The Finite Closure Band Theorem and the Continuum Pi Ceiling

What the question is really asking

The distinction is not between a good number and a bad number. The same finite configuration can support several different readouts. An occupied area, a radial slice, and a smooth-circle comparison do not ask exactly the same question. Treating their outputs as competing estimates of one supposedly corrected constant would lose the structure the model is meant to expose.

My starting point is a recurring finite closure process. The process must be admitted at its declared interface, remain within the relevant change conditions, and carry closure through its participating states. Only then do I compare the resulting family with the ideal continuum circle. The logical order is:

Ledger and gates → selected finite loop → native carrier readout → smooth shadow → continuum comparison.

This is why an orbital-style consequence appears in a paper about $\pi$. A recurring loop supplies the minimal testbed for closure. Showing how the model selects such a loop is relevant to motion, but it is not yet a complete treatment of orbital mechanics. The primary question here remains what finite recurring closure reports about its own geometry. Source: Appendix M, “Pi, finite closure” and “Minimal recurring-motion testbed.”

Admission, change, and closure have different jobs

The argument uses three already established RSI relations. They do not become three interchangeable ways of saying that a loop exists.

The declared geometry belongs in the carrier. The loop model therefore makes its interface participation explicit rather than obtaining a preferred answer by changing the meaning of recursion depth or weakening per-layer tension. The broader symbol definitions remain available in RSI Terminology and RSI Equations. Source: Appendix M, “RSI primitives, gate order.”

The minimal recurring-motion model

I use one anchor identity and one equal-scale orbiting identity. The candidate orbit consists of $N$ equal resolved turn-states. The state radius is $a$, the orbit radius is $R$, and the carrier choice is $\kappa$. The construction declares their relationship before evaluating closure.

\alpha=\frac{2\pi}{N}.

R=\kappa a,

d_N=2R\sin\!\left(\frac{\pi}{N}\right)=2\kappa a\sin\!\left(\frac{\pi}{N}\right).

For the literal equal-size tangent centerline reading, $\kappa=2$. The corridor-midline reading uses a smaller carrier choice; the source examines the interval $1.7\leq\kappa\leq2.0$. These choices identify the model being examined. They are not measurements showing that every physical orbit has one of these radii.

The finite turn step matters because consecutive resolved states overlap only through a particular region. A drawn ring does not establish how much of that overlap can carry the next state without introducing drift. That is the next part of the construction. Source: Appendix M, “Minimal recurring-motion testbed.”

How the model selects a loop

Local overlap and the usable core

Place two consecutive equal-radius states at $x=\pm d_N/2$. Inside their shared corridor lens, the local support profiles are:

s_1(x,y)=1-\frac{(x+d_N/2)^2+y^2}{a^2},
\qquad
s_2(x,y)=1-\frac{(x-d_N/2)^2+y^2}{a^2},

The profiles are zero outside that shared region. The overlap density is then:

O(x,y)=\sqrt{s_1(x,y)s_2(x,y)}.

Overlap alone is not the entire requirement. The source measures local imbalance relative to the support present:

B(x,y)=\frac{|s_1(x,y)-s_2(x,y)|}{s_1(x,y)+s_2(x,y)}.

It then imposes the no-drift condition against the half-step turn budget:

B(x,y)\le \tan\!\left(\frac{\alpha}{2}\right)=\tan\!\left(\frac{\pi}{N}\right).

That condition defines a usable subset of the shared lens:

\Omega_{\mathrm{usable}}(N,\kappa)
=
\left\{(x,y)\in\Omega_{\mathrm{lens}}:\;B(x,y)\le \tan\!\left(\frac{\pi}{N}\right)\right\}.

The weighted constructive-core fraction measures that subset using the local two-state overlap density:

f_{\mathrm{sel}}^{(w)}(N,\kappa)
=
\frac{\int_{\Omega_{\mathrm{usable}}} O(x,y)\,dA}{\int_{\Omega_{\mathrm{lens}}} O(x,y)\,dA}.

The distinction matters. The whole overlap region and the part usable for recurring closure are not automatically identical. Nor is the weighted fraction the same as a raw area fraction. I return to the unweighted alternative when testing robustness.

Carrier aperture and transfer

Each orbit state subtends an anchor-centered cap half-angle:

\beta=\arcsin\!\left(\frac{1}{\kappa}\right),

The effective geometric corridor is the spherical lens shared by the two caps. Its aperture fraction is:

c_{A_{\!P}}(N,\kappa)=\frac{\Omega_{\mathrm{lens}}(\beta,\alpha)}{4\pi}.

Here the source uses lens notation in two geometry-specific roles: the local two-state region in the core construction and the shared spherical-cap solid angle in the aperture fraction. They must be read in their declared contexts, not treated as numerically interchangeable areas.

The usable transfer factor also contains the adjacent-state inheritance factor $\lambda_N(\kappa)$, the local phase-smoothness factor $P_N=\cos(\pi/N)$, and the global loop-smoothness factor $L_N=\sin(\pi/N)/(\pi/N)$:

\eta_N(N,\kappa)=c_{A_{\!P}}(N,\kappa)\,f_{\mathrm{sel}}^{(w)}(N,\kappa)\,\lambda_N(\kappa)\,P_N\,L_N.

This is the model’s assembled transfer condition. Its components are not independent substitutes for one another. A reproduction must specify the same support, geometry, inheritance, and constructive-core choices rather than setting their product to obtain a desired step count.

The selector and first stable loop

The source gives the selector and loop change diagnostic as:

q_N=\frac{1}{N\,\eta_N(N,\kappa)},
\qquad
D^{\mathrm c}(N)=\frac{1}{\sqrt{N\,\eta_N(N,\kappa)}}.

The first stable loop is the smallest integer $N$ satisfying:

q_N<1
\qquad\Longleftrightarrow\qquad
N\,\eta_N(N,\kappa)>1.

Only after that selection do I calculate the native closure readout and its smooth shadow. “Before the shadow” does not mean the model forbids trigonometric notation: the step geometry and smoothness factors are already written above. It means the final continuum comparison is not used as the criterion for choosing the admitted loop. Source: Appendix M, “Minimal recurring-motion model and native loop selection.”

The occupied carrier and its readout

Let $U_N$ be the union of the $N$ resolved orbit-state discs. The carrier occupies the radial band bounded by $r_-=R-a$ and $r_+=R+a$. The first native readout normalizes the actual occupied area by that band’s squared-radius difference:

\Pi_{\mathrm{carrier}}(N,\kappa)=\frac{\operatorname{Area}(U_N)}{r_+^2-r_-^2}.

This readout begins from occupied support, not from an assumed circumference-to-diameter ratio. The source obtains the following closed form by integrating over one wedge and multiplying by $N$:

\Pi_{\mathrm{carrier}}(N,\kappa)
=
\frac{N}{2\kappa}\arcsin\!\left(\kappa\sin\frac{\pi}{N}\right)
+
\frac{N}{2}\sin\frac{\pi}{N}\sqrt{1-\kappa^2\sin^2\frac{\pi}{N}}.

The smooth finite-loop shadow is a different quantity:

The shadow is evaluated after selection. It gives a useful smooth comparison, but it does not become the native occupied-area readout merely because it lies closer to $\pi$. Source: Appendix M, “Native closure readouts and a worked example.”

A complete worked example

The source’s micro-compactness example chooses $\kappa=1.70$. With the specified weighted-core selector, its first selected loop is $N_\ast=39$. The reported selector values are $q_{39}\approx0.9953$ and $D^{\mathrm c}(39)\approx0.9976$, just inside the stated stable condition.

The ordering is the point:

These are reported values from the declared model, not measurements of a universal replacement for $\pi$. The carrier and shadow differ because they read the same selected loop in different ways. Source: Appendix M, worked micro-compactness example.

The formal argument

The following definitions, lemmas, reasons, theorem, proof, and corollary retain the source’s stated model. References point to the corresponding relations on this page; the equation labels use the Appendix M sequence.

\Pi_{\mathrm{carrier}}(N,\kappa)<\pi.

The native ceiling-gap law

The source expands the carrier readout for large $N$:

\Pi_{\mathrm{carrier}}(N,\kappa)
=
\pi
-
\frac{\pi^3(\kappa^2+1)}{6N^2}
-
\frac{\pi^5(3\kappa^4-10\kappa^2-1)}{120N^4}
+
O(N^{-6}).

It defines the native discreteness gap and its leading behavior by:

\Delta_{\pi}(N,\kappa):=\pi-\Pi_{\mathrm{carrier}}(N,\kappa).

