mersenneforum.org Proof of Goldbach Conjecture
 Register FAQ Search Today's Posts Mark Forums Read

 2007-11-30, 11:38 #1 vector     Nov 2007 home 318 Posts Proof of Goldbach Conjecture I think I proved the Goldbach conjecture, here is proof: This conjecture states that every even number greater than 2 can be expressed as the sum of two primes. This conjecture can be restated as: If X is even and Y is an integer then there exist prime numbers of the form X/2+Y and X/2-Y. A Diophantine equation can be generalized from this: (X/2-Y)(X/2+Y)=M; (X^2)/4-Y^2=M where M is a positive integer, and ((X^2)/4) is a fixed constant. The total number of positive M's is sqrt((X^2)/4). The chance that any of those M's is a semi-prime has a determined lower bound of ~1/4 for moderately sized M and increases as M gets larger. By finding M which is a semiprime the two primes that add to X can be found. Therefore the total number of 2 prime groups that sum to an even integer X has a lower bound of Floor[(sqrt((X^2)/4)/4], which is always greater than one for sufficiently large even integer. Last fiddled with by vector on 2007-11-30 at 11:46
 2007-11-30, 12:12 #2 akruppa     "Nancy" Aug 2002 Alexandria 2,467 Posts > The chance that any of those M's is a semi-prime has a determined lower bound of ~1/4 for moderately sized M and increases as M gets larger. What exactly do you mean by that? Alex
 2007-11-30, 12:21 #3 Brian-E     "Brian" Jul 2007 The Netherlands 326910 Posts You seem to be using a probability distribution argument. This would be sufficient to show that "almost all" even numbers are the sum of two primes, but it does not show the complete absence of counter-examples.
2007-11-30, 13:32   #4
vector

Nov 2007
home

52 Posts

Quote:
 Originally Posted by akruppa > The chance that any of those M's is a semi-prime has a determined lower bound of ~1/4 for moderately sized M and increases as M gets larger. What exactly do you mean by that? Alex
see: http://www.research.att.com/~njas/se...e?a=1358&fmt=5

The proof can be made deterministic by using theorem 3 from http://arxiv.org/PS_cache/math/pdf/0506/0506067v1.pdf
Using it the maximum distance between semi primes becomes 26.

2007-11-30, 14:36   #5
R.D. Silverman

Nov 2003

22×5×373 Posts

Quote:
 Originally Posted by vector see: http://www.research.att.com/~njas/se...e?a=1358&fmt=5 The proof can be made deterministic by using theorem 3 from http://arxiv.org/PS_cache/math/pdf/0506/0506067v1.pdf Using it the maximum distance between semi primes becomes 26.
It (the claimed proof) is codswallop.

There already exist probabilistic results regarding Goldbach. Look
up 'Goldbach exceptions'. For example, it is known that exceptions,
*if they exist* have asymptotic density 0. Indeed, the number of
possible primes less than P for which exceptions might exist is known
to be at most O(P^1/4+epsilon) for any epsion > 0. This does not
say whether any exceptions DO exist; merely that there can't be too many
if they do. The exponent 1/4 may have been improved since I last looked
at this problem.

The proofs of this and related results are sieve based and run into the
sieve parity problem & the fundamental lemma of the sieve. See
Halberstam & Richert's book.

 2007-12-01, 14:43 #6 vector     Nov 2007 home 52 Posts nevermind Last fiddled with by ewmayer on 2007-12-03 at 23:17 Reason: Don't worry - we didn't.

 Similar Threads Thread Thread Starter Forum Replies Last Post MattcAnderson MattcAnderson 17 2017-03-21 18:17 MattcAnderson MattcAnderson 1 2017-03-18 23:32 MattcAnderson MattcAnderson 3 2017-03-17 15:34 Patrick123 Miscellaneous Math 242 2011-03-15 14:28 Citrix Puzzles 3 2005-09-09 13:58

All times are UTC. The time now is 12:35.

Sun Apr 18 12:35:01 UTC 2021 up 10 days, 7:15, 0 users, load averages: 1.55, 1.46, 1.52