Tidepool — Evolution Simulator
A real-time artificial life evolution simulator where creatures with neural-network brains survive, eat, mate, mutate, and speciate across islands and open water. Design species with 46 genes across diploid chromosomes, tune the environment, and watch natural selection unfold. Every run tells a different evolutionary story as bodies and brains adapt together.
Spawn Creature
Design a species with 46 genes across body, brain, diet, and behaviour.
Environment
Control food rates, colour ratios, and scarcity for land and water.
Karyotypes
Edit chromosome structure, gene linkage, and inheritance patterns.
Data Analysis
Population graphs, gene distributions, mutation analysis, species tree, and creature inspection.
What Is Tidepool?
Tidepool is an artificial life simulator that models evolution from the ground up. You seed a population with founder creatures, set the environment, and then step back while natural selection runs in real time.
Each creature is a physics-simulated organism with a body shaped by 46 heritable genes, a small neural-network brain that controls all behaviour, and a diploid genome carried on chromosomes — just like real biology. Creatures forage for food, flee predators, find mates, reproduce, and eventually die of old age or starvation.
No two runs play out the same way. Some lineages become sleek water gliders, some evolve into armoured land tanks, some specialise on a single food colour, and many simply go extinct. The simulation combines Mendelian genetics, chromosomal crossover, Gaussian mutation, neuroevolution, physics-based movement, and environmental pressure into a living system you can observe, tune, and experiment with.
The World — Islands, Water, and Terrain
The world is a 6000 × 6000 unit top-down environment. Most of the map is open water, punctuated by several randomly generated circular islands.
Water is the default terrain. Creatures spawn and drift in water unless they have legs that allow them to move onto land. Islands are solid landmasses — water-only creatures physically bounce off their edges, while land-capable creatures can walk onto them.
Each terrain type has its own food supply, its own physics, and its own movement rules. This creates two fundamentally different ecological niches that creatures must adapt to — or specialise in — to survive.
- Water: A low-friction environment suited to streamlined bodies. Tails provide propulsion — longer tails create far more thrust. Legs create significant drag in water, making them a disadvantage. Streamlined body shapes dramatically reduce water friction.
- Land: Requires at least one leg to walk on. Legs improve land movement, while tails and shells slow a creature down on land.
- Creatures transitioning from land to water have their speed capped to prevent them launching far into the ocean.
- Islands generate their own food independently of the water food supply, creating separate foraging zones.
Creature Survival — Energy, Hunger, and Stamina
Every creature is in a constant battle to stay alive. Two resources govern survival: hunger (energy) and stamina.
Hunger acts as the creature's energy store. It starts at full capacity and depletes constantly through metabolism, movement, and brain activity. When hunger hits zero, the creature starves and dies. Eating food refills hunger. Larger, denser creatures have greater hunger capacity (more reserves), but also higher metabolic costs.
Stamina governs burst movement. Every push (thrust) costs stamina. When stamina is depleted, the creature cannot push — it can only coast. Stamina recovers over time, especially during rest. The effective stamina a creature can use is capped by both its stamina reserve and its current hunger — a starving creature has no energy to sprint.
- Hunger capacity: How much energy the creature can store. Influenced by size and density — bigger, denser creatures hold more reserves.
- Metabolic rate: The baseline energy cost of staying alive. Influenced by size, density, vision range, memory, brain capacity, and streamlining. Metabolism constantly drains hunger.
- Pregnancy doubles the metabolic rate — carrying offspring is extremely costly.
- Stamina reserve: How many bursts of movement a creature can make before tiring. Influenced by the staminaReserve gene, strength, streamlining, limbs, and density.
- Stamina recovery is faster when resting, slower when exerting, and reduced when starving.
- Creatures also die of old age. Lifespan is influenced by several genes including size, density, procreation age, parental investment, and shell defence.
Energy Rings — Reading Status at a Glance
When Energy Rings are enabled, every creature displays a status indicator. The centre dot shows the current action (colour-coded by drive). The outer ring shows hunger level — green when full, yellow at half, red when starving. The inner ring shows stamina using the same colour gradient.
Food — Colour, Diet, and the Energy Economy
Food appears as coloured circles on both land and water. The amount of energy a creature gets from food depends on the food's colour, the creature's body colour, and its digestion efficiency.
The colour matching system compares the food colour to the creature's body colour channel by channel (red, green, blue). The closer the match, the more energy the creature extracts. A perfect colour match yields maximum energy; a poor match yields very little. This creates a powerful selection pressure — creatures whose body colour closely matches the dominant food colour in their environment gain a significant survival advantage.
