Myriuna FAQ

Myriuna is a game. Myriuna Core is the engine underneath it.

Players create or choose actors, inhabit persistent worlds, and experience those worlds through generated prose. Myriuna Core governs the actors, world state, knowledge, relationships, rules, history, and consequences that keep the experience coherent.

No. The closest equivalent to the game master is the social contract defined by the module.

A module establishes the rules, customs, authority, social expectations, knowledge boundaries, and permitted consequences of its world. The language model performs inside that structure by rendering scenes, dialogue, interpretation, emotion, and atmosphere. It does not independently decide what reality is.

Yes. Myriuna can present three likely continuations, but those choices are invitations rather than boundaries.

Every participating entity—including the player character—is represented as an actor in the world. The model therefore does not treat the player as the person it must assist or automatically satisfy. It treats the player character as another person operating inside the setting.

You may attempt actions outside the suggested continuations, but success is not guaranteed. Your character remains subject to their abilities, circumstances, knowledge, relationships, authority, and the rules of the world.

Not in every module.

In a full simulation, every actor can receive model-driven turns. Most playable modules are expected to combine those turns with chance, dice, schedules, rules, and lighter background progression so the campaign can continue without generating a complete scene for every person during every loop.

Guaranteed actions create more influence over the world. By controlling which actors receive them and how often, a module can keep the player semi-central while allowing the wider society to continue changing.

Yes.

One universe running the core module as a text-first experience is designed to remain playable on one capable local machine. That core includes persistent world state, actors, bounded knowledge, relationships, history, freeform actions, and generated prose.

The complete product—with richer media, distributed model workers, advanced prerendering, and service-scale features—can use additional machines and services.

Myriuna is local-first because its worlds are intended to outlive the service that hosts them.

To keep that promise, the core text-first experience for one universe must remain capable of running on one machine. The wider product can expand across distributed local systems and compatible cloud services, but the world itself must not become dependent on those layers.

Early research also showed that cloud computing would consume the largest part of the development budget. Heavier generation and repeated testing therefore run primarily on local hardware during development.

Continuity created the requirement. Development cost reinforced the architecture.

Yes.

The architecture can use models and services running on local hardware, compatible cloud providers, or a mixture of both. A local model may be practical for sustained testing and heavy iteration, while a cloud model may offer stronger prose, additional capacity, or a specialised service that is impractical to run locally.

My current personal favourite for prose is ChatGPT.

Regardless of where the model runs, Myriuna Core remains responsible for the universe’s authoritative state.

Your world will be made available for download together with the software needed to continue running its core campaign locally.

The continuity promise applies to the world itself: its actors, history, relationships, knowledge, state, and unfinished stories. Every hosted convenience or future media service may not remain available forever, but the core text-first world should not become disposable.

A world should end because the story ends—not because the service ends.

No.

The first module is a broad original low-fantasy foundation centred on ordinary people, households, work, travel, customs, rumours, relationships, obligations, practical danger, and social consequence.

Magic may belong in later modules. It is deliberately absent from the first reference world so spectacle does not obscure whether the social and continuity systems actually work.

No. The core experience is text-first.

Voice, generated images, video, music, ambience, maps, and other presentation systems can extend the experience through local or cloud services. They render the same underlying world further, but they do not own its state or determine what is true.

World state first. Text renders the playable experience. Additional media render that experience further.

Myriuna is currently in pre-production, and no public release date has been set.

The next major milestone is the first real playtest: roughly fifty villagers inhabiting one persistent, feature-light world while every implemented mechanic passes through repeated actor loops.

Public availability should follow demonstrated continuity rather than an arbitrary date. The immediate test is whether the village can remember what happened, allow different actors to know different things, react differently, survive restart, and explain why later events occurred.

No. Myriuna is currently in pre-production and is not yet available for public playtesting or purchase.

There are no memberships or early-supporter packages available yet. Check back in the future as development progresses.

Development is currently funded primarily from my own savings. Supporting the project does not grant access to a build or reserve a place in future testing—but development does go through a shocking amount of coffee.

Buy the developer a coffee