home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!uknet!comlab.ox.ac.uk!oxuniv!loader
- From: loader@vax.oxford.ac.uk
- Newsgroups: sci.math
- Subject: Re: Chess Problem
- Message-ID: <1992Sep12.154632.8817@vax.oxford.ac.uk>
- Date: 12 Sep 92 14:46:32 GMT
- References: <BuFpLp.9nI@ecf.toronto.edu> <BuGA8t.DL8@cmptrc.lonestar.org>
- Organization: Oxford University VAX 6620
- Lines: 54
-
- In article <BuGA8t.DL8@cmptrc.lonestar.org>, carter@cmptrc.lonestar.org (Carter Bennett) writes:
- > rairan@ecf.toronto.edu (RAI Ranjan) writes:
- >>Chess Problem:
- >> Eight rooks are placed at random on a chessboard.
- >> What is the probability that no two rooks can attack one another?
- >
- > Ahoy!
- >
- > Oooh! I like these problems. Glad you sent it along!
- > First, let's check the number of distributions of 8 rooks on the board.
- > This is the combination calculation of 64 squares taken 8 at a time:
- >
- > / 64 \ - (64!) / ( (64 - 8)! 8! ) = 4426165368
- > \ 8 / -
- >
- > So there are 4426165368 possible ways to place the 8 rooks on the
- > board. My understanding is that no two rooks can attack each other
- > only when they are arranged such that they line up on one of the two
- > diagonals of the board. Meaning there are 2 ways out of 4426165368
- > possible setups that meet the criteria. Putting that in normalized odds:
-
- There's a few more than this, aren't there? All we need is exactly one rook on
- each row and one on each column, which gives us 8! possible ways...
- >
- > 2 / 4426165368 -> 1 chance in 2213082684
- >
- so isn't it 1 chance in (64!) / ((64-8)! 8! 8!)
-
- > Two REALLY NEAT THINGS about this problem -
- >
- > I've seen chess boards many times. We've seen all kinds of arrangements
- > of chess pieces. Just knowing how many possible arrangements of eight
- > pieces is still fascinatingly mind-boggling to me.
- >
- > BUT THE REALLY NEAT THING ABOUT THIS PROBLEM is that I was able to give
- > you integer answers to this problem because I had a calculator (HP48SX)
- > on hand that could handle them. dc(1) on Unix would do it, too. But
- > just a few short years ago, we would have had to settle for approximations
- > in exponential notation unless we actually multiplied 64x63x62...x56 out
- > by hand, and then dividing the result by 40320, again by hand!!! Too
- > many significant digits to keep track of on yesterday's calculators!
-
- THE REALLY NEAT THING about not having a calculator is that I get the answer as
- an expression that means something to me, not a just a sequence of digits...
-
- >
- > FOLLOW-UP QUESTION FOR EXTRA CREDIT:
- >
- > Determine the odds for doing the same with eight bishops! ;-)
- >
- > Carter R. Bennett, Jr. - Scientist No matter where you go...
- > carter@scilab.lonestar.org - home .../dev/tty!
- > carter@cmptrc.lonestar.org - work
- > KI5SR
-