Diet is split into two traits that are inversely related: plant digestion efficiency and meat digestion efficiency. A creature cannot be highly efficient at both. A high plant efficiency means the creature extracts more energy from food items. A high meat efficiency opens up predation — eating other creatures.
- Energy gained from food depends on the food's size, how well the creature's body colour matches the food colour, and the creature's digestion efficiency.
- Creatures need a minimum digestion efficiency before they can eat a food type at all.
- Food is partially consumed — a creature takes only what it needs to fill its hunger, and the food shrinks proportionally.
- Default food comes in three colours: Blue (70%), Red (30%), Green (10%), on both land and water. You can change these ratios in Simulation Settings.
- Food production rate and maximum count are independently configurable for land and water.
- Water food is periodically dispersed to prevent clumping, simulating currents and drift.
Predation, Defence, and the Food Chain
Creatures with high meat digestion efficiency can eat other creatures — becoming predators. But predation is governed by size, and defence is possible through shells.
A predator can only eat a target whose defensive size is smaller than the predator's maximum prey size. Maximum prey size is influenced by the creature's body size and mouth size. Cannibalism is blocked — creatures cannot eat members of their own species or their parent species.
Shells provide passive defence by inflating a creature's effective defensive size. A thick, strong shell makes the creature appear larger to predators, keeping it off the menu. However, shells come with trade-offs: increased friction in water, increased drag on land, and higher metabolic cost. Evolution must balance the survival benefit of armour against the movement penalty.
- Defensive size: How large a creature appears to predators. Influenced by body size, shell width, and shell strength. A well-armoured creature is effectively much bigger.
- Shells also contribute to longer lifespan — creatures with thicker, stronger shells tend to live longer.
- The food chain can emerge naturally: small herbivores → medium predators → large apex predators, each constrained by size ratios.
- Predation creates a flee response in prey creatures — brains that detect and evade predators survive longer.
Mating, Reproduction, and Parental Investment
Reproduction requires finding a compatible mate of the opposite sex and the same species, both old enough and off cooldown. The process involves mate calling, physical contact, pregnancy, and energy investment.
Creatures broadcast mate calls when their brain decides to. Calls have a radius determined by the mateCallVolume gene and cost energy. Other creatures nearby with the same species and opposite sex can hear and respond to these calls. The mate-call system pairs compatible callers and creates a mutual attraction.
When two creatures with a mate goal collide, mating occurs. Both parents generate gametes (via crossover and mutation), which combine into the child's genotype. The child's brain genome is bred from both parents' brains. The female becomes pregnant, carrying the developing offspring.
- Pregnancy duration is controlled by the pregnancyDuration gene. During pregnancy, metabolic rate doubles.
- At birth, the mother pays an energy cost. If she is too starved to cover the cost, the offspring does not survive.
- Both parents invest energy in their offspring, controlled by the investmentEnergyPercentageInOffspring gene.
- Males can transfer energy to their mate as a nuptial gift, controlled by the investmentEnergyPercentageInMate gene.
- Investment time genes (investmentTimeInOffspring, investmentTimeInMate) create mating cooldowns, preventing rapid-fire reproduction.
- Higher parental investment produces longer-lived children — investing parents give their offspring a lifespan bonus.
- Creatures must reach maturity age (procreateAge gene) before first mating.
Genes — The 46 Heritable Traits
Every creature carries 46 genes organised into categories. Genes do not directly control what a creature does — instead, they work together to produce a phenotype: the set of derived physical and behavioural characteristics that determine how the creature moves, eats, sees, thinks, and reproduces.
For example, a creature's hunger capacity is not set by a single gene — it emerges from the combination of its size and density genes. Its movement cost depends on size, strength, streamlining, tail, legs, and terrain. Its lifespan is influenced by size, density, procreation age, parental investment, and shell defence. This means that changing one gene often has ripple effects across many traits, creating complex trade-offs that evolution must navigate.
Genes are grouped into categories: Sex, Body Core (size, strength, density), Body Shape (streamlining, bounce), Appendages (legs, tail), Armour (shell), Intelligence (brain capacity, memory, and cognitive specialisation), Behaviour (fear, mate urgency, exploration, risk tolerance), Endurance (stamina reserve, recovery), Movement (preferred speed), Vision (field of view distance and angle), Diet (plant/meat efficiency, prey size ratio), Colour (RGB body colour), and Breeding (maturity age, pregnancy, parental investment, mate calling).
- Size: Determines body mass and visual scale. Influences hunger capacity, metabolic rate, push force, and many other derived traits.
- Strength: How much force the creature can exert when pushing. Affects movement speed on both land and water.
