mersenneforum.org 51*2^3774129-1 is not prime, but 51*2^5085142-1 is
 Register FAQ Search Today's Posts Mark Forums Read

 2014-02-11, 16:57 #1 Retep     Oct 2007 1438 Posts 51*2^3774129-1 is not prime, but 51*2^5085142-1 is The first find since several years: 51*2^3774129-1
2014-02-11, 17:39   #2
pepi37

Dec 2011
After milion nines:)

3·5·83 Posts

Quote:
 Originally Posted by Retep The first find since several years: 51*2^3774129-1
WOW
1.13M PRIME!

Congratulations!!!!

 2014-02-12, 01:49 #3 pepi37     Dec 2011 After milion nines:) 3·5·83 Posts it looks like your "prime" is not prime" 51*2^3774129-1 is not prime. LLR Res64: 992EC623E6652D30 Time : 30452.098 sec. [Elapsed time: 8.46 hours] Warning: This number is composite and will be deleted soon!
 2014-02-12, 02:42 #4 Kosmaj     Nov 2003 2·1,811 Posts Hi Retep, Can you check your log files for possible typos? I just started a double check on my machine. By the way, "51 3774129" is in one of the latest sieve files. Note that your submission was deleted and can be found here. Also, please check your lresults.txt and your working directory for any traces of errors. My double check is over, I got the same result like the machine on Top-5000: Code: 51*2^3774129-1 is not prime. LLR Res64: 992EC623E6652D30 Time : 5196.177 sec. Last fiddled with by Kosmaj on 2014-02-12 at 03:50
 2014-02-12, 09:44 #5 pinhodecarlos     "Carlos Pinho" Oct 2011 Milton Keynes, UK 33×132 Posts Do we need a double check effort for k=51? I see Retep reserved k=51 from 2500-3700k (completed) and from 3700k-4M (in progress).
2014-02-12, 12:13   #6
Thomas11

Feb 2003

76D16 Posts

Quote:
 Originally Posted by pinhodecarlos Do we need a double check effort for k=51?
Retep, could you post your lresults.txt file(s)? At least the part containing the "false" prime.

At first we should run double checks on the candidates immediately before and after the problematic case (e.g. exponents 3773995 and 3774142) and see if we get matching residues.

 2014-02-12, 16:45 #7 Retep     Oct 2007 32×11 Posts I'm using a PRPNet server since n=3.7M to get some help from outside. I run it with double checks, since I thought this would be a good idea at such an high level. For 51*2^3774129 there was one database entry with residue=992EC623E6652D30 and one entry (which was not submitted by one of my computers, by the way) with residue=PRIME. The server displayed it to be prime, which seems to be a bug of PRPNet, I'm in contact with the developer at the moment. Last fiddled with by Retep on 2014-02-12 at 16:45
 2014-02-12, 23:05 #8 Kosmaj     Nov 2003 2·1,811 Posts All right then, there is no prime found. The machine which returned "prime" is obviously not suitable for prime search (unstable, or with poor cooling) and it's a good idea to remove it from your search. Also, in spite of all double cheching by PRP net etc, it's recommended to double check potential primes on another trusted machine, and if possible using another software like pfgw (PRP confirmation is enough), or another version of LLR (32 vs. 64 bit). You can also post your candidate here and somobody will double check it for you. By the way, as Thomas suggested, can you kindly attach your k=51 results, all the way from n=2.5M if you have them handy, or at least the most recent ones. Thanks.
2014-02-13, 08:11   #9
Thomas11

Feb 2003

76D16 Posts

Quote:
 Originally Posted by Retep I'm using a PRPNet server since n=3.7M to get some help from outside. I run it with double checks, since I thought this would be a good idea at such an high level.
Since you run it with double checking: Are there any other non-matching residues?

 2014-02-15, 11:56 #10 Retep     Oct 2007 32×11 Posts I got ill at the same day I came to know 51*2^3774129-1 is composite, I am not at home so I don't have access to all the data at the moment. The results for n=3.5M up to n=3.7M can be found here: http://www.mersenneforum.org/showpos...&postcount=735 http://www.mersenneforum.org/showpos...&postcount=736 We have already eliminated one error in PRPNet, I will keep you informed about that. I remember I checked some residues at random and they were matching. I will check all of them via SQL when I am at home again
 2014-08-04, 21:28 #11 pinhodecarlos     "Carlos Pinho" Oct 2011 Milton Keynes, UK 33·132 Posts Peter Riesen "found" a prime for this k. Let's wait for its validation. http://primes.utm.edu/primes/page.php?id=118295

 Similar Threads Thread Thread Starter Forum Replies Last Post dabaichi News 561 2013-03-29 16:55 emily PrimeNet 3 2013-03-01 05:49 hhh Prime Cullen Prime 4 2007-09-21 16:34 jasong Data 7 2005-09-13 20:41 illman-q Miscellaneous Math 33 2004-09-19 05:02

All times are UTC. The time now is 14:14.

Fri May 29 14:14:54 UTC 2020 up 65 days, 11:47, 1 user, load averages: 2.56, 2.55, 2.49