Put and use conventional convergent indexing starting at zero; the same argument covers all the printed positive indices. The continued fraction complete quotients satisfy , . After the first step they exceed one, while their algebraic conjugates lie in . Indeed , and if with , then .
Write , where are integers and . This form follows by rationalization at each step: and , starting from . The two conjugate inequalities give
so . The standard convergent recursions also give
Substitute the displayed form of and compare rational and irrational coefficients. This yields and . Therefore the small norm residues from square-root continued fractions obey
The initial convergent satisfies the same relation with the usual . This avoids the insufficient estimate obtained by merely replacing the convergent error with .
For integer factorization, choose a factor base of small primes for which is a quadratic residue, together with to record negative residues; a prime dividing already supplies a factor. Test the small nonzero integers for being B-numbers. Complete factorization over the base records each prime exponent modulo two. Once enough relations are collected, linear algebra over can choose a nonempty subcollection whose exponent sum is even, including the sign. Then
If , one of or can yield a nontrivial factor. Relations with a noninvertible may already give a factor through its gcd. The small-residue bound makes smoothness more likely; it does not guarantee that any one dependency gives a nontrivial factor.