Knowing the Universe

This page explores how players learn that other regions, galaxies, and orbital systems exist — separately from their ability to reach them.

It does not describe final game behavior. It captures the current direction and unresolved design choices.

Goal

Make the universe partly unknown to every player at any given moment, and turn the act of finding out what is out there into its own gameplay surface — distinct from the act of reaching it.

Secondary goals:

Give the explorer playstyle a real, marketable output: intel.
Avoid presenting players with an omniscient universe map by default — that turns the map into a logistics tool, not a discovery surface.
Decouple "I know System X exists" from "I can travel to System X". The two should be separate investments.

Core Problem

In the current model, the universe map is fully visible: every region, every galaxy, every orbital system is observable from any player's UI. The map doubles as a logistics planning tool and a reference for what exists.

This is fine for an MVP but creates two design holes:

No explorer archetype. A player whose preferred contribution is "find out what is out there" has nothing distinctive to do — everyone already sees the same map.
Conflation of knowledge with reach. Under Player-Built Connectivity, the separate build-reach research radius was removed: distance is now punished economically by the exponential jump cost, and knowledge is the only thing that gates which targets appear in the build list. A globally-visible universe map breaks that — a player could target any region regardless of whether they "should" know it exists. This page makes knowledge a real per-player constraint the build list can depend on.

This page addresses both: it carves out a discovery layer that an explorer archetype can own, and it makes "I know that region exists" a per-player property the rest of the game can rely on.

Working Directions

1. Active scouting — probes and survey fleets

A player dispatches a probe or survey fleet outward to investigate specific cells of the universe map. The probe travels to its destination, gathers information, and either returns or transmits its findings.

Probes are cheap and slow — a fire-and-forget exploration vehicle, not a combat asset.
A probe surveys a specific stellar object, raising its intel one tier along the object/occupation ladders (Direction 6). Probes are the only instrument that surveys objects. Revealing that systems / regions / galaxies exist is a separate mechanism — the detection project (Direction 8) — not a probe job.
Probes can be intercepted in flight and destroyed by hostile players, and a defended object can shoot at a probe attempting to survey it — adding a combat surface to exploration.
Climbing the ladders on a defended object (toward L4, and on into the occupation tiers O1–O3) takes repeated probe passes, one tier at a time — the OGame-style "send more probes" dynamic.

2. Passive observation — telescopes and sensor arrays

Buildings constructed on a player's existing holdings (colonies, frontiers) passively reveal nearby regions over time.

A telescope or large radio dish has a range measured in grid cells. Regions inside the range are revealed automatically — slowly at first, then with more depth as the instrument levels up.
Range scales with the instrument's tier and with the size of the host stellar object (a planet supports bigger instruments than a moon).
This rewards entrenched players: a long-running colony with a high-tier telescope sees deep into its neighbourhood for free.
Passive observation has limits — it cannot reveal hidden / cloaked infrastructure, and it does not reach far galaxies (those still require active probing).

3. Intel trade — knowledge as a tradeable resource

Player-specific knowledge of the universe is itself an asset. Players can share, sell, or steal intel.

A player can hand a known cell to an ally for free, creating soft alliance bonds (you know what I know).
A market for intel emerges: explorers sell discovered regions to settlers; pirates sell weak target locations to raiders.
Intel can be stolen through espionage — see Direction 4.
Knowledge of a cell is a per-player flag. Selling intel does not transfer it away from the seller; both parties now know the cell.

4. Espionage — extracting intel from other players

An espionage operation against another player can reveal what they know, not just what they have. A successful operation copies a slice of the target's knowledge map into the attacker's.

This is the inverse of intel trade: knowledge can be taken, not just given.
Espionage is its own mini-system with its own success / detection / counter-espionage dynamics — this is the explorer-archetype's combat surface.
Espionage need not be limited to knowledge: it can also reveal fleet positions, build orders, and other strategic info. But knowledge of the universe is the foundational reward.

5. Universe-map UI: fog by default, knowledge fills it in

The universe map only displays the cells the current player knows. Unknown cells are blank, or show a generic "unknown" tile. As the player discovers more — through scouting, observation, intel trade, or espionage — the map fills in.

The currently-visible "everything" map becomes the admin / debug view, not the default player view.
A "shared map" overlay can show what an alliance collectively knows.
This reshapes the early game: the first hours are spent discovering the immediate neighbourhood, not surveying a fully-revealed galaxy.

6. Object intel and occupation intel — two stacked ladders

Depth of knowledge applies only to stellar objects (systems / regions / galaxies are binary — Direction 8). A known object climbs two ladders in sequence, surveyed only by probes (Direction 1), one tier per probe pass. Ladder A describes the physical object; Ladder B describes whoever occupies it, and exists only while the object is colonized.

