Hello and welcome to Dark Tunnels, a newsletter dedicated to exploring the emerging ecosystem of fully onchain games. Dark Tunnels is published by Always Scheming, a research consultancy focused on the future of interactive entertainment.

The following essay is Issue #10 of Dark Tunnels. Enjoy!

Recently, web3 gaming organization Game7 released their “State of Web3 Gaming Report.”

The report is notable for a number of reasons, not least of which being that it actually attempts to size the existing fully onchain games ecosystem as a separate, distinct segment from the rest of web3. [1] It’s also a thorough and sober overview of the current web3 gaming landscape. I highly recommend giving it a read.

One particular item that caught my eye was the genre breakdown within fully onchain games, below:

The first and most obvious takeaway is that Strategy is currently the dominant genre in fully onchain gaming.

I suspect that this fits most people’s priors, but it’s good to see data to back it up. Many of the most prominent games in the onchain gaming ecosystem are strategy games — titles such as Mithraeum, Citadel, and Dark Forest, to name a few. [2]

It makes sense that strategy games dominate: the genre typically requires fewer player actions (and therefore much lower transaction throughput) than, say, action or sports games, making it a good fit for the technical limitations imposed by building everything on a blockchain.

I believe strategy games will continue to maintain their lead in 2024. After all, there is likely to be some survivorship and confirmation bias involved. Given that everyone is still trying to figure out the best approach to onchain game development, I would not be surprised to see more developers borrow learnings from the most widely-known (relatively speaking) fully onchain games today. Many of these games’ creators have also been quite open about their approaches to development, with some even open-sourcing their code entirely.

However, the genre I am most focused on is Simulation, currently tied for third-most in Game7’s dataset at roughly 10% of all fully onchain titles. Examples of fully onchain simulation games that you can play today include Primodium, Land, Labor, and Capitol, and the recently updated Soccerverse.

In my view, simulation games are a great match for onchain gaming. Construction and management sims, city builders, tycoon games, sport management simulators, and so on all share a number of properties that make them well-equipped for a maximally onchain implementation.

To start, these games involve management of resources, economies, and flows of value. This is essentially crypto in a nutshell:

• Buildings or workers generate resources (token emissions; NFT mints)…

• …which are then converted into other resources (swapped; wrapped)…

• …or spent (burned; sent to a treasury; locked in a contract)…

• …on more buildings, more workers, or other upgrades (NFTs; SBTs).

“[T]hese games are all about iteration and planning for longevity…[they] speak to the very core of what blockchain is all about: value tracking, provenance, asset ownership.”

Land, Labor, and Capital - @LLCgame_

This type of gameplay also tends to be slower paced, with a certain amount of idle time inherent to the model. Players typically wait for resources to accumulate before spending them on upgrades. Consider the energy mechanic in Pirate Nation, or mining in Primodium, as two straightforward examples.

Factory sims like Primodium work especially well because they are all about optimization. In a fully onchain environment where every move may potentially have a cost associated with it, as well as some degree of permanence, optimization becomes very important.

One challenge that fully onchain simulation games will need to take into consideration is the tendency to not have clear goals or win conditions. Rarely will a simulation game ask you to best another player; rather, players are typically tasked with building something within the context of ongoing processes. Any combat or competition comes secondary to the ongoing management of the simulation.

Of course, this isn’t necessarily a bad thing. Players can create their own goals, or scenarios can be provided for them, either by developers or community builders.

Consider a game like Dwarf Fortress. The incredible fandom this game has amassed over the years despite having no explicit player objectives (and for many years, an ASCII-only frontend) speaks to the potential of open-ended simulations as long-lived sources of fun. [3]

Part of why Dwarf Fortress and other open-ended simulation games can be so broadly appealing is due to their malleability as narrative vehicles. Dwarf Fortress, RollerCoaster Tycoon, The Sims, and many other games in the simulation genre present players with intriguing emergent storytelling capabilities by providing just enough of a narrative foundation to get started.

From there, players begin to ascribe their own complex meaning and context to otherwise simple in-game events — what Rimworld developer Tynan Sylvester refers to as apophenia. In doing so, players become co-authors of a game’s story alongside the simulation itself.

In an onchain simulation game, these types of emergent stories are no longer destined to be fleeting moments experienced by individual players. Rather, they may now be captured immutably through community consensus, and immediately rendered visible to all other players. This allows onchain worlds to spawn more than simply stories, but also shared histories. [4]

Another challenge onchain simulation games will need to overcome is the inherent transparency of blockchain data. If the logic of the simulation is freely available for all to see, it becomes much easier to “solve,” resulting in min/max-ing or other potentially unwanted behaviors.

Heavily invested players will always try to uncover how a simulation game works — after all, optimization is part of the fun — but keeping the inner workings relatively opaque is probably your best bet as a developer. This takes on added importance in the hyper-financialized world of blockchain games.

There are multiple solutions available to obfuscate a simulation’s code. For example, the developers might run the simulation offchain (gasp!), only verifying its validity onchain via zero-knowledge proofs or similar tech. Taking this idea to its logical extreme, it’s even possible to create a simulation that the developers themselves don’t know the specifics of. [5] Whatever the case, this appears to be a solvable problem.

Of course, simply reproducing existing simulation game tropes will not be enough. Onchain games must lean into what a decentralized medium uniquely enables, rather than solely focusing on overcoming technical barriers in an effort to achieve what’s already been done without a blockchain.

Developers seeking to innovate in this genre would do well to consider strategies that leverage the characteristics that make onchain games unique. For example, simulation games may be well-suited for the human/bot dynamic of onchain gaming. Because the player’s primary tasks in most simulation games are all about managing and prioritizing, it’s very easy to envision a mix of human-driven and automated strategies coexisting.

Applied to a multiplayer onchain gaming scenario, the colonists, city planners, or tycoon managers of “traditional” games could easily be replaced by human actors or bots alike, with each managing their own slice of a larger shared simulation. While a developer might choose allow each actor to manually control every single move, this could lead to throughput challenges. Instead, one might imagine a type of prioritized behavior tree or stack-ranking of rules, such as the system utilized in Shoshin.

Simulation games also benefit from the clientless nature of onchain gaming. Typically, the core simulation logic itself is sufficiently abstract enough to accommodate a wide variety of possibilities for integration of third-party assets and integrations. New tilesets, custom units, or bespoke clients could be attached to any existing onchain simulation, enabling deep, mod-like customization.

Furthermore, game extensions can be easily created to enhance the simulation experience. Fans of simulation games are accustomed to having many different information panes and overlays at their disposal. In an onchain game, these sources of information could come from the game’s developers and community members alike, similar to tools like Mobalytics in League of Legends.

In practice, a lean development team could even forego building these quality-of-life improvements entirely, instead enabling (or even incentivizing) community builders to create their own extensions for the benefit of all. By doing so, developers can augment their own bandwidth while simultaneously adopting a community-first mindset and gathering player feedback early on.

1. Importantly, the full set of onchain games surveyed in the Game7 report totals less than 100, making it difficult to draw meaningful conclusions from such a small sample size. Nevertheless, it’s great to see this sort of data being collected, as it’s sorely needed!

2. Note that I do not have access to the full Game7 dataset, so in all cases where I ascribe a specific genre to an existing title, I am making assumptions. I think they’re all pretty safe assumptions, but if you believe I am mistaken, please let me know at DarkTunnelsNewsletter@gmail.com.

3. I’m probably going to write an entire edition dedicated to Dwarf Fortress and the lessons onchain game developers can learn from that game. If you’re building Dwarf Fortress onchain, get in touch!

4. For deeper thoughts on this topic, check out Large Lore Models by dmstfctn, Eva Jäger, and 

Maybe I’m projecting my own desires as a fan of simulation games, but I truly believe that this is an important genre to watch in the onchain games ecosystem moving forward.

And I’m not the only one. We have already seen many web3 developers (both fully onchain and “web2.5”) attempt to build simulation games. While the results have been mixed (though certainly no more or less so than the rest of web3 gaming), one of the biggest blockchain games to-date has come out of this genre: Sorare. There’s also been Zed Run, Sunflower Land, 

However, those simulation games that opt to build maximally onchain will be the ones best positioned to capitalize on the aspects uniquely enabled by blockchains, which is where true differentiation will occur.

Thanks for reading.