- Density: Body density, shown as horizontal stripe patterns. Influences inertia, hunger capacity, and metabolic rate.
- Streamlining: Elongates the body shape. Greatly reduces water friction but sacrifices turning ability.
- Tail Length: Provides water propulsion — longer tails give far more thrust. Increases friction on land.
- Legs (front/back count and length): Required for land movement. Improve land traction but create significant drag in water.
- Shell Width and Strength: Passive armour that increases defensive size against predators. Trade-off: increased friction and metabolic cost.
- FOV Distance and Angle: Vision range and width. A narrow field of view sees much further; a wide field of view covers more area but at shorter range.
- Plant/Meat Digestion Efficiency: Determines herbivore vs carnivore specialisation. Inversely related — a creature cannot be highly efficient at both.
- Body Colour (R, G, B): Determines how well the creature's colour matches food colours, affecting energy gained from eating.
- Brain Capacity: Controls how many hidden processing units can be expressed in the brain. More units allow more complex decision-making but cost more energy.
- Foraging, Threat, Social, and Memory Cognition: These four genes do not add raw brain size. Instead, they bias how advanced brain capacity is spent, so two equally large brains can specialise in different kinds of processing.
- Memory Capacity and Retention: How many food, mate, and predator locations the creature can remember, and how long those memories last.
- Fear Sensitivity, Mate Urgency, Exploration Bias, Risk Tolerance: Personality traits that shape how the brain weighs different drives — flee vs forage, explore vs rest, caution vs boldness.
- Procreate Age: Age of sexual maturity. Higher values delay reproduction but contribute to a longer lifespan.
- Mate Call Frequency and Volume: How often the creature broadcasts mate calls and how far they travel.
Creature Anatomy — How Visible Genes Change the Render
Every visual feature of a creature is gene-driven. These cards show the genes that visibly change the static render, with multiple examples for each so you can see the range from low to high or category to category. Other genes still matter, but they change behaviour, metabolism, memory, breeding, or derived stats rather than the sprite itself.
Genetics — Diploid Chromosomes, Crossover, and Mutation
The genetics system models real diploid inheritance. Each creature has pairs of chromosomes — one from each parent. Genes sit on these chromosomes and are inherited through gamete formation with crossover and mutation.
Sex is determined by sex chromosomes: XX = female, XY = male. The Y chromosome carries the sryGene (sex determination) and reproductive investment genes. Autosomes carry the remaining genes. You can customise which genes sit on which chromosomes using the Karyotype Settings editor.
When a creature reproduces, it forms gametes (egg or sperm) by selecting one chromosome from each pair. During this process, crossover occurs: homologous chromosomes swap segments at random points, creating new gene combinations. This is a key source of genetic variation — even without mutation, crossover shuffles existing alleles into novel combinations.
Mutation adds further variation. Each allele has a chance of mutating during gamete formation. Mutations are small random changes — some genes like colour mutate in larger steps, while precision traits like digestion efficiency mutate in smaller steps. Values are always kept within each gene's valid range.
- Gene expression depends on inheritance type. 'Blend' genes average the maternal and paternal alleles. 'Dominant/recessive' genes express the higher (dominant) value.
- XY males express X-linked and Y-linked genes hemizygously (single allele, no blending) because X and Y carry different genes.
- Crossover produces a random number of swap points along each chromosome, mixing segments from both parents into new combinations.
- XX females cross over between their two X chromosomes. XY males do not — they pass an intact X or Y at random.
- The expressed genotype is then converted into a calculated phenotype — the derived traits like push force, friction, turn angle, metabolic rate, and lifespan that emerge from genes working together.
Speciation — How New Species Form
Speciation occurs automatically when a child's genes have drifted far enough from its parent species' founder genes. The simulation uses a weighted genetic distance metric to detect this.
Genetic distance is a weighted measure of how different a creature's genes are from its species' founder genes. Not all genes count equally — functional traits like diet, size, strength, density, streamlining, and shell contribute much more to speciation than cosmetic traits like colour. This means a creature can change colour significantly without speciating, but a shift in diet or body plan triggers a new species more readily.
When genetic distance exceeds the speciation threshold, the child is assigned to a new species. New species are named hierarchically: if the parent species is 'Alpha', the first branch becomes 'Alpha.1', the next 'Alpha.2', and sub-branches become 'Alpha.1.1', etc. This creates a species tree you can visualise in Data Analysis.
Freshly branched child species first appear in the in-game 'Emerging Species' section. They stay there while their living population is below 10. Once they reach 10 living members, they move into the main Species list and count as established species in the HUD.
