Pith. sign in

REVIEW 3 major objections 5 minor 24 references

Certified Minimal-Prime Branch Closures for Odd Perfect Numbers

T0 review · 3 major / 5 minor · reviewed 2026-07-11 · grok-4.5

Pith's one-line read No odd perfect number can have smallest prime factor 5, 7, 11, 13 or 17; those five minimal-prime branches are exhausted by valuation balance, lower-prime avoidance, and frozen machine-checked certificates.

desk verdict Solid, scoped computer-assisted closures of five least-prime OPN branches, with the real risk being inventory completeness rather than the tail budget. read the letter →

arxiv 2607.04365 v1 pith:KXGYIVPL submitted 2026-07-05 math.GM

classification math.GM MSC 11A2511Y0511Y16
keywords oddperfectnumbersdivisorsumscyclotomicfactorsp-adicvaluationsminimalprimebranch-and-boundcertificationcomputer-assistedproof
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

An odd perfect number N would satisfy σ(N)=2N and would possess a smallest prime factor q. This paper proves that q cannot be any of 5, 7, 11, 13 or 17. The argument uses the exact q-adic valuation balance forced by σ(N)=2N together with the fact that every prime smaller than q is forbidden from the support of N and therefore cannot divide any divisor-sum factor. Those two constraints collapse each of the five branches into a finite first-input coverage split whose terminal leaves are either forced cofactors or pure cyclotomic equations; every leaf is then refuted or tail-controlled by an explicit, hash-locked certificate release. The result is deliberately scoped: it does not rule out odd perfect numbers altogether, and the remaining branches q=3 and q≥19 are left open. A sympathetic reader cares because the classical existence question is reduced, for the first time in certificate form, to two residual cases.

What carries the argument

The forced-or-pure cofactor mechanism: after the exact q-adic valuation balance and lower-prime avoidance reduce each branch to a first-input coverage split, every terminal reduced cofactor either forces a new support prime (then closed by a tail-envelope inequality) or collapses to a pure cyclotomic equation whose only solutions are excluded by the branch hypotheses.

What would settle it

Run the five released Python verifiers on the five frozen JSONL master bundles; any failure to print the exact terminal string “q=… branch inventory exhausted”, or any mismatch of the published SHA256 hashes, would falsify the claimed closures.

Watch

Extended reading notes

Core claim

Relative to the frozen certificate release C-small-2026-07 and its verifier contract, there is no odd perfect number whose least prime divisor lies in {5,7,11,13,17}. Equivalently, the five minimal-prime branch inventories are completely exhausted by forced-or-pure terminal records.

Load-bearing premise

The terminal leaves rely on certified bounds that every remaining support prime is at least 59 and that at most 18 such primes can still appear; if a configuration needed more large primes or a smaller prime outside the envelope, the abundance product would no longer stay strictly below 2.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proves a scoped, certificate-relative theorem: relative to the frozen release C-small-2026-07 and its verifier contract, there is no odd perfect number N with min{p:p|N} in {5,7,11,13,17}. The argument combines the exact q-adic valuation balance for σ(N)=2N with lower-prime avoidance (primes ℓ<q cannot divide any σ(p^e)), which forces a finite first-input coverage split by allowed orders modulo q under Euler form. Each child reduces to a forced cofactor (new support prime) or a pure exceptional row; terminals are closed by lower-prime refutation, pure Diophantine screens, endpoint checks, or tail-envelope abundance inequalities H(K_env)(B/(B-1))^M<2. The q=5 branch is developed in full detail (three children Q5E2/Q5E1/Q5N, finite windows, parametric Zsigmondy tails); q=7,11,13,17 are closed by the same forced-or-pure frontier and master-wrapper verifiers. The result does not address q=3 or q≥19 and does not claim nonexistence of odd perfect numbers.

