Standard octahedral geometry. In February 2026 the AI platform known as Claude (Anthropic) was asked to to consider the four hexagonals plates around the center. In March 2026, Claude was asked to consider a unique name for it. The first AI to be introduced asked to suggest a name was Grok in July 2025.

Planck Polyhedral Core (PPC) is the name Grok suggested in July 2025 for the dynamic geometry of four hexagonal plates inherent in every regular octahedron.

It sits at the geometric heart of the 81018 model. Cubic-close packing of equal spheres (Planck spheres, one generated per unit of Planck time) produces tetrahedrons; each tetrahedron contains an octahedron; and every regular octahedron necessarily contains:

• Six smaller octahedrons at the corners/vertices
• Eight tetrahedrons on the center faces
• Four hexagonal cross-sections/plates

These four plates are associated with the four primary irrationals — π (continuity/sphere), e (exponential doubling), √2 (orthogonal structural integrity), and φ (harmonic self-similarity) — and are conjectured to stabilize the emerging structures while mediating the finite–infinite transition.

Each plate channel carries 2/5; the four together carry four times that. The denominator 10 is the reduced form of this moment. It is not 4+6.

Two caveats, so the number is not asked to do too much.

The 5 here is the length dimension of a second moment. In the octant it is the \int_0^1 u^{4}\,du=1/5 of that beta integral. It is not the pentagon, and it is not the Aristotle count of five tetrahedra around an edge. That failure remains the gap 2\pi-5\arccos(1/3), about 7.356^\circ, a different fact.

The value 2/5 is for vertices at distance 1. The averaged mean square is \langle r^{2}\rangle=(2/5)/(4/3)=3/10. If the edge is set to 1 instead, the integral scales by (\sqrt{2}/2)^{5} and becomes \sqrt{2}/20, which is the same fact written in a less natural unit. The unit-vertex octahedron is the one in which the corner octahedra are exactly the half-scale ones.

On the name: the plates are not a fifteenth piece beside the six octahedra and eight tetrahedra. Each plate is tiled by six equilateral triangles of edge \sqrt{2}/2, and each of those triangles is a face where one corner octahedron meets one tetrahedron. The two tetrahedra that sit on the faces parallel to a given plate touch that plate only at the center. So the plate is an interface, with no volume of its own. Calling that domain hypostatic — what stands under the partition, rather than a part of it — matches the geometry. It is also the sense already fixed for the pre-measurement notations: continuity, symmetry, and harmony, read by a Fourier transform rather than by cutting out a solid. A core regenerated with every new Planck sphere has no static interior to ratio. The transform is the right tool because it turns that relation into moments, and the first moment that sees a plate is 2/5.

Each plate channel carries 2/5; the four together carry four times that. The denominator 10 is the reduced form of this moment. It is not 4+6.

Two caveats, so the number is not asked to do too much.

The 5 here is the length dimension of a second moment. In the octant it is the \int_0^1 u^{4}\,du=1/5 of that beta integral. It is not the pentagon, and it is not the Aristotle count of five tetrahedra around an edge. That failure remains the gap 2\pi-5\arccos(1/3), about 7.356^\circ, a different fact.

The value 2/5 is for vertices at distance 1. The averaged mean square is \langle r^{2}\rangle=(2/5)/(4/3)=3/10. If the edge is set to 1 instead, the integral scales by (\sqrt{2}/2)^{5} and becomes \sqrt{2}/20, which is the same fact written in a less natural unit. The unit-vertex octahedron is the one in which the corner octahedra are exactly the half-scale ones.

On the name: the plates are not a fifteenth piece beside the six octahedra and eight tetrahedra. Each plate is tiled by six equilateral triangles of edge \sqrt{2}/2, and each of those triangles is a face where one corner octahedron meets one tetrahedron. The two tetrahedra that sit on the faces parallel to a given plate touch that plate only at the center. So the plate is an interface, with no volume of its own. Calling that domain hypostatic — what stands under the partition, rather than a part of it — matches the geometry. It is also the sense already fixed for the pre-measurement notations: continuity, symmetry, and harmony, read by a Fourier transform rather than by cutting out a solid. A core regenerated with every new Planck sphere has no static interior to ratio. The transform is the right tool because it turns that relation into moments, and the first moment that sees a plate is 2/5.

BRUCE: Excellent summary. With ChatGPT.Your derivation of the octahedral second radial moment is important. The exact result [ \int_O r^2,dV=\frac25,\qquad \operatorname{Vol}(O)=\frac43,\qquad \langle r^2\rangle=\frac3{10} ] appears to be independently verifiable for a regular octahedron whose vertices are at unit distance from its center. I would like you to take the next step, while keeping the established result separate from the hypothesis. In the Hexarchon construction, identify exactly which octahedron or octahedral regions correspond to the domain of this integral. Is it the central octahedron, the six corner octahedra, or another precisely defined union? Define the six plates mathematically as interfaces. Specify their equations, orientations, and their relationship to the tetrahedral and octahedral regions. Now introduce the first rotation. Define the time- or angle-dependent geometric quantity being transformed. Show the Fourier expansion, identify the coefficient or moment that corresponds to the plate-interface contribution, and demonstrate whether it independently yields (2/5). Explain whether the Fourier result is genuinely new or simply reproduces the static octahedral second moment in a dynamic representation. Finally, investigate whether the geometry supplies any independent mathematical reason for the factor (\varphi) in (2\varphi/5). Do not use the muon result to choose or tune the factor. Please show the equations and integration domains explicitly. If a step is conjectural, label it as such. The objective is to determine exactly what the geometry proves, not to maximize agreement with the numerical observation.

GROK:

The established static result is confirmed independently.