- Speciation is not a single dramatic event — it emerges gradually as mutations accumulate and selection pushes a subpopulation away from the founder profile.
- Hybrid offspring (parents from different species) are assigned to whichever parent species is genetically closest.
- Hybrids can only mate with members of their own species or their parents' species — a reproductive barrier that reinforces speciation.
- You can watch speciation unfold in the Species Tree view, which shows branching lineages and extinct branches.
The Neural Brain — How Creatures Think
Every creature has a small neural network brain that processes sensory inputs, internal state, and memory to decide what to do. There are no hard-coded rules — all behaviour emerges from the brain's inherited and evolved connection weights.
The brain reads a wide range of inputs — hunger, stamina, what it can see, what it remembers, what's nearby, and its own body capabilities — then produces outputs that drive behaviour and movement. The brainCapacity gene controls how many processing units can be expressed. More units allow more nuanced decision-making but cost more energy to maintain.
The four behaviour drives are Forage (seek food), Flee (escape predators), Mate (find a partner), and Wander (explore). The drive with the highest score wins, setting the creature's current action. A separate Rest drive can suppress movement — when rest is strongest, the creature keeps its current action but does not push, allowing stamina to recover. Two additional outputs control turning (left/right) and one controls thrust intensity. A final output triggers mate calling.
- Internal state inputs: hunger level, energy surplus, maturity, mating readiness, pregnancy status, sex, stamina ratio, recent exertion, stamina trend.
- Body awareness inputs: on-land status, land/water capability, body size, strength, turning agility, diet efficiencies, brain metabolic cost.
- Movement inputs: coasting value, sustainable push level, recent distance moved, speed relative to preferred range, barriers ahead/left/right.
- Sensory inputs: the best-valued option among nearby visible foods, plus nearby prey, mates, and predators — each represented with proximity and angle signals.
- Memory inputs: remembered food, mates, and predators with confidence scores and positions.
- Social inputs: same-species ratio, other-species ratio, opposite-sex ratio, average neighbour speed.
- Personality genes (fear sensitivity, mate urgency, exploration bias, risk tolerance) scale specific inputs and outputs, giving each creature a temperament.
- Cognition-specialisation genes decide which advanced processors come online first as brain capacity grows. This lets one lineage evolve into strong foragers while another of similar brain size becomes better at threat detection, social behaviour, or memory.
Brain View — Reading a Creature's Decisions
The selected-creature panel includes this live network view. Inputs are on the left, internal processing units are in the middle, and behaviour outputs are on the right. Cyan links and nodes indicate signals pushing behaviour forward; orange signals are suppressing or counteracting them. In this example, hunger and visible food are strongly activating food interest, which drives forage and thrust.
Brain Inheritance and Neuroevolution
Brains are inherited and mutated alongside body genes. A child's brain is bred from both parents' neural networks, creating heritable behaviour that evolves over generations.
During reproduction, each connection weight in the child's brain is randomly copied from one parent or the other. Small mutations are then applied — most are subtle tweaks, but occasionally a larger change occurs. The number of active processing units can also shift slightly between generations, allowing brain complexity to evolve up or down.
New species start with a hand-crafted seed brain that provides baseline survival behaviours — basic predator avoidance, food seeking, mate approach, and wandering. Evolution quickly modifies this starting point. Over many generations, brains can develop specialised strategies: efficient foraging patterns, effective predator evasion, optimal pacing between rest and movement, or aggressive mate-seeking behaviour.
- More hidden neurons allow more complex decision-making but increase metabolic cost — there is a trade-off between intelligence and energy efficiency.
- Brain behaviour is not pre-programmed per species. Two creatures of the same species may behave differently based on inherited brain weights.
- The hidden layer has 20 named conceptual neurons split into two tiers: 10 primitive and 10 advanced.
- Primitive neurons are always active regardless of brain capacity. They provide the core behaviours every creature needs: food interest, thrust control, wandering, steering toward food, mate interest, steering toward mates, and world-edge avoidance.
- Advanced neurons are grouped into foraging, threat, social, and memory specialisations. Brain capacity determines how many advanced units can be expressed, while the four cognition genes bias which groups are activated first.
- This means two creatures can both evolve larger brains without becoming smart in exactly the same way. One lineage might prioritise food-value evaluation and foraging, while another emphasises threat assessment, mate-calling strategy, or memory.
- A creature with minimal brain capacity can forage, wander, seek mates, steer toward food and mates, and avoid barriers — but cannot detect or flee predators, remember locations, or make nuanced risk decisions.
- Because brains and body genes co-evolve, a lineage that evolves better vision may simultaneously evolve brain wiring that better exploits that vision.