Significance. If the inventories and wrappers are sound, this is a genuine finite exhaustion of five least-prime branches rather than another global bound on N, ω(N), or large prime factors. The contribution is methodological as well as numerical: it isolates the minimal-prime problem, gives a uniform valuation-and-avoidance mechanism, and ships a frozen, hash-pinned certificate release (JSONL bundles, Python verifiers, expected outputs, SHA256 hashes, Git tag C-small-2026-07). That reproducibility layer is a real strength for computer-assisted number theory. The result is correctly scoped and does not overclaim; closing these branches is a concrete advance even while q=3 and q≥19 remain open.

major comments (3)
  1. Theorem 1.1 / 14.1 and Criterion 7.6: the central claim is relative to the frozen inventories being complete and correctly labeled. The handwritten coverage splits (Prop. 5.4, Prop. 8.1) and terminal lemmas (6.2–6.3, 7.3) are standard, but the master wrappers only check that the listed records are present and arithmetically consistent; they do not independently discover missing children. An incomplete first-input split, missing pure-screen solution, or undelegated child archive would make the printed “inventory exhausted” output false while still satisfying local checks. The paper should state more explicitly what independent audit of inventory completeness is possible beyond re-running the supplied scripts on the supplied bundles.
  2. Definition 7.4 and the strict-frontier records (Tables 9–11, Prop. 13.1): every forced/pure tail uses B=59 and M=18. Lemma 7.3 is correct once those bounds are granted, but the manuscript does not give a self-contained derivation that, after the recorded envelope primes, every unresolved support prime is ≥59 and that at most 18 further slots remain. Remark 7.5 treats M as local release data; for a journal proof that is acceptable only if each terminal record’s justification for (B,M) is either handwritten or machine-checked against a stated ω(N) floor (Cor. 4.3). As written, the load-bearing step is partly opaque.
  3. q=5 pure and post-window closures (Thm. 9.32, Criteria 9.28–9.31; pure endpoint π=1249): the pure C2(π)=2 family and the four post-window pairs rely on Zsigmondy primitive divisors and endpoint inequalities checked only at the worst-case s. The argument is standard, but the paper should confirm that all exceptional Zsigmondy cases are excluded for the stated n (odd, n≥23 or ≥95) and that no additional post-window pairs appear in the release inventory beyond the four listed. The same completeness obligation applies to the finite windows in Q5E1 (π<1381) and Q5N (p<211).
minor comments (5)
  1. Table 2 and Prop. 5.4: the coverage split is correctly non-disjoint; a short explicit remark that multiple witnesses are allowed would help readers who expect a partition.
  2. Table 4 pure screens: the solution sets {(3,2)}, {(2,1)}, {(3,1)}, {(4,1)} are stated as certificate-verified; a one-line citation or elementary verification of each equation would make the pure filters self-contained without opening the JSONL.
  3. Section 7 / Table 7: the reproducibility table is excellent; adding the exact Python version and a one-line note that the verifiers are pure checkers (no branch discovery) would reduce ambiguity for auditors.
  4. Notation: envelope tuples allow repeated primes (e.g. (7,29,29)); a brief sentence that multiplicity encodes independent worst-case slots would clarify H(K_env) for readers unfamiliar with the device.
  5. References include several 2025–2026 preprints used only as background checks; flagging which are essential versus optional would help.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: branch closures rest on standard valuation identities plus independent finite certificate checks, not on self-definitional or fitted reductions.

full rationale

The derivation chain is: (i) Euler form and σ(N)=2N give the q-adic balance (Lemma 3.1) and cyclotomic input rule (Lemma 3.2); (ii) min{p:p|N}=q forces lower-prime avoidance (Lemma 4.2); (iii) order restrictions from Euler parity yield a finite first-input coverage split (Prop. 5.4 / 8.1); (iv) reduced cofactors either force a new support prime (Lemma 6.2), hit a lower prime (Lemma 6.3), or fall into a pure row; (v) terminal leaves are closed by frozen JSONL factorizations, pure-screen solution sets, endpoint checks, and exact rational tail inequalities H(Kenv)(B/(B-1))^M < 2 (Lemma 7.3), audited by master wrappers under Criterion 7.6. None of these steps defines a quantity in terms of the nonexistence conclusion, fits a parameter to the target and renames it a prediction, or imports uniqueness from the same author. The theorem is explicitly scoped relative to the external frozen release C-small-2026-07; that is honest computer-assisted scoping, not circularity. Completeness of the inventories is a correctness/audit risk (whether every coverage child is listed and correctly labeled), not a circular reduction of the claimed derivation to its inputs. Score 0.