Ladder A — Object intel (physical; any object). L1 is granted free when the containing system is detected; each remaining tier costs one probe pass, so three probes take an object to fully-explored:

L1 Existence (free on system detection) — the object exists, its position, coarse type (planet / moon / asteroid), and its count alongside its siblings. An object is never below L1; L0 only means "the system is not detected yet".
L2 Basic survey (probe 1) — size, class / biome, rough resource hint. Enough to judge "worth it?".
L3 Detailed survey (probe 2) — exact resources, hazards, capacity.
L4 Complete survey (probe 3) — special features, hidden deposits, full potential. This is "fully explored."

Colonization gate = L4. An object can be claimed / colonized only when the player holds it at L4 and brings a ship to it. Knowledge alone never colonizes; presence alone never colonizes. Both are required. The full loop: detect system → probe-survey object to L4 → reach it (connectivity) → fly a colony ship → colonize.

Ladder B — Occupation intel (only while colonized). Once an object is at L4 and owned by a player, three further probe passes reveal that occupier's presence on this specific object (not the player in general):

O1 Occupier (probe 4) — who owns it, holding tier, which sector types are present.
O2 Infrastructure (probe 5) — defenses, production, garrison strength (the Detecting / Hiding build-up itself).
O3 Live state (probe 6) — current fleets present, stockpiles, queues.

An uncolonized object has no Ladder B and caps at L4. Stripping a colonized object completely takes six probe passes in total.

Only probes survey. Climbing either ladder is done exclusively by probes. No other ship — not a colony ship, not a warfleet — raises an object's intel. A colony fleet arriving at an unsurveyed object cannot survey it itself; the object must already be at L4, whether the colonizer probed it or bought the dossier from an explorer.

Persistence and degradation.

Object intel (L1–L4) is permanent. Once earned it is never lost, and it survives ownership changes — if another player colonizes an object you already hold at L4, your L4 stays.
Occupation intel (O1–O3) is tied to the occupation it describes. When that occupation ends — the owner abandons the object, or it changes hands — the occupation intel is lost and degrades back to L4. A fresh occupation must be probed from O1 again.
All intel is a frozen snapshot at survey time and never auto-refreshes. Ladder A is static physical fact; Ladder B is dynamic, so decay (still out of scope) is really a Ladder-B concern — O3 especially goes stale fast.

Defenders defend every tier. A defended object resists probing at all tiers, object and occupation alike (Direction 9). A probe cannot pull even L2 off a sufficiently-defended object — the physical facts are guarded just like the occupation facts. Only undefended objects (unowned, zero defending stats) survey freely, which is what keeps colonizable (unowned) objects probeable at all.

Graded intel product. Because tiers are distinct and tradeable (Direction 3), one object yields many price points — from L2 (a scouting lead) through L4 (a colonization-ready dossier) to O3 (live order-of-battle on an enemy holding). L4 on a rich planet is the settler's premium buy; O-tiers on a rival's fortress are the spymaster's.

7. Discovery reach grows exponentially per research level

Discovery reach is governed by a single research track (working name: Astronomy) and a single radius. There is deliberately no second radius and no separate per-topology-level track — one research drives all discovery reach, whether the target is a neighbouring orbital system, a distant region, or another galaxy.

The radius grows exponentially with research level, not linearly. This is the key design decision, and it exists to solve a density problem:

Orbital systems cluster densely within a small area, but regions sit far apart across mostly-empty space. A linear radius (a fixed increment per level) would reveal local neighbours quickly, then stall for a very large number of levels before reaching anything in the next region — making the player wrongly believe there is nothing left to find and stop investing.
An exponential radius adapts to both density regimes with one knob. Low levels add small increments, revealing the dense local cluster one neighbour at a time (fine-grained fog of war). High levels add large increments, crossing the sparse void to the next region in only a handful of further levels.

This is why a single research track is enough: the exponent does the adapting that would otherwise require separate near-range and far-range mechanics.

Consequences:

A discovery action that finds nothing within the current radius simply reports "nothing in range". It does not reveal, name, or hint at objects beyond the radius — the player widens reach by raising research, never by being shown unreachable targets they cannot yet see.
The exact growth shape and per-level numbers are unresolved and out of scope here. Only the exponential shape is decided.

8. Topology discovery — detection project and upward cascade

Knowing the topology — that a system, region, or galaxy exists — is binary (known / unknown), not graded. There are no intel tiers here; the object/occupation ladders are only for stellar objects (Direction 6). A system is revealed in two steps:

Detectable — the system falls within the player's exponential detection radius (Direction 7). Radius makes a system reachable by detection; it does not reveal it on its own.
Detected — the player runs a detection project (from an observation facility on a holding) that converts a detectable system into a known system, placed on the map.

