(Assume that there is no friction and drag, and the ball follows the law of reflection.) Suppose we have an \(n\)-gon. There is a point-like ball at the midpoint of one of the sides of the \(n\)-gon (call that side Side A). It is launched to the midpoint of another side (call that Side B). Let \(k\) be the number of sides clockwise to Side A but counterclockwise to Side B (not including Side A and Side B). Define the function \(\delta_n(k)\) to be \(k\) if the ball bounces off every side of the \(n\)-gon before returning to the launch point, and otherwise \(0\). Find the closed form for the sum \[\sum^{n-2}_{k=0}{\delta_n(k)}\]
EDIT: Here is a hint: Euler's Totient Function
And if that doesn't make sense, here is a stepping stone: GCD. (Thanks Calvin L.)
Easy Math Editor
This discussion board is a place to discuss our Daily Challenges and the math and science related to those challenges. Explanations are more than just a solution — they should explain the steps and thinking strategies that you used to obtain the solution. Comments should further the discussion of math and science.
When posting on Brilliant:
*italics*
or_italics_
**bold**
or__bold__
paragraph 1
paragraph 2
[example link](https://brilliant.org)
> This is a quote
\(
...\)
or\[
...\]
to ensure proper formatting.2 \times 3
2^{34}
a_{i-1}
\frac{2}{3}
\sqrt{2}
\sum_{i=1}^3
\sin \theta
\boxed{123}
Comments
Are you bobthesmartypants? :P
Log in to reply
Yeah that's what I thought.
yep :P
I assume you are talking about a regular n-gon?
Log in to reply
yes.
A better hint would be Greatest Common Divisor, though it might give away too much.
Is the answer zero??
Log in to reply
no, sorry.
Consider a square. δ4(2)=2, because the ball will hit all sides before returning to its initial position. Also, δ4(0)=0, but that doesn't contribute to the sum. So k=0∑2δ4(k)=2 In general, k=0∑n−2δn(k)≥n−2 Another observation I made is that k=0∑p−2δp(k)=k=0∑p−2k=2(p−2)(p−1) for every prime number p. This is because the ball will never return to the initial position before hitting each side in such a p-gon, no matter which side you point the ball at.
Log in to reply
Sorry, but you answer is incorrect. Solving for the square does not solve it for every other n-gon.
HINT: Euler's totient function
Log in to reply
Yes, I already figured out I should use the Totient function. :) Also, I did not give an incorrect answer, because I did not yet give an answer… I just showed my observations. :)
I know that k=0∑n−2δn(k)=−ϕ(n)+k=1∑n−1{k0if gcd(n,k)=1otherwise but that's all I got so far. ;-) Can it get better than this?
Log in to reply
Log in to reply
k=1∑n−1{k0if gcd(n,k)=1otherwise is the sum of all values k smaller than n that are coprime to n. These values k seem symmetric about 2n. For example, if n=6 then k=1 or k=5. And (intuitively) this works for any n. So basically, the average for all values k is 2n. So the second term in my previous solution is equal to the average (2n) multiplied by the number of terms (ϕ(n)): k=0∑n−2δn(k)=ϕ(n)∗2n−ϕ(n)=ϕ(n)(2n−1) I'm guessing this is the answer? :) That was a very entertaining problem, thanks.
The second term: