GRAPH LAPLACIAN EIGENVALUE SPECTRA — ALL TESTED GEOMETRIES
================================================================================
Date: 2026-04-04
Author: Jonathan Shelton (theory), Claude (computation)
Status: COMPUTED — reconciliation needed with cipher eigenvalue counts
================================================================================


SPECTRA (from graph Laplacian L = D - A of convex hull)
================================================================================

  Tetrahedron (4V):   λ = {4.000}                    1 distinct. {2²}
  Octahedron (6V):    λ = {4.000, 6.000}             2 distinct. Ratio: 3/2
  Cube (8V):          λ = {2.764, 6.000, 7.236}      3 distinct. Ratio: φ²
  Icosahedron (12V):  λ = {2.764, 6.000, 7.236}      3 distinct. Ratio: φ²
  BCC trunc.oct (24V):λ = {2.764, 6.000, 7.236}      3 distinct. Ratio: φ²
  Cuboctahedron (12V):λ = {2.354, 4, 6, 7.646, 8}    5 distinct. Mixed
  Dodecahedron (20V): 19 distinct. φ ratio present (1.663)
  Rhombicuboct (24V): 23 distinct. Dense, mixed
  Icosidodec (30V):   29 distinct. Dense, phi-influenced
  C60 (60V):          59 distinct. Near-continuous

  KEY FINDING: Cube, Icosahedron, and BCC truncated octahedron
  share IDENTICAL eigenvalue spectra: {2.764, 6.000, 7.236}.
  Three different shapes, same eigenvalue fingerprint.
  The φ² ratio (7.236/2.764 = 2.618) is exact.

  RECONCILIATION NEEDED: The cipher previously used eigenvalue
  counts of 9 (BCC), 5 (FCC), 7 (HCP), 1 (Diamond). These may
  have counted with multiplicity. The DISTINCT eigenvalue count
  from the convex hull graph is 3 for BCC, not 9. The 9 may come
  from a different graph representation (e.g., including second
  neighbors or using the full periodic lattice vs single cell).
  This needs careful resolution.


COMBINED WITH HPC-032 RESONANCE DATA
================================================================================

  Geometry          Distinct_λ  Uniformity  Differentiation  Family
  ─────────────────────────────────────────────────────────────────
  Icosahedron            3        0.754        2.1x         phi
  Rhombicuboct          23        0.762        2.4x         mixed
  Icosidodec            29        0.581       23.4x         phi-dense
  C60                   59        0.577       25.2x         continuous
  Sphere              (∞)        0.555       17.4x         continuous
  Dodecahedron          19        0.511       21.7x         phi-dense
  BCC trunc.oct          3        0.393       16.2x         phi
  Cuboctahedron          5        0.294       19.8x         mixed

  The icosahedron with only 3 distinct eigenvalues achieves the
  HIGHEST uniformity of any geometry. Fewer distinct modes +
  phi-family ratios = optimal uniform distribution.


OUTPUT-AGNOSTIC. DATA SHOWS WHAT IT SHOWS.
================================================================================