Assumptions & free parameters 2 free parameters · 5 assumptions · 2 invented entities

The central claim rests on classical arithmetic (Euler form, multiplicativity of σ, lifting-the-exponent, Zsigmondy) plus the domain hypothesis that N is odd perfect, together with a small number of release-specific numerical budgets (B=59, M=18) that appear only in the terminal tail certificates. No new physical or algebraic entities are postulated; the certificate system is an engineering artifact, not an ontological invention.

free parameters (2)
  • tail-count budget M = 18
    Each strict-frontier tail_count_certificate carries M=18 remaining support-prime slots; the value is release data chosen so that the abundance product falls below 2, not derived from a universal bound on ω(N).
  • support-prime lower bound B = 59
    Terminal inequalities use B=59 for unresolved primes outside the envelope tuple; again release data attached to the frozen records rather than a theorem.
assumptions (5)
  • standard math Euler’s form: any odd perfect N = π^α M² with π ≡ α ≡ 1 (mod 4) and (π,M)=1
    Invoked throughout for parity of exponents and for the restriction on possible input orders (Lemma 5.3).
  • standard math σ is multiplicative and σ(N)=2N implies the exact q-adic valuation balance of Lemma 3.1
    Core identity driving every coverage split.
  • domain assumption Lower-prime avoidance: if ℓ < q = min S then v_ℓ(σ(p^e))=0 for every support prime power (Lemma 4.2)
    Follows immediately from the valuation balance once ℓ ∤ N; used to refute any cofactor divisible by a prime smaller than q.
  • standard math Zsigmondy’s theorem supplies a primitive prime divisor for a^n−1 when n is large enough and not exceptional
    Used in the parametric post-window criteria for the Q5E2 successor pairs.
  • ad hoc to paper The frozen certificate release C-small-2026-07 correctly enumerates every terminal leaf of the five branch inventories
    The mathematical claim is explicitly relative to this immutable release (Definition 2.2 and Criterion 7.6).
invented entities (2)
  • C-small-2026-07 certificate release (JSONL bundles + Python verifiers + SHA256 hashes) independent evidence
    purpose: Provides the finite, machine-checkable terminal inventory that converts the handwritten coverage splits into a closed branch theorem.
    The entity is an engineering artifact introduced by the paper; independent evidence consists of the public GitHub tag and the recomputable rational inequalities, but the completeness of the inventory is not proved by hand.
  • envelope tuple K_env and tail_count_certificate(B,M)
    purpose: Encodes worst-case abundance upper bounds for unresolved support primes after forced primes have been inserted.
    Defined in Definitions 7.2 and 7.4 solely for the certificate contract; not an independent mathematical object.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Certified Minimal-Prime Branch Closures for Odd Perfect Numbers." pith.science (2026). https://pith.science/paper/KXGYIVPL

@misc{pith2026260704365,
  author       = {Pith},
  title        = {Pith review of: Certified Minimal-Prime Branch Closures for Odd Perfect Numbers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KXGYIVPL}},
  note         = {Machine review of arXiv:2607.04365}
}
abstract

For an odd perfect number $N$, write $q=\min\{p:p\mid N\}$ for its smallest prime divisor. This paper proves a certified branch-closure theorem for the five minimal-prime branches $q=5,7,11,13,17$. The proof combines the exact $q$-adic valuation balance for $\sigma(N)=2N$ with lower-prime avoidance: primes below $q$ cannot occur in the support and therefore cannot divide any divisor-sum factor. These constraints reduce each branch to a finite first-input coverage split and then to terminal forced-or-pure cofactor records. The terminal records are checked by the frozen certificate release C-small-2026-07, consisting of JSONL certificate bundles, Python verifier scripts, expected terminal outputs, and SHA256 hashes. The $q=5$ branch is presented as the detailed audit model, while the branches $q=7,11,13,17$ are closed by the same forced-or-pure mechanism. The result is scoped: it does not prove nonexistence of odd perfect numbers, and the branches $q=3$ and $q\ge 19$ remain outside the paper.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

