Ordered Patch Theory

Original Task T-2 (v3.6.3 retitle): Entropic-Gravity Correspondence in OPT Codec Language Problem: The preprint describes gravity conceptually as “rendering cost” across the Markov Blanket, but does not deploy the available mathematics. The earlier title “Deriving General Relativity via Entropic Gravity” overstated the result — the appendix maps a dictionary between OPT’s codec language and Verlinde/Jacobson thermodynamic gravity rather than deriving GR from OPT primitives. Deliverable: A structural-correspondence dictionary recasting Verlinde’s exact mechanism and Jacobson’s Clausius-relation derivation in OPT’s codec vocabulary, with the load-bearing bridge assumptions explicitly named.

Closure status: STRUCTURAL DICTIONARY (v3.6.3 — formerly “Partially Resolved”). This appendix is not a derivation of Newtonian gravity or GR from OPT. It is a structural dictionary from OPT language into Verlinde/Jacobson-style thermodynamic gravity. Jacobson’s 1995 result derives the Einstein equation from entropy-area proportionality plus the Clausius relation \delta Q = T\, dS applied to all local Rindler horizons — those are powerful assumptions. Verlinde’s entropic-gravity argument is explicitly heuristic in the literature, deriving Newton-like gravity from holographic/thermodynamic assumptions rather than from microscopic first principles. T-2 imports the same assumptions in OPT codec language; the structural correspondence is real, but it inherits all of Verlinde’s and Jacobson’s bridging assumptions rather than replacing them with OPT-native derivations.

What is OPT-native: the rendering-entropy area bound S_{\text{render}}(A) \le |\partial_R A| \log q via the Markov-screening locality condition (§1.1) is a legitimate OPT-native result given the locality assumptions.

What is bridge-imported: mass-charge proportionality E_M \propto Q_M (Postulate §2.2); the Unruh formula at the codec boundary (§4.4); the radial entropy-gradient profile \propto Q_M/r^2 (Assumption T-2.A, §4.1); the Bekenstein-Hawking entropy assignment dS_{\text{render}} = c^3/(4G\hbar)\, dA (§5.1); the Einstein-Hilbert functional (§6.2). Newton’s inverse-square law and the Einstein field equations are recovered given these bridge assumptions; they are not derived independently.

v3.6.3 corrections in this appendix: (i) subtitle and §1 task description retitled from “Deriving General Relativity” to the structural-dictionary framing; (ii) §1.5 new “Units and dimensional conversion” subsection introduces \alpha (kg/bit) and the bits↔︎physical-entropy conversion S_{\text{phys}} = k_B \ln 2 \cdot S_{\text{render}} explicitly; (iii) §2.2 mass-charge bridge rewritten as an explicit Bridge Postulate with \alpha named upfront rather than deferred; (iv) §4.3 adds a constant-mismatch warning paragraph flagging the over-determination of \hbar_c (the appendix as written constrains \hbar_c = 4\pi from equipartition-vs-Unruh matching while also setting \hbar_c = l_{\text{codec}}^2 from natural units — these cannot both hold unless 4\pi = 1); (v) §4.5 adds a G_{\text{OPT}} dimensional-issue annotation flagging that c_{\text{codec}}^2/\log q has units m²/s² not m³/(kg·s²); (vi) §6.2 renames the Einstein-Hilbert functional from S_{\text{render}}[g] to I_{\text{grav}}[g] — that object is not entropy and using the same symbol obscures the distinction. The strong part (rendering entropy as boundary mutual information, §1.1) is preserved.

§1. Rendering Entropy — Formal Definition

The informal concept of rendering cost in opt-correspondences.md §7.1 (formerly preprint §7.2) is formalised here as rendering entropy, grounded in the area law established in §3.4 via the predictive cut entropy S_{\text{cut}}(A).

1.1 Definition

Let A \subset V be an observer patch on the substrate graph G, with boundary shell \partial_R A. The rendering entropy S_{\text{render}}(A, t) is formally defined as the boundary mutual information between the patch and the exterior:

S_{\text{render}}(A, t) := I\!\left(X_{\partial_R A} \,;\, X_{V \setminus A}\right)

If we assume the latent state Z_t acts as a sufficient statistic capable of capturing exactly the information X_{V \setminus A} reveals about X_{\partial_R A}, we posit this boundary correlation converges structurally to the codec’s internal conditional uncertainty: S_{\text{render}}(A, t) \sim H\!\left( X_{\partial_R A} \mid Z_t \right). The area bound follows from the structural Markov screening condition X_{A^\circ} \perp X_{V \setminus A} \mid X_{\partial_R A} established in §3.4 (preprint Eq. 7–8):

S_{\text{render}}(A, t) \leq \lvert\partial_R A\rvert \cdot \log q =: S_{\max}(A)

where q is the alphabet size of the local state space and |\partial_R A| is the number of boundary sites. If the substrate graph approximates a d-dimensional lattice, |\partial_R A| \sim \text{Area}(\partial A), confirming that S_{\text{render}} is an area quantity, not a volume quantity.

1.2 Local Rendering Entropy Density

For a continuous approximation (valid at scales much larger than the lattice spacing l_{\text{codec}} = 1/\sqrt{C_{\max}} — noting l_{\text{codec}} remains formally uninterpreted dimensionally as a spatial length until the explicit scaling identification in T-5):

S_{\text{render}}(A) = \int_{\partial A} s(x)\, dA

where s(x) [bits/area] is the local rendering entropy density at boundary point x. In the absence of sources, s(x) = (\log q)/l_{\text{codec}}^2 is uniform. A local concentration of predictive charge (see §2) perturbs s(x) away from this ground state, generating the entropy gradient that drives the entropic force.

1.5 Units and Dimensional Conversion (v3.6.3 addition)

T-2 routes between two unit systems and the OpenAI audit correctly noted that the earlier drafts did not make the conversions explicit at every step. The relevant primitives:

• Rendering entropy S_{\text{render}} is in bits (or nats, depending on the logarithm base). It is a mutual information quantity (§1.1).
• Physical / thermodynamic entropy S_{\text{phys}} is in J/K in SI. The conversion is:

S_{\text{phys}} \;=\; k_B \ln 2 \cdot S_{\text{render}} \tag{T2.1-5a}

where k_B is Boltzmann’s constant (1.381 \times 10^{-23} J/K). This conversion must appear everywhere S_{\text{render}} enters a thermodynamic equation alongside G, \hbar, c, k_B. Earlier drafts of this appendix conflated the two — using S_{\text{render}} directly in Clausius relations and Bekenstein-Hawking expressions where S_{\text{phys}} is required. The v3.6.3 corrections flag the affected locations (§5.1, §5.2) but do not yet rewrite the chain end-to-end; that is part of the T-5 reopening (separate Phase 3+ work).

• Predictive charge Q_M is in bits (mutual information; §2.1). It is not energy and not mass.
• Mass M_Q in kg is recovered via a bits-to-mass conversion factor \alpha:

M_Q \;:=\; \alpha\, Q_M \qquad \text{with} \quad [\alpha] = \text{kg/bit} \tag{T2.1-5b}

• Energy E_Q in J is then E_Q = \alpha Q_M c^2 (or \alpha Q_M c_{\text{codec}}^2 in codec units; §2.2).

The conversion factor \alpha is the bridging constant the mass-charge correspondence requires. It is not an OPT-native quantity — its value is fixed by the identification l_{\text{codec}} \to \ell_P in Appendix T-5 (currently blocked on T-2 + T-4 repairs per the appendix-corrections memo §4 priority queue). For T-2’s internal derivations, \alpha appears as an explicit constant of proportionality; for matching to standard physics, the empirical value of \alpha is constrained by the requirement that Newton’s G and the equivalence-principle conversion E = M c^2 recover their measured values.

Without \alpha named explicitly, the chain E_M = Q_M c_{\text{codec}}^2 (§2.2 below) is dimensionally invalid (it equates J on the left with bits·m²/s² on the right). With \alpha named, the correct form is E_M = (\alpha Q_M) c_{\text{codec}}^2 = M_Q c_{\text{codec}}^2, which is dimensionally consistent.

§2. Predictive Charge — The Codec Analogue of Mass

In Verlinde’s framework, mass M enters through the equipartition theorem applied to the holographic screen. OPT requires a codec-theoretic counterpart that is independently defined before any gravitational claim is made.

2.1 Definition

The predictive charge Q_M of a source region M \subset V is formally defined purely as the static spatial mutual information between M’s internal states and the observer’s Markov Blanket boundary over one codec cycle:

Q_M := I\!\left(X_M \,;\, X_{\partial_R A}\right)

We motivate an analogy to T-1 by mapping Q_M \approx R_{\text{req}}(h, D_{\min} \mid M) \cdot \Delta t. This approximation explicitly invokes a massive, unproven Stationary Ergodic Equilibrium Assumption: linking the temporal predictive rate (R_{\text{req}} \cdot \Delta t) directly to the static spatial boundary correlation (I). The exact conditions for this equality remain an open formal gap. Under this approximation, Q_M conceptually maps to the number of bits per codec cycle that the source M forces onto the observer’s boundary representation. This is the informational definition of mass: not inertia, not energy density per se, but mandatory predictive load.

