mersenneforum.org > Math Could a Distributed Computing approach help find the smallest Brier number?
 User Name Remember Me? Password
 Register FAQ Search Today's Posts Mark Forums Read

 2007-03-28, 02:45 #1 jasong     "Jason Goatcher" Mar 2005 350710 Posts Could a Distributed Computing approach help find the smallest Brier number? I ask this question in total ignorance of the math involved. I throw this out there just to see if people think DCing could help this problem. For those of you who don't know what a Brier number is, I'll explain. When you're searching for large prime numbers, there's a basic formula, k*b^n+c. For the most part, most people concentrate where b equals 2, and c is plus or minus 1. For k*2^n-1, there are k's where all n from 1 to infinity will give the equation a composite number, and it's the same thing for k*2^n+1, but usually with different numbers. The smallest number for k*2^n-1 is believed to be 509203, for +1, I think it's 78557(not sure). A Brier number is a number, k, where both k*2^n-1 and k*2^n+1 will yield composite numbers for all n from 1 to infinity. My question is,"Is it possible to use distributed computing to help solve the lowest Brier number problem in a timely fashion?" For the record, I know searching for primes below the smallest known Brier number is kind of idiotic since it's so unbelievably huge. I'm wondering if there is a possibility to do some sort of covering set research, in an attempt to find a smaller number. Last fiddled with by jasong on 2007-03-28 at 02:51
 2007-03-28, 03:30 #2 drake2   Aug 2005 1610 Posts The mathworld.com description for Brier Number currently says "The first few are 878503122374924101526292469, 3872639446526560168555701047, 623506356601958507977841221247, ... " The comment field of Sequence A076335 in the The On-Line Encyclopedia of Integer Sequences states "These are just the smallest examples known - there may be smaller ones." Don't one of these two statements need to be corrected or is there no rigor in the field of mathematics?
2007-03-28, 07:41   #3
Orgasmic Troll
Cranksta Rap Ayatollah

Jul 2003

10100000012 Posts

Quote:
 Originally Posted by drake2 The mathworld.com description for Brier Number currently says "The first few are 878503122374924101526292469, 3872639446526560168555701047, 623506356601958507977841221247, ... " The comment field of Sequence A076335 in the The On-Line Encyclopedia of Integer Sequences states "These are just the smallest examples known - there may be smaller ones." Don't one of these two statements need to be corrected or is there no rigor in the field of mathematics?
None at all. Try psychology, I hear those guys are sticklers.

2007-03-28, 16:22   #4
R.D. Silverman

Nov 2003

22×5×373 Posts

Quote:
 Originally Posted by drake2 The mathworld.com description for Brier Number currently says "The first few are 878503122374924101526292469, 3872639446526560168555701047, 623506356601958507977841221247, ... " The comment field of Sequence A076335 in the The On-Line Encyclopedia of Integer Sequences states "These are just the smallest examples known - there may be smaller ones." Don't one of these two statements need to be corrected or is there no rigor in the field of mathematics?
What needs correcting? I see nothing wrong with the statements.

2007-03-28, 17:05   #5
jasonp
Tribal Bullet

Oct 2004

1101110100102 Posts

Quote:
 Originally Posted by drake2 "The first few are 878503122374924101526292469, 3872639446526560168555701047, 623506356601958507977841221247, ... " The comment field of Sequence A076335 in the The On-Line Encyclopedia of Integer Sequences states "These are just the smallest examples known - there may be smaller ones." Don't one of these two statements need to be corrected or is there no rigor in the field of mathematics?
Read 'the first few' above as 'the first few we have found'.

jasonp

2007-05-29, 13:30   #6
robert44444uk

Jun 2003
Oxford, UK

1,933 Posts

Quote:
 Originally Posted by jasong I'm wondering if there is a possibility to do some sort of covering set research, in an attempt to find a smaller number.
You might want to check the Carlos Rivera site that talks about Briers and which also provides some useful correspondence by persons who have looked at strategies for smaller Briers in some depth.

http://www.primepuzzles.net/problems/prob_029.htm

I have been corresponding with Prof. Caldwell on techniques for Sierpinski/Riesel any base and he suggests there are alternatives to the traditional covering set approach (using primes modulo a certain base) which, in particular, look at known algebraic factoring methods to allow for a pseudo-cover to replace one or more of the primes in the cover, or will allow for a smaller covering set.

For example x^3 - a^3 = (x - a)(x^2 + ax + a^2)

When you set a=1, then this becomes x^3-1, which gives you a form of k*2^3-1, k*2^6-1, k^2^9-1...(pseudo order 3 base 2) as k*2^(3*integer) is still a cube if k is a cube.

Other useful factoring tools might be used to give pseudo cover, for both + and - cases. Aurifeuillian for example.

This type of strategy has not been investigated as far as I am aware for the Brier case.

But it would take a lot of brain power, which I am in limited supply, to investigate all the possibilities. But if it produced a smaller Brier, then it would be a breakthrough.

Last fiddled with by robert44444uk on 2007-05-29 at 13:45

 Thread Tools

 Similar Threads Thread Thread Starter Forum Replies Last Post a1call Miscellaneous Math 8 2018-05-21 16:25 mart_r Aliquot Sequences 6 2013-07-23 20:50 andi314 Factoring 50 2005-05-11 00:56 GP2 Lounge 2 2003-12-03 14:13 Fusion_power Puzzles 8 2003-11-18 19:36

All times are UTC. The time now is 15:08.

Tue May 18 15:08:41 UTC 2021 up 40 days, 9:49, 1 user, load averages: 1.50, 1.77, 1.91

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.