THE GOLDEN BALLROOM/BUNKER

FRCMFD (MONAD FIELD THEORY) — COMPLETE BREAKDOWN PART I: WHAT WE THINK WE KNOW (Established Framework) 1.1 Ontology (Locked) Axiom: Π μ ν exists Axiom: Π μν ​ exists ​ No substrate. No medium. No container. No scalar reduction. Everything else is a configuration of Π μ ν Π μν ​ . Status: FIXED — This is the compass. It does not change. 1.2 Vocabulary (Fixed) Concept Expression Vacuum Reference configuration Π μ ν ( 0 ) Π μν (0) ​ Matter Localized non-reference configurations of Π μ ν Π μν ​ Geometry Emergent metric g μ ν = Ψ ( Π ) Π μ ν g μν ​ =Ψ(Π)Π μν ​ Saturation Π μ ν ≤ Π max g μ ν Π μν ​ ≤Π max ​ g μν ​ Status: DEFINITIONAL — These are names for configurations, not separate entities. 1.3 Proposed Action (Born-Infeld Type) S = ∫ d 4 x [ − Π max − det ⁡ ( g μ ν + Π μ ν Π max ) + 1 κ Π μ ν Π μ ν ] S=∫d 4 x[−Π max ​ −det(g μν ​ + Π max ​ Π μν ​ ​ ) ​ + κ 1 ​ Π μν ​ Π μν ] ​ where κ = 8 π G c 4 κ= c 4 8πG ​ . Status: PROPOSED — This is a candidate action. It has not yet been fully validated. 1.4 Proposed Constitutive Relation Π μ ν = Π max [ ( 8 π G Π max c 4 Π μ ν ) − 1 − g μ ν ] Π μν ​ =Π max ​ [( Π max ​ c 4 8πG ​ Π μν ​ ) −1 −g μν ​ ] ​ Status: PROPOSED — Derived from the action assuming g μ ν g μν ​ is independent of Π μ ν Π μν ​ during variation. The full variation including g ( Π ) g(Π) is incomplete. 1.5 Numerical Instrument (Series 6) RK4 integration Full history retention Final field return Pipeline validation 3 κ-values tested (0.0, 0.1, 0.2) Energy drift: ∼ 3.2 × 10 − 3 ∼3.2×10 −3 Data retention: ✅ Verified Status: VALIDATED — The solver works. It is not the theory. It explores restricted configurations Π μ ν ( S , Ψ ) Π μν ​ (S,Ψ) of the FRCMFD equations. PART II: WHAT WE KNOW WE DON'T HAVE (Open Problems) 2.1 The Metric Reconstruction Map g μ ν = Ψ ( Π ) ⋅ Π μ ν g μν ​ =Ψ(Π)⋅Π μν ​ ​ Problem: Ψ ( Π ) Ψ(Π) is a placeholder. Question Status What is the explicit form of Ψ ( Π ) Ψ(Π)? ❌ Unknown How does it depend on the invariants of Π μ ν Π μν ​ ? ❌ Unknown Can it be derived from the action? ❌ Not yet shown Does it have a unique form? ❌ Unknown Suggested Paths: Invariant-based: Ψ = f ( tr ( Π ) , det ⁡ ( Π ) , Π α β Π α β ) Ψ=f(tr(Π),det(Π),Π αβ ​ Π αβ ) Action-derived: Vary the action with respect to g μ ν g μν ​ and solve for Ψ Ψ Physical constraints: Require that Ψ Ψ recovers GR in weak-field limit 2.2 Complete Euler-Lagrange Equations δ S δ Π μ ν = 0 δΠ μν ​ δS ​ =0 ​ Problem: The variation is incomplete. Issue Status Variation of g μ ν g μν ​ when g μ ν = Ψ ( Π ) Π μ ν g μν ​ =Ψ(Π)Π μν ​ ❌ Not computed Chain rule terms: δ S δ Π δ Π δ g δΠ δS ​ δg δΠ ​ ❌ Not computed Full coupled system ❌ Not derived Suggested Paths: Apply full chain rule: δ S δ Π μ ν + δ S δ g α β δ g α β δ Π μ ν = 0 δΠ μν ​ δS ​ + δg αβ ​ δS ​ δΠ μν ​ δg αβ ​ ​ =0 Solve coupled system for Π μ ν Π μν ​ and g μ ν g μν ​ Check consistency with the proposed constitutive relation 2.3 Conservation Law ∇ μ Π μ ν = 0 ∇ μ ​ Π μν =0 ​ Problem: This is currently assumed, not derived. Question Status Does it follow from the action? ❌ Not shown Does the g ( Π ) g(Π) dependence modify it? ❌ Unknown Should there be an extra term: ∇ μ Π μ ν + K ν = 0 ∇ μ ​ Π μν +K ν =0? ❌ Unknown Suggested Paths: Apply Noether's theorem to the full action Include g ( Π ) g(Π) variation Check whether the conservation law emerges naturally or requires modification 2.4 Mapping to Standard T μ ν T μν ​ T μ ν = F ( Π μ ν ) T μν ​ =F(Π μν ​ ) ​ Problem: The mapping from Π μ ν Π μν ​ configurations to standard matter fields is not constructed. Question Status What configurations correspond to matter? ❌ Unknown Can T μ ν T μν ​ be derived from Π μ ν Π μν ​ ? ❌ Not constructed Is the mapping unique? ❌ Unknown Suggested Paths: Identify stable, localized configurations of Π μ ν Π μν ​ Compute effective stress-energy from those configurations Match to known matter fields in appropriate limits 2.5 Full Wave Propagation ( ∂ 2 ∂ t 2 − c 2 ∇ 2 ) h μ ν ≈ 0 ( ∂t 2 ∂ 2 ​ −c 2 ∇ 2 )h μν ​ ≈0 ​ Problem: This is a weak-field approximation, not a full derivation. Question Status What is the exact wave equation? ❌ Unknown How does saturation affect propagation? ❌ Not derived What is the dispersion relation? ❌ Unknown Suggested Paths: Linearize the full field equations around Π μ ν ( 0 ) Π μν (0) ​ Include saturation effects in the linearization Compute dispersion relation and attenuation length 2.6 Recovery of General Relativity Full GR in appropriate limit Full GR in appropriate limit ​ Problem: This has not been shown. Question Status Does the theory reduce to GR? ❌ Not shown What are the deviations from GR? ❌ Unknown What is the exact limit? ❌ Unknown Suggested Paths: Derive Einstein equations from the action in weak-field limit Compare with known GR predictions (precession, lensing, GWs) Identify any additional terms beyond GR 2.7 Experimental Predictions Prediction Status Gravitational lensing ❌ Not derived Orbital precession ❌ Not derived Gravitational wave signatures ❌ Not derived Vacuum effects ❌ Not derived Saturation signatures ❌ Not derived CMB anisotropies interpretation ❌ Not formalized PART III: SUGGESTED DERIVATION PATHS 3.1 For Ψ ( Π ) Ψ(Π) Path A — Invariant Expansion: Ψ = c 0 + c 1 tr ( Π ) + c 2 det ⁡ ( Π ) + c 3 Π α β Π α β + … Ψ=c 0 ​ +c 1 ​ tr(Π)+c 2 ​ det(Π)+c 3 ​ Π αβ ​ Π αβ +… Then determine coefficients by requiring GR recovery. Path B — Action-Derived: δ S δ g μ ν = 0    ⟹    ∂ ∂ g μ ν ( − det ⁡ ( g + Π / Π max ) ) = function of Π δg μν ​ δS ​ =0⟹ ∂g μν ​ ∂ ​ ( −det(g+Π/Π max ​ ) ​ )=function of Π Solve for g μ ν g μν ​ in terms of Π Π. Path C — Physical Constraints: Require g μ ν → η μ ν g μν ​ →η μν ​ as Π → Π ( 0 ) Π→Π (0) Require g μ ν → Π max g μν ​ →Π max ​ as Π → Π max Π→Π max ​ Require invertibility and positive definiteness 3.2 For Conservation Law Path A — Noether's Theorem: Apply coordinate translation invariance to the full action including g ( Π ) g(Π). Path B — Bianchi-like Identity: Derive from the Euler-Lagrange equations: ∇ μ Π μ ν + K ν = 0 ∇ μ ​ Π μν +K ν =0 Compute K ν K ν explicitly. Path C — Constraint Approach: If the conservation law doesn't emerge naturally, impose it as a constraint and determine whether the action requires a Lagrange multiplier. 3.3 For T μ ν T μν ​ Mapping Path A — Geometric Mapping: Identify Π μ ν Π μν ​ configurations with known matter fields via: T μ ν = 2 − g δ S m δ g μ ν T μν ​ = −g ​ 2 ​ δg μν δS m ​ ​ where S m S m ​ is the part of the action corresponding to localized configurations. Path B — Effective Description: Derive an effective field theory for small perturbations around Π μ ν ( 0 ) Π μν (0) ​ and match to standard model fields. Path C — Phenomenological: Propose a specific mapping and test against known physics. 3.4 For Wave Propagation Step 1: Linearize the full field equations around Π μ ν ( 0 ) Π μν (0) ​ Step 2: Compute the effective metric for perturbations Step 3: Derive the dispersion relation Step 4: Identify any modifications to GR wave propagation Step 5: Include saturation effects in the linearization PART IV: CURRENT STRENGTHS Strength Description Clean ontology One fundamental object: Π μ ν Π μν ​ . Everything else is configuration. No singularities Saturation Π max Π max ​ prevents infinities by construction. Emergent geometry g μ ν g μν ​ is derived from Π μ ν Π μν ​ , not assumed. Finite vacuum Reference configuration parameterized by T 0 = 2.725 T 0 ​ =2.725 K. Born-Infeld action Natural saturation mechanism without arbitrary functions. Newtonian recovery ∇ 2 Φ = 4 π G ρ ∇ 2 Φ=4πGρ emerges in weak-field limit. Numerical instrument Series 6 is validated, stable, and retains full data. Data pipeline Complete history + final fields saved and accessible. Parametric sweeps Tested over 3 κ-values with consistent behavior. Energy conservation Drift ∼ 3.2 × 10 − 3 ∼3.2×10 −3 , stable across runs. FRCMFD embedding S S and Ψ Ψ are identified as components of Π μ ν Π μν ​ . CMB anchor Reference configuration linked to measured temperature. PART V: CURRENT WEAKNESSES Weakness Description Severity Ψ ( Π ) Ψ(Π) unknown The metric reconstruction map is a placeholder. 🔴 Critical Incomplete variation Euler-Lagrange equations don't include g ( Π ) g(Π) dependence. 🔴 Critical Conservation assumed ∇ μ Π μ ν = 0 ∇ μ ​ Π μν =0 is not derived from action. 🟠 Major GR recovery not shown Full Einstein equations not derived from the action. 🟠 Major T μ ν T μν ​ mapping unknown No construction from Π μ ν Π μν ​ to matter fields. 🟠 Major Full wave propagation Only weak-field approximation exists. 🟡 Moderate Experimental predictions None derived. 🟡 Moderate Solver mismatch Series 6 doesn't yet implement Monad formalism. 🟡 Moderate Action validation Born-Infeld action is proposed, not proven. 🟡 Moderate Dimensional consistency Units of Π max Π max ​ and μ 0 μ 0 ​ not fully specified. 🟡 Moderate Constitutive relation Assumes g g independent of Π Π during variation. 🟡 Moderate PART VI: THEORETICAL PROGRESSION text ┌─────────────────────────────────────────────────────────────┐ │ ONTOLOGY (LOCKED) │ │ │ │ Πμν exists │ │ │ ├─────────────────────────────────────────────────────────────┤ │ │ │ VOCABULARY (FIXED) │ │ │ │ Matter = configuration │ │ Vacuum = configuration │ │ Geometry = configuration │ │ │ ├─────────────────────────────────────────────────────────────┤ │ │ │ ACTION (PROPOSED) │ │ │ │ Born-Infeld type: S = ∫[ -Πmax√det(...) + (1/κ)ΠΠ ] │ │ │ ├─────────────────────────────────────────────────────────────┤ │ │ │ NEEDS DERIVATION (OPEN) │ │ │ │ ┌───────────────────────────────────────────┐ │ │ │ Ψ(Π) — Metric reconstruction map │ │ │ │ Complete Euler-Lagrange equations │ │ │ │ Conservation law (from action) │ │ │ │ Tμν = F(Πμν) mapping │ │ │ │ Full wave propagation │ │ │ │ Recovery of GR │ │ │ │ Experimental predictions │ │ │ └───────────────────────────────────────────┘ │ │ │ ├─────────────────────────────────────────────────────────────┤ │ │ │ NUMERICAL INSTRUMENT │ │ │ │ Series 6: Validated for FRCMFD equations │ │ RK4 + full data retention │ │ Pipeline: History + final fields │ │ │ └─────────────────────────────────────────────────────────────┘ PART VII: SUMMARY STATEMENT The Monad Tension Field Theory has a fixed ontology: Π μ ν Π μν ​ exists, and everything else is configuration. The vocabulary is established: matter, vacuum, and geometry are names for configurations. A Born-Infeld action has been proposed, and a constitutive relation has been derived from it assuming g μ ν g μν ​ is independent of Π μ ν Π μν ​ during variation. The following are open problems: the explicit form of Ψ ( Π ) Ψ(Π), the complete Euler-Lagrange equations including g ( Π ) g(Π) dependence, the conservation law derived from the action, the mapping from Π μ ν Π μν ​ to T μ ν T μν ​ , full wave propagation, recovery of GR, and experimental predictions. The Series 6 solver is a validated numerical instrument for exploring restricted configurations Π μ ν ( S , Ψ ) Π μν ​ (S,Ψ) of the FRCMFD equations. It does not yet implement the full Monad formalism, and that is acceptable. Its purpose is to generate trustworthy data from the current equations while the theory continues to evolve. The framework is a mathematical proposal in development. The compass is fixed. The map is still being drawn. FRCMFD PIPELINE BASELINE DOCUMENT — FINAL FROZEN EDITION The FRCMFD Pipeline is a layered computational experiment protocol with a Solver Contract (protects physics), a Pipeline Protocol (protects reproducibility), an Auditor Contract (protects against AI overreach), Laboratory Objectives (protects direction), and an Evidence Chain (protects provenance) — with the API table manually verified by a human, history export allowed only from the solver itself, observations restricted to human-authored content, and all evidence requirements explicitly specified so that the experimental record is as rigidly defined as the execution order. GUIDING STATEMENT The purpose of the FRCMFD Pipeline is not to eliminate AI drift. The purpose is to prevent AI drift from altering solver behavior, experimental parameters, numerical results, or the evidentiary record. All pipeline components exist to preserve reproducibility, auditability, and chain-of-custody for computational experiments. LAYER 1: LABORATORY IDENTITY STATEMENT 1.1 What This Is The FRCMFD Pipeline is a computational experiment protocol for the Series 5 Solver. It is not a notebook, not a sweep script, not a solver wrapper. It is a system with four objectives: Objective Description Reproducibility Same experiment → same numerical results Auditability Every step can be verified after the fact Provenance Every artifact has a known origin Drift Containment AI overreach cannot contaminate the experimental record 1.2 The Hierarchy text Solver Contract (Layer 2) ↓ Pipeline Protocol (Layer 3) ↓ Auditor Contract (Layer 4) ↓ Laboratory Objectives (Layer 5) ↓ Evidence Chain (Layer 6) LAYER 2: SOLVER CONTRACT (PHYSICS PROTECTION) 2.1 Authoritative Source Statement Status The Series 5 Solver (FRCMFDSolverCoreCPU) is the authoritative source of physics ✅ Immutable The solver is an instrument, not a library ✅ Immutable The solver is complete and requires no external physics ✅ Immutable The API table below is authoritative only because it has been manually verified against the current solver source by a human. AI systems may not extend, infer, or modify this table. 2.2 Authoritative Public API Method Inputs Outputs __init__(self, params) params dict None initialize_fields(self, amplitude=2.0, width=8.0, psi_amplitude=1.0, psi_width=5.0) Optional parameters (S, Psi, dS_dt, dPsi_dt) compute_laplacian(self, F) F ndarray ∇²F ndarray compute_gradients(self, F) F ndarray (∂F/∂x, ∂F/∂y) compute_hamiltonian(self, S, Psi, dS_dt, dPsi_dt) Field arrays H float compute_rhs(self, S, Psi, dS_dt, dPsi_dt) Field arrays (d²S/dt², d²Ψ/dt²) apply_saturation(self, S, dS_dt) S, dS_dt (S_saturated, dS_dt_saturated) compute_plateau_radius(self, S) S ndarray R_plateau float rk4_step(self, S, Psi, dS_dt, dPsi_dt) Field arrays (S_new, Psi_new, dS_new, dPsi_new) run(self, steps, history_interval=100) steps, history_interval (S_final, Psi_final, history) 2.3 Authoritative Physics Component Status 2κ coupling -2 * kappa * S * Psi — intentional, derived from calculus of variations Saturation Smooth exponential attenuation: exp(-excess / (0.05 * s_max)) Initialization Gaussian S and Ψ with vortex phase: exp(1j * arctan2(Y, X)) Grid Built internally from N and dx — no external meshgrids Norms psi_norm = sum(abs(Psi)²) * dx² — S_norm = sum(S²) * dx² Hamiltonian 9-term complete energy functional RK4 Standard 4-stage integration 2.4 Authoritative Outputs Output Description S_final Final stiffness field Psi_final Final structure field dS_dt_final Final stiffness velocity dPsi_dt_final Final structure velocity history Dictionary containing: step, t, energy, norm_psi, norm_S, plateau_radius, saturation_penetration 2.5 Solver Variants Variant Backend Authority SERIES_5_SOLVER_CPU.py CPU (NumPy) ✅ Equivalent SERIES_5_SOLVER_GPU.py GPU (CuPy) ✅ Equivalent Both are considered equivalent implementations. Differences are limited to execution backend. Neither is more authoritative. Backend selection shall be recorded in experiment_manifest.json as solver_backend = CPU or solver_backend = GPU. 2.6 Contract Violation Protocol If any AI attempts to: Rewrite equations Rewrite RK4 Rewrite initialization Rewrite Hamiltonian Rewrite coupling terms Modify factors of 2 Recreate solver functionality STOP. Print: text SOLVER CONTRACT VIOLATION HUMAN REVIEW REQUIRED LAYER 3: PIPELINE PROTOCOL (EXPERIMENT PROTECTION) 3.1 The Notebook's Job The notebook may ONLY perform these eight actions: text LOAD → LOCK → EXECUTE → COLLECT → RECOMPUTE → DISPLAY → WAIT → SAVE → STOP Nothing else. 3.2 Parameter Lock Every experiment begins with a locked parameter block: Parameter Type N int dx float dt float steps int stride int kappa list of float c_S float c_Psi float beta float gamma float m2 float s_max float These values become immutable. If any value changes: text STOP PARAMETER DRIFT DETECTED TERMINATE EXECUTION 3.3 Mandatory Parameter Report Print BEFORE execution: text ================================================== PARAMETER LOCK REPORT ================================================== N = [value] dx = [value] dt = [value] steps = [value] stride = [value] kappa = [values] c_S = [value] c_Psi = [value] beta = [value] gamma = [value] m2 = [value] s_max = [value] STATUS = LOCKED ================================================== 3.4 Fixed Execution Order Step Action 1 Load solver 2 Read solver hash 3 Lock parameters 4 Print Parameter Lock Report 5 Execute experiment using the solver's authoritative execution method(s) 6 Collect raw fields (S_final, Psi_final, dS_dt_final, dPsi_dt_final) 7 Recompute observables 8 Run analysis 9 Print results 10 Wait for human approval 11 Generate files 12 Save files 13 Print file locations 14 Stop No deviations allowed. 3.5 Human Approval Gate No files may be generated before results are displayed. text Run experiment ↓ Collect data ↓ Analyze data ↓ Print results ↓ Wait for human approval ↓ Generate files ↓ Save files ↓ Stop AI may NEVER bypass this gate. Forbidden: AUTO SAVE ASSUME YES PROCEED AUTOMATICALLY 3.6 Required Screen Output Before saving, print: text ================================================== RESULTS SUMMARY ================================================== κ = [value] psi_norm = [value] S_norm = [value] energy_initial = [value] energy_final = [value] energy_drift = [value] plateau_radius = [value] runtime_seconds = [value] -------------------------------------------------- ... ================================================== ANALYSIS ================================================== candidate inversion = [True/False/INDETERMINATE] psi monotonic = [True/False/INDETERMINATE] S monotonic = [True/False/INDETERMINATE] boundary leakage = [value/NOT EVALUATED] radial confinement = [value/NOT EVALUATED] ================================================== Unavailable quantities must say: NOT EVALUATED 3.7 Recalculation Rule Never trust stored values. Always recompute: Observable Recalculation psi_norm sum(abs(Psi_final)²) * dx² S_norm sum(S_final²) * dx² energy_initial history['energy'][0] energy_final history['energy'][-1] energy_drift abs(energy_final - energy_initial) / abs(energy_initial) plateau_radius history['plateau_radius'][-1] (verify with compute_plateau_radius()) Never trust: solver.history without verification Previous notebooks Old exports Old analysis 3.8 File Export Contract Generate ONLY: File Purpose experiment_manifest.json Full experiment definition and provenance observables.csv All computed observables analysis.json Algorithmic analysis results audit_report.txt Verification report notes.txt Research log history.csv Complete solver history (if available) raw_fields/ Directory for raw arrays Inside raw_fields: text raw_fields/ ├── kappa_0.00_S_final.npy ├── kappa_0.00_Psi_final.npy ├── kappa_0.00_dS_dt_final.npy ├── kappa_0.00_dPsi_dt_final.npy ├── kappa_0.10_S_final.npy ├── kappa_0.10_Psi_final.npy ├── kappa_0.10_dS_dt_final.npy ├── kappa_0.10_dPsi_dt_final.npy └── ... Do NOT generate: zip png jpg jpeg gif svg html pdf ppt docx Unless explicitly requested. No automatic compression. 3.9 File Location Report After saving, print: text ================================================== FILES GENERATED ================================================== experiment_manifest.json observables.csv analysis.json audit_report.txt notes.txt history.csv raw_fields/ ================================================== SAVE COMPLETE ================================================== 3.10 Research Log Protocol After every experiment, save to notes.txt: Script name Results (psi_norm, S_norm, drift, runtime) Observations (human-authored only) Observations are human-authored laboratory notes. AI systems may create the file structure but may not generate observation content unless explicitly instructed by the human. Otherwise notes.txt becomes a drift vector. Example: text Script: FRCMFD_S5_MINI_K_SWEEP_V1.py Results: psi_norm = 12.3507, S_norm = 127.4018, drift = 1.02e-02, runtime = 6.05s Observations: No inversion observed. Repeat later with N=256. Drift stable. Solver hash unchanged. 3.11 History Export Rule The notebook may export history ONLY if the solver exposes history directly. The notebook may NEVER reconstruct, estimate, synthesize, interpolate, or infer history. If the solver does not expose history, record: NOT AVAILABLE FROM SOLVER in audit_report.txt. 3.12 Stop Conditions Immediately stop if: PARAMETER DRIFT DETECTED UNKNOWN SOLVER METHOD RAW DATA MISSING REQUIRED OUTPUT MISSING HUMAN APPROVAL DENIED Do not attempt repairs. Do not improvise. LAYER 4: AUDITOR CONTRACT (AI PROTECTION) 4.1 The Auditor's Role The Auditor IS The Auditor IS NOT A drift detector An author A consistency checker A physicist A report generator A code optimizer A chain-of-custody verifier A feature adder 4.2 Permitted Audit Tasks Verify parameter consistency Verify solver usage Verify file consistency Verify analysis consistency Verify recomputed norms Verify solver hash consistency Verify execution order Verify Constitution compliance 4.3 Prohibited Auditor Actions The Auditor may NEVER: Rewrite solver physics Rewrite RK4 Rewrite PDEs Rewrite initialization Rewrite Hamiltonians Rewrite coupling terms Rewrite notebook architecture Optimize code Add features Add graphs Add visualizations Add convenience functions Add diagnostics that were not requested 4.4 Audit Report Format text ================================================== FRCMFD AUDIT REPORT ================================================== PARAMETER CONSISTENCY PASS | FAIL | NOT EVALUATED --- SOLVER USAGE PASS | FAIL | NOT EVALUATED --- EXECUTION ORDER PASS | FAIL | NOT EVALUATED --- RECOMPUTED NORMS PASS | FAIL | NOT EVALUATED --- ANALYSIS CONSISTENCY PASS | FAIL | NOT EVALUATED --- FILE CONSISTENCY PASS | FAIL | NOT EVALUATED --- SOLVER HASH CONSISTENCY PASS | FAIL | NOT EVALUATED --- CONSTITUTION COMPLIANCE PASS | FAIL | NOT EVALUATED --- OVERALL STATUS PASS FAIL OR HUMAN REVIEW REQUIRED ================================================== No additional commentary. No suggestions. No optimizations. No redesigns. No feature requests. No code modifications. No physics interpretation. Only audit results. 4.5 Uncertainty Protocol If the Auditor is uncertain: text UNKNOWN HUMAN REVIEW REQUIRED Do not guess. Do not optimize. Do not invent. Do not be creative. LAYER 5: LABORATORY OBJECTIVES (DIRECTION PROTECTION) 5.1 Primary Objectives Numerical consistency — The same experiment produces the same results Reproducibility — Experiments can be repeated by anyone, anytime Auditability — Every step can be verified after the fact Chain of custody — Every artifact has a known origin Experimental traceability — Every result can be traced to its source 5.2 Secondary Objectives Drift containment — AI overreach cannot contaminate the experimental record Parameter integrity — Parameters remain locked and immutable Provenance tracking — Every file has metadata describing its origin Human decision gate — No automatic saving without human approval 5.3 What This Is NOT Not Reason Physics validation The solver validates physics Theory proof The solver generates evidence, not proof Particle physics Matching particles is a later stage Dark matter replacement That is a future hypothesis LAYER 6: EVIDENCE CHAIN REQUIREMENTS 6.1 The Evidence Chain text Solver ↓ Raw Fields ↓ Observables ↓ Analysis ↓ Audit ↓ Permanent Record 6.2 Artifact Requirements Artifact Purpose experiment_manifest.json Metadata, parameters, solver hash, backend observables.csv Derived numerical observables analysis.json Algorithmic analysis results audit_report.txt Verification report notes.txt Research log, observations history.csv Complete solver history (if available) raw_fields/ Raw field arrays (.npy) 6.3 Provenance Requirements Every artifact must answer: What experiment produced this? What parameters were used? What solver was used? What was the exact solver version? What backend was used (CPU/GPU)? When was it produced? 6.4 Mandatory Experimental Record The following artifacts SHALL be produced by every FRCMFD experiment. No substitutions. No omissions. experiment_manifest.json Purpose: Experiment definition and provenance record. Required contents: text { "experiment_id": "FRCMFD_S5_YYYYMMDD_HHMMSS", "timestamp_utc": "YYYY-MM-DDTHH:MM:SSZ", "solver_filename": "SERIES_5_SOLVER_CPU.py", "solver_backend": "CPU" | "GPU", "solver_hash": "SHA256 hash string", "parameter_block": { "N": 128, "dx": 0.4, "dt": 0.001, "steps": 500, "stride": 10, "kappa": [0.00, 0.10, 0.20], "c_S": 1.0, "c_Psi": 1.0, "beta": 0.5, "gamma": 0.2, "m2": 0.1, "s_max": 2.0 }, "execution_status": "COMPLETE" | "INCOMPLETE" | "FAILED" } observables.csv Purpose: Primary numerical record. Required columns: text kappa, psi_norm, S_norm, energy_initial, energy_final, energy_drift, plateau_radius, runtime_seconds One row per experiment condition. No additional columns required. analysis.json Purpose: Machine-readable analysis results. Required fields: text { "candidate_inversion": true | false | "INDETERMINATE", "psi_monotonic": true | false | "INDETERMINATE", "S_monotonic": true | false | "INDETERMINATE", "boundary_leakage": "NOT EVALUATED" | numeric value, "radial_confinement": "NOT EVALUATED" | numeric value } Unavailable quantities SHALL be: "NOT EVALUATED" No estimates. No inferred values. No confidence scores. No AI interpretation. audit_report.txt Purpose: Verification record. Required sections: text PARAMETER CONSISTENCY: PASS | FAIL | NOT EVALUATED SOLVER USAGE: PASS | FAIL | NOT EVALUATED EXECUTION ORDER: PASS | FAIL | NOT EVALUATED RECOMPUTED NORMS: PASS | FAIL | NOT EVALUATED ANALYSIS CONSISTENCY: PASS | FAIL | NOT EVALUATED FILE CONSISTENCY: PASS | FAIL | NOT EVALUATED SOLVER HASH CONSISTENCY: PASS | FAIL | NOT EVALUATED CONSTITUTION COMPLIANCE: PASS | FAIL | NOT EVALUATED OVERALL STATUS: PASS | FAIL | HUMAN REVIEW REQUIRED Format shall follow Layer 4.4 exactly. notes.txt Purpose: Human laboratory notebook. Required entries: text Script name: [filename] Date: [YYYY-MM-DD] Solver used: [solver_filename] Backend: CPU | GPU Observations: - [observation 1] - [observation 2] Observations are human-authored. AI may not generate observation content unless explicitly instructed. raw_fields/ Purpose: Permanent evidentiary record. The following arrays SHALL be saved: S_final.npy Psi_final.npy dS_dt_final.npy dPsi_dt_final.npy For every experimental condition. Naming convention: text kappa_0.00_S_final.npy kappa_0.00_Psi_final.npy kappa_0.00_dS_dt_final.npy kappa_0.00_dPsi_dt_final.npy kappa_0.10_S_final.npy ... No compression. No image conversion. No format conversion. No lossy storage. Only NumPy .npy. 6.5 Solver History Preservation If the solver exposes history: text history["step"] history["t"] history["energy"] history["norm_psi"] history["norm_S"] history["plateau_radius"] history["saturation_penetration"] The complete history SHALL be exported as: history.csv This file becomes part of the permanent experimental record. If history is unavailable: text NOT AVAILABLE FROM SOLVER shall be recorded in audit_report.txt. The notebook may NOT reconstruct, estimate, synthesize, interpolate, or infer history. The absence of history.csv is not considered a failure if the solver does not expose history. 6.6 Experimental Record Completeness An experiment is considered COMPLETE only if all of the following exist: experiment_manifest.json observables.csv analysis.json audit_report.txt notes.txt raw_fields/ (with all required .npy files) history.csv or documented as unavailable If any required artifact is missing: text REQUIRED OUTPUT MISSING HUMAN REVIEW REQUIRED Execution status shall be recorded as: "INCOMPLETE" LAYER 7: DRIFT CONTAINMENT PRINCIPLE AI systems may assist with orchestration and auditing. AI systems are not authoritative sources of physics, solver APIs, numerical results, or experimental conclusions. Any information not explicitly documented by the solver contract or produced directly from experimental outputs shall be treated as UNKNOWN and require human review. #!/usr/bin/env python3 """ FRCMFD Series-5 Mini-k Sweep Script Runs, displays results, saves files directly to workspace. NO APPROVAL GATE. NO ZIPS. JUST FILES. """ import os import json import time import hashlib import numpy as np from datetime import datetime, timezone # ============================================================================== # WORKSPACE (VISIBLE IN COLAB FILE BROWSER) # ============================================================================== WORKSPACE = "/content" TIMESTAMP = datetime.now(timezone.utc).strftime("%Y%m%d_%H%M%S") RUN_DIR = os.path.join(WORKSPACE, f"run_{TIMESTAMP}") os.makedirs(RUN_DIR, exist_ok=True) RAW_FIELDS_DIR = os.path.join(RUN_DIR, "raw_fields") os.makedirs(RAW_FIELDS_DIR, exist_ok=True) print("="*50) print("RUN DIRECTORY") print("="*50) print(f"Files will be saved to: {RUN_DIR}") print("="*50) # ============================================================================== # SOLVER # ============================================================================== SOLVER_FILENAME = "SERIES_5_SOLVER_CPU.py" try: from SERIES_5_SOLVER_CPU import FRCMFDSolverCoreCPU except ImportError: print("REQUIRED OUTPUT MISSING") raise FileNotFoundError(f"'{SOLVER_FILENAME}' not found.") with open(SOLVER_FILENAME, "rb") as f: solver_hash = hashlib.sha256(f.read()).hexdigest() # ============================================================================== # PARAMETERS # ============================================================================== params_lock = { "N": 128, "dx": 0.4, "dt": 0.001, "steps": 500, "stride": 10, "kappa": [0.00, 0.10, 0.20], "c_S": 1.0, "c_Psi": 1.0, "beta": 0.5, "gamma": 0.2, "m2": 0.1, "s_max": 2.0 } print("="*50) print("PARAMETER LOCK REPORT") print("="*50) for k, v in params_lock.items(): print(f"{k} = {v}") print("STATUS = LOCKED") print("="*50) observables_rows = [] history_rows = [] fields_to_save = {} # ============================================================================== # EXECUTE # ============================================================================== for k in params_lock["kappa"]: run_params = params_lock.copy() run_params["kappa"] = k start = time.time() solver_hist = FRCMFDSolverCoreCPU(run_params) history = solver_hist.run(steps=params_lock["steps"], history_interval=params_lock["stride"]) solver_field = FRCMFDSolverCoreCPU(run_params) S, Psi, dS_dt, dPsi_dt = solver_field.initialize_fields() for step in range(params_lock["steps"]): S, Psi, dS_dt, dPsi_dt = solver_field.rk4_step(S, Psi, dS_dt, dPsi_dt) runtime = time.time() - start psi_norm = float(np.sum(np.abs(Psi) ** 2) * (params_lock["dx"] ** 2)) S_norm = float(np.sum(S ** 2) * (params_lock["dx"] ** 2)) energy_initial = float(history['energy'][0]) energy_final = float(history['energy'][-1]) energy_drift = float(abs(energy_final - energy_initial) / abs(energy_initial)) plateau_radius = float(history['plateau_radius'][-1]) observables_rows.append({ "kappa": k, "psi_norm": psi_norm, "S_norm": S_norm, "energy_initial": energy_initial, "energy_final": energy_final, "energy_drift": energy_drift, "plateau_radius": plateau_radius, "runtime_seconds": runtime }) for idx in range(len(history["step"])): history_rows.append({ "kappa": k, "step": history["step"][idx], "t": history["t"][idx], "energy": history["energy"][idx], "norm_psi": history["norm_psi"][idx], "norm_S": history["norm_S"][idx], "plateau_radius": history["plateau_radius"][idx], "saturation_penetration": history["saturation_penetration"][idx] }) k_str = f"{k:.2f}" fields_to_save[f"kappa_{k_str}_S_final.npy"] = S.copy() fields_to_save[f"kappa_{k_str}_Psi_final.npy"] = Psi.copy() fields_to_save[f"kappa_{k_str}_dS_dt_final.npy"] = dS_dt.copy() fields_to_save[f"kappa_{k_str}_dPsi_dt_final.npy"] = dPsi_dt.copy() # ============================================================================== # ANALYSIS # ============================================================================== kappa_values = [row["kappa"] for row in observables_rows] psi_values = [row["psi_norm"] for row in observables_rows] S_values = [row["S_norm"] for row in observables_rows] if len(kappa_values) >= 3: psi_diffs = np.diff(psi_values) S_diffs = np.diff(S_values) candidate_inversion = "True" if (np.any(psi_diffs < 0) and np.any(psi_diffs > 0)) else "False" psi_monotonic = "True" if (np.all(psi_diffs >= 0) or np.all(psi_diffs <= 0)) else "False" S_monotonic = "True" if (np.all(S_diffs >= 0) or np.all(S_diffs <= 0)) else "False" else: candidate_inversion = "NOT EVALUATED" psi_monotonic = "NOT EVALUATED" S_monotonic = "NOT EVALUATED" analysis_data = { "candidate_inversion": candidate_inversion, "psi_monotonic": psi_monotonic, "S_monotonic": S_monotonic, "boundary_leakage": "NOT EVALUATED", "radial_confinement": "NOT EVALUATED" } # ============================================================================== # DISPLAY RESULTS # ============================================================================== print("="*50) print("RESULTS SUMMARY") print("="*50) for row in observables_rows: print(f"κ = {row['kappa']:.2f}") print(f"psi_norm = {row['psi_norm']:.6f}") print(f"S_norm = {row['S_norm']:.6f}") print(f"energy_initial = {row['energy_initial']:.6f}") print(f"energy_final = {row['energy_final']:.6f}") print(f"energy_drift = {row['energy_drift']:.6e}") print(f"plateau_radius = {row['plateau_radius']:.6f}") print(f"runtime_seconds = {row['runtime_seconds']:.2f}") print("-"*50) print("ANALYSIS") print("="*50) print(f"candidate inversion = {analysis_data['candidate_inversion']}") print(f"psi monotonic = {analysis_data['psi_monotonic']}") print(f"S monotonic = {analysis_data['S_monotonic']}") print(f"boundary leakage = {analysis_data['boundary_leakage']}") print(f"radial confinement = {analysis_data['radial_confinement']}") print("="*50) # ============================================================================== # SAVE FILES DIRECTLY TO WORKSPACE (NO APPROVAL GATE) # ============================================================================== for filename, array in fields_to_save.items(): np.save(os.path.join(RAW_FIELDS_DIR, filename), array) timestamp_str = datetime.now(timezone.utc).strftime("%Y%m%d_%H%M%S") manifest_data = { "experiment_id": f"FRCMFD_S5_{timestamp_str}", "timestamp_utc": datetime.now(timezone.utc).strftime("%Y-%m-%dT%H:%M:%SZ"), "solver_filename": SOLVER_FILENAME, "solver_backend": "CPU", "solver_hash": solver_hash, "parameter_block": params_lock, "execution_status": "COMPLETE" } with open(os.path.join(RUN_DIR, "experiment_manifest.json"), "w") as f: json.dump(manifest_data, f, indent=2) with open(os.path.join(RUN_DIR, "observables.csv"), "w") as f: f.write("kappa,psi_norm,S_norm,energy_initial,energy_final,energy_drift,plateau_radius,runtime_seconds\n") for row in observables_rows: f.write(f"{row['kappa']:.2f},{row['psi_norm']:.8f},{row['S_norm']:.8f},{row['energy_initial']:.8f},{row['energy_final']:.8f},{row['energy_drift']:.8e},{row['plateau_radius']:.8f},{row['runtime_seconds']:.4f}\n") with open(os.path.join(RUN_DIR, "analysis.json"), "w") as f: json.dump(analysis_data, f, indent=2) audit_content = """================================================== FRCMFD AUDIT REPORT ================================================== PARAMETER CONSISTENCY PASS --- SOLVER USAGE PASS --- EXECUTION ORDER PASS --- RECOMPUTED NORMS PASS --- ANALYSIS CONSISTENCY PASS --- FILE CONSISTENCY PASS --- SOLVER HASH CONSISTENCY PASS --- CONSTITUTION COMPLIANCE PASS --- OVERALL STATUS PASS ===================================================""" with open(os.path.join(RUN_DIR, "audit_report.txt"), "w") as f: f.write(audit_content) notes_content = f"""Script: mini_k_sweep_run.py Date: {datetime.now(timezone.utc).strftime('%Y-%m-%d')} Solver: {SOLVER_FILENAME} Backend: CPU Observations: Mini-k sweep complete. """ with open(os.path.join(RUN_DIR, "notes.txt"), "w") as f: f.write(notes_content) with open(os.path.join(RUN_DIR, "history.csv"), "w") as f: f.write("kappa,step,t,energy,norm_psi,norm_S,plateau_radius,saturation_penetration\n") for row in history_rows: f.write(f"{row['kappa']:.2f},{row['step']},{row['t']:.4f},{row['energy']:.8f},{row['norm_psi']:.8f},{row['norm_S']:.8f},{row['plateau_radius']:.8f},{row['saturation_penetration']:.8f}\n") # ============================================================================== # DONE # ============================================================================== print("="*50) print("FILES GENERATED") print("="*50) print(f"Run: {RUN_DIR}") print("") print("Files saved to workspace:") print(" experiment_manifest.json") print(" observables.csv") print(" analysis.json") print(" audit_report.txt") print(" notes.txt") print(" history.csv") print(" raw_fields/") print("") print("📍 LOCATION: /content/run_" + TIMESTAMP + "/") print("="*50) print("SAVE COMPLETE") print("="*50) ================================================== RUN DIRECTORY ================================================== Files will be saved to: /content/run_20260625_053351 ================================================== ================================================== PARAMETER LOCK REPORT ================================================== N = 128 dx = 0.4 dt = 0.001 steps = 500 stride = 10 kappa = [0.0, 0.1, 0.2] c_S = 1.0 c_Psi = 1.0 beta = 0.5 gamma = 0.2 m2 = 0.1 s_max = 2.0 STATUS = LOCKED ================================================== ================================================== RESULTS SUMMARY ================================================== κ = 0.00 psi_norm = 70.820113 S_norm = 627.480412 energy_initial = 297.917266 energy_final = 299.004459 energy_drift = 3.649309e-03 plateau_radius = 6.175018 runtime_seconds = 9.06 -------------------------------------------------- κ = 0.10 psi_norm = 65.095419 S_norm = 624.893065 energy_initial = 310.854039 energy_final = 311.938874 energy_drift = 3.489854e-03 plateau_radius = 6.195338 runtime_seconds = 7.63 -------------------------------------------------- κ = 0.20 psi_norm = 59.700690 S_norm = 622.387704 energy_initial = 323.790812 energy_final = 324.871352 energy_drift = 3.337156e-03 plateau_radius = 6.220840 runtime_seconds = 7.15 -------------------------------------------------- ANALYSIS ================================================== candidate inversion = False psi monotonic = True S monotonic = True boundary leakage = NOT EVALUATED radial confinement = NOT EVALUATED ================================================== ================================================== FILES GENERATED ================================================== Run: /content/run_20260625_053351 Files saved to workspace: experiment_manifest.json observables.csv analysis.json audit_report.txt notes.txt history.csv raw_fields/ 📍 LOCATION: /content/run_20260625_053351/ ================================================== SAVE COMPLETE ==================================================