3
$\begingroup$

The following question was asked at MSE without any solution:

Show that the equation $x^3+y^3+z^3-2xyz=1$ have infinitely many integer solutions $(x,y,z)$.

A more general question was also proposed at MSE:

For which positive integers $N$ does the equation $x^3+y^3+z^3-2xyz=N$ have infinitely many integers solutions $(x,y,x)$?

Some references were given, like the solutions of $x^3+y^3+z^3=nxyz$.

Usually cubic diophantine equations are very hard to deal with.

Does anyone know how any of these two problems could be solved?
Is the second problem (on the equation $x^3+y^3+z^3-2xyz=N$) open?

Any help would be appreciated.

$\endgroup$
4
  • 1
    $\begingroup$ I am guessing the original intention was to state a cubic norm form equation, where the norm form is the principal norm form of a cubic field (so that it represents one). Then the result would be obvious from the fact that the unit group of a ring of integers in a cubic field is always infinite. The second question seems very hard. For instance, it is not known exactly which integers are sums of three cubes. $\endgroup$ Mar 24, 2018 at 14:59
  • $\begingroup$ @StanleyYaoXiao: The OP and you claim that there are infinitely many integer solutions of $x^3+y^3+z^3−2xyz=1$. It does not seem to be true. Are there integer solutions except $(1,1,1), (1,0,0),(13,13,−23)$ (and permutations of those)? The question itself is a bad copy of the question from MSE linked there. I voted to close. $\endgroup$
    – user6976
    Mar 25, 2018 at 13:53
  • $\begingroup$ @Mark Sapir I don’t claim that this equation has infinitely many solutions, but say $x^3 + 2y^3 + 4z^3 - 6xyz = 1$ definitely has infinitely many solutions, since the left hand side is the principal norm form of the cubic field $\mathbb{Q}(\sqrt[3]{2})$, and there is a non-trivial unit group which acts on the solutions, generating infinitely many. This was the intent of my first comment $\endgroup$ Mar 25, 2018 at 16:06
  • $\begingroup$ @StanleyYaoXiao: OK, I understand your comment now. It is easy to see that if 1 is replaced by, say, 19, then there are infinitely many integer solutions of the original equation. $\endgroup$
    – user6976
    Mar 25, 2018 at 16:11

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.