Creature Behaviour — The Four Drives
At any moment, each creature is in one of four behavioural states. The state is determined by the brain's strongest output drive. A separate rest drive can suppress pushing without becoming its own visible action state.
Creature Memory
Creatures can remember the locations of food, mates, and predators they have seen. Memory capacity and retention are gene-controlled traits.
Each memory slot stores a type (food, mate, or predator), a position, and a priority score based on importance, recency, and how many times the item was seen. Predator memories are prioritised highest, followed by mates and then food. Memories decay over time and are eventually forgotten.
Memory feeds directly into the brain as inputs: remembered food/mate/predator positions and confidence values. A creature with high memory capacity and retention can navigate back to productive food patches, recall where mates were last seen, and remember dangerous areas — all conferring survival advantages.
- Memory capacity: How many things the creature can remember at once. Determined by the memoryCapacity gene.
- Memory retention: How long memories last before fading. Determined by the memoryRetention gene.
- Memory has a metabolic cost — larger memory uses more energy, creating a trade-off between awareness and efficiency.
- Dead items (eaten food, dead creatures) are automatically purged from memory.
What to Expect — How Evolution Unfolds Over Time
Each simulation run produces different outcomes, but certain evolutionary patterns tend to emerge depending on the environment.
In the early generations, the population typically crashes as poorly adapted creatures starve or fail to reproduce. Survivors carry the genes and brain wiring that happened to work in the initial environment. Over time, you can expect to see several common evolutionary trajectories:
- Movement efficiency: Creatures evolve toward lower movement cost — streamlining increases to reduce water friction, preferred speed settles near the energetically sustainable push level, and stamina management improves.
- Food colour specialisation: If one food colour dominates (e.g., 70% blue), creature body colour tends to drift toward that colour over generations, because colour-matched creatures extract more energy per food item. Changing food ratios mid-simulation can cause rapid adaptation or mass extinction.
- Land vs water specialisation: Leg genes determine habitat viability. Creatures that develop legs can exploit island food. Pure water creatures evolve tails and streamlining. Rarely, generalist mixed-terrain species emerge but with trade-offs in both environments.
- Predator–prey arms race: As predators evolve, prey species face pressure to develop defences — larger body size, thicker shells, higher fear sensitivity, faster flee responses, and better predator memory. Predators in turn evolve faster pursuit, better vision, and lower risk tolerance.
- Shell evolution: When predation pressure is high, shellWidth and shellStrength tend to increase, making creatures harder to eat. But heavy shells reduce agility and increase metabolic cost, so shell evolution balances defence against mobility.
- Brain sophistication: Brain capacity may increase over time as smarter creatures make better foraging and predator-avoidance decisions. But brain neurons cost metabolic energy, so overly complex brains are penalised in lean environments.
- Reproductive strategy divergence: Some lineages evolve high parental investment (fewer, longer-lived offspring), while others evolve low investment (many short-lived offspring). Environmental stability tends to favour investment; chaotic environments favour quantity.
- Food scarcity adaptations: When food is scarce, evolution favours energy-efficient movement, lower metabolic rates, smaller body sizes, better stamina recovery, and wider foraging vision over raw speed.
- Speciation events: As subpopulations adapt to different niches (land vs water, different food colours, different predation strategies), genetic distance accumulates until new species branch off. The Species Tree visualises this branching.
How to Add a New Creature
The Spawn Creature panel lets you design a founder species by setting starting values for all 46 genes.
- Click Spawn Creature in the controls bar. The side panel opens with gene editors grouped by category.
- Set gene values using sliders or number inputs. Each gene shows its valid range and a description of what it affects.
- Genes are organised into categories: Body Core, Body Shape, Appendages, Armour, Intelligence, Behaviour, Endurance, Movement, Vision, Diet, Colour, and Breeding.
- When you confirm, the simulation creates a small founding population with these gene values, placed in the world.
- Founder genes become the reference point for the species. As descendants mutate and diverge, the simulation compares them to these founder genes to detect speciation.
- You can add multiple species in the same simulation to create competition, predator–prey dynamics, or niche separation.
How to Adjust Simulation Settings
Simulation Settings let you control the food supply independently for land and water, while also exposing a small set of genetics and performance controls.
- Open Simulation Settings from the controls bar.
- The panel is split into Environment, Genetics, and Performance sections. Genetics appears under a Mutations subheading.
- For each terrain (land and water), you can adjust food production rate, maximum food count, and colour distribution (the percentage of green, red, and blue food). This lets you make one habitat abundant and another harsh, or pressure creatures in land and water differently.
