1

Prove that there exists a positive integer $N$ such that there are at least $2005$ ordered pairs $(x,y),$ of non-negative integers $x$ and $y,$ satisfying $x^2 + y^2 = N.$

I have that the $sqrt(N)$ would be the lcm of the hypotenuses of the first $2005$ Pythagorean triples. Since there are infinite amounts Pythagorean triples, there would be $2005$ pairs of $(x, y).$ But, I don't know if this proves what the question asks for completely.

Jyrki Lahtonen
  • 140,891
qs13
  • 344

3 Answers3

2

In this paper you can see that the number $r_2(n)$ of integral solutions to $x^2+y^2=n$ is exactly $4(d_1(n)-d_3(n))$ (provided that $n$ can be written as a sum of two squares) where $$d_i(n)=\sum\limits_{\substack{d\mid n\\d\equiv i\pmod{4}}}1$$ Now if we take $n$ to be a product of primes of the form $4m+1$, then $d_3(n)=0$ and $r_2(n)=4d_1(n)$. Hence $r_2(N)\geq2005$ for sufficiently large $N$.

James Moriarty
  • 6,460
  • 1
  • 20
  • 39
  • Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nunc quis sagittis dui, sit amet lobortis ipsum. Pellentesque facilisis mi dolor, eget vulputate ante luctus in. Ut pretium erat sem, vulputate cursus nisi tempor non. Sed sed magna lacinia, iaculis lorem sit amet, faucibus velit. Sed sed sagittis odio. Praesent nulla ex, bibendum eu euismod ac, lacinia vel augue. Proin auctor ut ligula et posuere. Sed porttitor dolor eget tellus lobortis feugiat. Ut viverra pulvinar arcu, sit amet aliquet felis. Maecenas sagittis varius nisi et venenatis. – Jeff Hanks May 23 '22 at 20:08
  • there are no solutions, to your problem, where 1≤n≤107 1 ≤ n ≤ 10 7 . – Jeff Hanks May 23 '22 at 20:09
1

Consider the line $\ell$ through $(0,1)$ and $(p/q,0)$, where $p/q\in\mathbb{Q}$ is a rational number greater than $1$ written in lowest terms. $\ell$ intersects the unit circle at ${\left(\frac{2pq}{p^2+q^2},\frac{p^2-q^2}{p^2+q^2}\right)}$, which corresponds to the Pythagorean triple $(2pq,p^2-q^2,p^2+q^2)$. In particular, if $q=1$ and $p$ is even, these form the points ${\left(\frac{2p}{p^2+1},\frac{p^2-1}{p^2+1}\right)}$.

So, you can create as many rational points on the unit circle as you want, say $2005$ of them. Multiply each solution by the common denominator of each point and you're done.

Integrand
  • 8,078
1

Not a 'real' answer, but it was too big for a comment.

I wrote and ran some Mathematica code:

In[1]:=ParallelTable[
 If[TrueQ[Length[Solve[{x^2 + y^2 == n}, {x, y}, PositiveIntegers]] >=
     2005], n, Nothing], {n, 1, 10^7}]

In the code, I am looking for the number of solutions that the command Solve[] finds, by using the command Length[]. Than I only return a solution for $\text{n}$ that satisfy the condition that the number of solutions is bigger or equal to $2005$, where I solved the equation $x^2+\text{y}^2=\text{n}$ for positive integers $x$ and $\text{y}$.

Running the code gives:

Out[1]={}

This shows that there are no solutions, to your problem, where $1\le\text{n}\le10^7$.

Note: this does not imply that there is no solution but only that there is no relatively 'small' solution.

Jan Eerland
  • 29,457