home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!wupost!usc!news.service.uci.edu!unogate!mvb.saic.com!cpva!singerm
- From: singerm@cpva.saic.com
- Newsgroups: sci.math
- Subject: More Fermat - sort of
- Message-ID: <13682.2b50d05a@cpva.saic.com>
- Date: 11 Jan 93 01:41:46 PST
- Organization: Science Applications Int'l Corp./San Diego
- Lines: 48
-
-
- Here are some results of some work I have done on something I call the
- min function. It is an attempt to evaluate min(n) = the minimum number
- of 'perfect' nth powers required to sum to another perfect nth power:
-
-
- min(n)
- -------
- \
- \ n n
- / A = A for any positive, integral value of n.
- / i min(n)+1
- -------
- i = 1
-
-
- Theorem: min(n) is an increasing function.
-
- Discussion: Certain discrete values for min(n) are known:
-
- 1 1 1
- min(1) = 2 example: 2 + 3 = 5
-
- 2 2 2
- min(2) = 2 example: 3 + 4 = 5
-
- 3 3 3 3
- min(3) = 3 example: 3 + 4 + 5 = 6
-
- 4 4 4 4
- min(4) = 3 95800 + 217519 + 414560 = 422481
-
- 5 5 5 5 5
- min(5) <= 4 27 + 84 + 110 + 133 = 144
-
- I also have results for some fractional and negative exponents. It is my
- hope that by extending the domain of min(n) in this way I can get a better
- feel for the behavior of the function. This has proven correct, but it
- still hasn't led me to a general approach - YET.
-
- I believe this is enough information to give you the idea of what I've
- been working on. Any thoughts/recommendations/help will be welcomed!
-
- And, of course, Fermat's Last Theorem will be nothing more than a simple
- corollary of this theorem.
-
- Mark Singer
- singerm@cpva.saic.com
-