Research notes on every problem, and a library of the papers behind them. Built from the open erdos repository.
Updated
On the Equation Phi(n) = Phi(n+1)
theorem_1_1: Proves that the reciprocal sum of the integers n with phi(n) = phi(n+1) is less than 7.8358.
Paul Kinlaw, Mitsuo Kobayashi, and Carl Pomerance, On the Equation , Acta Arithmetica 196 (2020), no. 1, 69--92, DOI 10.4064/aa190627-20-1; published online 15 June 2020.
Edition guard. The copy read for this card is the Online First PDF, not the final-paginated issue copy. It contains local article pp. 1--24 (physical pp. 1--24) followed by an abstract-only physical p. 25. The first article page visibly bears the local label [1]. The final issue span 69--92 is bibliographic metadata and is not used to locate claims in that copy. The original retrieval time of that copy is unknown. It prints "© Instytut Matematyczny PAN," followed by three asterisks in place of the year on its first page, and the publisher's record offers the PDF to subscribers only and carries no CC-BY label (https://www.impan.pl/get/doi/10.4064/aa190627-20-1), every other right reserved.
Put
Theorem 1.1, on local p. 1, proves
The introduction, local p. 1, says that it is still not known whether is infinite, and recalls that Bayless and Kinlaw (its reference [1]) had bounded the reciprocal sum by 441702 and conjectured that it is less than 2; Theorem 1.1 improves that upper bound. The proof uses the exact computation of up to , an averaging argument that limits the odd member of each pair, and, by local p. 2, further techniques for the range ; the count of 10,755 solutions in the computed range is given in Section 4.1, on local p. 11. A finite computation and a convergent reciprocal-sum bound do not decide infinitude.
Section 4, local pp. 11--23, proves Theorem 1.1 by splitting the reciprocal sum into the small range through , a middle range through , and the remaining large range. The last page combines the three bounds as .
Problem 1003 asks whether is infinite. The theorem bounds the reciprocal sum of and is consistent with either answer; the paper proves neither infinitude nor finiteness, and its introduction states that the question is open.
Bears on. #1003 (the problem page cites Theorem 1.1, a bound on the reciprocal sum of the solutions of ; it does not decide whether there are infinitely many solutions).
Results to transcribe.
- Theorem 1.1 (local p. 1): the reciprocal sum over the solutions of is less than 7.8358.
Living verification. Needs review. The final bibliographic identity, selected-edition map, theorem statement, computation cutoff, and proof endpoints were checked against the adopted metadata and selected Online First PDF. No complete proof is supplied, reconstructed, or independently certified here.
No file of this source is held: no license on record permits its redistribution, and the card cites the edition it names above.