0
$\begingroup$

$$k*100x^2+20x=100y^2+20y$$ I have the following equation where $k $ and $y $ and $x $ are all positive intergers. Also $k\gt1$ and $k\neq x\neq y $ Is there any solutions to this equation?

$\endgroup$
  • 3
    $\begingroup$ Sure $$x = y = 0$$ Also, $$k = 1, x = y$$ $\endgroup$ – Moo May 31 '18 at 21:55
  • $\begingroup$ I forgot to say positive $\endgroup$ – PintOfMilk May 31 '18 at 21:55
  • $\begingroup$ @moo see my edited response $\endgroup$ – PintOfMilk May 31 '18 at 21:58
  • 1
    $\begingroup$ Well $k = 37; x = 1; y=6$ is a solution. I have found that for there to be a solution we must have $y = x + 5j$. And so $(k-1)x^2 = 10xj + 25j^2 + j$ but... If we set $x = 1$ we can set $j$ to any integer value. But I don't see how to solve in general. $\endgroup$ – fleablood May 31 '18 at 23:24
3
$\begingroup$

Yes. The MATLAB code below

f_L = @(x, k) k * 100 * x * x + 20 * x;
f_R = @(y) 100 * y * y + 20 * y;

LIMIT = 100;
i = 1;

for k = 2:1:LIMIT
    for x = 1:1:LIMIT
        for y = 1:1:LIMIT
            if k ~= x && k ~= y && x ~= y
                if f_L(x, k) == f_R(y)
                    arr(i, 1) = k;
                    arr(i, 2) = x;
                    arr(i, 3) = y;
                    i = i + 1;
                end
            end
        end
    end
end

produces $$\begin{array}{|c|c|c|} \hline k & x & y \\\hline 19 & 3 & 13 \\ 21 & 7 & 32 \\ 28 & 14 & 74 \\ 29 & 8 & 43 \\ 37 & 1 & 6 \\ 59 & 6 & 46 \\ 73 & 2 & 17 \\\hline \end{array}$$

| cite | improve this answer | |
$\endgroup$
2
$\begingroup$

Final answer. For any $x$ solve (if possible) $10xj + 25j^2 + j\equiv 0 \mod x^2$.

Then $y = x + 5(j + mx^2)$ and $k = \frac {10x(j+mx^2) + 25(j+mx^2)^2 + (j+mx^2)}{x^2} + 1$ will be solutions.

Why? Well see my working below.

====== working below (or old answer) =====

Divide everything by $20$.

So $k*5x^2 + x = 5y^2 + y$

So $x - y = 5y^2 - k5x^2$. Assuming $x\ne y$ then

$x-y = 5(y^2 - x^2) - (k-1)5x^2$

$1 = 5(y + x) - 5\frac {(k-1)x^2}{x-y}$

So $x-y|5(k-1)x^2$ but assuming $k > 1$ then $x-y \not \mid (k-1)x^2$ so, as $5$ is prime $5|x-y$ and $x \equiv y \mod 5$.

So $y = x + 5j$ for some positive $j$.

$k*5x^2 + x = 5(x + 5j)^2 + x + 5j$

$k*5x^2 = 5x^2 + 50xj + 125j^2 + 5j$

$(k-1)x^2 = 10xj + 25j^2 + j$

Welp... if we set $x=1$ we can get an infinite number of solutions.

$k = 10j + 25j^2 + j + 1$

So $x = 1$ and $k=37;y=6$ and $k= 123; y = 11$ etc. will be solutions.

.... otherwise....

Let $x = x$ (as Laurie Anderson used to say)

Then solve for $10j + 15j^2 + j \equiv 0 \mod x^2$ if possible.

Let $k - 1 = \frac {10j + 15j^2 + j}{x^2}$. That Will give us solutions.

If $x = 2$ then $j$ can be any multiple of $4$ of $3 \mod 4$.

And if $x = 3$ we need $(k-1)9 = 31j + 25j^2$ and ...$31j + 25j^2 \equiv 4j-2j^2\mod 9$ so ... $j = 2+ 9m$ is a solution (as are $j = 9m$).

$x = 3; y = 13; k = \frac {162}9 + 1= 19$.

Ex: $x = 2$ then $4(k-1)= 21j + 25j^2$ so $21j + 25j^2 \equiv j + j^2 \equiv 0 \mod 4$ and so

$j = 3$

$x = 2; y=17; k= \frac {63 + 225}4 + 1 = 73$

.....

$37*100*1^2 + 20*1 = 100*6^2 + 6*20$

$3720 = 3600 + 120$. Check

$123*100*1^2 + 20 = 100*11^2 + 11*20$

$12320 = 12100 + 220$. Check.

$19*100*3^2 + 20*3 = 100*13^2 + 13*20$

$19*9*100 + 60 = 16900 + 260$

$17160 = 17160$. Check.

$x = 2; y = 17;k=73$ and

$73*100*2^2 + 2*20 = 100*17^2 + 17*20$

$29240 = 28900 + 340$. Check.

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ thanks for your answer! $\endgroup$ – PintOfMilk May 31 '18 at 23:38
  • $\begingroup$ (All variables are positive intergers) The original problem was proving that $100x^2+20x\neq m! $ for all values of $x$ and $m$. Stupidly I thought someone like this. $\endgroup$ – PintOfMilk Jun 1 '18 at 0:07
  • 1
    $\begingroup$ $100x^2 + 20x =x*4*5(5x + 1)$ Consider the prime factorization of $x$ and of $m!$. $\endgroup$ – fleablood Jun 1 '18 at 0:27
  • 1
    $\begingroup$ $x$ and $5x+1$ are relatively prime. Let $p$ and $q$ be primes larger than $x$ but less than $m$ then $p$ and $q$ divide $5x + 1$ but if $x > 5$ then $5x + 1 < pq$. $\endgroup$ – fleablood Jun 1 '18 at 0:36
  • 1
    $\begingroup$ No, I meant less than $m$. $m!$ must be divisible by every number less than and including $m$. The larger $m$ is the more incrediably unlikely the factors $x, 4,5$ and $5x +1$ are going to cut it. If $p, q$ are prime and $x < p < m$ and $x < q < m$ then $pq|5x + 1$ and if $p,q > 5$ that is impossible. It'll take a small amount of parameter setting to show that for most $m$ there will have to be such primes and to make exceptions for the very small values of $m$ where this isn't true but that's a solid game plane. $\endgroup$ – fleablood Jun 1 '18 at 4:49
0
$\begingroup$

Definitely. To do this easily, simply set x=0, making the value of k irrelevant. So you now have $100y^2+20y=0$. This can obviously be simplified to $5y^2+y=0$. Factoring out y, we get $y(5y+1)=0$, making $y=-\frac{1}{5}$ and $y=0$ solution. Since $y\neq x$, $y=-\frac{1}{5}$

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ this is a nice solution however are all the variables are positive intergers. $\endgroup$ – PintOfMilk May 31 '18 at 22:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.