Every rule earns its place. This page states each rule and then explains why it exists — what problem it solves and what it protects the game from.
Rule: All gameplay happens on a single full-screen command center. There are no Research, Fleets, or Reports nav tabs.
Rationale: The classic games this draws from buried key information in menus. Hiding fleets in a list makes the map feel dead and makes attacks unreadable. By putting ships on the map, reports in a news ticker, and research behind a single Tech number, the whole strategic situation is legible at a glance — which is what makes a turn feel fast and a decision feel informed.
Rule: Gold is the only spendable resource. Factories cost 200; Tech Labs cost 250. A Factory produces gold or ships per the planet's focus slider; ship output can accumulate fractionally.
Rationale: A single currency keeps the mental model tiny — no juggling metal, crystal, gas, and population. The deliberate omission of extra resources is a feature: depth comes from where and when you build, not from bookkeeping. Playtesting showed a pure ship-rush with no economy went 1-for-8, which is intentional — skipping economy should be weak, not free.
Balance tension caught in testing: a planet with zero Factories produces zero ships regardless of the slider, so an aggressive player who skipped economy had literally no fleet. A proposed fix — a small base ship trickle on any owned world so "attack" is never a dead end — is documented but held pending a follow-up simulation, precisely because it risks speeding up snowballing on large maps.
Rule: Tech Labs generate hidden research (1 unit per lab per cycle — a confirmed original-game rule). Crossing a cost threshold raises your Tech Level (0–15). Early levels are cheap; later levels ramp up.
| To reach Tech | Research needed | …10 | Research needed |
|---|---|---|---|
| 1 | 5 | 10 | 140 |
| 2 | 10 | 11 | 168 |
| 3 | 18 | 12 | 198 |
| 4 | 28 | 13 | 230 |
| 5 | 40 | 14 | 265 |
| 6–9 | 55 → 115 | 15 | 303 |
Rationale: Research used to be a top-level resource, which added HUD clutter for something the player mostly wanted expressed as "how far can I reach and how hard do I hit." Folding it into one Tech number keeps the HUD calm. The cost curve is front-loaded so that a slightly-too-far planet takes a handful of cycles to reach, not ~20 — an explicit fix for early playtests where the enemy homeworld often wasn't even reachable until the game was nearly over. Later levels still ramp so a decisive tech lead is genuinely earned.
Rule: Range and speed are measured in hops, where one hop is the typical neighbor spacing.
range (hops) = 1 + techLevel
speed (hops/cycle) = 1 + techLevel × 0.5
travelTime = ceil(distance ÷ speed)Rationale: Raw pixel distances are meaningless to a player. Hops make reachability intuitive: "I can reach the next world, then hop from there." Range grows one full hop per Tech level so high tech means longer single-flight jumps that skip intermediate worlds. Speed grows at only half a hop per level on purpose — if speed equaled range, every legal max-range flight would resolve in exactly one cycle, erasing incoming-fleet warnings and any chance to intercept. At half-speed scaling, a long jump takes two cycles, which preserves the warning window and defensive counterplay.
Rule: Combat is a sequence of probabilistic ship-vs-ship rounds. Each round the loser drops one ship; when a side hits zero, the other holds or captures the planet.
attackerWinChance = clamp(0.5 + techDifference × (0.4 / 15), 0.10, 0.90)Rationale: Equal tech is a 50/50 exchange, so numbers matter most — a bigger fleet reliably beats a smaller one, which keeps the strategy about economy and positioning rather than luck. Tech tilts the odds but is clamped between 10% and 90%: a tech lead is a real edge, never an auto-win, so an underdog with enough ships always has a puncher's chance. The maximum swing of ±0.40 at a full 15-level gap is tuned to feel decisive without being oppressive.
Rule: Every commander starts with a 16-ship garrison on their homeworld. Defenders on their own homeworld fight as if they had +3 Tech.
Rationale: Homeworlds originally started with zero ships, which meant a single early fleet could end a player before they'd done anything — a feel-bad, unreadable loss. The starting garrison does double duty: it gives homeworlds real defense and gives every commander enough force to make an aggressive move in the first few cycles instead of waiting to build an economy. The +3 defensive tech bonus ensures a same-size stack can't flip a homeworld on a coin toss; taking one requires a genuinely committed force — often ships pooled from several worlds at once, since garrison attrition caps how much a single planet can throw.
Rule: The win condition is last commander standing — capture every rival homeworld. There is no score timer or dominance-percentage shortcut.
Rationale: This is faithful to the classic Strategic Commander lineage. Early builds risked 300-cycle stalemates, so pacing additions (a dominance-fraction win and a hard cycle cap) were trialed — and then removed once garrison attrition proved it could keep endgames decisive on its own. The philosophy: solve runaway games with a real mechanic (homeworlds bleed attackers, so a stalled front eventually breaks) rather than an artificial trigger that ends a game the players didn't finish. Difficulty tuning pulls AI aggression forward so games resolve in target windows — roughly 13–23 min on Tiny and 20–40 min on Medium at standard timers.
Rule: Named planets always carry the same rating and multipliers. Generated names for big maps are drawn by tier.
Rationale: Randomly assigning a legendary name to a poor planet destroys the mental map players build across games. When Mars is always an excellent balanced world and Europa is always tech-friendly, "I'm racing you to Mars" carries real, learnable meaning. Fixed identities turn the galaxy into a place you get to know rather than noise you re-learn every match. See the full catalog →
Rule: Commander Level (from XP) is permanent and never decreases; it unlocks cosmetics only. Rank/ELO is competitive and moves only in ranked online matches. Sandbox and debug earn neither.
Rationale: "Level = your progress, Rank = your skill." Keeping the two separate means grinding never buys power — there is explicitly no pay-to-win — so a match between a Level 50 and a Level 5 is decided purely by play. ELO changes are gated to ranked so casual and solo games stay pressure-free, and demotion protection plus the softer "Rank adjusted" wording keep competitive play from feeling punishing.