Article View: rec.puzzles
Article #2675Re: A question on the game "SET"
From: Andrew C. Plotkin
Date: Tue, 12 Jan 1993 18:23
Date: Tue, 12 Jan 1993 18:23
57 lines
2753 bytes
2753 bytes
Excerpts from netnews.rec.puzzles: 10-Jan-93 A question on the game "SET" Mark Kobayashi@hpindda.c (696) > I thought someone here might know the answer.... > I bought the card game "SET" (matching off 3 cards at a time with > similar or unique combinations). > My question - what is the maximum number of cards which can > be put down which would not contain any "set"? (or, what > number of cards would guarantee at least 1 set?) Ha -- I just subscribed to rec.puzzles for the sole purpose of asking that question :-) Since you seem to have left out the game rules, let me do so: Each card shows a figure. The figure has N independent attributes; each attribute has 3 possible values. So there are 3^N possible cards. The deck contains exactly one of each. In the deck sold, N=4, so there are 81 cards. The attributes are shape (diamond / oval / squiggle) shade (solid / striped / hollow) color (red / green / blue) number (one copy / two copies / three copies) Three cards form a "set" if: For every attribute, either all three cards' values match, or the three cards' values are all different. So, for the standard deck (N=4): (one solid red diamond) (two solid red diamond) (three solid red diamond) is a set. So is (one solid blue diamond) (two solid green diamond) (three solid red diamond) And (one hollow blue diamond) (two striped green oval) (three solid red squiggle) Three identical cards would form a set, except that there is only one of each card in the deck. But (one solid blue oval) (two solid green diamond) (three solid red diamond) is not a set; it fails for the attribute of shape. Again, the question is: what is the maximum number of cards you can lay out without any three of them forming a set? I've thought about this for a few days now. My first guess was 2^N, but that's wrong; I found arrangements of 9 cards with no set in the N=3 case, and 18 cards for N=4. Bonus question (easier, I have an answer and proof): How does the answer change if you allow repeated cards? BTW, it's a very cool game, and only costs $10 or $12. The game is just laying out cards and trying to find sets as fast as possible. --Z
Technical Details
Message-ID:
<cfIkmBu00gpIQH02s8@andrew.cmu.edu>Path:
arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!news.sei.cmu.edu!bb3.andrew.cmu.edu!crabapple.srv.cs.cmu.edu!andrew.cmu.edu!ap1i+References:
<5240016@hpindda.cup.hp.com>Show Full Headers
Complete Article Headers
Organization: School of Computer Science, Carnegie Mellon, Pittsburgh, PA Path: arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!news.sei.cmu.edu!bb3.andrew.cmu.edu!crabapple.srv.cs.cmu.edu!andrew.cmu.edu!ap1i+ Newsgroups: rec.puzzles Message-ID: <cfIkmBu00gpIQH02s8@andrew.cmu.edu> Date: Tue, 12 Jan 1993 13:23:09 -0500 From: "Andrew C. Plotkin" <ap1i+@andrew.cmu.edu> Subject: Re: A question on the game "SET" In-Reply-To: <5240016@hpindda.cup.hp.com> References: <5240016@hpindda.cup.hp.com> Lines: 57