home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math
- Path: sparky!uunet!wri!news
- From: roach@bikini.wri.com (Kelly Roach)
- Subject: Re: u(v^n)w prime puzzle
- Message-ID: <1992Aug19.221946.29768@wri.com>
- Sender: news@wri.com
- Nntp-Posting-Host: bikini.wri.com
- Organization: Wolfram Research, Inc.
- References: <1992Aug19.141957.17410@ecrc.de>
- Date: Wed, 19 Aug 1992 22:19:46 GMT
- Lines: 67
-
- In article <1992Aug19.141957.17410@ecrc.de> jeanmarc@ecrc.de (Jean-Marc
- Andreoli) writes:
- > Let x_n be the number u(v^n)w.
- > hence x_{n+1} - 10^r x_n = (vw - wz)
- > More generally, we have x_{n+1} = a x_n + b
- > and we have to prove that such a sequence is either constant or
- > contains at least one composite, for any integral values of
- > x_0,a,b (with a>1).
- > ...
- > QED
-
-
- Looks like you've got it. Not the proof I had in mind,
- but still, I think it is correct. Perhaps I'll still post my
- solution Friday so that others understand what I was hinting
- at.
-
-
-
- > Now, I propose the following question:
- >
- > Is it true that any sequence x_{n+1} = a x_n + b is either
- >constant or contains *infinitely many* composite ? *almost
- >only* composite (i.e. it contains only a finite number of primes) ?
-
-
- I think I can answer the first question quickly.
- I will just dismiss the cases a=-1,0,1 to the readers.
- Certainly, x_0=1 with x_{n+1}=-x_n is pretty boring.
- Also, we may suppose (a,b)=1 since (a,b)!=1 is not
- interesting. My restriction |a|>1 makes |x_n|<=1 for
- all n impossible. If necessary drop some of the first
- few terms of {x_i} and renumber the sequence so that we
- have |x_0|>1 and also (x_0,a)=1. The latter relation is
- trivial to achieve because (a,b)=1. If you doubt I
- can make |x_0|>1, note |a|>1 and see (2) below.
- Since |x_0|>1 and also (x_0,a)=1, I can pick a
- prime p such that
-
- (1) p | x_0 and p does not divide a
-
- We see by the difference equation x_{n+1} = a x_n + b
- that
-
- (2) x_n = a^n*x_0 + b*Sum[a^i,{i,0,n-1}]
-
- Now this Sum is a geometric series and by (1)
-
- a^i (mod p)
-
- is a periodic function in i. You can read my solution
- to the u(v^n)w puzzle on Friday to find out that
-
- (3) Sum[a^i,{i,0,n-1}] = 0 (mod p) if p(p-1) | n.
-
- or prove it yourself. Hence, by (1), (2), and (3), we
- have
-
- (4) p | x_n if p(p-1) | n
-
- By |a|>1 and (2), it is clear that |x_n| -> infinity,
- so it follows by (4) that infinitely many of the x_n
- are composite.
-
-
-
-
-