Hello and welcome to Dark Tunnels, a newsletter dedicated to exploring the emerging ecosystem of fully onchain games. Dark Tunnels is published by Always Scheming, a research consultancy focused on the future of interactive entertainment.

The following essay is Issue #10 of Dark Tunnels. Enjoy!

Recently, web3 gaming organization Game7 released their “State of Web3 Gaming Report.”

The report is notable for a number of reasons, not least of which being that it actually attempts to size the existing fully onchain games ecosystem as a separate, distinct segment from the rest of web3. [1] It’s also a thorough and sober overview of the current web3 gaming landscape. I highly recommend giving it a read.

One particular item that caught my eye was the genre breakdown within fully onchain games, below:

The first and most obvious takeaway is that Strategy is currently the dominant genre in fully onchain gaming.

I suspect that this fits most people’s priors, but it’s good to see data to back it up. Many of the most prominent games in the onchain gaming ecosystem are strategy games — titles such as Mithraeum, Citadel, and Dark Forest, to name a few. [2]

It makes sense that strategy games dominate: the genre typically requires fewer player actions (and therefore much lower transaction throughput) than, say, action or sports games, making it a good fit for the technical limitations imposed by building everything on a blockchain.

I believe strategy games will continue to maintain their lead in 2024. After all, there is likely to be some survivorship and confirmation bias involved. Given that everyone is still trying to figure out the best approach to onchain game development, I would not be surprised to see more developers borrow learnings from the most widely-known (relatively speaking) fully onchain games today. Many of these games’ creators have also been quite open about their approaches to development, with some even open-sourcing their code entirely.

However, the genre I am most focused on is Simulation, currently tied for third-most in Game7’s dataset at roughly 10% of all fully onchain titles. Examples of fully onchain simulation games that you can play today include Primodium, Land, Labor, and Capitol, and the recently updated Soccerverse.

In my view, simulation games are a great match for onchain gaming. Construction and management sims, city builders, tycoon games, sport management simulators, and so on all share a number of properties that make them well-equipped for a maximally onchain implementation.

To start, these games involve management of resources, economies, and flows of value. This is essentially crypto in a nutshell:

• Buildings or workers generate resources (token emissions; NFT mints)…

• …which are then converted into other resources (swapped; wrapped)…

• …or spent (burned; sent to a treasury; locked in a contract)…

• …on more buildings, more workers, or other upgrades (NFTs; SBTs).

“[T]hese games are all about iteration and planning for longevity…[they] speak to the very core of what blockchain is all about: value tracking, provenance, asset ownership.”

Land, Labor, and Capital - @LLCgame_

This type of gameplay also tends to be slower paced, with a certain amount of idle time inherent to the model. Players typically wait for resources to accumulate before spending them on upgrades. Consider the energy mechanic in Pirate Nation, or mining in Primodium, as two straightforward examples.

Factory sims like Primodium work especially well because they are all about optimization. In a fully onchain environment where every move may potentially have a cost associated with it, as well as some degree of permanence, optimization becomes very important.

One challenge that fully onchain simulation games will need to take into consideration is the tendency to not have clear goals or win conditions. Rarely will a simulation game ask you to best another player; rather, players are typically tasked with building something within the context of ongoing processes. Any combat or competition comes secondary to the ongoing management of the simulation.

Of course, this isn’t necessarily a bad thing. Players can create their own goals, or scenarios can be provided for them, either by developers or community builders.

Consider a game like Dwarf Fortress. The incredible fandom this game has amassed over the years despite having no explicit player objectives (and for many years, an ASCII-only frontend) speaks to the potential of open-ended simulations as long-lived sources of fun. [3]

Part of why Dwarf Fortress and other open-ended simulation games can be so broadly appealing is due to their malleability as narrative vehicles. Dwarf Fortress, RollerCoaster Tycoon, The Sims, and many other games in the simulation genre present players with intriguing emergent storytelling capabilities by providing just enough of a narrative foundation to get started.

From there, players begin to ascribe their own complex meaning and context to otherwise simple in-game events — what Rimworld developer Tynan Sylvester refers to as apophenia. In doing so, players become co-authors of a game’s story alongside the simulation itself.

