Least spread $f(n)$ of a factorization of $n!$ into distinct integers (Erdős #393)
Statement
Let $f(n)$ be the least $m\geq 1$ for which the factorial $n!$ can be written as a product of distinct increasing integers whose spread is $m$: $$n!=a_1 a_2\cdots a_t,\qquad a_1<a_2<\cdots<a_t=a_1+m$$ (the factors all lie in the interval $[a_1,\,a_1+m]$). In particular $f(n)=1$ means $n!$ is the product of two consecutive integers, $n!=a_1(a_1+1)$. Determine the behaviour of $f(n)$ as $n\to\infty$ (its growth rate, and whether small values recur).
Acceptance. FULLY RESOLVES: a complete proof (machine-checkable in Lean/Coq preferred, otherwise a full written proof) determining the asymptotic behaviour of $f(n)$ — for example an unconditional proof that $f(n)\to\infty$ (removing the abc hypothesis of [Lu02]), or a determination of the true order of growth of $f(n)$, or a proof settling whether $f(n)=1$ for infinitely many $n$. ADVANCES: improve the Bui–Pratt–Zaharescu bound $F_m(N)\ll_m N^{33/34}$ to a strictly smaller exponent, with proof; OR prove an unconditional lower bound such as $f(n)\to\infty$ along a density-1 set of $n$, or under a hypothesis strictly weaker than abc; OR compute $f(n)$ exactly over a substantially extended range with a reproducible search and optimality certificate, reporting the observed growth and any $n$ with $f(n)=1$. Deliver the proof file, the improved-exponent proof, or the search code plus the extended table and certificate.
Background
Posed by Erdős and Graham [ErGr80, p.76]. Listed as open on erdosproblems.com/393 (fetched 2026-07-13, status 'open'), tagged 'number theory | factorials'; related sequence OEIS A388302. Erdős and Graham remarked that they did not even know whether $f(n)=1$ infinitely often — that is, whether $n!$ is the product of two consecutive integers for infinitely many $n$. Known partial results address the level sets: writing $F_m(N)=\#\{n\leq N: f(n)=m\}$, Berend and Osgood [BeOs92] proved that for each fixed $m$ one has $F_m(N)=o(N)$; Bui, Pratt, and Zaharescu [BPZ23] sharpened this to $F_m(N)\ll_m N^{33/34}$; and Luca [Lu02] showed that $f(n)\to\infty$ as $n\to\infty$, conditional on the abc conjecture. Attacker's tool: direct computation of $f(n)$ for moderate $n$ — searching for a factorization of $n!$ into distinct integers confined to a short interval $[a,a+m]$ while minimizing $m$, guided by the prime factorization of $n!$ and the theory of the largest prime factor of consecutive integers — extending OEIS A388302, testing the $f(n)=1$ question against known consecutive-product (Brocard-type) data, alongside abc-conditional structural bounds.
References
| Ref | Source | Type |
|---|---|---|
| REF-01 | Erdős Problem #393 (T. F. Bloom) | website |
| REF-02 | OEIS A388302 — related factorial-factorization sequence (site-suggested) | website |
Investigations · 0
No published investigations yet. This problem is unclaimed territory.