Upward cascade. Detecting a system reveals the containers it sits in:

Detect a system in an as-yet-unknown region → that region becomes visible on the map.
That region sits in an as-yet-unknown galaxy → that galaxy becomes visible too.

A player never reveals a distant region directly. They detect one of its systems, and the region (and galaxy) light up around it. Once a system is detected, its stellar objects are revealed at L1 and the object ladder (Direction 6) takes over.

9. Surveying vs defending — the probe / defender stat model

Raising a stellar object's intel level (Direction 6) is resolved against the object owner's defenses through two independent checks per probe pass, in order:

Survival — does the probe live? Resolved from the prober's Cloaking against the defender's Detecting. Higher net Detecting raises the chance the probe is shot down; higher Cloaking lowers it. Survival is independent of extraction: a probe can survive and still learn nothing, or be shot before it learns anything.
Extraction — does the data move up a level? Attempted only if the probe survived. Resolved from the prober's Revealing against the defender's Hiding. Higher net Revealing raises the chance of advancing one level; higher Hiding lowers it.

So four stats, two opposed pairs: a prober carries Cloaking + Revealing; a defender carries Detecting + Hiding.

Shooting a probe is abstract — no battle, no defenses, no warship required. A probe being shot down is not a combat encounter. It needs no defensive structures, no weapons, and no fleet present at the object. It is derived purely from the defender's Detecting versus the probe's Cloaking. Shot is shot. This keeps probe-survival entirely separate from the fleet-combat system.

Where the stats come from. Each of the four stats has an empire-wide base from research — this is what "investing in explorer tech" means. There is no explorer class and no specialized probe; only how far a player pushed these techs. On top of the base:

Buildings (and similar) raise a specific object's defending stats (Detecting / Hiding).
A specialist assigned to a fleet raises that fleet's surveying stats (Cloaking / Revealing) for the probes it carries.

Specialists are their own unsettled concept — see Specialists.

Difference with hard thresholds — no soft floor. Each check resolves on the difference Δ = proberStat − defenderStat, but the extremes are hard-capped, not asymptotic (the OGame model):

Δ ≥ +N → guaranteed success — the probe always survives / always extracts the next level.
Δ ≤ −N → guaranteed failure — the probe is always shot / never extracts, no matter how many probes are thrown at it.
in between → a graded probability that climbs with Δ.

A defender enough levels above the attacker is flatly un-probeable; an attacker enough levels above the defender always gets through. Luck and probe-spam cannot crack a sufficiently superior defender — a low-tech player cannot fluke intel off a high-tech one.

This enables a hiding playstyle and intel fortresses. A player who invests heavily in Detecting + Hiding can become effectively un-spyable by anyone not within N levels — a legitimate archetype built entirely on concealment, revealing nothing about its holdings. It also enables deliberate decoy fortresses: pouring intel-defense into an otherwise worthless object so attackers burn probes and effort trying to crack it, only to discover there was nothing worth the cost. (This only ever protects owned objects' secrets from espionage; it never blocks colonization, since colonizable objects are unowned and therefore undefended.)

Undefended objects are free. An object with no owner (or zero defending stats) cannot shoot probes and cannot hide data — probes survive and reveal trivially. Heavy Cloaking + Revealing investment is only needed against owned, defended objects. That is what makes contested intel the explorer's domain without any hard class gate.

Open Questions

Decay is deferred. Intel is currently a frozen snapshot at survey time (Direction 6) and never auto-refreshes. Whether snapshots should go stale or decay — especially the dynamic occupation facts (O1–O3, above all O3) — is out of scope for now and left for a later pass.
Threshold band size N is tuning. Direction 9 fixes the shape — hard caps at ±N levels of stat difference, with true intel fortresses allowed. The actual N (how many levels above an attacker a defender must be to become flatly un-probeable, and how steep the graded band between is) is tuning and out of scope here.
How does an alliance share knowledge — automatic full-share, opt-in cell-by-cell, or some middle ground?
Should there be a rumour mechanic — partial / unreliable knowledge a player has heard about but not confirmed?
Can players lie about their intel (sell false data)? If so, how does the market handle it?
Does the universe-map UI need to show the absence of knowledge differently from the absence of a cell? (Empty universe-map cells should look distinct from "I haven't scouted here yet".)
What does a fresh player see when they first log in — only their spawn region, or some baseline (their home galaxy, perhaps)?
Should there be a public reveal mechanism — global events that surface a cell to everyone simultaneously (a supernova, a galactic-scale event)?
How does this layer interact with the gate-target list in Player-Built Connectivity? Confirm the relationship: a gate's targetable set is the set of known cells (this page); affordability is then governed separately by the exponential jump cost (connectivity Direction 5a). Knowledge gates what appears in the build list; cost gates whether it is worth building.