\Delta_{\pi}(N,\kappa)
\approx
\frac{\pi^3(\kappa^2+1)}{6N^2}.

The coefficient depends on the declared carrier choice. The $1/N^2$ law concerns the large-$N$ comparison; it does not identify every finite loop with its leading asymptotic term. Source: Appendix M, “Finite Closure Band Theorem and the continuum-pi ceiling.”

Robustness: what changes and what persists

A single example can illustrate a result without showing how it behaves under a changed modeling choice. The source therefore varies the selector, introduces a second native readout, and changes the constructive-core weighting. Those variations have distinct jobs.

The weighted-core witness family

The labels identify the source’s selector examples. In particular, “cosmology edge-hugging” is not a statement that this table validates a cosmological model. The reported bands are:

3.12834\lesssim \Pi_{\mathrm{carrier}}\lesssim 3.13440,

3.13820\lesssim \Pi_{\mathrm{shadow}}\lesssim 3.14016.

The step count moves with the declared choice. The proposed stable object is the banded readout structure, not one integer that every application must reproduce.

A second native readout

The occupied-area readout is not the only way to inspect the selected carrier. The source also integrates the radial support thickness over a turn. Within a local wedge:

\ell_N(\theta)=2\sqrt{a^2-R^2\sin^2\theta},\qquad |\theta|\le \frac{\pi}{N}.

The corresponding slice readout is:

\Pi_{\mathrm{slice}}(N,\kappa)=\frac{1}{4a}\int_0^{2\pi}\ell_N(\theta)\,d\theta
=
N\int_0^{\pi/N}\sqrt{1-\kappa^2\sin^2\theta}\,d\theta.

For the weighted-core selected family, the source reports:

3.13176\lesssim \Pi_{\mathrm{slice}}^{(w)}\lesssim 3.13584.

Its asymptotic expansion is:

\Pi_{\mathrm{slice}}(N,\kappa)
=
\pi
-
\frac{\pi^3\kappa^2}{6N^2}
+
\frac{\pi^5(4\kappa^2-3\kappa^4)}{120N^4}
+
O(N^{-6}).

The slice and area readouts concern the same finite carrier, but they normalize different geometric information. Agreement about the ceiling does not make their finite values identical.

A second constructive-core family

The weighted core uses overlap density. The second family removes that weighting and counts only the area of the usable no-drift subset:

f_{\mathrm{sel}}^{(a)}(N,\kappa)=\frac{\operatorname{Area}(\Omega_{\mathrm{usable}})}{\operatorname{Area}(\Omega_{\mathrm{lens}})}.

Replacing the weighted fraction by this hard-area fraction changes the transfer calculation. Over the same carrier interval, the source reports representative selected pairs $(N_\ast,\kappa)=(52,1.7)$, $(61,1.8)$, $(71,1.9)$, and $(82,2.0)$ and the following bands:

\begin{aligned}
3.13416 &\lesssim \Pi_{\mathrm{carrier}}^{(a)} \lesssim 3.13775,\\
3.13606 &\lesssim \Pi_{\mathrm{slice}}^{(a)} \lesssim 3.13852,\\
3.13968 &\lesssim \Pi_{\mathrm{shadow}}^{(a)} \lesssim 3.14082.
\end{aligned}

The new family changes the first admitted counts. It does not make a previously selected loop acquire a new readout by fiat; selection must be performed for the new core definition.

Fixed-carrier smoothing and the combined bands

The source also gives numerical monotonicity witnesses at fixed $\kappa$. The entries below follow $N=20,40,60,80$ in that order. They are readout witnesses at those counts, not a declaration that every listed count is the first admitted loop.

Across the two constructive-core families, the combined reported bounds are:

\begin{aligned}
3.12834 &\lesssim \Pi_{\mathrm{carrier}} \lesssim 3.13775,\\
3.13176 &\lesssim \Pi_{\mathrm{slice}} \lesssim 3.13852,\\
3.13820 &\lesssim \Pi_{\mathrm{shadow}} \lesssim 3.14082.
\end{aligned}

The source states the following ordering for its selected families:

\Pi_{\mathrm{carrier}}\;<\;\Pi_{\mathrm{slice}}\;<\;\Pi_{\mathrm{shadow}}\;<\;\pi

That is a comparison within the declared families, not a claim about every conceivable selector or every physical system. The robustness result gains its force from changing specific model components while retaining the same directional structure. Source: Appendix M, “Robustness and convergence of the finite closure family.”

