home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!darwin.sura.net!news.Vanderbilt.Edu!vuse.vanderbilt.edu!necs!waknispv
- From: waknispv@necs.Vanderbilt.EDU (Prashant V. Waknis)
- Subject: A sequence matching puzzle
- Message-ID: <BzzDto.28M@vuse.vanderbilt.edu>
- Sender: waknispv@necs (Prashant V. Waknis)
- Nntp-Posting-Host: necs
- Organization: School of Engineering, Vanderbilt University
- Date: Mon, 28 Dec 1992 17:56:11 GMT
- Lines: 42
-
- Consider any two pairs of numbers. (n1, n2) and (n3, n4).
- Let (n1 * n2)/ (n1 +n2) = (n3 *n4) / (n3 + n4) = C.
-
- For example, n1 = 12, n2 = 60, n3 = 15, n4 = 30.
-
- Now, look at the following chart.
-
- 12 | 0 12 24 36 48 60 ....
- 60 | 0 60 ....
- -------------------------------------------
- 15 | 0 15 30 45 60 ....
- 30 | 0 30 60 ....
-
- Can you see how the chart is made? Basically, starting with 0, each
- the sequences are constructed.
- Now try to match the numbers (on the right of the |s) that are above
- the horizontal line to the once that are below the horizontal line.
- The rule is the following.
- Assume that the numbers start appearing on the scene from left to right (in
- increasing order)
-
- So, in the above example,
- all zeros appear first, get matched to other zeros above/below the line.
- 12 comes, waits till 15 comes (for 3 time units), they match up.
- 24 comes, waits till 30 comes (for 6 time units), and matches to one
- of the 30s.
- Remaining 30 waits for 6 units, and matches to 36.
- 45 waits for 48 (3 units) and they match up.
- All 60s match.
- The cycle repeats.
-
- My hypotheis is, in this matching, the max. waiting period (in this case 6)
- can't exceed C ( = (n1 * n2)/ (n1 +n2), in this case 10).
-
- Is this hypothesis right? Can somebody prove/disprove it?
-
-
- -- Prashant
- (waknispv@vuse.vanderbilt.edu)
-
-
-
-