Wiki
Wiki

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

Updated


Statement

There exists a two-distance set T⊂R64T\subset\mathbb R^{64} with ∣T∣=352|T|=352 such that every partition of TT into subsets of diameter strictly less than diam⁡(T)\operatorname{diam}(T) has at least 7171 parts.

Source: Jenrich--Brouwer, published PDF, Theorem 1, p. 3. The preceding paragraph establishes the stronger local property that every smaller-diameter subset of this TT has at most five points. Thus the counting bound is ⌈352/5⌉=71\lceil352/5\rceil=71.

Since 71>64+171>64+1, rescaling TT to diameter one disproves the diameter-one question in dimension 64. This is an upper bound on the first failing dimension, not a proof that dimension 64 is the first one.

Proof pointer and dependencies

The proof is in sections 2--4 and the paragraph preceding Theorem 1, pp. 1--3. Its graph representation uses Bondarenko's G2(4)G_2(4) construction; the graph structure and equitable partition use the graph-theoretic sources cited there. On p. 3 the authors form a nonzero vector orthogonal to the 352 selected points, placing them in a copy of R64\mathbb R^{64}. The smaller-diameter subset bound then gives the stated partition obstruction.

This is a proof pointer and brief outline. A complete rewritten proof and independent review of the representation, clique bound, partition data, and external graph inputs remain outstanding. The computational qualifications of Jenrich's separate solo manuscript must not be substituted for the published paper's argument or used as its theorem locator.

Bears on

  • E0505: a published counterexample in dimension 64, independent of the unreviewed public dimension-63 claims.