Two falsifiable prediction laws

Separation between the native readouts

The first comparison here stays between the two native readouts:

\Delta_{\mathrm{dual}}(N,\kappa):=\Pi_{\mathrm{slice}}(N,\kappa)-\Pi_{\mathrm{carrier}}(N,\kappa).

Using their expansions gives:

\Delta_{\mathrm{dual}}(N,\kappa)
=
\frac{\pi^3}{6N^2}
-
\frac{\pi^5(6\kappa^2+1)}{120N^4}
+
O(N^{-6}).

The leading separation is independent of $\kappa$, even though the higher-order correction retains carrier dependence. The source normalizes that separation as:

H(N,\kappa):=\frac{6N^2\left(\Pi_{\mathrm{slice}}(N,\kappa)-\Pi_{\mathrm{carrier}}(N,\kappa)\right)}{\pi^3}.

H(N,\kappa)\to 1
\qquad\text{for the selected finite closure family.}

The representative selected loops give reported values of $H$ around 0.994–0.998. This is the source’s model calculation. A physical application would still need an independently specified relation between the measured system and these native quantities.

The gap to the continuum ceiling

The second normalized comparison uses the occupied-area deficit:

G(N,\kappa):=
\frac{6N^2\left(\pi-\Pi_{\mathrm{carrier}}(N,\kappa)\right)}{\pi^3(\kappa^2+1)}.

G(N,\kappa)\to 1
\qquad\text{for the selected finite closure family.}

The source names a finite selected loop reaching or exceeding $\pi$, or failure of the stated convergence behavior, as a challenge to the present theorem form. The declared carrier and selector matter to such a test. Changing them after seeing the result would change the claim being tested.

For a useful examination, identify which question is in dispute: the selection rule, a component of transfer, the geometric readout, the proof chain, the numerical evaluation, or the proposed physical application. A disagreement with one of those does not automatically answer all the others. Conversely, an internally successful calculation is not by itself an independent experiment. Source: Appendix M, dual-readout separation and “Falsifiable predictions and practical implications.”

What follows, and where the result stops

The source separates three layers. First are the inherited RSI gates, closure rule, and admission floor. Second is the declared one-anchor, one-orbit construction with its selector, readouts, finite-band theorem, robustness treatment, and convergence laws. Third are the further questions that this construction opens rather than settles.

The positive result is not merely “finite things approximate a circle.” The model asks which recurring loop is admitted before the final circle comparison, and then asks what that finite support actually reports. The native carrier, the slice readout, and the smooth shadow remain distinguishable. Their relation explains why a useful continuum description need not be the first object in the physical account.

The orbital consequence is likewise specific. The appendix develops recurring orbital-style closure within this testbed. It does not supply a complete account of every orbit, a universal replacement for celestial mechanics, or a completed dark-sector result. The later orbital program has its own work to do.

The source explicitly leaves open whether a deeper admissibility law would select one universal finite constant, whether a stronger form of domain independence can be proved, and how to extend the orbital consequences into a full RSI treatment. Those are further questions about the family, not reasons to replace it with one attractive decimal.

My interpretation is that finite closure should precede its continuum representation in the physical explanation. The theorem’s mathematical scope remains the specified model. This page does not change the definition of $\pi$, identify the illustrative bands with measured universal constants, or transfer the theorem’s status automatically to an empirical application. The broader distinction is developed in Carrier and Readout. Source: Appendix M, interpretation, claims ladder, and conclusion.

Terms and notation used here

Finite closure band

The family of finite recurring loops selected in the declared model, together with the bounded native readouts reported for those loops. It is not one universal measured constant.

Selected finite loop

The smallest integer step count satisfying the declared selector condition for a specified carrier choice.

Elios admission floor

The minimum admitted interface in the RSI account. Admission at the floor is distinct from selecting a persisting recurring loop.

Existence gate, S

The ratio E_A/A_P used to evaluate admission at the declared carrier.

Change gate, D_c

The ratio E_F/E_A used to evaluate change conditions. The loop model also writes its selector-related diagnostic as D^c(N).

Anchor and field budgets, E_A and E_F

The source-defined RSI energy accounts. They retain their separate roles in admission, change, and closure.

