Game Design Document (High-Level)

Working Title

TBD

TODOs

If you want next steps, we can: Split this into Design / Systems / Technical docs Turn this into a vertical slice plan Add a combat system spec Add a world knowledge data schema Just tell me where you want to go next.

Vision Statement

Create an isometric, systems-driven RPG that blends free exploration with contextual engagement modes (combat, conversation, mini-games). The world exists independently of the player, while the player character maintains memories and beliefs about the world that evolve through exploration, dialogue, and discovery.

The game prioritizes:

  • Strong systemic foundations
  • Cohesive visual fidelity (stylized, not hyper-realistic)
  • Seamless transitions between modes of play
  • Long-term scalability toward complex simulations (tactical combat, world knowledge, factions)

Core Inspirations

  • Isometric tactical RPGs (e.g. MYZ-style engagement flow)
  • Simulation-driven worlds (e.g. Project Zomboid)
  • Narrative RPGs with strong world-building
  • RTS-style world awareness (fog-of-war reimagined as knowledge)

Core Gameplay Pillars

1. Engagement-Based Gameplay

Gameplay is structured around Engagement Modes, not scenes or levels.

At any given time, the game is in exactly one engagement mode, which defines:

  • How player input is interpreted
  • How time flows
  • Which systems are active
  • Which entities may act

Exploration is the default engagement; others are entered contextually.

2. Persistent World, Persistent State

  • The world simulation never resets when modes change
  • Entities persist across modes
  • Combat, conversation, and exploration all operate on the same world state

The difference between modes is rules, not reality.

3. World Knowledge & Memory

The player does not merely “see” the world — they know it to varying degrees.

The map reflects the character’s knowledge, not absolute truth.

Engagement Modes

Exploration Mode (Default)

  • Real-time movement
  • Continuous time
  • Isometric camera (constrained)
  • Environmental interaction
  • Detection and stealth systems active

Triggers:

  • Player movement
  • Navigation
  • Discovery

Combat Engagement (Tactical)

  • Triggered by proximity, detection, or player action
  • World remains the same; rules change
  • Grid overlay appears
  • Turn-based or AP-based action system
  • Line-of-sight, cover, and positioning emphasized

Variants:

  • Seamless combat (no scene transition)
  • Region-limited combat zones

Conversation Engagement

  • Triggered by NPC interaction
  • Time frozen or abstracted
  • Dialogue system becomes primary input handler
  • Skill checks, branching outcomes, and information discovery

Conversation can:

  • Reveal rumors or locations
  • Update world knowledge
  • Unlock quests or engagements

Mini-Game / Special Engagement

  • Custom rule sets
  • Temporary override of most systems
  • Examples: hacking, puzzles, rituals, games of chance

Developer / Debug Mode

  • Free camera (fly / inspect)
  • No effect on world state
  • Debug overlays (grid, AI state, nav, memory layers)

World Knowledge System

Concept

The world exists independently of what the player knows.
The player maintains a Knowledge Model separate from the World Model.

Knowledge can be:

  • Gained through exploration
  • Gained through information (dialogue, notes, maps)
  • Outdated, incomplete, or incorrect

Knowledge States (Per Region / Location)

Each location has a knowledge level:

  • Unknown

    • Not shown on map
    • No name or indication

  • Rumored

    • Approximate location hinted
    • Shown as a circle, silhouette, or annotation
    • May be inaccurate

  • Known (Unvisited)

    • Name and approximate location known
    • Still visually obscured

  • Visited

    • Fully revealed map geometry
    • Basic understanding of layout

  • Familiar

    • Player understands danger, shortcuts, NPC routines

  • Outdated

    • Knowledge exists, but world may have changed

Knowledge Acquisition

Physical Discovery

  • Exploration
  • Line-of-sight
  • Time spent in region increases fidelity

Informational Discovery

  • NPC dialogue
  • Notes, books, rumors
  • Quest briefings
  • Environmental storytelling

Information-based discovery never reveals full geometry — only beliefs.

Map Representation

  • The map is a diegetic object (represents the character’s understanding)
  • Visual cues reflect certainty:
    • Blur
    • Opacity
    • Hand-drawn markings
    • Question marks
  • Knowledge layers can be toggled (player-facing or dev-facing)

Movement & Space

Dual Coordinate System

  • World Space (continuous, float)
  • Grid Space (discrete, integer)

Explore Mode:

  • World space movement
  • Grid position tracked passively

Combat Mode:

  • Grid rules enforced
  • World space used for animation and presentation

Camera System

Player Camera

  • Isometric perspective
  • Constrained rotation / zoom
  • Context-sensitive framing

Developer Camera

  • Detached
  • Free movement
  • Entity inspection
  • Raycast selection

Camera never modifies world state.

Systems Architecture (Conceptual)

Separation of Concerns

  • World State: Pure data (entities, positions, stats)
  • Systems: Logic that mutates world state
  • Presentation: Engine nodes, rendering, animation
  • Engagement Controller: Governs active mode and input routing

The engine scene graph is not the source of truth.

Progression & Scaling

Short-Term

  • Small, dense environments
  • Few engagement types
  • Minimal AI complexity

Mid-Term

  • Larger regions
  • Multiple combat variants
  • Rich dialogue and knowledge interplay

Long-Term

  • Faction intelligence
  • Conflicting world beliefs
  • Large-scale simulations
  • Potential multiplayer asymmetry (different knowledge per character)

Non-Goals (Early Development)

  • Photorealistic rendering
  • Massive open worlds
  • Console certification
  • Full engine replacement

Design Principles

  • Systems over scripts
  • Knowledge over visibility
  • Modes over scenes
  • World truth ≠ player belief
  • Finish small games before scaling up

Open Questions

  • Single-player only, or multiplayer later?
  • Fully turn-based or hybrid real-time/turn-based combat?
  • How unreliable can information be?
  • How much automation vs player intent in engagement triggers?

Summary

This design emphasizes:

  • Seamless transitions between gameplay modes
  • A living world independent of player awareness
  • Player knowledge as a first-class mechanic
  • An architecture that supports gradual ownership of systems

The goal is not to build everything at once, but to build a foundation that can grow.