24 extracted references · 6 linked inside Pith

  1. [1]

    Deza, Elena , title =

  2. [2]

    2022 , note =

    Cai, Tianxin , title =. 2022 , note =

  3. [3]

    Mathematische Zeitschrift , volume =

    Gr. Mathematische Zeitschrift , volume =. 1952 , pages =

  4. [4]

    , title =

    Nielsen, Pace P. , title =. Mathematics of Computation , volume =. 2007 , pages =. math/0602485 , archivePrefix =

  5. [5]

    , title =

    Nielsen, Pace P. , title =. Mathematics of Computation , volume =. 2015 , pages =

  6. [6]

    Odd perfect numbers are greater than \(10^

    Ochem, Pascal and Rao, Micha. Odd perfect numbers are greater than \(10^. Mathematics of Computation , volume =. 2012 , pages =

  7. [7]

    Mathematics of Computation , volume =

    Goto, Takeshi and Ohno, Yasuo , title =. Mathematics of Computation , volume =. 2008 , pages =

  8. [8]

    , title =

    Iannucci, Douglas E. , title =. Mathematics of Computation , volume =. 1999 , pages =

Show all 24 references
  1. [9]

    , title =

    Iannucci, Douglas E. , title =. Mathematics of Computation , volume =. 2000 , pages =

  2. [10]

    International Journal of Number Theory , volume =

    Zelinsky, Joshua , title =. International Journal of Number Theory , volume =. 2019 , pages =. 1810.11734 , archivePrefix =

  3. [11]

    Integers , volume =

    Zelinsky, Joshua , title =. Integers , volume =. 2021 , pages =. 1810.13063 , archivePrefix =

  4. [12]

    Integers , volume =

    Zelinsky, Joshua , title =. Integers , volume =. 2023 , pages =

  5. [13]

    Integers , volume =

    Bibby, Samuel and Vyncke, Pieter and Zelinsky, Joshua , title =. Integers , volume =. 2021 , pages =. 1908.09420 , archivePrefix =

  6. [14]

    and Hansen, C

    Clayton, G. and Hansen, C. S. , title =. Integers , volume =. 2023 , pages =. 2303.11974 , archivePrefix =

  7. [15]

    , title =

    Stone, A. , title =. Integers , volume =. 2024 , pages =

  8. [16]

    , title =

    Voutier, Paul M. , title =. 2024 , eprint =

  9. [17]

    Integers , volume =

    Zelinsky, Joshua , title =. Integers , volume =. 2025 , pages =

  10. [18]

    2025 , eprint =

    Zelinsky, Joshua and Zhang, Kyle , title =. 2025 , eprint =

  11. [19]

    2025 , eprint =

    Ross, Tyler and Shen, Zhongyan and Cai, Tianxin , title =. 2025 , eprint =

  12. [20]

    , title =

    Alekseyev, Max A. , title =. 2026 , eprint =

  13. [21]

    2026 , eprint =

    Cheng, Kaimin and Zhang, Ke , title =. 2026 , eprint =

  14. [22]

    A note on the \(q\)-adic valuation of \( _k(n)\) , year =

    Bordell. A note on the \(q\)-adic valuation of \( _k(n)\) , year =. 2605.17872 , archivePrefix =

  15. [23]

    Monatshefte f

    Zsigmondy, Karl , title =. Monatshefte f. 1892 , pages =

  16. [24]

    and Vandiver, Harry S

    Birkhoff, Garrett D. and Vandiver, Harry S. , title =. Annals of Mathematics , volume =. 1904 , pages =

Pith tools

Reviewed July 11, 2026 · model on record in the stance chip above.