2.2 Proportionality to Inertial Mass — Bridge Postulate (v3.6.3 sharpened)

Bridge Postulate T-2.M (Mass-Charge Identification). For a macroscopically stable source satisfying the Stability Filter, the predictive charge Q_M (in bits) is proportional to the inertial / gravitational mass M (in kg) via the bits-to-mass conversion factor \alpha introduced in §1.5:

M_Q \;:=\; \alpha\, Q_M, \qquad [\alpha] = \text{kg/bit} \tag{T2.2-1}

The total energy bound within the region is then:

E_M \;=\; M_Q\, c_{\text{codec}}^2 \;=\; \alpha\, Q_M\, c_{\text{codec}}^2 \tag{T2.2-2}

Why this is a bridge postulate, not a derivation (v3.6.3 explicit framing). The identification M_Q = \alpha Q_M is not derived from OPT primitives — it is a structural assumption that an independent quantity (the mutual information between source M and Markov-blanket boundary, measured in bits) maps proportionally to the standard-physics quantity (inertial mass, measured in kg). Earlier drafts wrote E_M = Q_M c_{\text{codec}}^2 directly, which is dimensionally invalid (J on the left vs bits·m²/s² on the right) and obscured the load-bearing assumption. The v3.6.3 reformulation puts \alpha in front so the bridge is visible at every step.

The conventional relativistic correspondence E = Mc^2 is recovered by setting c_{\text{codec}} = c (the empirical speed of light) and M = M_Q = \alpha Q_M. The equivalence principle (inertial mass = gravitational mass) is then additionally assumed — OPT does not derive the equivalence principle; it inherits it from standard physics as part of the bridge postulate.

The value of \alpha is fixed by matching Newton’s G to its empirical value via the identification l_{\text{codec}} \to \ell_P in Appendix T-5 (currently blocked on T-2 + T-4 repairs per the appendix-corrections memo §4 priority queue).

§3. The OPT–Verlinde Dictionary

Before deploying the mathematics, we make explicit the translation between Verlinde (2011) [38] and OPT. This prevents the derivation from inheriting assumptions of standard entropic gravity that OPT has not earned.

§4. Derivation of Newton’s Inverse-Square Law

We execute Verlinde’s exact three-step mechanism — screen entropy, equipartition, entropic force — entirely within OPT’s codec language.

4.1 Codec Surface Gravity and Boundary Temperature

Consider a spherical Markov Blanket of radius r enclosing a source of predictive charge Q_M. At each boundary point x \in \partial A, we structurally map the classical scalar potential gradient to the outward entropy gradient, defining the codec surface gravity:

\kappa(x) := c_{\text{codec}}^2 \cdot \partial_n \log s(x)

where c_{\text{codec}} is the maximum causal propagation speed in the rendered patch (identified with c in opt-correspondences.md §7.1, formerly preprint §7.2), and \partial_n is the outward normal derivative.

Assumption T-2.A (Radial entropy profile). The entropy perturbation profile of an isotropic predictive charge Q_M is radially symmetric with gradient proportional to Q_M/r^2. This is structurally equivalent to the Newtonian potential gradient; it is imported as a structural input, not derived from OPT primitives. The subsequent recovery of Newton’s law is therefore a conditional derivation contingent on this assumption, not a closed derivation.

Under Assumption T-2.A, an isotropic source Q_M at the origin reduces \kappa to:

\kappa = \frac{Q_M c_{\text{codec}}^2}{4\pi r^2 \cdot s_0}

where s_0 = (\log q)/l_{\text{codec}}^2 is the ground-state rendering entropy density.

The codec boundary temperature is:

T_{\text{codec}} = \frac{\hbar_c \,\kappa}{2\pi k_B}

where \hbar_c = 1/C_{\max} is the minimum quantum of informational action — the codec analogue of the reduced Planck constant.

4.2 Step 1 — Number of Bits on the Screen

For a spherical boundary of radius r with surface area 4\pi r^2:

N = \frac{4\pi r^2}{l_{\text{codec}}^2} \cdot \log q = S_{\max}(r)

4.3 Step 2 — Equipartition Determines T_{\text{codec}}

By the equipartition theorem applied to the N independent codec modes on the screen:

Q_M c_{\text{codec}}^2 = \tfrac{1}{2} N k_B T_{\text{codec}}

Solving for the temperature:

T_{\text{codec}} = \frac{2 Q_M c_{\text{codec}}^2}{N k_B} = \frac{Q_M c_{\text{codec}}^2 l_{\text{codec}}^2}{2\pi r^2 k_B \log q}

Consistency Constraint: Equating this equipartition temperature with the Unruh temperature derived in §4.1 (T_{\text{codec}}^{\text{Unruh}} = \frac{\hbar_c Q_M c_{\text{codec}}^2 l_{\text{codec}}^2}{8\pi^2 k_B r^2 \log q}) imposes a strict formal constraint \hbar_c = 4\pi. In the natural codec units adopted in §4.5 (c_{\text{codec}} = 1), this requires \hbar_c / l_{\text{codec}}^2 = 4\pi. In physical units, this is equivalent to the constraint on C_{\max} noted in §7.2, and is resolved in T-5.

v3.6.3 CONSTANT-MISMATCH WARNING. The appendix as written over-determines \hbar_c. Three constraints appear in §4: (i) the equipartition-vs-Unruh matching here gives \hbar_c = 4\pi; (ii) the natural-units choice (c_{\text{codec}} = 1) gives \hbar_c / l_{\text{codec}}^2 = 4\pi; (iii) the substitution in §4.5 boxed formula sets \hbar_c = l_{\text{codec}}^2. These cannot all hold simultaneously unless 4\pi = 1, which is false. The appendix marks the resolution as “deferred to T-5,” but T-5 is currently REOPENED (appendix-corrections memo §2.7, blocked on T-2 + T-4 repairs). At v3.6.3 the cleanest reading is: only one of the three constraints can be load-bearing in any given calculation; the appendix needs to be rewritten end-to-end with explicit unit-system choices and a single canonical relationship between \hbar_c, l_{\text{codec}}, and c_{\text{codec}}. Until that rewrite lands, the §4 derivation of Newton’s law should be read as structurally indicative rather than dimensionally closed — the inverse-square form is recovered from the radial-flux assumption (Assumption T-2.A) under any choice of constants; the value of G_{\text{OPT}} depends on which constraint resolution is adopted and is not yet pinned down.

4.4 Step 3 — Entropy Change for the Test Patch

A test patch of predictive charge m_p displaced by \Delta x toward the source changes its overlap with the boundary representation. We explicitly import the Unruh effect formula as a structural correspondence at the codec boundary:

\Delta S_{\text{render}} = \frac{2\pi k_B m_p c_{\text{codec}}}{\hbar_c} \cdot \Delta x

(Note: Because we are importing this Lorentz-symmetry formula rather than deriving it from the lattice, the subsequent force derivation serves strictly as a consistency check of this mapping.)

4.5 Step 4 — The Entropic Force

Verlinde’s entropic force formula F = T_{\text{codec}} \cdot \Delta S_{\text{render}}/\Delta x gives:

F = T_{\text{codec}} \cdot \frac{2\pi k_B m_p c_{\text{codec}}}{\hbar_c} = \frac{2 Q_M c_{\text{codec}}^2}{N k_B} \cdot \frac{2\pi k_B m_p c_{\text{codec}}}{\hbar_c} = \frac{4\pi Q_M m_p c_{\text{codec}}^3}{N \hbar_c}

Substituting N = 4\pi r^2 \log q / l_{\text{codec}}^2, and substituting \hbar_c = l_{\text{codec}}^2 alongside an explicit bits-to-mass dimensional conversion parameter mapping \alpha: \alpha is the bits-to-mass conversion factor with dimensions [\alpha] = \text{kg}/\text{bit} (in SI units), to be fixed by the identification l_{\text{codec}} \to \ell_P in T-5.

\boxed{F_r \propto -\frac{G_{\text{OPT}}\, (\alpha Q_M)\, (\alpha m_p)}{r^2} \qquad \text{with} \quad G_{\text{OPT}} = \frac{c_{\text{codec}}^2}{\log q}}

Restoring the preprint’s notation \lambda = G_{\text{OPT}}/(4\pi), this mathematically aligns with the entropic-force law of opt-correspondences.md §7.1 (formerly preprint §7.2 Eq. 15): F_r = -\lambda m Q_M / (4\pi r^2). Newton’s inverse-square law is recovered as a structural correspondence, up to the dimensional conversion factor \alpha^2; its explicit evaluation is deferred to T-5.

v3.6.3 DIMENSIONAL ISSUE WARNING. The boxed expression G_{\text{OPT}} = c_{\text{codec}}^2 / \log q is dimensionally invalid. Newton’s gravitational constant has units of \mathrm{m^3\,kg^{-1}\,s^{-2}}. The right-hand side c^2 / \log q has units of \mathrm{m^2/s^2} (since \log q is dimensionless). No manipulation of the Planck length alone can fix this — the expression is missing factors of \hbar_c (Planck-action-analogue), l_{\text{codec}} (Planck-length-analogue), and \alpha (kg/bit conversion). A dimensionally plausible structural form would look more like:

G_{\text{OPT}} \;\sim\; \frac{c_{\text{codec}}^3\, l_{\text{codec}}^2}{\hbar_c\, \log q}

possibly with additional factors of \alpha^{-2} depending on how predictive charges are converted to masses. The full reconciliation requires the constant-mismatch resolution flagged in §4.3 above plus a careful end-to-end unit audit of §4. This is part of the T-2 repair queue (appendix-corrections memo §4 priority queue item #3); at v3.6.3 the boxed formula is preserved for traceability but should be read as not yet dimensionally closed. The structural claim — that Newton’s inverse-square form is recoverable from the radial-flux assumption + equipartition + Unruh under suitable unit identifications — survives the dimensional issue; the numerical value of G_{\text{OPT}} does not follow from this calculation as written.

§5. Deriving the Einstein Field Equations

Newton’s law (§4) establishes the static, weak-field limit. To recover full general relativity, we follow Jacobson’s (1995) thermodynamic method: impose the Clausius relation \delta Q = T\,\delta S on the rendering entropy for every local Rindler-like horizon in the codec.

5.1 Setup — Local Rindler Horizons in the Codec

Consider any point p in the rendered spacetime. The codec’s causal structure defines a local Rindler horizon \mathcal{H} — the boundary of the past of a uniformly accelerating observer within the codec. The key ingredients are:

• Rendering entropy of \mathcal{H}: We formally explicitly import the Bekenstein-Hawking entropy assignment mapping the area law directly: dS_{\text{render}} := \frac{c_{\text{codec}}^3}{4G_{\text{OPT}}\,\hbar_c}\, dA Note: This specific coefficient maps the area bound proportionally tracking S_{\text{render}} \propto A, but the exact numeric constant here is a direct definition imported matching standard physics natively, rather than an algebraic derivation strictly extracted from the pure codec bound.

• Codec surface gravity \kappa: At the local Rindler horizon, \kappa = c_{\text{codec}}^2/l_\mathcal{H}. The codec temperature is T_{\text{codec}} = \hbar_c \kappa/(2\pi).

• Heat flux \delta Q: The predictive charge flux through dA in proper time d\tau is: \delta Q_{\text{pred}} = T^{\text{pred}}_{\mu\nu}\, k^\mu k^\nu\, dA\, d\tau where T^{\text{pred}}_{\mu\nu} is the predictive stress-energy tensor and k^\mu is the null generator of \mathcal{H}.

5.2 The Clausius Relation

The Clausius relation \delta Q_{\text{pred}} = T_{\text{codec}}\, \delta S_{\text{render}} applied to every local Rindler horizon gives:

T^{\text{pred}}_{\mu\nu}\, k^\mu k^\nu = \frac{c_{\text{codec}}^3}{4\pi G_{\text{OPT}}\,\hbar_c} \cdot \kappa\, \theta_{\mu\nu} k^\mu k^\nu

where \theta_{\mu\nu} = \nabla_\mu k_\nu + \nabla_\nu k_\mu is the expansion tensor of the null congruence. To proceed with Jacobson (1995), we must assume that the codec scales structurally satisfying the generic proportional bounds \delta S_{\text{render}} \propto \delta A mapping evenly across all local horizons. Applying the Raychaudhuri equation, the null energy condition T^{\text{pred}}_{\mu\nu} k^\mu k^\nu \geq 0, integration over the null surface, and the contracted Bianchi identity:

\boxed{G_{\mu\nu} + \Lambda g_{\mu\nu} \propto \frac{8\pi G_{\text{OPT}}\,\hbar_c}{c_{\text{codec}}^3}\, T^{\text{pred}}_{\mu\nu}}

Subject to the imported Bekenstein-Hawking coefficient (§5.1) and the proportionality assumption \delta S \propto \delta A, Jacobson’s derivation produces the Einstein field equations in OPT codec language with coupling constant 8\pi G_{\text{OPT}}\hbar_c/c_{\text{codec}}^3. The cosmological constant \Lambda arises identically as the mapping constant of integration of the Clausius relation — natively mapping to the ground-state rendering entropy density s_0 tracking the vacuum codec.

The stress-energy tensor T^{\text{pred}}_{\mu\nu} is the predictive stress-energy: the distribution of predictive charge density and flux across the rendered spacetime. In the Newtonian limit for pressureless matter, T^{\text{pred}}_{00} = Q_M/V and all other components vanish, recovering §4.

§6. Gravitational Curvature as Rate-Distortion Overflow

The closure criterion for T-2 requires a formal proof that gravitational curvature is the codec’s resistance to rendering information exceeding the rate-distortion equilibrium. §5 provides the Einstein equations; this section makes that identification precise.

6.1 The Rate-Distortion Localization Hypothesis

From T-1, the Stability Filter imposes a global boundary conditional threshold R_{\text{req}}(D_{\min}) \leq B_{\max} = C_{\max} \cdot \Delta t. Rate-distortion mappings in AIT are formally global process ensembles. Defining a strictly local predictive constraint requires explicitly extending the formalism (e.g. spatial ergodic sub-ensemble averages), deferred formally to T-5. For the purposes of this structural sketch, we treat local curvature as reflecting the local density of rate-distortion overflow, with the formal justification deferred to T-5.

6.2 Curvature as Codec Resistance — The Formal Identification (v3.6.3 functional renamed)

To map the rendering-entropy bound functionally onto G_{\mu\nu}, we construct a formal structural identification with the Einstein-Hilbert action. v3.6.3 renaming: the resulting functional is not rendering entropy — it is an Einstein-Hilbert-style action-like object — and the earlier drafts’ use of S_{\text{render}}[g] as the symbol for both quantities was a category confusion. The v3.6.3 form uses I_{\text{grav}}[g] for the gravitational-action functional to keep it distinct from the entropy S_{\text{render}}(A) defined in §1:

I_{\text{grav}}[g] \;:=\; \frac{1}{4 G_{\text{OPT}}} \int R\sqrt{-g}\, d^4x \tag{T2.6-1}

This is a bridge postulate imported to match the standard Einstein-Hilbert form (modulo standard constants and boundary terms which are absorbed into G_{\text{OPT}} here and would need restoration for a closed derivation). It is not the rendering entropy of §1, and the appendix’s earlier conflation of the two obscured this. Subject to this bridge identification, standard variational calculus gives:

\frac{\delta I_{\text{grav}}}{\delta g_{\mu\nu}} \;\propto\; \left(G_{\mu\nu} + \Lambda g_{\mu\nu}\right) \tag{T2.6-2}

The Einstein field equations (§5.2) recast as a variational equilibrium of I_{\text{grav}}:

\frac{\delta I_{\text{grav}}}{\delta g_{\mu\nu}} \;\propto\; \frac{1}{2 T_{\text{codec}}}\, T^{\text{pred}}_{\mu\nu} \tag{T2.6-3}

v3.6.3 FUNCTIONAL-RENAME NOTE. Earlier drafts used S_{\text{render}}[g] to denote both (a) the rendering entropy as boundary mutual information S_{\text{render}}(A) = I(X_{\partial_R A}; X_{V\setminus A}) in §1 and (b) the Einstein-Hilbert-style action functional \int R\sqrt{-g}\,d^4x above. These are different mathematical objects — the first is a scalar mutual-information quantity (bits) at a fixed spacetime instant; the second is a functional of the metric tensor (J·s in physical units after dimensional restoration). The v3.6.3 form uses I_{\text{grav}}[g] for the latter. The structural-correspondence claim of §6.2 is that variations of I_{\text{grav}} encode the same information as Clausius variations of rendering entropy at the local Rindler horizons of §5 — that is, \delta I_{\text{grav}} \leftrightarrow T_{\text{codec}}\, \delta S_{\text{render}} on-shell — but this is the postulate, not a derivation. Treating the action as the entropy directly (as earlier drafts did by symbol-reuse) is not a valid identification.

This defines the extremal rendering condition: the metric configuration that minimises the rendering entropy cost given T^{\text{pred}}_{\mu\nu} is exactly the one satisfying Einstein’s equations.

Formal statement of the partial closure mapping.

Under this identification, the Einstein tensor G_{\mu\nu} arises as the metric variation of the gravitational-action functional I_{\text{grav}}[g] (Eq. T2.6-2), which corresponds on-shell to Clausius variations of rendering entropy at the local Rindler horizons of §5 (\delta I_{\text{grav}} \leftrightarrow T_{\text{codec}}\, \delta S_{\text{render}} — the §6.2 bridge postulate, not a metric derivative of S_{\text{render}} itself, which carries no metric dependence). Conceptually, curvature encodes the codec’s second-order resistance to metric perturbation: it is large where additional boundary bits must be allocated to accommodate local predictive charge density.

§7. Event Horizons as Codec Saturation Points

Note: The following analysis treats R_{\text{req}}(p, D_{\min}) as a well-defined local quantity; this requires the Localization Hypothesis of §6.1 and is therefore heuristic pending T-5.

7.1 The Saturation Condition

An event horizon forms where R_{\text{req}}(p, D_{\min}) = B_{\max} exactly — the boundary at which the Stability Filter is saturated. For a spherically symmetric source of predictive charge Q_M, setting R_{\text{req}}(r_S) = B_{\max} and solving:

r_S = \frac{G_{\text{OPT}}\, Q_M}{c_{\text{codec}}^2}

This is OPT’s native Schwarzschild radius. The standard general-relativistic result is r_S = 2GM/c^2, which differs by a factor of 2. This factor-of-2 discrepancy is not derived from OPT primitives; matching the classical result would require either Q_M = 2M (an ad-hoc identification) or a proper treatment of the near-horizon geometry that produces the factor naturally. We do not impose this matching; instead, we note the factor-of-2 as an open discrepancy that may be resolved by a full near-horizon analysis.

Inside r_S, \Delta R(p) > 0 at every point: the codec is in permanent overflow. The interior of a black hole is the region where the Stability Filter irrecoverably fails — not a location in physical space, but a topological boundary of the codec’s representational capacity.

7.2 Hawking Radiation as Codec Boundary Leakage

At the horizon r = r_S, the codec temperature with \kappa = c_{\text{codec}}^4/(4G_{\text{OPT}} Q_M) gives:

T_H = \frac{\hbar_c\, c_{\text{codec}}^4}{8\pi k_B G_{\text{OPT}}\, Q_M}

This reproduces the standard Hawking temperature in structural form. Matching to the physical value requires \hbar_c c_{\text{codec}}^4/G_{\text{OPT}} = \hbar c^3/G, which fixes C_{\max} in terms of fundamental constants — introducing a tension with T-1’s treatment of C_{\max} as a free empirical parameter. Resolution is deferred to T-5.

7.3 The Information Paradox and the Page Curve

Hawking’s 1976 paradox [104] arises only under the assumption that the render must preserve unitarity across a substrate-level loss event. Under OPT, no such loss occurs: the substrate is independent of the codec’s representational state, and the render-side irretrievability of trans-horizon detail is the same Fano-bounded structure that handles every other codec horizon (preprint §3.12). The paradox dissolves at the render layer; what remains is the technical question of how the codec’s boundary state evolves during evaporation.

The Page curve [105]. For an evaporating black hole that begins in a pure state, unitarity requires the von Neumann entropy of the radiation S_{\text{rad}}(t) to rise from zero, peak at the Page time t_P \approx t_{\text{evap}}/2, and return to zero by t_{\text{evap}}. From within OPT, this curve is a direct prediction of the codec’s bandwidth budget: the saturation boundary at r_S holds at most S_{BH} distinguishable bits (§7.1 above), so the radiation’s information content cannot exceed the boundary’s running capacity. The peak-and-fall structure is precisely what one expects when bandwidth bound at the saturation surface is gradually re-allocated to the asymptotic patch.

Quantum-extremal-surface / island prescription. Penington [106] and Almheiri–Hartman–Maldacena–Shaghoulian–Tajdini [107] recover the Page curve in semiclassical gravity by computing the von Neumann entropy of an asymptotic radiation region R as

S(R) = \min_{\mathcal{I}} \left[\, \frac{\text{Area}(\partial \mathcal{I})}{4 G \hbar} + S_{\text{matter}}(R \cup \mathcal{I}) \,\right]

where the minimisation is over candidate “island” regions \mathcal{I} inside the black hole. Before the Page time, the trivial island (\mathcal{I} = \emptyset) is dominant and S(R) rises; after the Page time, a non-trivial island emerges and S(R) falls.

OPT structural reading. This is the form expected from Appendix P-2’s QECC layer. Theorem P-2b establishes that the codec’s boundary representation must satisfy the Knill–Laflamme conditions; the island prescription is the renormalised continuum analogue of the discrete Ryu–Takayanagi formula proved in P-2d (S_{\text{vN}}(\rho_A) \leq |\gamma_A|\log\chi). The “island” is the part of the codec’s interior representation reconstructible from the radiation via the QECC; its emergence at the Page time is the codec’s re-encoding of bound interior bits into the asymptotic patch as the horizon shrinks. OPT does not derive the islands construction de novo — that derivation lives in semiclassical gravity — but it predicts its structural form: any consistent treatment of evaporation must operate at the QECC layer of the codec rather than on a substrate-level wave function.

Falsification footprint. A sustained empirical Page-curve violation — super-thermal information leakage from an evaporating black hole that cannot be embedded in any QECC structure — would falsify P-2’s QECC layer and the structural account given here. Conversely, a clean derivation of the islands construction from substrate-level unitarity without an effective error-correcting code would weaken (though not strictly falsify) OPT’s structural-confirmation reading.

7.4 Complementarity and Firewalls

The black-hole complementarity hypothesis (Susskind, Thorlacius & Uglum, 1993) and the AMPS firewall argument [108] both target the consistency of OPT-style frame-relative descriptions. We register their OPT reading.

Complementarity as frame-relative codec description. The infalling and asymptotic frames carry codec representations of the same boundary information, related by the Asymmetric One-Way Holography map (preprint §3.12). The render does not require both frames to be simultaneously consistent in a frame-independent substrate; what it requires is that each frame’s codec satisfies the Stability Filter for the trajectories it actually executes. Complementarity therefore costs nothing extra in OPT — it is the expected consequence of the codec being frame-relative all the way down, and is structurally aligned with the relativity-of-simultaneity reading carried by Rovelli’s RQM (preprint §7 quantum residue; opt-correspondences.md §6).

Firewalls as predicted local QECC failure. Almheiri, Marolf, Polchinski & Sully [108] argue that the conjunction of (i) horizon smoothness, (ii) unitary evaporation, and (iii) effective field theory outside the horizon is mutually inconsistent: at least one must fail, and the smoothness assumption is the typical casualty, yielding a “firewall” at the horizon. From within OPT, the trilemma is structural rather than substrate-level: smoothness is a property of the codec’s QECC layer (P-2), not of the substrate; unitary evaporation holds at the render layer (§7.3 above); and effective field theory is the codec’s macroscopic compression. A firewall is therefore what the infalling observer would encounter if the QECC layer fails locally at the horizon — i.e., if the codec’s error-correction tolerance is exceeded by trans-horizon noise. This is a predicted failure mode of the codec, not a paradox within it.

Quantitative prediction (heuristic). Whether actual black holes exhibit firewalls is an open empirical question; OPT’s structural account predicts that horizons of high Q_M — where the QECC has more error-correcting room per boundary site — should be smoother than horizons of low Q_M, with the cross-over set by the codec’s local noise threshold rather than by Planck-scale substrate physics. A definite cross-over scale Q_M^\star is not predicted by T-2 alone; it requires the local QECC tolerance derived in P-2 combined with a substrate noise model, both of which remain open formal edges (P-2 closure).

§8. Cosmological Constant as Vacuum Rendering Cost

The cosmological constant \Lambda appears in §5.2 as the integration constant of the Clausius relation. The vacuum state of the codec is not empty: it is the ground-state configuration of rendering entropy with uniform density s_0 = (\log q)/l_{\text{codec}}^2. The associated vacuum predictive stress-energy is:

T^{\text{vac}}_{\mu\nu} = -\frac{\Lambda\, c_{\text{codec}}^4}{8\pi G_{\text{OPT}}\,\hbar_c}\, g_{\mu\nu}

In OPT, \Lambda > 0 corresponds to a de Sitter codec geometry — the codec’s ground state is an accelerating expansion. Qualitatively, this is an expected structural rationalization: the Stability Filter preferentially selects configurations where Forward Fan branches are maximally separated (cosmological expansion increases the informational distance between branches, reducing the rate of accidental causal recoupling). This framework provides a qualitative explanation for the sign of \Lambda, though deriving its extraordinarily small, quantitative observed limits is deferred to the physical constants recovery in T-5.

§9. Closure Summary and Open Edges

T-2 Deliverables — Partially Resolved (Structural Mapping)

1. Rendering entropy formalised. S_{\text{render}}(A) defined via bounding mutual information. Area law confirmed; local density s(x) defined.

2. Newton’s law mapped. F_r = -G_{\text{OPT}} Q_M m / r^2 recovered via Verlinde’s mechanism, contingent upon importing the Unruh boundary assumption.

3. Einstein equations mapped. G_{\mu\nu} + \Lambda g_{\mu\nu} \propto T^{\text{pred}}_{\mu\nu} aligns with Jacobson’s Clausius method, contingent upon horizon-saturation and Einstein-Hilbert functional assumptions.

4. Closure criterion satisfied as mapping. G_{\mu\nu} \propto \delta I_{\text{grav}} / \delta g_{\mu\nu}, with \delta I_{\text{grav}} \leftrightarrow T_{\text{codec}}\, \delta S_{\text{render}} on-shell (§6.2 bridge postulate; S_{\text{render}} itself carries no metric dependence). Curvature is structurally identified, via this on-shell Clausius correspondence, with the codec’s mapped resistance to rate-distortion overflow. \blacksquare

5. Event horizons. r_S = G_{\text{OPT}} Q_M / c_{\text{codec}}^2 derived as the codec saturation point. Hawking temperature recovered from boundary thermodynamics.

6. Information paradox dissolved at render layer (§7.3). The Hawking paradox [104] resolves structurally: no substrate-level loss; render-side irretrievability is the §3.12 Fano-bounded asymmetric holography. The Page curve [105] and the islands / quantum-extremal-surface results [106, 107] are read as structural confirmation of P-2’s QECC layer rather than as derivations independent of it.

7. Complementarity and firewalls (§7.4). Complementarity is the expected consequence of frame-relative codec description under §3.12. AMPS firewalls [108] are interpreted as a predicted local QECC failure mode rather than a paradox; a quantitative cross-over scale Q_M^\star depends on the open closure of P-2.

Remaining open edges

• T-3 (MERA Tensor Networks) now has a sharper target: the tensor network upgrade of Z_t is required to convert S_{\text{render}} from a classical area law into the Ryu-Takayanagi holographic entropy bound. The Jacobson derivation here is the intermediate floor.

• T-5 (Constants Recovery) depends on T-2: G_{\text{OPT}} = c_{\text{codec}}^2 / \log q (dimensionally open — see the §4.5 v3.6.3 warning) must be matched to the empirical G via the l_{\text{codec}} \to l_P identification. This would constrain the codec lattice spacing to the Planck length; the corresponding structural inequality T-5a.1 was demoted at v3.6.11 pending the joint T-2 / T-5 dimensional rewrite.

• Quantum gravity (open): Deriving the exact Einstein field equations from Active Inference — rather than from Jacobson’s thermodynamic method — remains a profound open challenge. The tensor-network upgrade (T-3) and the ADH quantum error correction path (P-2) are the next formal steps.

• de Sitter extension (open): The derivation in §5 follows Jacobson and applies cleanly to asymptotically flat and AdS geometries. Extending to dS/CFT — consistent with the observed positive \Lambda — requires the open mathematical extension noted in preprint §8.3 item 4.

This appendix is maintained as part of the OPT project repository alongside theoretical_roadmap.pdf. References: Verlinde (2011) [38], Jacobson (1995), Bekenstein (1981) [40], Almheiri-Dong-Harlow (2015) [42].