Packing the Squares

Geometry Level 5

What is the perimeter of the smallest rectangle that can contain all the squares with side length of 1 n \LARGE \frac{1}{n} .

Note: n N n∈\mathbb{N} and n 1 n \neq 1


The answer is 3.214.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

2 solutions

Michael Mendrin
Jun 25, 2015

1) The problem should ask for the perimeter of the rectangle with the smallest area, instead of the ambiguously worded, "smallest rectangle".

2) Again, to reduce ambiguity, it should be stated that this is a packing problem where one of each such square must fit inside the rectangle of least area without overlapping.

3) The answer is the lower limit, but proof is lacking that all such squares with sides 1 n \frac {1}{n} can actually fit inside a rectangle with this area. Lacking such proof, maybe the problem should be reworded to ask for the lower limit?

I gave my answer by taking the chance that the correct answer is the lower limit, without knowing how I could pack all those squares inside such a rectangle.

Digvijay Singh
May 19, 2015

Let S 1 \large S_{1} and S 2 \large S_{2} be the sides of the rectangle

Then S 1 \large S_{1} is the sum of largest two squares sides summed together

S 1 = 1 2 + 1 3 = 5 6 \large S_{1}=\frac{1}{2}+\frac{1}{3}=\frac{5}{6}

and area A \large A is the sum of the area of all the squares

A = n = 2 1 n 2 = π 2 6 1 \large A=\displaystyle \sum_{n=2}^\infty\frac {1}{n^2}=\frac {\pi^2}{6}-1

The side S 2 = A S 1 = π 2 6 1 5 6 = π 2 6 5 \large S_{2}=\frac{A}{S_{1}}=\frac{\frac {\pi^2}{6}-1}{\frac{5}{6}}=\frac{\pi^2-6}{5}

So, the perimeter = 2 ( S 1 + S 2 ) = 2 5 π 2 11 15 3.214 \large =2(S_{1}+S_{2})=\frac{2}{5}\pi^2-\frac{11}{15}\approx\boxed{3.214}

Relevant .

Pi Han Goh - 5 years, 11 months ago

Your question didn't say that the squares mustn't overlap. And why must S 1 S_1 be the sum of the largest TWO squares only? Why not three? Or four, and so on?

And even if you're right, you have not shown that such a rectangle exist. How do you confine these squares into a fixed dimensions? Do you arrange them like in Fibonacci Blocks ?

Pi Han Goh - 6 years ago

Log in to reply

One bound to the answer is dependent on one rectangle side, S 1 S_{1} , to be the sum of the largest two squares sides summed together

The other rectangle side, S 2 S_{2} , depends on the order of remaining squares. However, we know the total area, A, of the squares

So the other ideal rectangle side length can be found

Digvijay Singh - 6 years ago

Log in to reply

You didn't answer my question, I asked "why must it be sum of largest TWO squares only?"

Just because you said it exist does not mean it exist. I could say I can fit seven squares of sides 1 , 1 , 2 , 3 , 5 , 8 , 13 1,1,2,3,5,8,13 inside a rectangle, but I still need to show a geometric construction of the arrangements. Do you have a rough sketch of a geometric figure to that shows the arrangements of such squares in the fixed rectangle?

Pi Han Goh - 6 years ago

Log in to reply

@Pi Han Goh Would you please check the last section of this

Digvijay Singh - 5 years, 11 months ago

Log in to reply

@Digvijay Singh Wow, interesting! Do you have the visual proof of the arrangements of the blocks?

Pi Han Goh - 5 years, 11 months ago

Log in to reply

@Pi Han Goh nah... I don't... but, if there is no arrangment of such squares, why would they post such question on wikipedia?

Digvijay Singh - 5 years, 11 months ago

Log in to reply

@Digvijay Singh Yea, Wikipedia is pretty much reliable. I'm under the impression that you can construct the rectangle.

Pi Han Goh - 5 years, 11 months ago

Log in to reply

@Pi Han Goh Like Fibonacci Blocks?

Digvijay Singh - 5 years, 11 months ago

Log in to reply

@Digvijay Singh I doubt it. Fibonacci sequence is the subset of of harmonic sequence.

Pi Han Goh - 5 years, 11 months ago

Log in to reply

@Pi Han Goh umm.. okay... now what..? Is this question good to be posted? Or still faulty?

Digvijay Singh - 5 years, 11 months ago

Log in to reply

@Digvijay Singh Nope. It's all good. But I think a complete solution is very difficult to write out.

Pi Han Goh - 5 years, 11 months ago

Log in to reply

@Pi Han Goh all right.. I've posted a similar question - Packin' Squares . Check it out

Digvijay Singh - 5 years, 11 months ago

Log in to reply

@Digvijay Singh Yay! Now I'm gonna spend hours on that problem!

Pi Han Goh - 5 years, 11 months ago

@Pi Han Goh I have to believe that there must be some infinite subset of the natural numbers A such that i = 0 1 a i 2 = π 2 6 1 \displaystyle \sum_{i=0}^{\infty} \frac{1}{a_i^2}=\sqrt{\frac{π^2}{6}-1} , leading to a minimal perimeter of 4 π 2 6 1 3.212 4\sqrt{\frac{π^2}{6}-1}\approx3.212

Garrett Clarke - 5 years, 11 months ago

Log in to reply

@Garrett Clarke I'm certain that I would like to see your proof! ¨ \ddot\smile

Pi Han Goh - 5 years, 11 months ago

Log in to reply

@Pi Han Goh Haha what I wouldn't give to be able to supply it!

Garrett Clarke - 5 years, 11 months ago

@Digvijay Singh @Digvijay Singh Which last section??? Could you please check again??? There are only Series and Integral representations of zeta(2) in the last section...............

Aaghaz Mahajan - 2 years, 7 months ago

@Pi Han Goh @Pi Han Goh I still have a doubt so as to why it must be the sum of largest TWO squares??? Could you please explain it sir???

Aaghaz Mahajan - 2 years, 7 months ago

Log in to reply

@Aaghaz Mahajan You just reiterated what I said.

I asked the author why if the two sides of the rectangle be S 1 S 2 S_1 \leq S_2 , then S 2 S_2 must have the sum of lengths of the 2 largest squares.

Pi Han Goh - 2 years, 7 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...