-up if we put lots of CPU power on this project?


Short answer: SGS tasks grow at a much slower rate than other projects and it will never make it onto the T5K list no matter how many tasks are run.

Unless we re-sieve for a new n. We do not plan on doing this until we exhaust the current sieve file, which will take years.
18) Message boards : Number crunching : Grace Period list for the apps. (Message 127124)
Posted 3 days ago by Profile Michael GoetzProject donor
marmot wrote:
Are the statements "(up to X days)" in the preferences section the list of all the grace periods beyond the deadline for each project? (ie: "Deadline: 7 days (up to 30 days)")


No. Well, yes. It's complicated.

The actual grace period is 25 hours. However, we use trickles to inform the server daily about the task's progress. If your computer is progressing, the server will extend the deadline from the nominal 7 days as needed. If your computer stops making progress on the task, the server stops extending the deadline.

Increasing the deadline is identical to increasing the grace period. It's the same mechanism: the deadline recorded in the database is increase without the deadline on the host changing. Your computer still thinks the deadline is 7 days. The server thinks it's more. (There's no way for us to adjust the deadline on your computer, unfortunately.)

For example, Woodall LLR has 1 day completed, estimates 13 days left with a deadline of Feb 26th; it'll need a 6 day grace period to complete that (although the WUProps hours will still count).


As long as your computer is progressing each day, once you get within one week of the deadline (the server's version of the deadline), the server will increment the deadline to keep it one week in the future. As long as you're making progress, you'll never have less than 7 days until the deadline. Until, of course, you hit the 30 day maximum deadline.

Bottom line: If your computer can finish within the maximum deadline and the computer keeps reporting progress to the server, you're good.
19) Message boards : Number crunching : 2019 Tour de Primes (Message 127120)
Posted 3 days ago by Profile Michael GoetzProject donor
mattozan wrote:
Boy, it sure is a quiet mountain this year!


Everybody listen up:

Mattozan has now found two primes during the mountain stage. (Only one is visible as I write this.)

You can draw your own conclusions. :)

P.S. There's also a mega prime today.
20) Message boards : Number crunching : 2019 Tour de Primes (Message 127115)
Posted 3 days ago by Profile Michael GoetzProject donor
Boy, it sure is a quiet mountain this year!


Don't forget the pipeline.

Aha! So this is what we need to win, huh? ;P


Well played sir.
Good luck everyone.
We have had days where no primes were found at all. If by chance that happens, will it be extended another 24 hours, pick another day, or forget it all together?


That's an interesting question, but since there's already a posse of primes today it's not a question we need to answer.

If we needed a rule to cover that situation, I suppose we should go into sudden death: if there's no primes during the Mountain Stage, the first prime following the Mountain Stage wins.

EDIT: Never has there been a single day during TdP when there were no primes discovered since we've been recording the discovery dates. That takes us back to 2014. (Prior to 2014, we went by the date a prime was reported to T5K.) And yes, this includes February 29th, 2016.


Next 10 posts
[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2019 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 6.10, 5.57, 4.79
Generated 23 Feb 2019 | 9:22:34 UTC