A very interesting problem (1 Viewer)

fishy89sg

Member
Joined
Feb 20, 2006
Messages
674
Gender
Male
HSC
2007
obviously the answer is 0 because you cant put 12 marbles in a circle
 

KFunk

Psychic refugee
Joined
Sep 19, 2004
Messages
3,323
Location
Sydney
Gender
Male
HSC
2005
This kind of question is nasty, I don't think you can solve it through methods other than sheer force. It's not too disimilar to asking how many distinct ways there are to arrange n identical objects, e.g. for 4 objects you have groups of 4, 3+1, 2+2, 2+1+1, 1+1+1+1... which is much the same same deal as the partition function which you will see, if you click the link, isn't the simplest of things.
 

Affinity

Active Member
Joined
Jun 9, 2003
Messages
2,062
Location
Oslo
Gender
Undisclosed
HSC
2003
80 should be the answer, you don't need brute force for this... think burnside+frobenius+cauchy
 
Last edited:

hyparzero

BOS Male Prostitute
Joined
Sep 10, 2005
Messages
246
Location
Wankersville
Gender
Female
HSC
2006
Affinity said:
80 should be the answer, you don't need brute force for this... think burnside+frobenius+cauchy

opps..
wait 80
sorry
I don't think the question is based number theory as Burnside Lemma implies. the BFC is used if we assume the circle to be a 3D object, (one where we can walk around)

But in this case, its not.
 

Affinity

Active Member
Joined
Jun 9, 2003
Messages
2,062
Location
Oslo
Gender
Undisclosed
HSC
2003
80 should be the answer, you don't need brute force for this... think burnside+frobenius+cauchy

EDIT: I mistakenly place 157 there.. It's 80

Here's how you do it:

You consider the spots aroud the circle as distinct, and look at how it behaves under rotations

Burnside's lemma states that the number of orbits in a permutation group is equal to the average number of fixed points under each permutation...
applied here, it would mean that:

The number o different placements(orbits) is equal to the average number of placements which does not change(these are fixed points) under different amounts of rotation(there are 12 possibilities)


For rotation of 0 degrees, every placement is a fixed point: there are 12C = 924 of them
For rotation of 30 and 330 degrees there are no configurations which can be fixed
for rotation of 60 and 300 degrees, there are 2 configurations which are fixed for each (4 total) (both alternating white and black, one starting with white, the otehr starting with black)
and the others:
90,270: none
120:240: 4C2 = 6 each
150,210 : none
180: 6C3 = 20

And then just apply the lemma:

number of distinct placements = (1/12)*(924 + 2 + 6 + 20 + 6 + 2) = 80

(I said 157 first time coz I made a mistake and added another 924 there :S)
 
Last edited:

Affinity

Active Member
Joined
Jun 9, 2003
Messages
2,062
Location
Oslo
Gender
Undisclosed
HSC
2003
hyparzero said:
I don't think the question is based number theory as Burnside Lemma implies. the BFC is used if we assume the circle to be a 3D object, (one where we can walk around)

But in this case, its not.
all you need is a permutation group G (here, rotation) and a G-set X (here, the different arrangements, where each spot on the circle is considered distinct)

an elementary application:

how many ways can you arrange 12 people in a circle?

Rotation of 0 degrees: all 12! configurations are fixed
Other roations: none

So there are 12!/12 = 11! ways agreeing with folklore
 

Affinity

Active Member
Joined
Jun 9, 2003
Messages
2,062
Location
Oslo
Gender
Undisclosed
HSC
2003
If anyone cares, for general n.



(D is the set of divisors of n greater than 2. The first formlua applies if n is even, the second when n is odd)
 
Last edited:

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Top