The diophantine equation (y+q₁)(y+q₂) (y+q_m) = f(x)
The diophantine equation \((y+q 1)(y+q 2) \cdots (y+q m) = f(x)\)
Consider the diophantine equation \[ (y + q_1)(y + q_2) \cdots (y + q_m) = a_0x^n + a_1x^{n-1}+ \cdots + a_n,\tag{1} \] where \(n, m\geq 2\), \(a_0, a_1, \ldots, a_n\), \(q_1,q_2,\ldots, q_m\) are fixed integers, and \(x\) and \(y\) are unknown integers. This equation has been studied in many papers. The aim of the author is to prove the following theorem. For any positive integers \(k, p_1, p_2 ,\ldots, p_m\) let \[ \lambda_k (p_1 ,\ldots, p_m) = \sum_{\substack{ i_1,\ldots,i_k\\ 1\leq i_1<\cdots<i_k\leq m}} p_{i_1} p_{i_2}\cdots p_{i_k} \] and \(\lambda_0(p_1,\ldots, p_m) = 1\). The function \(\lambda_k (p_1,\ldots, p_m)\) is called elementary symmetric polynomial. For the polynomial \(f(x) = x^m+a_1 x^{m-1}+ \cdots + a_m\) with positive integral coefficients, denote by \(\zeta(f (x)) \) the least positive integer such that \[ \zeta(f (x)) > \max\{\lambda_i (p_1 ,\ldots, p_m)- a_i, a_2, a_3, \ldots, a_m\} \] for all \(i\geq 2\), and all positive integers \(p_1, p_2, \ldots, p_m\) with \(a_1 = p_1 + p_2 +\ldots + p_m\). \textbf{Theorem 1.} Let \(m\geq 2, q_1, q_2 ,\ldots, q_m\) be positive integers with \(q_1, q_2 \ldots, q_m\) and let \(f(x) = x^m + a_1 x^{m-1} +\cdots + a_m\) be a monic polynomial of degree \(m\) with positive integer coefficients such that one root of \(f (x)\) is not an integer. Consider the binary diophantine equation \[ (y + q_1)(y + q_2)\cdots (y + q_m) = f (x).\tag{2} \] Then any positive integral solution \((x, y)\) of equation (2) satisfies \[ x < q_m + \zeta(f(x+q_m)) \] and \[ y < \max \{\zeta (f (x + q_m)), f (\zeta (f (x)))^{1/m}- q_1\}. \] To prove Theorem 1, the author needs to prove the following theorem: \textbf{Theorem 2.} Let \(f (x) = x^m + a_1 x^{m-1} + \cdots + a_m\) be a polynomial with positive integral coefficients. Then there exists a positive integer \(n\) such that \(n\geq\zeta(f(x))\), and \(f(n)\) can written as a product of at least \(m\) divisors which are greater than \(n\) if and only if all roots of \(f (x)\) are integers.
- A note on the integer solutions ofhyperelliptic equations
- A quantitative version of Runge's theorem on diophantine equations
- Corrections to "A quantitative version of Runge's theorem on diophantine equations" (Acta Arith. 62 (1992), 157-172)
- ESTIMATES FOR THE SOLUTIONS OF CERTAIN DIOPHANTINE EQUATIONS BY RUNGE'S METHOD
- scientific article; zbMATH DE number 3278404 (Why is no real title available?)
- On the Diophantine equation F(x)=G(y)
This page was built for publication: The diophantine equation \((y+q_1)(y+q_2) \cdots (y+q_m) = f(x)\)
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q519911)
