Wiki
Wiki

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

Updated


Claim. Let V(x)=A(x)V(x)=A(x) count the integers n≤xn\le x that are sums of two squarefull (powerful) numbers. Corollary 2 of V. Blomer and A. Granville, Estimates for representation numbers of quadratic forms, Duke Math. J. 135 (2006), no. 2, 261–302, states that for some A∈RA\in\mathbb R

x(log⁡log⁡x)A(log⁡x)1−2−1/3≪V(x)≪x(log⁡log⁡x)22/3−1(log⁡x)1−2−1/3,\frac{x(\log\log x)^A}{(\log x)^{1-2^{-1/3}}}\ll V(x)\ll \frac{x(\log\log x)^{2^{2/3}-1}}{(\log x)^{1-2^{-1/3}}},

as recorded on the source card blomer_2006_estimates_representation_numbers_quadratic_forms. Since 1−2−1/3≈0.2063<1/21-2^{-1/3}\approx0.2063<1/2, the lower bound alone gives A(x)log⁡x/x→∞A(x)\sqrt{\log x}/x\to\infty, so A(x)A(x) is not asymptotic to c x/log⁡xc\,x/\sqrt{\log x} for any c>0c>0. This disproves Problem 1081 independently of [[problems/diophantine_problems/E1081/claims/1981_01_01_odoni|Odoni's earlier disproof]], and it fixes the order of A(x)A(x) up to powers of log⁡log⁡x\log\log x.

Depends on. No other wiki page; the claim rests on the cited paper.

Acceptance. Refereed: the paper appeared in the Duke Mathematical Journal, volume 135 (2006), published online on 1 November 2006 according to its Crossref record. Not reviewed: the site's curator credits Odoni with the disproof and cites this paper for the sharpest estimate of A(x)A(x), which is not a credit for the disproof.