Botlings Swarm Guide — Routing, Density & Swarm Health

A hundred bots is not a workforce — it is a traffic system, and Botlings treats it like one: routes are roads, junctions are intersections, and jams are the tax on sloppy geometry. This Botlings swarm guide covers route painting, density rules, junction discipline, and the diagnostics that keep the swarm flowing at every scale.

Route Painting Fundamentals

Routes are painted lines the swarm follows, and the fundamentals are geometric: one bot, one loop, one direction. The first-week discipline from the beginner guide holds at every scale — loops that don't overlap, junctions that don't queue, and every drop point adjacent to a pickup so the loop closes cleanly. The most common paint error is the shared segment: two loops borrowing the same corridor, which works at ten bots and gridlocks at sixty. Paint as if the swarm you have is half the swarm you'll have.

Loop shape matters more than loop length. Tight loops (short, direct, few turns) out-earn sprawling ones per bot even when the sprawling loop touches more resources, because turns cost time and junctions cost risk. The strong shapes: the simple ring around a district, the out-and-back on a spine, the figure-eight through two clusters with one controlled crossing. The weak shapes: anything with a crossing you didn't design, any route with a dead-end leg, any loop whose return path shares more than a junction with its outbound.

Paint direction is a free efficiency. All loops should circulate the same rotational direction on shared corridors — counter-flow segments (one loop's outbound on another's return) halve effective corridor capacity and double collision idling. A factory whose routes all rotate the same way flows like traffic on a roundabout; a factory of mixed rotations flows like an intersection with no signals.

Density And The Swarm Economy

Swarm density is bots per tile of route, and it has a healthy band: dense enough that throughput saturates the consumers, sparse enough that no segment queues. Below the band, consumers starve and the fix is manufacturing; above it, corridors jam and the fix is geometry or duplication. The diagnostic is queue-watching — any bot waiting on any segment means that segment is over density, and the answer is never "faster bots," it is "more route."

Duplication is the density release valve. When a loop saturates, paint its twin — same stations, offset corridor — and the swarm self-balances between them. Duplicate loops look redundant and are actually the cheapest throughput in the game: no new stations, no new logic, just paint. The rule of thumb: one loop per roughly thirty bots of demand, and re-watch the junctions after every duplication, because twins meet somewhere and that meeting point inherits both flows.

Junction discipline is where factories live or jam. A junction is any shared tile or crossing, and the discipline set: minimize count (redesign routes to share less), separate grades where the map allows (over/under crossings don't interact), never park logic at a junction (gates at crossings combine conditions with collisions), and give every junction an empty buffer tile on each approach so waiting bots stack off the flow. Botlings' jams are honest — they only happen where two routes were painted to fight.

The Mesh Era Transition

Past a hundred bots, the swarm guide and the automation guide converge: the mesh decouples routes from stations, and route painting shifts from point-to-point lines to station-to-mesh drops. Paint loops that feed the nearest mesh node and draw from the nearest mesh node, and the factory's routing complexity collapses — every loop becomes short, every long-haul becomes the mesh's problem, and heavies inherit the spine.

The mesh-era swarm splits into castles and couriers: district castles (dense local loops serving cluster consumers) and mesh couriers (the long-haul routes that move pooled resources between districts). Couriers get the heavies and the direct spines; castles get the standards and the tight rings. The split is what makes hundred-bot factories readable — you stop managing routes and start managing two route categories.

Mesh-era health checks remain the same two numbers: zero idle (the rebellion currency) and zero queue (the jam currency). A hundred-bot factory passing both checks is breathing — and the vacation test from the automation guide is the graduation exam: walk away, come back, and the swarm that didn't notice you is the swarm you built correctly.

The Swarm As A Living System

Past the mechanics, a Botlings swarm behaves like a single organism with moods, and reading the moods is the veteran's real skill. A content swarm hums — loops flow, junctions breathe, idle counters sit at zero — and the hum is visible: bots in motion everywhere, none clustered, none wandering. An anxious swarm shows the early tells — micro-queues at one junction, a handful of wanderers, buffers trending wrong — and the tells arrive minutes before any system actually fails. The operators who read moods fix factories for the price of a glance; the operators who read alarms fix them for the price of a rebellion.

The swarm's moods map to the three currencies of the whole guide series: flow (queues and jams), supply (starvation and buffer trends), and attention (idle and awakening). Each currency has its own gauge and its own maintenance move — geometry for flow, buffers for supply, work for attention — and mood-reading is just triage: which gauge is drifting, which move applies, and whether the drift is a local fix or a symptom of scale. The beautiful property of Botlings' design is that the three currencies never fail simultaneously in a well-built factory; one always drifts first, and the drift names the move.

The endgame of swarm skill is delegation: the factory reaches a size where you stop steering routes and start steering policies — duplication thresholds, staffing ratios, retire triggers — and the swarm steers itself. Every guide on this wiki is a policy in disguise: the ratios, the checks, the ladder rungs. Learn them as rules first, then as gauges, and finally as the policies your factory runs while you watch your own hundred bots hum.

FAQ

How do I fix a jam in Botlings?

Jams are geometry, never luck: find the shared segment, and either duplicate the loop on an offset corridor or redesign one route to remove the crossing. Adding bots to a jam makes it worse; removing the conflict makes it gone.

How many bots should be idle?

Zero, always — idle is both wasted throughput and awakening pressure. The ten-minute checklist exists because idle clusters compound into rebellion; a factory passing zero-idle is profitable and obedient.

Do bots collide or wear out?

Bots don't collide destructively — they queue, which costs time instead of units. Wear exists as aging: long-service bots accumulate awakening pressure faster, which is why retire triggers auto-convert the elders.

Is there a max swarm size?

No hard cap — the practical ceiling is your routing quality. Factories past a few hundred bots exist and run clean on mesh-era castle-and-courier patterns; the ceiling is readability, not the game. In Botlings, a swarm of any size flows if its routes were painted for the swarm it became rather than the swarm it started as.

What one habit most improves swarm health?

The ten-minute zero-idle walk — physically touring the factory on a timer and treating every idle bot as a leak. Idle is the currency of both waste and awakening; the walk costs minutes and pays in throughput, obedience, and the quiet confidence that your Botlings factory is exactly as healthy as it looks.