Wiki
Wiki

Research notes on every problem, and a library of the papers behind them. Built from the open erdos repository.

Updated


Statement

Theorem 1.3 (p. 2, quoted). "Let c0=0.70881678090424862707121c_0=0.70881678090424862707121. Then we have π(m+n)≤π(m)+π(n)\pi(m+n)\le\pi(m)+\pi(n) for all integers m≥n≥2m\ge n\ge2 with n≥c0m/log⁡2mn\ge c_0m/\log^2m."

The paper presents it (p. 2) as a refinement of Panaitopol's range π(m)≤n≤m\pi(m)\le n\le m, display (1.6).

Proof pointer

Section 5, pp. 5--6. The theorem is Proposition 5.1 with k=2k=2, a1=1a_1=1, a2=2.85a_2=2.85, α2=38 099 531\alpha_2=38\,099\,531, ε=0.70863503301170907614119\varepsilon=0.70863503301170907614119, β2=14 000 264 036 190 262\beta_2=14\,000\,264\,036\,190\,262, γ2=23\gamma_2=23 and c=c0c=c_0, the values of α2\alpha_2 and β2\beta_2 taken from Axler's earlier paper (its reference [2], Corollary 3 and Theorem 2). This gives the inequality for m≥14 000 264 036 190 263m\ge14\,000\,264\,036\,190\,263 and n≥c0m/log⁡2mn\ge c_0m/\log^2m. For smaller mm one has c0/log⁡2m≥1/1950c_0/\log^2m\ge1/1950, and the claim follows from Theorem 1.1. The printed proof (p. 6) says the values are substituted "into Proposition 3.1" [sic]; the substitution described is into Proposition 5.1.

Read depth

Claims checked: the statement, Proposition 5.1 and the parameter values were read clause by clause on the pages of the copy named on the source card. The proof was read but not checked, and the constants were not recomputed. Nothing here is independently reviewed.

Dependencies

  • Proposition 5.1 (p. 5).
  • Theorem 1.1 (p. 2), for m≤14 000 264 036 190 262m\le14\,000\,264\,036\,190\,262.
  • Explicit bounds for π(x)\pi(x) from C. Axler, Integers 18 (2018), Paper No. A52 (see its card).

Source. Christian Axler, "Some Results on a Conjecture of Hardy and Littlewood," arXiv:1909.12625v2 (2019), the edition read for the source card.

Bears on

  • Problem 855: with integers X≥Y≥2X\ge Y\ge2 the two arguments, the theorem proves the problem's inequality whenever Y≥c0X/log⁡2XY\ge c_0X/\log^2X, so any pair of integers violating it has Y<c0X/log⁡2XY<c_0X/\log^2X. It says nothing about that remaining region and does not decide the problem.