Skip to main content
27 events
when toggle format what by license comment
Jan 27 at 17:53 history edited Max Alekseyev CC BY-SA 4.0
Code updated with safe iterator aborting per https://github.com/sagemath/sage/issues/41015
Oct 8, 2025 at 11:44 history edited Max Alekseyev CC BY-SA 4.0
a few minor speed-up improvements to the code
Oct 7, 2025 at 15:48 history edited Max Alekseyev CC BY-SA 4.0
Sage code bugfixed and improved
Oct 6, 2025 at 21:26 history edited Max Alekseyev CC BY-SA 4.0
updated with thread-safe code
Oct 6, 2025 at 18:31 history edited Max Alekseyev CC BY-SA 4.0
code updated a bit
Oct 6, 2025 at 15:59 history edited Max Alekseyev CC BY-SA 4.0
deleted 11 characters in body
Oct 6, 2025 at 15:48 history edited Max Alekseyev CC BY-SA 4.0
edited body
Oct 6, 2025 at 15:36 history edited Max Alekseyev CC BY-SA 4.0
added 9 characters in body
Oct 6, 2025 at 15:23 history edited Max Alekseyev CC BY-SA 4.0
deleted 1 character in body
Oct 6, 2025 at 15:01 history edited Max Alekseyev CC BY-SA 4.0
typo corrected; added 15 characters in body
Oct 6, 2025 at 14:52 history edited Max Alekseyev CC BY-SA 4.0
added 2646 characters in body
Oct 5, 2025 at 17:10 comment added Max Alekseyev @TerryTao: I've added a draft sequence oeis.org/draft/A389482 (and its sister A389483) Please update as suitable.
Oct 4, 2025 at 22:33 comment added Terry Tao @MaxAlekseyev I've created a separate community answer to collate these new results.
Oct 4, 2025 at 22:17 comment added Max Alekseyev @TerryTao: Indeed, I also thought about such a sequence (and a sister one with no prime power restriction). I will add it after a bit more computing.
Oct 4, 2025 at 22:13 comment added Terry Tao It seems like there is a non-trivial but finite OEIS sequence in the making, namely the set of $n$ which are prime powers (A000961) for which $L_n$ is highly abundant: we know the sequence has all the prime powers up to $67$, together with $81$, $83$, $89$, with the next potential element being $121$. On the other hand, the existing constructions should be able to block out large regions from this sequence; for instance, my original example blocks out $[199999, 479^2-1]=[199999,229440]$. Determining this sequence completely via crowdsourcing seems feasible.
Oct 4, 2025 at 21:43 history edited Max Alekseyev CC BY-SA 4.0
added 454 characters in body; added 5 characters in body; deleted 112 characters in body; added 85 characters in body; added 4 characters in body
Oct 4, 2025 at 1:00 comment added GH from MO @TerryTao $L_{71}$ and $L_{73}$ are also not higly abundant...
Oct 4, 2025 at 0:59 comment added GH from MO @MaxAlekseyev Yes. I updated my post accordingly.
Oct 4, 2025 at 0:57 comment added Max Alekseyev @GHfromMO: I've got same result about $L_{79}$ with the ratio M/L = 2^5 * 3 * 5 * 11 * 67^-1 * 79^-1.
Oct 4, 2025 at 0:54 comment added Terry Tao It's remarkable that the OEIS data was almost maximally misleading as to the truth of the OP's question.
Oct 4, 2025 at 0:52 comment added GH from MO @TerryTao $L_{79}$ is also not higly abundant. I will update my post ("second response") shortly.
Oct 3, 2025 at 21:52 comment added Terry Tao Nice! So only eight candidates remaining now... (and I really doubt 81 is a strong contender...)
Oct 3, 2025 at 21:37 comment added GH from MO @TerryTao If I did not make a mistake, $L_{97}$ is not highly abundant. See the new section in my response.
Oct 3, 2025 at 21:31 comment added Terry Tao In fact, since $L_n$ only varies at prime powers, there are only 13 remaining candidates for the minimal counterexample $n$: $67,71,73,79,81,83,89,97,101,103,107,109,113$. Anyone want to place some bets?
Oct 3, 2025 at 21:07 comment added Terry Tao Nice! According to the OEIS lists provided in the comments, $L_n$ is highly abundant for $n \leq 66$, so it is now not inconceivable that with a bit more cleverness and computation (maybe involving integer programming?) one could actually find the minimal counterexample.
Oct 3, 2025 at 20:00 history edited Max Alekseyev CC BY-SA 4.0
deleted 41 characters in body
Oct 3, 2025 at 19:44 history answered Max Alekseyev CC BY-SA 4.0