Moria (1975, PLATO / University of Illinois)

Author(s): Kevet Duncombe and Jim Battin

Sources: Wikipedia: Moria (1975 video game) · HowToMakeAnRPG: Moria, the first first-person multiplayer RPG · The CRPG Addict: Game 121, Moria (1975) · Wikipedia: Moria (1983 video game) · A History of Moria (umoria.org)

Category: inspiration, Batch 10 (gameplay-mechanics mining). One shared, persistent, procedurally generated dungeon that many players inhabit at once, with a first-person maze view and cooperative parties built into the core loop. Serves: Tunnel Rat (assigned by Josh; see cart map)

Disambiguation (read this first)

This entry is the PLATO Moria (~1975) by Kevet Duncombe and Jim Battin: a multiplayer, first-person, persistent-world dungeon crawler that ran on the PLATO time-sharing system. It is a different game from the 1983 Moria by Robert Alan Koeneke, which is the single-player Rogue-like written at the University of Oklahoma (VMS BASIC in 1981, rewritten in VMS Pascal in 1983 on the VAX-11/780). Koeneke’s Moria was ported to C as UMoria by James E. Wilson in 1987 and became the basis of Angband. The two share only a Tolkien-derived name. They have no common code, no common design, and no shared lineage. Everything below is about the PLATO game.

What it is

Moria is a dungeon crawl that up to ten players explore at the same time inside one shared world. The dungeon is generated dynamically as players move, and the world persists: a character’s stats, gold, position, and progress are saved between sessions, so logging back in resumes where the player left off. It began as a top-down two-dimensional crawl inspired by the PLATO game dnd; during development Jim Battin suggested converting it to a first-person 3D view, drawing on other early PLATO 3D titles (Panther, Spasim, Airfight). The name was suggested by dnd developer Dirk Pellett.

Mechanics

• First-person wireframe maze. Movement is shown as a first-person wireframe corridor view rendered in a small viewport near the center of PLATO’s 512×512 display, with the player’s coordinates shown alongside for orientation.
• A shared persistent world. Characters are saved between sessions, and each character’s status and location are restored on return. Play is an early persistent multiplayer world in the MMO sense, years before the term.
• Procedural dungeons, generated on the fly. Cell data (wall, passage, secret passage) is computed per X,Y coordinate as players advance. The world spans five dungeons (Forest, Desert, Cave, Mountain, Ocean), each reported at 60 levels deep, so depth is effectively the difficulty axis.
• Four stats that grow through use. Cunning, Piety, Valor, and Wizardry are chosen at creation and rise through the matching activity: fighting raises Valor, trickery raises Cunning, and so on. Additional tracked resources include Vitality (100 hit points), Gold (starting 150), and Food and Water.
• Aging and permadeath with an heir. Time passes as the character ages (roughly one in-world month per eight minutes of play), starting at age 13. Death is permanent, but a player can pass items to an heir, the next character, softening the loss.
• Guilds and long-horizon advancement. Reaching a stat value of 20 lets a character join a guild (the Brotherhood, the Circle of Wizards, the Union of Knights, the Thieves Guild), each granting special abilities (Brotherhood members heal; Circle of Wizards members teleport). Higher rank requires pushing stats to 30/40/50 and donating large sums of gold, with guild banks that store wealth and items and pass them on at death.
• Quasi-real-time combat menu. Encounters outside towns run in rounds with a menu of responses (Fight, Trick, Pray, Cast, Evade, Bribe, Run), each drawing on a relevant skill. There is time pressure: waiting without acting can get the character killed. Combat state persists, so fleeing and returning does not reset the fight.
• Cooperation as a first-class system. Parties of up to ten travel together. Players can talk to their party, whisper to a single ally, or yell for help even when ungrouped, and other players can respond immediately; a “run to help” command sends a player to an ally in trouble. Cooperation is designed into the core loop and grows more valuable as the dungeon deepens and monsters get lethal, though solo play remains possible.

Mineable patterns

• One shared, persistent world as the spatial container. A single generated dungeon that many operators occupy at once, with per-character progress saved across sessions, is the reusable frame. State lives with the world and the character, and survives any single play session.
• Procedural generation keyed to coordinates. Deriving each cell from its X,Y position lets a large world exist without storing it, and lets depth stand in for difficulty.
• First-person maze rendering in a small viewport. A wireframe corridor view drawn in a bordered window, with coordinates shown beside it, is a compact way to present 3D space on a character-cell display.
• Cooperation built into the core loop. Party travel, targeted and broadcast messaging, and a “run to help” response make other players a mechanic. Difficulty that scales with depth turns cooperation from optional into pressure.
• A long-horizon advancement economy. Guilds gated by stat thresholds and very large gold sinks give the persistent world a multi-session goal structure, with inheritance to blunt permadeath.
• A fixed combat verb menu under time pressure. A small, stable set of responses (fight/trick/pray/cast/evade/bribe/run) that each read a different stat keeps encounters legible while rewarding character build.

KN-86 resonance

Keep this at the concept level. The fit is with shared-world and persistence ambitions, and this stops short of designing a specific cart.

• Moria is a clean historical reference for any multiplayer or persistent-world direction on KN-86: one generated world, many participants, character progress that survives across sessions. That rhymes with Universal Deck State (progress that persists across carts) and with the World Engine’s model of a single coherent generated world.
• Its coordinate-keyed procedural dungeon and depth-as-difficulty axis echo the World Engine’s approach of generating a world and letting operators lens into it, without prescribing how a KN-86 world does so.
• Guilds gated by stat thresholds and heavy gold sinks are an early example of a reputation-and-credits advancement economy over a persistent world, adjacent to KN-86’s credits and reputation model.
• The first-person wireframe maze in a small viewport is a concept-level reference for what a line-and-vector rendering pass can put on the 128×75 amber cell grid (see the retro-technique roadmap in docs/influences/synthesis-atari-graphics.md), noted here as a possibility only.

Sources & verification

Consulted: Wikipedia: Moria (1975 video game) (authors, PLATO, ~1975 start with 1978 copyright and 1984 update, up to ten players, dynamic generation, first-person wireframe, the dnd inspiration, and Dirk Pellett naming it); HowToMakeAnRPG’s Moria article (the four stats, guilds and their abilities, the combat menu, the five dungeons at 60 levels each, gold and food/water, the postage-stamp viewport); The CRPG Addict’s Game 121 playthrough (persistence across sessions, aging and the heir mechanic, party chat / whisper / yell / run-to-help, quasi-real-time combat, and the observation that solo play was viable); and, for the disambiguation, Wikipedia: Moria (1983 video game) plus umoria.org’s history.

Confidence: high on the identity and headline design (multiplayer, persistent, first-person wireframe, procedural, cooperative, on PLATO) and on the disambiguation from the 1983 Koeneke Moria. High on the disambiguation facts (Koeneke, University of Oklahoma, VMS BASIC 1981 then VMS Pascal 1983, UMoria by James E. Wilson 1987, basis of Angband). Medium on exact numeric details (60 levels per dungeon, the precise aging rate, the 1,000,000-gold guild donations): these come from secondary playthrough writeups of a living, updated PLATO game whose numbers shifted between its 1978 and 1984 revisions. One point to state plainly: cooperation was central to the design and strongly rewarded, but it was not a hard gate; the game could be, and was, played solo. Not confirmed: any relationship to KN-86 beyond the conceptual analogy drawn here.