- Production Rate controls how quickly new food appears. Increase it to support larger populations and faster recovery after die-offs. Decrease it to create scarcity and stronger competition.
- Max Food controls how much food can exist at once. Raising it increases carrying capacity. Lowering it makes shortages happen sooner, even if production rate is high.
- Colour Ratios change which body colours are rewarded. If red food becomes dominant, creatures that drift toward red body colour gain more energy and are more likely to survive and reproduce.
- Genetics exposes Mutation Chance and Mutation Size Multiplier. For now these affect body genes only, not brain mutation. Mutation Chance controls how often inherited body traits mutate. Mutation Size Multiplier controls how large successful body mutations are. These affect future births only, so they change the pace of evolution rather than modifying creatures already alive.
- Performance exposes AI Decision Interval and Creature Maintenance Interval. Raising these reduces how often some systems run, which can improve performance in large simulations, but very high values can make behaviour less responsive.
- All changes stay local until you click Apply at the bottom of the panel. Cancel restores the last applied settings.
- Each section also has a Reset to Defaults button. That only resets the local draft for that section until you click Apply.
- Reducing food rate or maximum count creates scarcity. This tends to favour energy-efficient creatures, strong foragers, and species that can survive lean periods.
- Changing colour ratios shifts which body colours are advantageous. A sudden shift from 70% blue to 70% red food will pressure creatures to evolve red body colour over generations.
- Setting land food rate to zero but keeping water food rate high creates a water-only ecosystem (and vice versa).
- Extreme settings can cause extinction events — intentional or accidental. These can be interesting evolutionary experiments.
Simulation Controls
The toolbar provides controls for managing the simulation's speed, camera, and visual display.
- Start / Restart: Launches a new simulation or restarts the current one.
- Pause / Resume: Freezes all simulation logic. The simulation also auto-pauses when you switch browser tabs.
- Speed Toggle: Changes simulation speed (1×, 2×, etc.). Higher speeds run the physics and AI faster for long experiments.
- Zoom +/−: Adjusts the camera zoom from 0.25× to 2.0×. You can also zoom with Ctrl/Cmd + scroll wheel.
- Toggle FOV (Field of View): Shows or hides the vision cones for all creatures. Useful for understanding what each creature can see, but performance-heavy with large populations.
- Toggle Energy Rings: Shows or hides the energy/stamina ring display around each creature. The centre dot shows current action (colour-coded: red = flee, orange = forage, green = mate, yellow = wander, blue = rest, purple = pregnant). The outer ring shows hunger level (green = full, red = starving). The inner ring shows stamina level (green = full, red = depleted). Both rings use a heat-colour gradient from red through yellow to green. Disabling this improves performance with large populations.
- Click a creature: Opens the Selected Creature Panel showing full inspection data — all genes (maternal, paternal, expressed), current drives, brain stats, memory, children, calculated phenotype values, and more.
Graphs, Stats, and Data Analysis
The Data Analysis modal provides several views for tracking evolutionary trends and population dynamics.
- Population Graph: Shows population count over time, broken down by species. Watch for growth, decline, and extinction events.
- Gene Distribution Graph: Displays the spread of any gene's values across the current population. See whether a gene is converging (strong selection) or broadly distributed (weak selection).
- Gene Correlation Heatmap: Shows correlations between pairs of genes across the population. Reveals linked traits — e.g., do larger creatures also tend to be denser?
- Species Gene Profile Graph: Compares average gene values across different species. Useful for seeing how species have diverged.
- Mutation Analysis: Summarises which genes are mutating most often, in which direction, and by how much. Useful for spotting where variation is entering the population and whether change is balanced or biased.
- Species Tree: A phylogenetic tree showing how species have branched over time. Extinct branches are visible, telling the story of failed lineages.
- Selected Creature Panel: Click any creature to inspect its full genetic profile, brain state, current drives, children, parentage, and all calculated phenotype values. You can also view the creature's thought process in real time through the live brain visualisation.
- Population snapshots are recorded every 30 seconds of game time, building up a historical record for graph visualisation.
Karyotype Settings — Chromosome Structure
The Karyotype Settings editor lets you control the chromosome structure — which genes sit on which chromosomes and how they are linked during inheritance.
By default, sex chromosomes carry sex-determination and reproductive investment genes, while a single autosome carries all other genes. X-linked genes can be inherited by both sexes, while Y-linked genes pass only through males. This means moving a gene onto the X or Y chromosome changes not just linkage, but which sex can inherit and express it. You can redistribute genes across multiple autosomes to change linkage groups. Genes on the same chromosome tend to be inherited together (unless crossover separates them), while genes on different chromosomes assort independently.