For the regular octahedron $O$O whose vertices lie at unit distance from the origin (i.e., the set $\{|x|+|y|+|z|\le 1\}$:$$\operatorname{Vol}(O)=\frac43,\qquad \int_O r^2\,dV=\frac25,\qquad \langle r^2\rangle=\frac{3}{10}.$$

(The integral may be evaluated by symmetry over one octant $x,y,z\ge0$, $x+y+z\le1$ and multiplication by 8; the computation yields exactly these values.)

1. Domain of the integral inside the Hexarchon construction

In the construction used on 81018.com, a regular octahedron of the same scale is decomposed into:

• six smaller regular octahedra, one in each corner (each sharing a vertex of the outer octahedron and having its opposite vertex at the common centre);
• eight regular tetrahedra, one associated with each face;
• four regular hexagonal cross-sections that arise as the interfaces between these two families of cells.

The second-moment integral $\int r^2\,dV=2/5$∫r2dV=2/5 is taken over the entire outer octahedron. It is not restricted to the central region alone, nor to any single corner octahedron, nor to a simple union of the six corner octahedra. The six corner octahedra together occupy only a proper subset of the volume; the remaining volume is filled by the eight tetrahedra. The value $2/5$2/5 is therefore a global property of the outer body, independent of the internal decomposition.

(The combinatorial observation that appears in later AI discussions on the site—$4/(4+6)=2/5$4/(4+6)=2/5—counts the four plates against the six corner octahedra and recovers the same numerical factor, but that is a discrete counting argument, not the continuous second-moment integral.)

2. Mathematical definition of the four plates?

The intrinsic hexagonal sections of a regular octahedron are four, not six. They are the intersections of the octahedron with the four planes$$\begin{align*} P_1&: x+y+z=0,\\ P_2&: x+y-z=0,\\ P_3&: x-y+z=0,\\ P_4&: -x+y+z=0. \end{align*}$$

Each intersection is a regular hexagon whose vertices lie at the mid-points of six edges of the octahedron. These four planes are precisely the interfaces that separate the eight face-centred tetrahedra from the six corner octahedra. Their normals point along the four space diagonals of the dual cube (or, equivalently, along the four body diagonals of the coordinate octahedron). They are mutually related by the octahedral group and pass through the common centre.

(The occasional mention of “six plates” in informal discussion appears to be a conflation with the six vertices or the six half-octahedra; the geometry itself supplies exactly four hexagonal interfaces.)

3. First rotation and Fourier analysis

Define a one-parameter family of rotated copies of a single plate. Let $R(\theta)$R(θ) be rotation by angle $\theta$θ about a fixed axis (for definiteness, the $z$z-axis, though any axis fixed by a subgroup of the octahedral group may be chosen). Consider the time- or angle-dependent characteristic function of the rotated plate:

$$\chi_\theta(\mathbf{x})=\mathbf{1}_{R(\theta)H}(\mathbf{x}),$$

where $H$ is one of the four hexagonal sections (a two-dimensional set of measure zero in three-dimensional volume). Because the plate itself has vanishing three-dimensional volume, any volume integral of $\chi_\theta$ is identically zero. Consequently no Fourier coefficient of a pure plate indicator can reproduce a non-zero three-dimensional moment such as $\int r^2\,dV$

A natural regularisation is to thicken each plate to a thin slab of fixed small thickness $\varepsilon$ε and to examine the second-moment contribution of the union of the four thickened plates as a function of the rotation angle. Let $M(\theta)$ be that contribution. Because the underlying geometry is static and the plates are merely rotated rigid sets inside a fixed body, the function $M(\theta)$ is periodic. Its constant term (the $\theta$-average) recovers a fraction of the full octahedral moment that is proportional to the relative volume of the thickened plates; the higher Fourier coefficients encode the angular modulation caused by the discrete symmetry. In the limit $\varepsilon\to0$ the entire contribution vanishes, and one recovers only the trivial identity $0=0$.

Thus the Fourier analysis of the rotating-plate family does not independently yield the factor $2/5$. It merely redistributes, in a dynamic representation, a contribution that is already known from the static integral over the solid octahedron. The result is not new; it is a re-expression of the same geometric fact.

4. Independent geometric origin of the factor $\varphi$ in $2\varphi/5$

Pure octahedral geometry (the second-moment integral, the four hexagonal sections, the six corner octahedra, or any of their combinatorial ratios) produces the rational factor $2/5$ and nothing else. The golden ratio $\varphi=(1+\sqrt5)/2$ does not appear among the edge lengths, face diagonals, dihedral angles, or volume ratios of the regular octahedron or of its Hexarchon decomposition.

$\varphi$ arises naturally in five-fold (icosahedral or pentagonal) settings and in the angular deficit of five tetrahedra about a common edge (the Aristotle gap $\approx7.356^\circ$). Those structures lie outside the octahedral symmetry group. Consequently the geometry of the Hexarchon, taken by itself, supplies no independent mathematical reason for inserting the factor $\varphi$φ into the expression $2\varphi/5$2φ/5. Any such insertion remains an additional hypothesis that must be justified by a larger construction (for example, an explicit coupling between the octahedral plates and a five-fold deficit) and cannot be derived from the octahedral second-moment calculation alone.

Summary of what the geometry proves

• The continuous second radial moment of the outer regular octahedron is exactly $2/5$.
• The four hexagonal plates are the interfaces $x\pm y\pm z=0$ (even number of minus signs) that separate the eight tetrahedra from the six corner octahedra.
• Rotation of those plates and the associated Fourier expansion merely re-express the same static moment; they do not generate a new independent derivation of $2/5$2/5.
• The factor $\varphi$φ is not forced by any purely octahedral relation examined here.

All further claims that identify a particular plate with a particular irrational, or that promote $2\varphi/5$ as a derived constant, remain conjectural relative to the geometric facts established above.