In an onchain simulation game, these types of emergent stories are no longer destined to be fleeting moments experienced by individual players. Rather, they may now be captured immutably through community consensus, and immediately rendered visible to all other players. This allows onchain worlds to spawn more than simply stories, but also shared histories. [4]

Another challenge onchain simulation games will need to overcome is the inherent transparency of blockchain data. If the logic of the simulation is freely available for all to see, it becomes much easier to “solve,” resulting in min/max-ing or other potentially unwanted behaviors.

Heavily invested players will always try to uncover how a simulation game works — after all, optimization is part of the fun — but keeping the inner workings relatively opaque is probably your best bet as a developer. This takes on added importance in the hyper-financialized world of blockchain games.

There are multiple solutions available to obfuscate a simulation’s code. For example, the developers might run the simulation offchain (gasp!), only verifying its validity onchain via zero-knowledge proofs or similar tech. Taking this idea to its logical extreme, it’s even possible to create a simulation that the developers themselves don’t know the specifics of. [5] Whatever the case, this appears to be a solvable problem.

Of course, simply reproducing existing simulation game tropes will not be enough. Onchain games must lean into what a decentralized medium uniquely enables, rather than solely focusing on overcoming technical barriers in an effort to achieve what’s already been done without a blockchain.

Developers seeking to innovate in this genre would do well to consider strategies that leverage the characteristics that make onchain games unique. For example, simulation games may be well-suited for the human/bot dynamic of onchain gaming. Because the player’s primary tasks in most simulation games are all about managing and prioritizing, it’s very easy to envision a mix of human-driven and automated strategies coexisting.

Applied to a multiplayer onchain gaming scenario, the colonists, city planners, or tycoon managers of “traditional” games could easily be replaced by human actors or bots alike, with each managing their own slice of a larger shared simulation. While a developer might choose allow each actor to manually control every single move, this could lead to throughput challenges. Instead, one might imagine a type of prioritized behavior tree or stack-ranking of rules, such as the system utilized in Shoshin.

Simulation games also benefit from the clientless nature of onchain gaming. Typically, the core simulation logic itself is sufficiently abstract enough to accommodate a wide variety of possibilities for integration of third-party assets and integrations. New tilesets, custom units, or bespoke clients could be attached to any existing onchain simulation, enabling deep, mod-like customization.

Furthermore, game extensions can be easily created to enhance the simulation experience. Fans of simulation games are accustomed to having many different information panes and overlays at their disposal. In an onchain game, these sources of information could come from the game’s developers and community members alike, similar to tools like Mobalytics in League of Legends.

In practice, a lean development team could even forego building these quality-of-life improvements entirely, instead enabling (or even incentivizing) community builders to create their own extensions for the benefit of all. By doing so, developers can augment their own bandwidth while simultaneously adopting a community-first mindset and gathering player feedback early on.

1. Importantly, the full set of onchain games surveyed in the Game7 report totals less than 100, making it difficult to draw meaningful conclusions from such a small sample size. Nevertheless, it’s great to see this sort of data being collected, as it’s sorely needed!

2. Note that I do not have access to the full Game7 dataset, so in all cases where I ascribe a specific genre to an existing title, I am making assumptions. I think they’re all pretty safe assumptions, but if you believe I am mistaken, please let me know at DarkTunnelsNewsletter@gmail.com.

3. I’m probably going to write an entire edition dedicated to Dwarf Fortress and the lessons onchain game developers can learn from that game. If you’re building Dwarf Fortress onchain, get in touch!

4. For deeper thoughts on this topic, check out Large Lore Models by dmstfctn, Eva Jäger, and 

Maybe I’m projecting my own desires as a fan of simulation games, but I truly believe that this is an important genre to watch in the onchain games ecosystem moving forward.

And I’m not the only one. We have already seen many web3 developers (both fully onchain and “web2.5”) attempt to build simulation games. While the results have been mixed (though certainly no more or less so than the rest of web3 gaming), one of the biggest blockchain games to-date has come out of this genre: Sorare. There’s also been Zed Run, Sunflower Land, 

However, those simulation games that opt to build maximally onchain will be the ones best positioned to capitalize on the aspects uniquely enabled by blockchains, which is where true differentiation will occur.

Thanks for reading.