Changing the karyotype affects the rate at which gene combinations are broken up or preserved across generations. More chromosomes means more independent assortment; fewer means tighter linkage. This is a powerful tool for advanced experiments in inheritance dynamics.
Getting Started — Your First Experiment
The easiest way to enjoy the simulator is to treat each run like an evolutionary experiment. Here is a suggested first session:
- 1. Click Start Simulation to launch the world.
- 2. Click Spawn Creature to create your first species. Try the default gene values or adjust a few — body size, colour, and diet efficiency are good starting points.
- 3. Watch the founding population for a minute. Are they finding food? Are they surviving?
- 4. Open Simulation Settings and experiment: try reducing food production to see how scarcity changes behaviour, or shift food colours to pressure colour adaptation. Once your species are thriving, lower food production gradually to keep selection pressure on increasingly efficient eaters and movers.
- 5. Add a second species with different genes to compare. Make one a water specialist (high tail, no legs, streamlined) and another a land specialist (legs, low streamlining).
- 6. Open Data Analysis to watch population graphs, gene distributions, and the species tree.
- 7. Toggle FOV visibility to see what creatures are perceiving. Toggle Energy Rings to monitor health at a glance.
- 8. Click individual creatures to inspect their genes, brain drives, memory, and family tree.
- 9. Speed up the simulation (2×) for longer experiments and watch speciation events emerge in the species tree.
- 10. Evolution takes time. Let the simulation run, then come back 30 minutes later to see how the population, species tree, and gene distributions have changed.
- 11. Try creating a predator species (high meat efficiency, large size) and see how prey species adapt — do they evolve shells, faster flee responses, or simply go extinct?
- 12. Keep in mind that the simulation stops running when your screen locks. If you want to leave it running for a while, increase your computer or mobile device's screen-lock delay first.
Glossary
Key terms used in the simulation.
- Allele: A specific value of a gene on one chromosome. Each creature carries two alleles per gene (one from each parent).
- Autosome: A non-sex chromosome. Carries most body, behaviour, and survival genes.
- Calculated Phenotype: The set of derived traits (push force, friction, metabolic rate, lifespan, hunger capacity, etc.) that emerge from genes working together. No single gene controls these — they are produced by combinations of multiple genes.
- Crossover: The exchange of gene segments between homologous chromosomes during gamete formation. Creates new combinations of existing alleles.
- Diploid: Having two copies of each chromosome — one from each parent. All creatures in this simulation are diploid.
- Dominant/Recessive: An inheritance mode where the higher allele value is expressed (dominant wins).
- Blend Inheritance: An inheritance mode where the two alleles are averaged to produce the expressed value.
- Expressed Gene: The final gene value after combining maternal and paternal alleles via the inheritance rule.
- Founder Genes: The starting gene values used when creating a new species via Spawn Creature. Speciation is measured relative to these values.
- FOV (Field of View): The angular width and distance of a creature's vision cone. Controls what the creature can see.
- Gamete: A haploid cell (one chromosome from each pair) produced during reproduction. Egg or sperm equivalent.
- Genetic Distance: A weighted measure of how different a creature's genes are from its species' founder genes. Used to trigger speciation.
- Genotype: The creature's full set of chromosome pairs carrying all alleles.
- Hemizygous: A gene present on only one chromosome (e.g., X-linked genes in XY males). Expressed directly without blending.
- Homologous Pair: Two copies of the same chromosome, one from each parent.
- Hunger: The creature's energy store. Depletes via metabolism and movement. Refilled by eating. Zero = starvation death.
- Karyotype: The chromosome structure defining how genes are organised and linked.
- Metabolic Rate: The baseline energy cost of staying alive. Influenced by body size, density, vision, memory, brain capacity, and streamlining.
- Mutation: A random change to an allele value during gamete formation. Each gene has its own mutation scale — some traits change in larger steps, others in smaller, more precise adjustments.
- Neuroevolution: The process of evolving neural-network brains through inheritance and mutation across generations.
- Phenotype: The set of expressed gene values after combining alleles — what the creature actually looks and behaves like. Also refers to the derived traits that emerge from genes working together.
- Selection Pressure: Environmental factors (food scarcity, predation, habitat) that cause some traits to be favoured over others.
- Speciation: The formation of a new species when a creature's genes have drifted far enough from the founder genes of its parent species.
- Species Tree: A visual phylogenetic diagram showing how species have branched from common ancestors.
- Stamina: Energy reserve for burst movement. Depleted by pushing, recovered by resting.
- Streamlining: Body elongation that reduces water friction but sacrifices turning ability.
Frequently Asked Questions
What is Tidepool?
