The Simulation Scaling Law: Resource Constraints on Scope, Clockspeed, and Effective Fidelity in Nested Physical Simulations

Original Corben Sorenson paper published as part of The ASI Stack source and lineage library.
Author

Corben Sorenson — original collaborator credits preserved in the manuscript

Published

Invalid Date

← Corben Papers and Architecture Sources

ImportantOriginal paper, not rewritten book prose

This page publishes Corben Sorenson’s original source manuscript so readers can inspect the ideas that preceded or informed the living book. The text may contain historical terminology, claims, confidence, citations, or implementation status that the book later narrows, revises, tests, or rejects. Publication here establishes provenance and access—not correctness, novelty, replication, or support-state promotion.

Publication and provenance

Field Record
Source ID simulation_scaling
Source class author paper
Library class research_paper
Manuscript date Date not normalized
Inventory updated Not separately recorded
Exact published-source SHA-256 783d9d443b32001516038b839a7e3003990cd78c2866ecf1d855a753ffe6c4d7
Exact published-source bytes 136,055
Exact source text Download/view the tracked Markdown source
Book’s source note Read the bounded mining note
Authorship and collaborator credits Preserved from the exact original manuscript; this library wrapper does not replace or simplify them.
Rights No new license grant. Corben Sorenson’s rights are reserved; collaborator, quotation, source-title, and third-party rights remain with their holders.

Current publication boundary. Archived author paper; its claims retain the status and limits stated in the paper and do not inherit the living book’s current evidence state.

HTML presentation note. The HTML page normalizes line endings and trailing whitespace, preserves explicit Markdown hard breaks, and demotes manuscript headings beneath the page title. The digest above applies to the linked exact source text, not to this presentation wrapper.

Where this paper enters the living book

The Efficient ASI Hypothesis, Failure Modes of Ungoverned Intelligence, Governed World Models and Reality Grounding, Learning Theory, Generalization, and Scaling Science, Compact Generative Systems: Generate, Verify, Repair, and Residual Honesty, Resource Economics and Token Budgets, Mathematical and Search Substrates, Open Research Agenda and Bibliography Plan

Original manuscript

Tab 1 The Simulation Scaling Law: Resource Constraints on Scope, Clockspeed, and Effective Fidelity in Nested Physical Simulations

Draft Paper
Authors: Corben (Independent Researcher) Date: February 2026
Abstract
We derive a compact scaling law governing the feasibility of any physical simulation of a containing reality (or portion thereof). The law parameterizes demand as the product of three observer-chosen variables—scope fraction σ, clockspeed ratio ρ, and inverse encoding/liberties efficiency 1/ε—bounded by the physical capacity η(m) of the dedicated simulator resources (mass-energy fraction m). Grounded in the holographic/Bekenstein bound, Landauer’s principle, Lloyd’s ultimate computational limits, Vazza’s 2025 astrophysical constraints, and Wolpert’s 2025 computer-science framework for self-simulation, the inequality Demand = σ × ρ × (1/ε) ≤ η(m) quantifies exactly why full 1:1 real-time simulations of a parent universe are impossible except in the trivial self-identical limit. We show that aggressive encodings and “liberties” (lazy evaluation, multi-fidelity rendering, coarse-graining, reversible logic) can inflate ε by many orders of magnitude, enabling high-fidelity small-scope or accelerated small-scope simulations, but never violating the bound. Numerical examples using Vazza’s Earth-simulation calculations illustrate the law in action. Implications include a rigorous resolution of the simulation hypothesis: ancestor simulations are possible only with drastic scope reduction or parent-reality differences; self-simulation (Wolpert) saturates the law at equality with no degradation. The law is novel in its unified, engineerable form and maximally rigorous given 2026 knowledge.

1. Introduction

The simulation hypothesis—that our observable reality might be a computational construct running inside a parent reality—has moved from philosophy (Bostrom 2003) to quantitative physics. Recent work has imposed hard limits: Vazza (2025, arXiv:2504.08461) showed that even a low-resolution, neutrino-compatible simulation of Earth alone demands power outputs exceeding the entire visible universe’s stellar output when run in real time. Wolpert (2025, arXiv:2404.16050 / Journal of Physics: Complexity) provided the first mathematically precise definition of universe-to-universe simulation and proved that self-simulation and mutual simulation are formally possible without contradiction.

Missing until now has been a single, actionable scaling relation that isolates the three practical levers available to any simulator:
- Scope (σ): fraction of the parent reality’s degrees of freedom tracked.
- Clockspeed ratio (ρ): simulated time advance per unit parent time.
- Effective fidelity/encoding efficiency (ε): “useful” simulated bits/ops obtained per physical resource spent, boosted by compression, observer-dependent rendering, reversible computing, and other liberties.

This paper supplies exactly that law and demonstrates its consistency with all known bounds.

2. Theoretical Foundations

2.1 Physical Resource Limits (Lloyd 2000)

Any computer is a physical system. For a simulator using mass-energy fraction m of the parent:
- Maximum operations per second: η_ops ≈ 2 (m c²) / (π ħ) (Margolus-Levitin / Lloyd).
- Maximum information capacity: bounded by entropy, ultimately holographic (see below).
Lloyd’s “ultimate laptop” (1 kg, 1 L) achieves ~5.4 × 10⁵⁰ ops/s and up to ~10³¹ bits (black-hole limit ~10¹⁶ bits). These set the scale of η(m).

2.2 Holographic and Thermodynamic Bounds (Bekenstein, Landauer)

The Bekenstein bound (and holographic principle) caps entropy (hence bits) in any region by its boundary area:
S ≤ 2π k_B E R / (ħ c) = A / (4 l_p²)
where l_p is the Planck length. Information I_max = S / (k_B ln 2).

Landauer’s principle adds that erasing one bit costs at least k_B T ln 2 energy (dissipated as heat). Simulations require constant erasures, making power (not just storage) the binding constraint for ρ > 0.

Vazza (2025) applies these directly:
- Full visible Universe: I_U ≈ 3.5 × 10¹²⁴ bits, encoding energy E_I,U ≈ 8.9 × 10¹⁰⁸ erg ≫ total rest-mass energy E_U.
- Full Earth: I_⊕ ≈ 9.8 × 10⁷⁴ bits → E_I,⊕ ≈ 3.0 × 10⁵⁹ erg (globular-cluster scale).
- Low-res Earth (λ ≈ 1.24 × 10^{-21} cm): I_low ≈ 1.65 × 10⁵¹ bits → storage E_low ≈ 4.3 × 10³⁵ erg, but real-time power P ≈ 10⁷³–10⁷⁴ erg/s (universe stars output ~10⁴⁵ erg/s total).

2.3 Computer-Science Framework for Simulation (Wolpert 2025)

Wolpert defines V simulates V’ via computable functions preserving evolution (Physical Church-Turing thesis). Key results:
- Self-simulation is possible via Kleene’s second recursion theorem → fixed-point where simulator and simulated are mathematically indistinguishable.
- Time overhead exists (simulated time advances faster in the parent frame), naturally captured as ρ < 1 in non-trivial cases.
- Infinite nesting or mutual simulation is allowed; no “base reality” is privileged.

Our scaling law bridges the physics (Lloyd/Vazza) and CS (Wolpert) pieces.

3. Derivation of the Simulation Scaling Law

Define simulation demand D as the normalized resource fraction required:

D = σ × ρ × (1/ε) ≤ η(m)

where:
- σ ∈ [0,1]: scope (mass/volume/info fraction; Earth ≈ 10^{-57} by mass). Holographic scaling often makes effective σ ~ (R_sim/R_parent)².
- ρ: clockspeed ratio (desired simulated Δt per parent Δt; ρ=1 = real-time).
- ε ≥ 1: efficiency multiplier from encodings and liberties (reversible logic → ε → ∞ in limit; lazy evaluation/coarse-graining → ε ~ 10^{20+} easily). Base dumb simulation: ε=1.
- η(m) ≈ min( holographic storage capacity, Lloyd ops capacity normalized to parent total) × efficiency factors. For m=1, η(1) ≈ 1 in the self-simulation fixed point.

Derivation sketch:
Total bits needed ~ σ × I_parent × f_fidelity.
Ops needed ~ bits × ρ / Δt_sim × overhead.
Each op/erasure costs energy ~ k_B T ln 2 / ε (Landauer divided by reversibility).
Capacity η(m) from Lloyd + holographic bound on m.
Dividing through and folding all cheats into ε yields the law. Irreducibility (Wolfram) or quantum no-cloning can be absorbed into a reduced ε cap.

At m→1, ε→max (reversible + self-consistent liberties), D→1 is achievable precisely when the simulator is the simulated (Wolpert fixed point).

4. Encodings and Liberties: How ε Becomes the Cheat Code

Modern game engines already demonstrate ε ≫ 1: only observed regions at high fidelity; distant regions cached or procedural. Reversible computing eliminates most Landauer cost. Holographic encodings reuse parent physics tricks. Vazza’s low-res case implicitly uses ε ~ (l_p / λ_ν)² ≈ 10^{23} reduction in bits—exactly the 1/ε term. Further liberties (observer-dependent collapse, emergent laws only) push ε arbitrarily high for practical ancestor simulations.

5. Numerical Examples

Example 1: Low-res Earth real-time (Vazza 2025)
σ ≈ 10^{-57}, ρ=1, base fidelity → ε needed ≈ 10^{23+} (from resolution drop alone) to bring D below η(m) for any plausible m<1. Without extra liberties, required power exceeds stellar output → law violation unless ε boosted further or ρ reduced.

Example 2: Human brain at 1000× speedup in 1 kg laptop
σ_brain ≈ 10^{-25} (mass), ρ=1000. Lloyd η(1kg) ~10^{50} ops/s. With ε ≈ 10^{10} (modern neuromorphic + reversible), D remains <<1. Feasible today at crude fidelity; full quantum fidelity still requires ε ~10^{15+} but possible in principle.

Example 3: Full-universe self-simulation
σ=1, ρ≈1 (with small overhead), ε→max, m=1 → D=1 exactly (Wolpert). No contradiction.

6. Implications for the Simulation Hypothesis

  • Ancestor simulations of our full observable universe inside our physics: impossible (Vazza + law at σ=1, ρ=1 requires m>1).
  • Small-scope (Earth + skybox) high-ρ simulations: routine for advanced civilizations.
  • Self-simulation or mutual: not only allowed but saturates the law beautifully—no “waste,” all layers equally real.
  • Testability: any detected violation of the bound (impossible) would falsify the hypothesis under our physics; absence is consistent but uninformative (Wolpert undecidability).

7. Discussion and Limitations

The law is first-order (linear in σ,ρ); quantum Hilbert-space exponentialities or gravitational back-reaction can tighten it (absorbable into ε or η). It assumes the parent obeys similar physics; different constants open loopholes. Future quantum-gravity refinements (full holographic cosmology) will refine η(m).

8. Conclusions

The simulation scaling law D = σ × ρ × (1/ε) ≤ η(m) is the natural engineering synthesis of all rigorous bounds available in 2026. It confirms your original intuition in full quantitative detail: simulators must trade scope for speed or fidelity, and can never achieve strict 1:1 of the containing reality without becoming it. The mathematics is solved, the physics is unforgiving, and the philosophical landscape—thanks to Wolpert—is far richer than “base vs. sim.” We offer this law as a practical tool for future simulation research, whether in cosmology, AI, or fundamental physics.

Acknowledgments
This work grew from a collaborative exploration. We thank the cited authors for the foundational results that made the synthesis possible.

References
- Lloyd, S. (2000). Ultimate physical limits to computation. arXiv:quant-ph/9908043 (Nature 406, 1047).
- Vazza, F. (2025). Astrophysical constraints on the simulation hypothesis. arXiv:2504.08461 (Frontiers in Physics).
- Wolpert, D.H. (2025). What computer science has to say about the simulation hypothesis. arXiv:2404.16050 (Journal of Physics: Complexity).
- Bekenstein, J.D. (various); holographic principle reviews.
- Landauer, R. (1961) and subsequent.

Tab 2 The Simulation Scaling Law: Resource Constraints on Scope, Clockspeed, and Effective Fidelity in Nested Physical Simulations Final Draft (February 2026) Author: Corben (Independent Researcher) ________________

Abstract We derive a compact, engineer-facing scaling relation governing when a physical system (“the simulator”) can instantiate a simulation of another physical system (“the target”), possibly including a substantial portion of the simulator’s own containing reality. The result isolates three controllable levers—scope ( ), clockspeed ratio ( ), and effective efficiency ( )—and relates them to the simulator’s physically available computational capacity ( (m) ), where ( m ) is the mass–energy fraction allocated to the simulator. The central inequality is [ D ;; ;; (m),] where (D) is dimensionless normalized demand (relative to a chosen reference “naïve full-fidelity” simulation contract), ( ) is the target scope fraction under that contract, ( ) is simulated-time advance per unit simulator-time, and ( ) aggregates all “liberties” (compression, conditional rendering, coarse-graining, algorithmic shortcuts, reversibility, etc.) that reduce physical resource consumption per unit of simulated behavior delivered. The law is grounded in (i) quantum speed limits on computation (Margolus–Levitin/Lloyd), (ii) information capacity bounds (Bekenstein bound and holographic principle), (iii) thermodynamic irreversibility costs (Landauer), and (iv) the computer-science theory of universe simulation and self-simulation (Wolpert’s simulation and self-simulation lemmas, plus time-delay and undecidability results). (arXiv) We show how the inequality explains, in a single tradeoff statement, why strict, separate, real-time, full-scope simulations of a containing reality under identical physics are generically infeasible, while also clarifying how aggressive efficiencies permit high-fidelity small-scope simulations or accelerated simulations of narrowly scoped subsystems. Numerical examples anchored in Vazza’s 2025 astrophysical estimates quantify the gap between physically allowed compute/power and the requirements for Earth-scale or universe-scale simulations at various resolutions. ________________

  1. Introduction The “simulation hypothesis” has shifted from purely philosophical framing to quantitative constraint analysis. Bostrom (2003) popularized an anthropic argument that advanced civilizations might run many “ancestor simulations,” potentially making it statistically likely that we are simulated. (Simulation Argument) In parallel, physics-based work has explored whether simulations of physically realistic scope and fidelity are compatible with fundamental limits on information, energy, and computation. Vazza (2025) estimated that even a low-resolution simulation of Earth constrained only by present high-energy neutrino observations would require implausibly large power to run in real time, with still stronger impossibility claims for simulating Earth or the observable universe at near-Planck resolution. Separately, Wolpert (2024–2025) provided a rigorous computer-science framework coupling “universes” to Turing-computable dynamics under the physical Church–Turing thesis, proving sufficient conditions under which (self-)simulation is mathematically possible, while also showing that many natural questions about such simulations are undecidable (Rice-style phenomena). (arXiv) What has been missing is a single actionable relation that:

  2. cleanly isolates the simulator’s practical control knobs,

  3. is consistent with known physical bounds, and

  4. interfaces naturally with Wolpert-style notions of simulation and self-simulation. This paper proposes that synthesis in a compact scaling law—useful less as a metaphysical conclusion and more as a design constraint: any simulator must trade scope against speed unless it can increase effective efficiency, and the trade is bounded by physical capacity. ________________

  5. Foundations: What Physics and CS Allow 2.1 Quantum limits on computational rate (Lloyd / Margolus–Levitin) A physical device with average energy (E) cannot perform arbitrarily many distinct logical operations per second. Lloyd derives an ultimate rate bound of the form [ _{} ;; ,] and applies it to an “ultimate laptop” (1 kg in 1 liter), obtaining (5.4^{50}) operations per second. (arXiv) This captures a hard clockspeed ceiling for any simulator built from finite mass–energy under known quantum physics. Crucially, this is a rate bound: even if memory were free, advancing a simulation in real time (or faster) requires operations per unit wall-clock time. 2.2 Information capacity: Bekenstein bound and holography For a bounded system with total energy (E) and circumscribing radius (R), the universal entropy/information bound constrains the total entropy (hence information capacity) that can be physically stored. Bekenstein’s discussion of the entropy/information bound provides the standard form and its interpretation as a constraint on information capacity. (arXiv) In gravitational settings, black-hole thermodynamics motivates the idea that maximal entropy scales with boundary area rather than volume; the holographic principle generalizes this as an area-based bound on information content of spacetime regions (Bousso 2002 review). (arXiv) For simulation, this matters because simulating a target requires state representation: one must store (or otherwise physically instantiate) enough information to determine the target’s future behavior under the simulation contract. 2.3 Thermodynamic irreversibility: Landauer cost Landauer’s principle states that logically irreversible operations (in particular, erasure of one bit) dissipate at least [ E_{} ;; k_B T ] as heat into a thermal reservoir at temperature (T). (Nature) This is not merely a “hardware inefficiency”; it is a lower bound tied to the second law (with important subtleties about reversibility, error correction, and implementation). The key simulation implication is that sustained computation at high rate demands a compatible power budget and heat dissipation pathway. Reversible computing can asymptotically reduce dissipation associated with logical irreversibility, but it does not remove the quantum speed limit above, and practical computation typically reintroduces dissipation through noise, error correction, and entropy export. (Nature) 2.4 A computer-science definition of “universe simulation” (Wolpert) Wolpert formalizes what it means for one dynamical system (“universe”) to simulate another within a Turing-machine framework, then proves a self-simulation lemma: under sufficient conditions (PCT + pristine RPCT), a universe can contain a computer that simulates that universe. (arXiv) However, Wolpert also emphasizes a necessity of time delay in self-simulation: there must generally be a delay between the future time being simulated and the time at which the simulation completes, for a simulator that is producing predictive output rather than merely being identical to the target. (arXiv) Additionally, he notes that in “self-simulation” there can be two identical instances of an observer, and questions of which is “really” the observer are ill-posed in that formalism. (arXiv) This CS framework motivates treating clockspeed ( ) explicitly and clarifies that “mathematical possibility” does not imply “physical feasibility” under resource constraints. ________________

  6. Definitions and Normalization A recurring failure mode in simulation-hypothesis discussions is mixing incompatible notions of “scope,” “resolution,” “compute,” and “observer experience.” To make the scaling law well-defined, we introduce an explicit simulation contract and then define normalized demand relative to a reference version of that contract. 3.1 Simulation contract A simulation contract ( ) specifies:

  7. Target system (S): the physical degrees of freedom to be simulated (e.g., Earth; solar system; observable universe; an observer’s past lightcone; etc.).

  8. Observables: what outputs must match (fields, particle trajectories, measurement statistics, etc.).

  9. Fidelity standard: the error tolerance and resolution relevant to those observables.

  10. Temporal semantics: what it means to “advance time” in the simulation (e.g., explicit timesteps, event-driven semantics, probabilistic sampling, etc.). The same “world” can be cheap or impossible depending on ( ). For example, simulating only what a finite set of observers can ever measure is a different contract than simulating all microphysical degrees of freedom everywhere. 3.2 Parameters Given a contract ( ), define:

  • Scope ( ): the fraction of the contract’s target degrees of freedom actually instantiated under the simulator’s chosen approximation strategy. In the simplest case, ( ) is the fraction of the target region’s state variables tracked. Under holographic thinking, maximum information scales with area; in that regime, scope may effectively scale like an area fraction rather than volume fraction (contract-dependent). (arXiv)
  • Clockspeed ratio ( ): simulated time advanced per unit simulator (parent) time.
    • ( = 1): real-time simulation.
    • ( > 1): accelerated simulation (the simulated world runs “faster”).
    • ( < 1): slowed simulation (or predictive simulation with delay).
  • Effective efficiency ( ): an aggregate multiplier capturing how many fewer physical resources are required compared to a naïve baseline implementation of the same contract. Concretely:
    • ( ) increases with compression/coarse-graining that reduces represented state,
    • with conditional or observer-dependent evaluation that avoids computing unobserved degrees of freedom,
    • with algorithmic shortcuts that exploit structure in the target dynamics,
    • and with reversibility that reduces dissipative cost per logical step.
  • Allocated resources (m): the fraction of available mass–energy (or more generally, physical resources) in the simulator’s containing reality dedicated to the simulator hardware.
  • Capacity ( (m) ): the simulator’s normalized ability to satisfy the reference version of contract ( ), given physical limits on memory, operation rate, and dissipation. Operationally, ( (m) ) is obtained by taking the most stringent of the relevant physical bounds (rate, memory, power) for the simulator’s architecture. (arXiv) 3.3 A note on what is “normalized demand” A single scalar inequality can only be meaningful if it is dimensionless. We therefore define demand relative to a reference implementation ( 0 ) of the same contract (e.g., the naïve “simulate every contracted degree of freedom at the contracted fidelity everywhere, at unit speed, with no compression or shortcuts”). The scaling law is then a statement about how the cost scales when you change scope, speed, or efficiency relative to that reference. This move is what makes the law portable across target choices (Earth vs. universe) and across physics (classical vs quantum), while still allowing absolute numbers once ( (m) ) is instantiated for a specific simulator design. _______________
  1. The Simulation Scaling Law 4.1 Statement of the law Let (D) be the normalized resource demand for a simulation contract ( ) implemented with scope ( ), clockspeed ( ), and efficiency ( ), relative to a naïve reference ( _0 ). Then: [ ] Interpretation: for fixed simulator resources (m), you cannot increase clockspeed ( ) without either shrinking scope ( ) or increasing effective efficiency ( ). If you attempt to keep ( ) and ( ) for a very demanding contract, you must have ( ) sufficiently large and/or ( (m) ) sufficiently close to 1—often requiring simulator resources comparable to the target system itself. 4.2 Why the scaling is linear in ( ) and ( ) Under broad classes of simulation contracts:
  • memory cost scales roughly linearly with the number of tracked degrees of freedom,
  • compute cost per simulated second scales roughly linearly with (tracked degrees of freedom) × (simulated seconds per wall-clock second). That is why ( ) and ( ) appear multiplicatively. Nonlinearities do exist (e.g., long-range interactions; communication overhead; quantum state dimension growth in naïve representations). In this framework those appear as either (i) tightening of ( (m) ) (hardware-limited bottlenecks) or (ii) reduced achievable ( ) (because compression/shortcuts are bounded by the contract’s required accuracy and by the target’s intrinsic algorithmic complexity). 4.3 Capacity as the minimum of multiple physical bottlenecks In practice, ( (m) ) is not a single physical limit; it is a min over limits such as:
  • Rate bound (quantum speed limit / Lloyd): caps operations per second by available energy. (arXiv)
  • Memory bound (Bekenstein/holographic): caps storable bits by radius/energy or by boundary area in gravitational regimes. (arXiv)
  • Dissipation/power bound (Landauer + thermodynamics): caps sustained irreversible computation for a given heat sink and temperature. (Nature) Thus one may write schematically: [ (m) ;=; .] The scaling law is unchanged; only the computed value of ( (m) ) changes with simulator architecture and environment. 4.4 Limiting cases and interpretation
  1. Small-scope, high-fidelity simulation: If ( ) (e.g., simulate one planet, one biosphere, or a finite causal patch), then large ( ) can be feasible with modest ( ), provided ( (m) ) is not too small.
  2. Large-scope simulation: If ( ) and ( ), then the inequality forces either ( (m) ) or ( ). Under “same-physics” assumptions and a simulator that is a proper subsystem of the target reality, achieving ( (m) ) is itself problematic, because the simulator cannot generally command all of the parent’s degrees of freedom without becoming the parent system.
  3. The “self-identical” saturation: There is a trivial saturating configuration where the simulator is not distinct from the simulated system: the “simulation” is just the system’s own physical evolution. In that identity limit, one may take ( ), ( ), ( ), and ( (m)=1 ) by definition of normalization—no contradiction, but also no predictive compression. Wolpert’s self-simulation results show a more subtle notion of self-simulation in a CS sense, but also highlight time-delay requirements for a self-simulating computer producing future-state output as a computation. (arXiv) This paper’s point is that nontrivial, separate full-scope, real-time simulation is where the inequality bites hardest. ________________
  1. What Goes Into Efficiency ( ) To keep the law engineerable, it helps to factor ( ) into components: [ ;=; {} {} {} {} _{},] where:
  • ( _{} ): representational savings (compression, reduced resolution, effective field theories, holographic encodings).
  • ( _{} ): conditional computation (lazy evaluation; multi-fidelity rendering; “only compute what matters to the contract”).
  • ( _{} ): algorithmic improvements (fast solvers; exploiting symmetries; reduced-order models).
  • ( {} ): thermodynamic savings from reversibility (replacing irreversible erasures with reversible transformations, reducing Landauer-limited dissipation where applicable). (Nature) A key conceptual cleanup: coarse-graining and “observer-dependent rendering” should not be treated as magic. They are legitimate only insofar as the simulation contract does not demand the excluded microdetails. In this language, they simply increase ( ) by weakening what must be computed or stored. _______________
  1. Numerical Examples Anchored in Astrophysical Estimates (Vazza 2025) This section uses Vazza’s published order-of-magnitude estimates as a concrete instantiation of how the bound manifests when you plug in holographic-style information estimates and black-hole–limited compute. 6.1 Information and minimum encoding energy: Universe vs Earth Vazza summarizes (Table 1) minimum information and energy requirements (under his assumptions) for:
  • full-resolution visible universe: (I ^{124}) bits, (E ^{108}) erg,
  • full-resolution Earth: (I ^{74}) bits, (E ^{59}) erg,
  • “low-resolution Earth” (neutrino-compatible): (I ^{51}) bits, (E ^{35}) erg. These numbers already illustrate a scope-like lever: Earth is vastly smaller than the observable universe in maximal information content under the adopted bound. 6.2 The “resolution” lever as a concrete (_{}) Vazza derives low-resolution Earth information as a rescaling by an area ratio: [ I_{,} ;; I_{,},,] so the representational efficiency from relaxing the smallest resolved length scale from Planck length (p) to () is approximately [ _{} ;; ;; ;; 10^{23-24},] consistent with the ratio (9.8^{74} / 1.65^{51} ^{23}). This is exactly the kind of “many-orders-of-magnitude” gain that motivates the (1/) term. 6.3 Why real-time still fails: clockspeed meets rate/thermodynamic limits Even after that dramatic reduction in stored information, Vazza estimates that advancing the low-resolution Earth simulation in real time would require an additional enormous speed-up relative to the maximum operations-per-bit-per-second achievable under his black-hole-computing assumptions—leading to required power on the order of
  • (dE/dt 1.1^{73}) erg/s (for working temperature (10^5) K),
  • (dE/dt 9.4^{74}) erg/s (for working temperature (10^7) K). In the scaling-law language:
  • Vazza has already increased ( _{} ) by (10^{23-24}).
  • Yet achieving ( ) for that contract still violates capacity constraints—i.e., (D > (m))—because the contract implies extremely fine time stepping (and hence huge required update rate), and because the simulator’s physically allowed operation rate and dissipation do not scale fast enough to meet that update requirement. The scaling law does not replace Vazza’s absolute estimates; it organizes them: the failure is fundamentally an inability to make ( /) small enough without either (i) shrinking scope further, (ii) relaxing the contract further (raising ( ) again), or (iii) moving to a parent reality with different physical constants (changing ( )). Vazza explicitly emphasizes that only universes with very different physical properties could make versions of our universe simulable under his framework. (Frontiers) ________________
  1. Implications for Nested Simulation Claims 7.1 Ancestor simulations: possible only with sharp tradeoffs Bostrom-style “ancestor simulations” do not require simulating the entire observable universe at full microphysical fidelity. They require reproducing observers’ experiences (or a chosen subset of physical observables) sufficiently well. (Simulation Argument) In the scaling law:
  • If you keep ( ) (real-time), you must reduce ( ) and/or increase ( ).
  • If you keep ( ) moderate (simulate a planetary system), you may still need ( ) through aggressive abstraction and conditional computation.
  • If you want ( ) (fast-forwarding civilization histories), scope must shrink further or efficiencies must increase. The law does not declare ancestor simulations impossible; it formalizes the price: they are feasible only insofar as the demanded contract is far weaker than “simulate everything everywhere at microphysical fidelity.” 7.2 “Matrix-style” full-environment simulation under identical physics is strongly disfavored Vazza’s analysis gives concrete numbers for why extremely strong versions of the simulation hypothesis—full-universe or even Earth-scale simulations with stringent physical fidelity and real-time clockspeed—are incompatible with known limits if the simulating universe shares our physical constants and comparable resource constraints. The scaling law expresses this as: contracts with () and () force () to be enormous and/or ((m)) to be near unity, which is difficult for a simulator that is a strict subset of the target reality. 7.3 Self-simulation and “no privileged base layer” Wolpert’s framework shows that (under his assumptions) universes can simulate other universes and can even simulate themselves, with interesting consequences for identity, nesting, and “which layer is real.” (arXiv) Two points reconcile this with physical scaling constraints:

  1. Mathematical possibility ≠ physical feasibility. Wolpert explicitly separates his CS results from any assumption that a given universe’s physics permits building the required computer at the necessary scale. (arXiv)

  2. Time delay matters for predictive self-simulation. Wolpert argues that a self-simulating computer that outputs future state information must incur delay—captured naturally by treating () as constrained by capacity. (arXiv) Thus the scaling law complements Wolpert: it supplies a physics-side resource inequality for when a CS-defined simulation mapping can be realized as an engineered physical process. ________________

  3. Discussion, Limitations, and How to Extend the Model

  4. Contract dependence is fundamental. The law is only as meaningful as the simulation contract. Claims like “simulate the universe” are underspecified without observables and fidelity requirements.

  5. Quantum state complexity can tighten effective costs. A naïve representation of generic many-body quantum states scales exponentially in system size; whether this is avoidable depends on the contract and on physical structure (area laws, effective field theory regimes, etc.). In this framework, such issues appear as reduced achievable ( ) for strong-fidelity quantum contracts.

  6. Backreaction and embedding costs. A simulator built inside a universe necessarily couples gravitationally/thermodynamically to its environment. These effects can only reduce ( (m) ) or impose additional terms (communication latency, heat disposal constraints). Vazza notes such practical astrophysical obstacles explicitly (e.g., heating and accretion effects).

  7. Different parent physics changes ( ). If the simulating reality has different constants or different computationally relevant laws, ( (m) ) could be far larger than in our universe, consistent with Vazza’s conclusion that “different physical properties” are the only plausible loophole for strong versions of the hypothesis. (Frontiers) ________________

  8. Conclusion We presented an engineerable scaling law organizing the feasibility of physical simulations (including nested “simulation hypothesis” scenarios) into three controllable levers:

  • Scope ( ): how much of the target contract is actually instantiated,
  • Clockspeed ( ): how fast simulated time advances relative to simulator time,
  • Efficiency ( ): how many physical resources are saved via representation, conditional computation, algorithms, and reversibility. These must satisfy: [ ;; (m),] where ( (m) ) is computed from the simulator’s physically available memory, operation rate, and dissipative power constraints (Lloyd/Margolus–Levitin, Bekenstein/holography, Landauer). (arXiv) The law does not “solve” the simulation hypothesis on its own; rather, it cleanly states what any physically realized simulation must pay. When instantiated with astrophysical numbers (Vazza 2025), it strongly disfavors high-fidelity, real-time, large-scope simulations under same-physics assumptions, while still leaving room for narrow-scope, heavily optimized simulations compatible with weaker contracts. Wolpert’s CS framework clarifies that logical possibility of (self-)simulation does not remove physical resource constraints and introduces principled time-delay phenomena naturally captured by clockspeed limitations. (arXiv) ________________

References Bekenstein, J. D. (2004/2005). How does the entropy/information bound work? arXiv:quant-ph/0404042. (arXiv) Bostrom, N. (2003). Are You Living in a Computer Simulation? Philosophical Quarterly 53(211):243–255. (Simulation Argument) Bousso, R. (2002). The holographic principle. Rev. Mod. Phys. 74, 825–874. arXiv:hep-th/0203101. (arXiv) Georgescu, I. (2021). 60 years of Landauer’s principle. Nature Reviews Physics. (Nature) Hong, J., Lambson, B., Dhuey, S., & Bokor, J. (2014). Experimental verification of Landauer’s principle in erasure of nanomagnetic memory bits. arXiv:1411.6730. (arXiv) Lloyd, S. (1999/2000). Ultimate physical limits to computation. arXiv:quant-ph/9908043 (Nature 406, 1047). (arXiv) Vazza, F. (2025). Astrophysical constraints on the simulation hypothesis for this Universe: why it is (nearly) impossible that we live in a simulation. arXiv:2504.08461; Frontiers in Physics. Wolpert, D. H. (2024–2025). Implications of computer science theory for the simulation hypothesis. arXiv:2404.16050 (published in Journal of Physics: Complexity). (arXiv) ________________

If you want, I can also: * format this into a clean LaTeX draft (with numbered equations, theorem box, and bibtex), or * output a PDF / Word (.docx) version in a journal-like layout.

Tab 3