Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Round table

  1. Mar 20, 2005 #1

    T@P

    User Avatar

    two people are playing the amazingly complex and strategy filled game that has no name.

    the way you play is like this: you take an ideal and round table, and alot of ideal circular coins (normal coins, little cylinders, nothing crazy). each player then (by turn) puts a coin onto the table. the loser is the one that cant put another coin on to the table so that its not on top of any other coins. also, you cant move the coins once they are placed. you cant put the coins on the table sideways. nothing funny.

    question is, is there a strategy (to win) for the first player or the second player?

    major hint. there is. :rofl: otherwise it would be a boring question
     
  2. jcsd
  3. Mar 20, 2005 #2
    Player 1 puts a coin in the center of the table, and then he mirrors every move by player 2 by a move the same distance from the center but exactly opposite. This is just like the winning strategy for eot-cat-cit!
     
  4. Mar 21, 2005 #3

    T@P

    User Avatar

    too fast!! lol exactly right. what i like about the puzzle is it complete lack of details - and the totally general solution. any way i didnt read the eot posts, so ill trust you on that. maybe i should...
     
  5. Mar 22, 2005 #4
    great :) really cool puzzle!
     
  6. Mar 22, 2005 #5
    player one could pick a table the size of his coin :P
     
  7. Mar 22, 2005 #6

    T@P

    User Avatar

    hehe actually a table that was even a little less than twice the radius of his coin would work :)
     
  8. Mar 22, 2005 #7
    What if the table is an equilateral triangle?
     
  9. Mar 22, 2005 #8

    T@P

    User Avatar

    hmmm. triangles dont have and "even" radial symmetry...

    this means you cant find "the" spot to put your coin and copy his/her moves. unless im wrong.

    you could attempt going for the height and trying to fill it up, and then follow symmetry along that axis, but it would then depend on who puts the last coin in the height. (depends on the height)

    well thats all applying the same basic idea as in the round one. there maybe a new idea to use here. did you have anything in mind bicycletree?

    the equilateral triangle idea would be a good one for three people. that way you kick out the loser (the one who goes second) and then the last two face off in the round table game... but its predetermined. and the other two may "team up" against the third (this is why more than 2 player games are hard to analyze this way).

    anyone else have some brilliant ideas?
     
  10. Mar 22, 2005 #9
    No, I just tossed it out there. I don't know how to solve it. Maybe if you looked at small triangles and considered all possible types of moves, some strategy might appear.
     
  11. Mar 22, 2005 #10
    If the triangle is just large enough to fit a central coin and three other coins in the corners, then putting the first coin in the center loses, so a perfect strategy can't always be based around putting a coin in the center.
     
  12. Mar 23, 2005 #11

    T@P

    User Avatar

    very true. only if one were to have a winning strategy, it must depend on the first persons ability to go first, other wise you have the old white wins in double chess trick. therefore, if the first player does not make a *unique* move, or one that the second player can copy and end up where they started, you can say that the strategy does not exist for that player. (if you know the double chess trick, this should make sense. to those who dont, in a nutshell, it is like this: assume you play chess in where a player goes twice in one turn. you must go twice. the question is, who wins? and the answer is like this: assume black has a winning strategy. then white moves the knight in and out an suddenly black plays first (white did nothing on the first turn) and you can think of white as black and the other way too. that means that white should win, but we assumed black does. contradiction, therefore white *has a winning strategy*. im not sure anyone knows what it is though...
     
  13. Mar 23, 2005 #12
    No, therefore black does not have a winning strategy.
     
  14. Mar 23, 2005 #13
    Who gets to choose the table? Player 1 or 2 (or 3?)
     
  15. Mar 23, 2005 #14

    T@P

    User Avatar

    haha i guess player one would always choose the round table, 2 the triangular and 3 the four dimensional torus table.

    and jimmysnyder, we proved that black cant have a winning strategy, which more or less entails that it cant win (if played correctly etc.) because if it did have a strategy, white would simply *become* black and then win with the smae strategy.
     
  16. Mar 23, 2005 #15
    If I was player 2 and could choose the table, I would choose a size, of any shape table, that was too small for any "ideal." Then player 1 (or 3) would be unable to play, thus, forfeiting the game. Player 2 wins. yeh!
     
  17. Mar 23, 2005 #16

    T@P

    User Avatar

    i think the 19 dimensional torus would be better, since the coint would keep falling off...
     
  18. Mar 24, 2005 #17
    T@P,

    Your proof is a good one. It starts out with "assume black has a winning strategy" and by "reductio ad absurdum" proves that statement false. In other words "Black does not have a winning strategy."

    However, this is not the same thing as "White has a winning strategy." It may be that neither has a winning strategy. In any case, you have not proved that White has a winning strategy, only that Black doesn't. If you don't see this, perhaps you have not accounted for the fact that Black might be able to force a draw.

    Again, your proof that Black has no winning strategy is a good one. My only issue is with your statement that I quoted before "therefore white *has a winning strategy*."
     
  19. Mar 24, 2005 #18

    T@P

    User Avatar

    you raise a good point. however, i read the proof in some bigwig book and im sure its not just a figment of my imagination. so theres that justification ;)

    however, i think i mis-spoke. the idea is that you assume *black will WIN* not black has a strategy. you prove that black *cannot* win (if everyone plays their best). clearly white wins then, which proves the existance of a strategy. i hope i explained it better now :)
     
  20. Mar 24, 2005 #19
    Statements like "Black will win" or "White will win" are almost certainly not true. Either side can probably lose deliberately even if there is a winning strategy.
     
  21. Mar 24, 2005 #20
    The game might end in a draw.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Round table
Loading...