Question But Not A Doubt 1

Find a suitable $x$ such that $a_{1}^2+a_{2}^2=x^2$ then the problem reduces to $x^2+a_{3}^2+...+a_{m}^2=a_{m+1}^2+...+a_{n}^2$ then find a suitable $y$ such that $a_{m+1}^2+a_{m+2}^2=y^2$. So, doing this process repeatedly on both the L.H.S and the R.H.S until you reach a point where your equation will be of the form: $a^2+a_{m}^2=a_{n}^2$. Now, you know $a_{1},...,a_{m-1},a_{m+1},...,a_{n}$ because we have been substituting suitable values. So, we will have an equation like this: $a^2+a_{m}^2=a_{n}^2$. Now, we can easily solve this using pythagorean triples formula, because we know $a$. Hence we have found a suitable solution.

Uhh, this is a pretty vague question I believe.

What is $m$? What is $n$? Must $a_i$ be an integer? Can it be zero? If it is then I can always make a trillion solutions just by saying $a_i$ = 0 for all $i$ ranging from 1 to $n$.

Even if the numbers must be integers and positive, by letting $n$ divisible by 2, I can make infinitely many solutions. (How, I'll leave you to it.)

So, the takeaway? Restrictions!

P/S: Good ideas though, maybe you can capitalize on it by forcing $a_i$ to be all different; that way you can make a pretty fun problem.

Yeah. yes ai is an integer. No, the question is not to find the number of solutions, but one such solution. m,n can be any positive integers you want. I just want a general solution.

Sorry. All numbers are distinct

But there is a very easy method!!!

There is no general formula, there are infinite solutions. So, maybe I should tell my method now!!!

Just us pythagorean triplet formula

You posted a problem related to this right?

That is just awesome!!!

You do have brilliant thinking skills on brilliant!!!

maybye I will eplain you with an eample. Ok?!

Example: Find one 5-tuple $(a,b,c,d,e)$ such that, $a^2+b^2+c^2=d^2+e^2$

Solution: Find a suitable $x$ such that $a^2+b^2=x^2$. One such solution is $a=3, b=4, x=5$. So, the problem reduces to $5^2+c^2=d^2+e^2$. Let $e=13$ then we have $c^2-d^2=144$. So, $(c+d)(c-d)=144$ which implies that $c=37, d=3$. So, one solution set is $(a,b,c,d,e)=(3,4,37,35,13)$

I am sorry I made a mistake when writing the general solution. This solution is only properly correct

I was in a hurry to post that solution. So a small error occured.

@Mohammed Imran I will decide to move to this comment for the sake of clarity.

Again, I have still not understood how you would choose the suitable $x$. State your ideas here, if you have any.

We want one solution to the equation: $a_{1}^2+a_{2}^2+...+a_{m}^2=a_{m+1}^2+...+a_{n}^2$.

Now, choose a suitable $x$ such that $(a_{1},a_{2},x)$ forms a Pythagorean triple. Now, the problem reduces to finding one solution to $x^2+a_{3}^2+..+a_{m}^2=a_{m+1}^2+...+a_{n}^2$. Now, choose a suitable $x_{1}$ such that $(x,a_{3},x_{1})$ forms a Pythagorean triple. Keep doing this process on both sides until we land on an equation of the form $a^2+a_{m}^2=b^2+a_{n}^2$. Now, since we have been substituting values for $a_{1},a_{2},..,a_{m-1}$, we know the values of $a,b$. So, we can rewrite the equation as $a^2-b^2=a_{n}^2-a_{m}^2$ and solve the equation by "Simon's Factorisation". So, we have found a solution set.

Steven Jim, do you have any comments??

Steven Jim, if you are interested for general divisibility criteria of odd numbers, please have a look at my note: "A Question But Not A Doubt 2"