Another generalization of a theorem of Baker and Davenport (Q2406376): Difference between revisions

A set \(\{a_1,\dots,a_m\}\) of positive, pairwise distinct integers is called a Diophantine \(m\)-tuple, if \(a_ia_j+1\) is a perfect square for all indices \(i\neq j\). Recently, the first author et al. [Trans. Am. Math. Soc. 371, No. 9, 6665--6709 (2019; Zbl 1430.11044)] showed that no Diophantine quintuple exists. Given a Diophantine triple \(\{a,b,c\}\), then this triple can always be extended to a quadruple by adding the number \[d^+=a+b+c+2abc+2\sqrt{(ab-1)(ac-1)(bc-1)}.\] A Diophantine quadruple \(\{a,b,c,d\}\) is called regular, if for \(d>\max\{a,b,c\}\) we have that \(d=d^+\). The main conjecture concerning Diophantine \(m\)-tuples is that every Diophantine quadruple is regular. In the paper under review, the authors show that every Diophantine quadruple which contains \(\{1,b\}\) with \(b-1\) is a prime power is regular.