Interface carrier, A_P

The declared geometric and participation carrier. The aperture fraction belongs to this carrier construction.

Turn-state count, N

The finite integer number of equal resolved turn-states in the candidate loop.

State radius, a

The radius assigned to each equal-scale resolved state in the testbed.

Carrier choice, κ

The dimensionless ratio R/a. The source examines weighted and hard-area examples over 1.7 ≤ κ ≤ 2.0.

Orbit radius, R

The declared centerline radius R = κa in the loop model, not recursion depth.

Step angle, α, and separation, d_N

The angle 2π/N between adjacent turn-states and the corresponding center separation.

Support profiles, s_1 and s_2

The two local profiles defined within the shared corridor lens; the source sets them to zero outside that shared region.

Overlap density, O

The square root of the product of the two local support profiles, used to weight their shared support.

Imbalance ratio, B

The absolute profile difference divided by their sum, used in the local no-drift condition.

Usable core, Ω_usable

The subset of the shared lens satisfying the no-drift condition.

Weighted constructive-core fraction

The usable overlap-density integral divided by the full shared-lens overlap-density integral.

Hard-area constructive-core fraction

The usable area divided by the shared-lens area, without the overlap-density weighting.

Aperture fraction, c_AP

The participating spherical-cap lens solid angle divided by 4π. This is distinct from the planar core-area fraction.

Inheritance factor, λ_N

The adjacent-state overlap inheritance factor in the transfer product. Its role must remain specified when reproducing the selector.

Smoothness factors, P_N and L_N

The local phase and global finite-loop factors cos(π/N) and sin(π/N)/(π/N).

Transfer factor, η_N

The product of aperture, usable-core fraction, inheritance, and the two smoothness factors.

Native selector, q_N

The quantity 1/(Nη_N). The source selects the smallest N for which q_N < 1.

Occupied carrier, U_N

The union of the finite resolved orbit-state discs used in the area-based readout.

Native carrier readout, Π_carrier

The occupied area divided by r_+² − r_-². It is not the smooth-shadow readout.

Slice readout, Π_slice

The source's normalized integral of radial support thickness over a turn, a second native readout of the same selected carrier.

Smooth shadow, Π_shadow

The later comparison N sin(π/N), evaluated after selection rather than used to choose the loop.

Continuum ceiling, π

The ideal-circle constant approached from below by the specified family. The theorem does not change its mathematical definition.

Native gap, Δ_π

The difference π − Π_carrier for the declared model and carrier.

Dual-readout gap, Δ_dual

The difference Π_slice − Π_carrier; its leading asymptotic term is independent of κ.

Normalized gap quantities, G and H

The two source-defined convergence diagnostics. G normalizes the gap to π; H normalizes the separation between the native readouts.

Questions

Sources, reading, and citation

The source treatment

Matthew Tripp Zejda. “The Finite Closure Band Theorem and the Continuum pi Ceiling.” Appendix M of Resolution State Interface (RSI): Introduction to Universal Relativity. The public source edition used here is available through the Zenodo book record, version v5, DOI 10.5281/zenodo.19176988.

Open the book and appendix source record.

The record supplies a book PDF, a source project, and the separate appendix source. The DOI identifies that RSI book record, not a separately accepted journal article for this theorem. The page follows this specified source treatment; it does not claim that the cited book version is the latest edition of all RSI research.

Where each part comes from

The source’s eleven subsections cover, in order: the pi question; RSI primitives and explanatory order; the minimal recurring-motion testbed; native loop selection; the native readout and worked example; the theorem and ceiling; robustness and convergence; physical interpretation; prediction laws; scope and significance; and the conclusion. The equation sequence reproduced here corresponds to Appendix M’s 36 relations. The definitions, four lemma reasons, central proof, corollary, and robustness proposition retain their source identities.

The source’s figures can be examined in the linked treatment. This page develops the mathematical and explanatory sequence without turning those source illustrations into new experimental evidence.

Suggested citation

Zejda, Matthew Tripp. 2026. “The Finite Closure Band Theorem and the Continuum pi Ceiling.” Appendix M in Resolution State Interface (RSI): Introduction to Universal Relativity, source edition v5. Zenodo book record. DOI: 10.5281/zenodo.19176988.

Related reading

Back to the beginning