Maths Olympiad Prep

Track / Stage 6 / 321 of 400 #1321 of 1964

Problem 1321

National olympiad, first round
Geometry Difficulty 6.5 Find the answer

Does there exist on the Cartesian plane a convex 20232023-gon with vertices at integer points, such that the lengths of all its sides are equal?

Proposed by Anton Trygub

The source for this one didn't record the answer, so there is nothing to check what you type against. Work it on paper and mark yourself against the solution below.

Official solution

To determine whether there exists a convex 20232023-gon with vertices at integer points and all sides of equal length, we need to analyze the possible values of the squared side length ss.

1. **Checkerboard Coloring Argument for Odd ss:**
- Suppose the common side length is s\sqrt{s}, where ss is odd.
- Color the lattice points in a checkerboard pattern, where each point is either black or white, and adjacent points have different colors.
- In a convex 20232023-gon, each vertex must be connected to two other vertices, forming sides of the polygon.
- Since ss is odd, the distance between any two adjacent vertices (which is s\sqrt{s}) must connect points of different colors.
- However, since 20232023 is odd, there would be an odd number of sides, implying an odd number of color changes. This is impossible because the polygon must return to its starting point, requiring an even number of color changes.
- Therefore, no such polygon exists if ss is odd.

2. **Row Coloring Argument for s2(mod4)s \equiv 2 \pmod{4}:**
- Suppose s2(mod4)s \equiv 2 \pmod{4}.
- Color the rows of the lattice alternately black and white.
- For any side of length s\sqrt{s}, the squared distance s=Δx2+Δy2s = \Delta x^2 + \Delta y^2 must hold.
- Given s2(mod4)s \equiv 2 \pmod{4}, both Δx\Delta x and Δy\Delta y must be odd (since the sum of two squares is 2(mod4)2 \pmod{4} only if both squares are 1(mod4)1 \pmod{4}).
- This implies that any side of the polygon connects points in different rows, hence different colors.
- Again, since 20232023 is odd, there would be an odd number of color changes, which is impossible for a closed polygon.
- Therefore, no such polygon exists if s2(mod4)s \equiv 2 \pmod{4}.

3. **Scaling Argument for s0(mod4)s \equiv 0 \pmod{4}:**
- Suppose s0(mod4)s \equiv 0 \pmod{4}.
- Then both Δx\Delta x and Δy\Delta y must be even (since the sum of two squares is 0(mod4)0 \pmod{4} only if both squares are 0(mod4)0 \pmod{4}).
- This means we can write Δx=2a\Delta x = 2a and Δy=2b\Delta y = 2b for some integers aa and bb.
- The side length s\sqrt{s} can be scaled down by a factor of 22, resulting in a new polygon with side length s4\sqrt{\frac{s}{4}}.
- Repeating this process, we eventually reduce the problem to one of the previous cases (either ss becomes odd or s2(mod4)s \equiv 2 \pmod{4}).
- Since we have already shown that no such polygon exists in those cases, it follows that no such polygon exists if s0(mod4)s \equiv 0 \pmod{4} either.

Since we have exhausted all possible cases for ss, we conclude that no convex 20232023-gon with vertices at integer points and all sides of equal length exists.

Source: NuminaMath-1.5, licensed Apache-2.0. Statement and solution reproduced as published; topic, difficulty and ordering added by this site.