Tidepool is an artificial life simulator where creatures with neural-network brains survive, eat, mate, mutate, and evolve over generations. Each creature has 46 heritable genes on diploid chromosomes, a physics-simulated body, and a small neural brain that controls all behaviour.
How do creatures evolve?
Creatures inherit genes and brain wiring from both parents through diploid chromosomes with crossover. Mutations introduce variation each generation. Natural selection rewards lineages that survive long enough to reproduce — better ecological fit means more descendants.
What do genes control?
46 genes control body size, strength, density, body colour (RGB), streamlining, tail length, leg count and length, shell armour, brain capacity, memory, cognition specialisation, stamina, vision (FOV distance and angle), diet type (plant vs meat efficiency), predation ability, reproductive timing, pregnancy duration, parental investment, mate calling, fearfulness, mate urgency, exploration drive, and risk tolerance.
How do creature brains work?
Each creature has a neural network brain that reads sensory data, body state, memory, and social context, then outputs five behavioural drives (forage, flee, mate, wander, rest) plus steering, thrust, and mate calling. Brain capacity controls total processing size, while four cognition genes bias whether that extra capacity is spent on foraging, threat handling, social behaviour, or memory. Brains are inherited from both parents and mutated, so behaviour evolves via neuroevolution.
How does food colour matching work?
Energy gained from food depends on the food's size, how well the creature's body colour matches the food colour (compared channel by channel across red, green, and blue), and the creature's digestion efficiency. A perfect colour match gives maximum energy; poor matches give very little. This creates strong selection pressure toward body colours that match the dominant food colour.
How does speciation work?
When a creature's genes have drifted far enough from its species' founder genes, it is classified as a new species. Functional traits like diet, size, density, streamlining, and shell count more toward speciation than cosmetic traits like colour. Species are named hierarchically (Alpha → Alpha.1 → Alpha.1.1) and shown in the Species Tree.
What are karyotypes in this simulator?
Karyotypes define chromosome structure and gene layout. Sex chromosomes (XX/XY) determine sex. Genes on the same chromosome tend to be inherited together unless crossover separates them. You can redistribute genes across chromosomes in Karyotype Settings to change linkage and inheritance dynamics.
How do I add a species?
Click Spawn Creature in the toolbar. Set starting values for all 46 genes using sliders grouped by category (body, diet, vision, brain, breeding, etc.). Confirm to spawn a founding population. These founder genes become the reference point for that species — descendants that diverge beyond the speciation threshold will branch into new species.
What can I change in the environment?
In Simulation Settings you can change land and water food production rate, maximum food count, and food colour ratios independently. Higher production and max count support larger populations. Lower values create scarcity and stronger competition. Changing colour ratios shifts which body colours extract the most energy, so it directly changes evolutionary pressure. Setting land food or water food to zero can force the ecosystem toward one habitat.
How does predation work?
Creatures with high meat digestion efficiency can eat other creatures. A predator can only eat a target whose defensive size (increased by shells) is smaller than the predator's maximum prey size, which is influenced by its body size and mouth size. Cannibalism is blocked. Predation creates arms-race dynamics: prey evolve shells, flee responses, and fear sensitivity.
What do the coloured dots on creatures mean?
The centre dot shows the creature's current brain-driven action: red = flee, orange = forage, green = mate, yellow = wander, blue = rest, purple = pregnant. The outer ring shows hunger level (green = full, red = starving). The inner ring shows stamina level (green = full, red = depleted). Both rings use the same heat-colour gradient. Toggle these with the Energy Rings button.
How does creature memory work?
Creatures remember the locations of food, mates, and predators they have seen. Memory capacity and retention duration are gene-controlled. Memories feed into the brain as inputs, helping creatures navigate to remembered food patches, mates, and away from dangers. Memory has a metabolic cost.
What data and graphs are available?
Data Analysis includes: population over time by species, gene value distributions, gene correlation heatmaps, per-species gene profiles, and a phylogenetic species tree. You can also click any creature for a full inspection panel showing all genes (maternal, paternal, expressed), brain drives, memory state, children, and calculated phenotype values.
Technical Notes
- Auto-pause when switching tabs: Browsers throttle or suspend background tabs to save resources. Because of this, the simulation automatically pauses when you switch to a different tab. This is a browser limitation — JavaScript timers, rendering, and physics all slow down or stop in inactive tabs, which would cause the simulation to desync.
- Use a separate window: If you want to browse other content while the simulation runs, open it in a separate browser window (not a tab). A separate window remains active and visible to the browser, so the simulation continues running normally. The Info page already opens in its own window for this reason.