8 Lonely Rooks

Submitted by kurtgodden on Wed, 07/02/2008 at 11:09am.

Chess is a mental playground for mathematicians.  How many ways can you place 8 rooks on a chessboard such that none of them attacks another?  This is similar to the 8 Lonely Queens problem that I discussed in another blog.  For the rooks, there is only one solution for approximately every 109,776 different ways of placing the rooks on the board.  However, if you drew the conclusion that there must not be very many solutions, you would be wrong.  There are, in point of fact, 40,320 solutions to the rook problem.

This is easy to calculate.  There are 8 ways of placing the first rook on file a.  There are 7 ways of placing the second rook on file b.  You only have to be careful not to place it on the same rank as the first rook.  Thus, there are 8 * 7 = 56 different ways of placing the first two rooks on files a and b.  With each rook on the board, the number of valid ways of placing the next rook is decreased by one.  Continuing in this way there are 8 * 7 * 6 * 5 * 4 * 3 * 2 * 1 = 40,320 solutions.

The information in the first two paragraphs implies that there must be an amazingly large number of ways to place the 8 rooks onto the board.  Again, the math is simple.  There are 64 different squares on which to place the first rook.  There are 63 other squares where you can place the second rook.  Continuing with this pattern, there is the honking huge number of 64 * 63 * 62 * 61 * 60 * 59 *58 *57 = 178,462,987,637,760 different ways to place the 8 rooks on the chessboard. 

But don’t honk just yet because I have deceived you in a sense.  Consider the accompanying picture that shows the rooks all on a diagonal.  If you rearrange the actual rooks in some way, but still keeping them on the same diagonal (e.g. swap the two rooks on the first two ranks), you would still have the same picture.  If we eliminate all permutations due to swapping, then our number is significantly smaller – a mere 4,426,165,368.

Dividing this number by the number of rook solutions, and then rounding, gives us the 109,776 number cited in the first paragraph.  That’s still a pretty large haystack to bury each solution in.

Returning to those solutions, a great many of them are either rotational or reflective transformations of the others.  If we consider any such transformations as being equivalent then our 40,320 solutions reduces to 5,282 distinct solutions to the 8 Lonely Rooks problem – still a large number, especially when compared with the 12 distinct solutions for the 8 Lonely Queens problem. 

But this should not be overly surprising since the queens problem involves the additional constraint that the queens cannot be on the same diagonal, which greatly reduces the number of possible solutions.

Your homework is to compile the 5,282 distinct rook solutions and publish them on the web.  This assignment is not as hard as it might seem if you are a programmer.  Extra credit will be given for identifying the 12 solutions that are configurationally identical to the 8 distinct queen solutions.


 

Comments:

by estevon - 11 months ago
Maine United States
Member Since: Oct 2008
Member Points: 1479

How would the ANSWER would be if NO Rooks is NOT touching each other Dialongany,and not attacking each other=Adjacents squares..=The # of solutions.=How many?

by santiR - 15 months ago
outside Washington D.C. United States
Member Since: Apr 2008
Member Points: 1004
no, dragonknightx.
by Ellbert - 16 months ago
Baltimore United States
Member Since: Jan 2008
Member Points: 151
I learned something I did not know.Thank you Kurtgodden.
by taxman224201 - 16 months ago
Brawley, ca United States
Member Since: May 2008
Member Points: 21
do you write these yourself? whoever writes these stories has to have some degree of genius on their side
by Gogetax - 16 months ago
The Shadowlands International
Member Since: Mar 2008
Member Points: 684
Does playing chess require great skills of maths?
by TimMoroney - 16 months ago
Michigan United States
Member Since: Feb 2008
Member Points: 77
I have never encountered a treatise of an 8 rook puzzle before. Though I imagine it is the mathematical foreground that keeps it out of everyday coffee-clubhouse discussions. I suppose another reason could be the potential practical application for the concepts involved in the queen problem. Understanding a trick for keeping your queen safe from an enemy queen during a time struggle can prove somewhat useful. Studying the 8 lonely rooks might have the same practical application of preparing opening theory in case they ever allow pawns to move backwards.

Personally, I find it fascinating. Think of all the artistic designs that could be created. Actually, this gives me a thought I may have to follow up on...

And, without applying too much mental strain, could I take a stab at the knight puzzle and suggest the answer is 32?
by Phobetor - 16 months ago
Eindhoven Netherlands
Member Since: Aug 2007
Member Points: 1202

I wonder how many mathematicians play chess, and how many chess players are at least good at maths. I like both maths and chess alot, and I know quite some chess players who are good at maths too.

It's an interesting read, but most of it is fairly simple for me Tongue out Still, how did you get to the number 5,282 from taking all "double" from the 40,320? You can't just divide by 8, because some are in some way symmetric. But the fraction of 40,320 / 5,282 = 7.63347... seems somewhat "random" to me. So I guess you just somehow manually deleted all the doubles? 

 

It's also nice to ask people how many knights you can place on a board without any of them attacking the other. It's quite simple, but some people would work out difficult placements and patterns, ending up with "I don't know". In fact, if you count the number of distinct ways to place as many knights as possible on a board, without attacking eachother, you'd only get 1! But that surely doesn't mean that it's 12 times easier than the 8-queens problem... Wink


by Omicron - 16 months ago
Buenos Aires Argentina
Member Since: Apr 2008
Member Points: 192
This is interesting... I like how you think of it at first and it seems hard to calculate, but in truth is fairly simple. thanx for sharing.
by ADK - 16 months ago
Santa Clarita, CA United States
Member Since: Aug 2007
Member Points: 16233

NICE read!!! (AGAIN)

ADK


 

Add your comment:

Join Chess.com for free to add your comment! Already a member? Then login now to comment.