Skip to content

Link OEIS sequences for problems 308, 829, 929 - #402

Open
hjyuh wants to merge 1 commit into
teorth:mainfrom
hjyuh:patch-1
Open

hjyuh wants to merge 1 commit into
teorth:mainfrom
hjyuh:patch-1

Conversation

@hjyuh

@hjyuh hjyuh commented Sep 6, 2026

Copy link
Copy Markdown

Title: Link OEIS sequences for problems 308, 829, 929

Same pattern as #388: adds OEIS identifiers to three entries in data/problems.yaml that currently carry only "possible", keeping the "possible" flag. Each link is to a sequence that is exactly the object in the problem statement (or its inverse), not a related one, and each was checked by independent computation.

Problem 308 -- smallest integer not a sum of distinct unit fractions with denominators ≤ N
Added: A101877

$f(N)$ in the problem is the inverse of A101877 ("a(n) = least k such that some set S ⊆ {1,…,k} has Σ 1/s = n"): $f(N)$ = least $n$ with $A101877(n) > N$. Terms $1, 6, 24, 65, …$ are the values of $N$ at which $f$ increments.

Verification: computed $f(N)$ for $N \le 24$ by exact subset-sum over $\mathrm{lcm}(1..N)$: $f = 2$ for $N<6$, $3$ for $6 \le N < 24$, $4$ at $N=24$ — i.e. increments at $N = 1, 6, 24$, matching $A101877(1..3)$. The next increment at $65$ is A101877(4) (Croot's paper cited on the problem page uses the same value).

Problem 829 -- $1_A * 1_A(n)$ for $A$ = cubes
Added: A025455, A025468

A025455 = number of partitions of $n$ into 2 positive cubes (unordered); A025468 = same with distinct cubes. The convolution in the problem is the ordered count, $= 2\cdot A025455(n) - [,n = 2m^3,]$; the growth question is unaffected by the normalisation.

Verification: recomputed both sequences for $n \le 6000$; they match the OEIS DATA lines term-for-term (offset 0), including the first values $\ge 2$ at $n = 1729, 4104$.

Problem 929 -- $S(k)$, least $x$ such that a positive-density set of $n$ has $n+1,…,n+k$ all divisible by primes $\le x$
Added: A058989, A048670, A049300

Whether $n+1,…,n+k$ are all covered by primes $\le p_m$ depends only on $n \bmod \prod_{p\le p_m} p$, so "positive density" is equivalent to "at least one $n$", and $S(k)$ is the inverse of A058989 ("largest number of consecutive integers each divisible by a prime ≤ the n-th prime"): $S(k)$ = least $p_m$ with $A058989(m) \ge k$. A048670 (Jacobsthal function at primorials) $= A058989 + 1$ and A049300 gives the starting points; both are cross-referenced from A058989.

Verification: brute-force longest covered run over one full period for the first 9 primes gives $1, 3, 5, 9, 13, 21, 25, 33, 39$, matching A058989(1..9).

AI disclosure
Automated search with claude for efficiency,

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant