Ask why. Follow the evidence.
Model notes
How the simulation represents a system, and where it simplifies the science.
The simulation
Lean models wherever they’re honest, brute-force simulation only where chaos makes the outcome unknowable. The kernel can replay a whole history, nebula to lifeform, in under a minute. That’s what lets every forged system be checked, sealed and rebuilt anywhere from its seed and its command log.
Time, at four scales
The disk evolves in epochs of thousands of years. The collisions run as a Kepler-orbit encounter model over millions of years, the window that gets a full N-body integrator in the production kernel. The worlds run as coupled box models, a million years a step, for four billion years. Life runs on hazards inside open windows. The player sees all of it as turns.
Hash-chained contracts
- Seed
- WorldSeed
- PlanetSeed
- BiosphereSeed
- LineageSeed
Each stage hands the next a versioned, hashed record. The same seed and the same choices give the same system, the same lineage, the same name, on any machine. That is what makes a forged system something you keep, replay and share.
Reference checks
The production kernel is tested against the open codes the field uses as oracles: REBOUND for orbits, DustPy and TwoPopPy for dust, VPLanet for interiors and tides, ExoPlaSim for climate. Equations are ported from the papers, never the code.
| Module | Model | After |
|---|---|---|
| Cloud | Tracer parcels circularise at r = j²/GM; Jeans mass, virial ratio and mass-to-flux decide collapse, a pair, or a brown dwarf. | standard collapse theory |
| Gas disk | α-disk, viscous diffusion solved implicitly, external photoevaporation and an inner UV switch. | Lynden-Bell & Pringle 1974; Clarke 2001 |
| Dust | Two-population pebbles, fragmentation-limited Stokes number, radial drift, a cooling snow line with vapour recondensation. | Birnstiel 2012; Drążkowska 2017 |
| Planetesimals | Streaming-instability threshold on dust-to-gas ratio and Stokes number; oligarchic embryo spacing. | standard |
| Pebble accretion | Accretion efficiency from mass and Stokes number; pebble isolation mass blocks the inward flux. | Ormel & Liu 2018; Bitsch 2018 |
| Migration | Type I and II with gap inertia; resonance capture for converging pairs; giant pairs reverse. | Tanaka 2002; Walsh 2011 |
| Assembly | Kepler orbits plus Monte Carlo encounters; merge, scatter, eject; the giant instability at 60–100 Myr. | Avdellidou 2024; Izidoro 2024 |
| Interior | Parameterised convection, radiogenic heating, a regime classifier (mobile, episodic, stagnant, heat-pipe), a dynamo from core heat flux. | Foley & Smye 2018 |
| Volatiles | Outgassing, silicate and seafloor weathering, water exchange, escape limited by energy or diffusion with dynamo shielding. | Walker–Hays–Kasting |
| Climate | Energy balance with CO₂ and H₂O, ice–albedo bistability, runaway threshold, faint young Sun, XUV decline. | Goldblatt 2013; Gough 1981; Ribas 2005 |
| Impacts | Calibrated flux with an accretion tail, crater scaling, atmosphere erosion, volatile delivery, sterilisation thresholds. | Marchi 2014; Sleep & Zahnle |
| Life gate | Hazard rates for ponds, vents and sub-ice oceans; a prebiotic window that fades with age; first life sampled from the seed. | Benner 2020; Pearce 2017; Kipping 2020 |