Fully onchain games are the most ambitious and visionary symphony in Ethereum’s pursuit toward decentralized applications. Infinite games, autonomous worlds, digital physics — how have these captivating and enigmatic narratives evolved? In the ever-shifting landscape of market dynamics, how have the grand narratives of large-scale decentralized application ecosystems been inherited, transformed, and reimagined?

Looking back to 2021, the NFT market was at its peak, and the metaverse remained a topic of widespread enthusiasm and debate. Dark Forest, an experimental game pioneering zk-snarks and fully onchain mechanics, was beginning to take shape. At the same time, Gubsheep was actively sharing his vision of permissionless interoperability and infinite game.

The concept of infinite game describes game that never truly end, capable of persisting indefinitely. Given that Vitalik Buterin himself was a Dark Forest player, one could reasonably infer that Ethereum’s more recent advocacy for the Infinite Garden philosophy has drawn inspiration from the principles of infinite game.

In August 2022, Ludens (from Lattice team) published the article “Autonomous Worlds (Part 1),” signaling the quiet arrival of a new era. An autonomous world refers to a world built on the blockchain substrate, constrained by digital physics encoded in smart contracts. The immutability of blockchain ensures that the inhabitants of this world possess greater autonomy. The narrative of autonomous worlds is both an inheritance and an expansion of the narrative of infinite game, with the shift in perspective from constructing games to constructing worlds, which is a significant and positive step forward.

Both the metaverse and autonomous worlds aim to construct a digital realm distinct from the physical world. However, their core visions differ greatly. The metaverse emphasizes a digital avatar that allows individuals to immerse themselves in a three-dimensional virtual environment, focusing on real-time experiential engagement. In contrast, autonomous worlds stress the immutability of the world’s substate, enhancing the autonomy of its digital inhabitants.

At the end of 2022 and the beginning of 2023, the Lattice team released MUD v1 and v2. MUD reduces the complexity of building Ethereum apps with a tightly integrated software stack. Developers only need to focus on the underlying data structures and the rendering of the top layer, while MUD engine provides solutions for the entire middle layer technology stack. This allows application developers to focus on the design and development of the application. At the same time, the Starknet ecosystem also introduced the Dojo engine, aiming to build a similar tool on the Cairo VM that lowers the development barrier for developers.

Ludens once shared a widely circulated quote in a previous talk: “Don’t worry about gas fees. Don’t worry about making money. Living in the future.” The launch of Redstone L2 at the end of 2023 can be seen as a response to the notion of “Don’t worry about gas fees,” as Redstone offers very low gas fees. Additionally, it provides excellent RPC services and a MUD indexer, making it very developer-friendly for applications built on the MUD engine.

MUD engine and Redstone L2 attracted many community developers to build their applications based on MUD. From 2023 to mid-2024, the ecosystem was quite vibrant. However, by the end of 2024, it seems that this positive momentum did not continue. I think there are several key reasons for this.

1. The resources available to community developers of fully onchain games are very limited. As far as I know, Redstone did not provide cash grants to developers within its ecosystem. Many developers, motivated by passion or drawn to the vision of autonomous worlds, were forced to suspend their development due to a lack of necessary funding. In contrast, the Starknet ecosystem offers greater support for developers, leading some teams working on fully onchain games to shift towards Starknet.

2. The narrative vision of autonomous worlds is grand, but it lacks an executable plan for building applications. Over the past two years, few teams have consistently persisted in exploring and implementing their development philosophy.

3. The models for cold-starting and commercializing fully onchain games / autonomous world projects are still to be explored.

In August 2024, the Lattice team’s Sky Strife project announced the establishment of an independent game studio, but after that, the Sky Strife Twitter account stopped posting new updates. By the end of 2024, Biomes was included as part of the megaETH ecosystem project @0xMegaMafia. In February 2025, the founder of the Promodium project, Tax Cuts, announced the dissolution of the development team and that no further development would be carried out on the project.

There are still some positive signals worth noting. The MUD codebase remains active with updates, and the improvement of the tech stack lays a solid foundation for future development. MUD’s deep collaboration with CCP Games signals the long-term development and progress of the Eve Frontier project. The DFArchon community is also continuing the development of the Dark Forest community rounds, with Dark Forest MUD v3 expected to launch its public beta next month. The Biomes team is also continuing to optimize and update the new game interface. Many other teams are also pushing forward with their development and operations. These movements indicate that, despite market turbulence, there are still teams holding on to their ideals and pushing the ecosystem forward.

As the crypto market continues to struggle, fully onchain games and autonomous worlds ecosystems may face even more challenging tests. However, true innovation often emerges in adversity. In the future, we may witness new technological breakthroughs, more sustainable business models, and even stronger community collaboration, driving the evolution of this narrative once again.

The midgame of the autonomous worlds is far from over; the second half is just beginning.