mersenneforum.org (https://www.mersenneforum.org/index.php)
-   No Prime Left Behind (https://www.mersenneforum.org/forumdisplay.php?f=82)
-   -   n=333333 off the list (https://www.mersenneforum.org/showthread.php?t=10274)

 Mini-Geek 2008-05-13 12:17

n=333333 off the list

[URL]http://primes.utm.edu/primes/lists/all.txt[/URL]
All k*2^333333-1 primes are now off the Top 5000 list. #5001 is 28782838101*2^333333-1 (archivable for being an arithmetic progression). I remember reading some posts between Jasong and Gary before NPLB started about Jasong wanting to have a "I hate n=333333" project to make the bottom of the list not boring, and Gary saying he had something coming up.
In any case, the bottom of the Top 5000 list is no longer boring.
:party:
p.s. it will be #5002 once I submit my prime, and probably drop a lot from several prime submissions...why did they have to do maintenance right when I found a prime? Oh well.
:party: again.

 henryzz 2008-05-13 16:18

it still is partly boring becausee of the number of primes found for k*2^333666-1:smile:

 gd_barnes 2008-05-13 17:26

Yep, this is the project that was in the works that helped knock off the n=333333 primes. I hadn't planned it for that reason but it was a nice side benefit of it.

Don't worry, the n=333666 primes will soon go away also.

The best way to knock off fixed-n-searches with HUGE k-values is to do small-k-value searches across a moderate range of n-values just above the fixed-n-values. The sieving is a little less efficient but the LLRing is much faster.

Editor note :-) : I never said anything about 'hating' the n=333333 primes. I should say that Jasong came up with that verbiage. They may have been somewhat of an annoyance but I think it's important to respect what the leader of an effort is after (unless they are poaching other projects). MooMoooo did a very nice job of organizing that project and they were a little unlucky not to find a huge twin prime, which was their objective. I believe the consensus on that project now is that the effort will search across a wide range of n-values with smaller k-values in the future but that is not set in stone yet.

In the mean time, PrimeGrid has somewhat of an 'annoying' effort to do a similar thing with n=666666. At least it takes ~8 times as long to find an n=666666 prime vs. an n=333333 prime so they are trickling in quite slowly right now.

Gary

 All times are UTC. The time now is 16:13.