home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math
- Path: sparky!uunet!wri!news
- From: roach@bikini.wri.com (Kelly Roach)
- Subject: Re: Big primes puzzles
- Message-ID: <1992Aug15.010448.9747@wri.com>
- Keywords: Ackermann function/Puzzle/Primes
- Sender: news@wri.com
- Nntp-Posting-Host: bikini.wri.com
- Organization: Wolfram Research, Inc.
- References: <1992Aug13.050021.737@magnus.acs.ohio-state.edu>
- Date: Sat, 15 Aug 1992 01:04:48 GMT
- Lines: 151
-
-
-
- A solution to the puzzle concerning the Ackerman
- function which was posted in article
- <1992Aug13.050021.737@magnus.acs.ohio-state.edu>
- wjcastre@magnus.acs.ohio-state.edu (W.Jose Castrellon G.):
-
-
- >=================================================
- >A(0,m) = m+1
- >A(n+1,0) = A(n,1)
- >A(n+1,m+1) = A(n, A(n+1,m))
- >=================================================
-
- >=================================================
- >1. Find a closed form for A(c,n), c in {1,2,3,4}
- >2. Decide whether A(4,4) is a prime or not.
- >=================================================
-
-
- SOLUTION (SHORT VERSION):
-
-
- A[1,m] == m+2
- A[2,m] == 2*m+3
- A[3,m] == 2^(m+3)-3
-
- A[4,m] == 2^2^...2^16-3
- \______/
- m 2's
-
- A[4,4] == 2^2^2^2^16-3
-
- 2 | 2^2^16-2
- 3 | 2^2^2^16-4
- 12 | 2^2^2^16-4
- 13 | 2^2^2^2^16-2^4
- 13 | A[4,4]
- A[4,4] is not prime.
-
-
- SOLUTION (LONG VERSION WITH COMMENTARY):
-
-
- We get
-
- A[1,0] = A[0,1] = 1+1 = 2
- A[1,m+1] = A[0,A[1,m]] = A[1,m]+1, hence
- A[1,m] = 2,3,4,5,6,... for m=0,1,2,3,4
- A[1,m] = m+2
-
- A[2,0] = A[1,1] = 1+2 = 3
- A[2,m+1] = A[1,A[2,m]] = A[2,m]+2, hence
- A[2,m] = 3,5,7,9,11,... for m=0,1,2,3,4
- A[2,m] = 2*m+3
-
- A[3,0] = A[2,1] = 2*1+3 = 5
- A[3,m+1] = A[2,A[3,m]] = 2*A[3,m]+3, hence
- A[3,m] = 5,13,29,61,127,... for m=0,1,2,3,4
- A[3,m] = 2^(m+3)-3
-
- A[4,0] = A[3,1] = 2^(1+3)-3 = 13
- A[4,m+1] = A[3,A[4,m]] = 2^(A[4,m]+3)-3, hence
- A[4,1] = 2^(13+3)-3 = 2^16-3
- A[4,2] = 2^(2^16-3+3)-3 = 2^2^16-3
- A[4,3] = 2^(2^2^16-3+3)-3 = 2^2^2^16-3
- A[4,m] = 2^2^...2^16-3
- \______/
- m 2's
-
- Rigorous proof of the last lines of the four groups of
- equations above requires mathematical induction. These
- four proofs are easy to do. Discovery of the formulas
- to be proven is what is important and there should be
- enough information above to suggest to the reader how
- we arrived at them.
-
-
- >=================================================
- >2. Decide whether A(4,4) is a prime or not.
- >=================================================
-
-
- This is an application of Fermat's Little Theorem
- which says that
-
- a^(p-1) == 1 (mod p)
-
- if p is prime and (a,p)==1. By part 1 of the puzzle
- we get
-
- A[4,4] == 2^2^2^2^16-3
-
- Now if A[4,4] is not prime (as we hope since proving
- A[4,4] prime seems nearly hopeless) then
-
- 2^2^2^2^16 = 3 (mod p)
-
- for some prime p. If 3 is a power of 2 (mod p), then
-
- 3 = 2^x (mod p)
-
- for some 0<=x<=p-1. So,
-
- 2^2^2^2^16 = 3 = 2^x (mod p) and
- 2^(2^2^2^16-x) = 1 (mod p)
-
- if (p-1) divides 2^2^2^16-x. Recursing downward, we see
- that if q is a prime dividing (p-1), we need to have
-
- 2^2^2^16 = x (mod q)
-
- Perhaps the reader now sees the pattern. Each recursion
- will stip off a "2^".
- Trying out the values p=2,3,5,7,11 and determining
- their success or failure to prove p | A[4,4] using our
- method above will be left to the reader. We skip straight
- to p=13 which happens to work. We get
-
- 2^2^2^2^16 = 3 = 2^4 (mod 13)
-
- So we need
-
- 2^2^2^16 = 4 (mod 12)
-
- Hence we need 2^2^2^16 = 4 = 2^2 (mod 3). We are now in
- good position to work backwards.
-
- 2^2^16-2 = 0 (mod 2) is clear
-
- 2^(2^2^16-2) = 1 (mod 3) by the previous line and F.L.T.
- with p=3
-
- 2^2^2^16 = 2^2 = 4 (mod 3) by the previous line
-
- 2^2^2^16 = 2^2 = 4 (mod 12) by the previous line and observing
- 4 divides both 2^2^2^16 and 4
-
- 2^(2^2^2^16-4) = 1 (mod 13) by the previous line and F.L.T.
- with p=13
-
- 2^2^2^2^16 = 2^4 = 3 (mod 13) by the previous line
-
- A[4,4] = 0 (mod 13) by the previous line and our result
- A[4,4]=2^2^2^2^16-3
-
- Thus A[4,4] is divisible by 13. Since A[4,4] is bigger
- than 13, A[4,4] is not prime.
- The method we used can be used to prove non-primality
- of many other large numbers.
-
-