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 #9 of Dark Tunnels. This edition, along with the prior edition (#8: Finding Forever(Games)), are centered on the idea of the “forever” game and its place in the fully onchain games discourse. Enjoy!

In the podcast miniseries Gamecraft, legendary games investor Mitch Lasky provides a useful framework for evaluating what he and his Benchmark Capital colleague Blake Robbins refer to as “forever games.”

The whole nine-episode run of the podcast is well worth a listen, regardless of your level of familiarity with the topics presented. Mitch is an excellent storyteller and shares all sorts of fascinating anecdotes from his time as an operator and investor in the games industry.

What makes this specific episode (“Episode 004 - The Forever Games”) particularly salient to our discussion of onchain games is the introduction of Mitch’s “five attributes of long-term engagement.”

“I really do believe that there are attributes of these forever games…whether or not a game has the bone structure, essentially, to become a forever game”

Mitch Lasky, Gamecraft (Episode 004 - The Forever Games)

While successful forever games don’t need to be top-tier in every single category, they must (according to Lasky) be strong in at least one or two of the following areas:

• Balance

• Competition

• Community

• Eventfulness

• Player Investment

Given the aspiration that many onchain builders have expressed around creating truly immortal games, I thought it’d be useful to unpack this framework in greater detail. Let’s tease apart these defining characteristics and examine some of the considerations that go into each one:

Lasky uses “Balance” to describe games in which all players have a fair shot at winning the game. Interestingly, Lasky also makes a connection to the balance between players’ boredom and anxiety, otherwise known as “flow.”

Normally, balance in a multiplayer experience is achieved through a continual series of updates, patches, and stat changes to the core game, shifting the “meta” from one weapon, character, or team composition to another.

In single player games, this amounts to a vacillation between difficult challenges (tricky puzzles, overwhelming boss fights, etc.) and easier or slower-paced interludes (cutscenes, narrative exposition, exploration, and so on).

In all cases, balance is often specific to each individual player, requiring careful planning and extensive playtesting in order to deliver the optimal experience.

When applied to onchain games, a few challenges immediately spring to mind. For one, the idea of continuous balance updates and game patches is fundamentally at odds with immutable smart contracts.

“If any entity can change the rules of an on-chain game, it becomes a player so powerful that no mechanics other than the mechanics of changing the rules really matter.”

Bohdan Melnychuk, Founder - Mithraeum

Of course, a developer’s first game need not be their forever game. Pivots and iteration are inevitable. However, while it is possible to create updated contracts and then migrate your game’s community from the old version of the game to a new one, this is likely easier said than done.

Another challenge to creating a balanced forever game comes with the addition of bots. How can any game ensure balance when humans and bots operate in the same arena?

One potential solution to this is to make the game inherently social, leaning heavily into the most “human” elements of gaming: activities like negotiation, social deduction, cooperation, and the like. Yet even these areas may be susceptible to competitive imbalance over the long run. Early experiments with Meta’s CICERO AI and turn-based strategy game Diplomacy give us an idea of just how convincing non-human players may soon become.

Note also that balance will always be a moving target in a game or virtual world that is composable and extensible. Even if the base set of rules is perfectly balanced, some players may still gain an edge through the use of bots or other automations, more efficient UIs, information marketplaces (e.g. Nightmarket in Dark Forest), and so on.

Competition encompasses more than just player-versus-player games. Yes, competing against others is a key element, but there are many other components that go into fostering healthy competition. [1]

Matchmaking, lobby composition, player ranking systems, leaderboards, and other features such as these must all be taken into consideration. Some of this overlaps with the previous discussion around balance, but other aspects exist, too.

For example, is the game’s competition direct (i.e. one player’s performance impacts that of another) or indirect (e.g. high score leaderboards, time trials, community goals, etc.)? Where do players encounter competitors? How is competition rewarded at various skill levels? What are the stakes of the competition?

In both onchain and traditional games, competition can take many forms. The way competition functions in-game is likely similar in either scenario, but the means by which it is achieved is likely more challenging in a fully onchain environment.

For example, I’m not aware of any games-specific matchmaking algorithms that operate entirely onchain, but that seems like a nontrivial technical challenge. [2] There is no trusted moderator or centralized service pairing opponents of similar skill levels together. As such, skilled players will almost always seek out weaker or more vulnerable opponents, as opposed to equally skilled competitors. When surveying the current crop of onchain games available today, many have opted for a gigantic shared “world map” or global view as a sort of proxy for matchmaking, as this format lends itself well to transparency and ease of understanding.

Given the inherent and inevitable financialization of onchain games, it is likely that competition will often be realized in the form of high stakes, skill-based gaming. I’m speculating here, but my assumption is that this form of competition will allow players to self-select based on risk tolerance and personal liquidity, rather than forcing matches based on skill level or tenure. I offer no judgment on the validity of this approach (I can see both pros and cons to this form of competition), but would simply point out that it is fundamentally different from traditional gaming and should be planned for accordingly by both players and developers.

This one is pretty straightforward: the ability to form relationships with other people who play the game. We’ve all heard of or experienced our own tales of meeting people, building friendships, or even falling in love in the context of virtual worlds. This is frequently referenced as a positive emergent phenomenon of gaming’s continued proliferation and is easy to understand in the context of forever games.

Social bonds are extremely sticky and can help to keep players retained and engaged far longer than they might otherwise be. Personally, I probably would have (rage)quit playing fantasy football or Call of Duty years ago were it not for the group of friends I play with. I’m sure many readers will have similar examples from their own lives.

In blockchain gaming more broadly, community-building seems to happen largely via established social media channels (Twitter, Discord, Telegram, etc.), rather than through in-game features. That said, there are definitely products out there seeking to bring this behavior onchain, such as Summon.xyz.

Because of the interoperability of tokenized assets and the composability of smart contracts, it is also possible that social systems or community platforms traditionally contained within a game client — features like clans, friends lists, group chats, and so on — will never need to be developed from scratch by onchain game studios. Rather, products and protocols like the aforementioned Summon, Farcaster, Lens, or perhaps even gaming guilds might one day fulfill those needs.

Eventfulness represents perhaps the largest departure from the traditional model of gaming. In the pre-Internet premium games era of $60 boxes, whatever developers put on the game disc or cartridge was all there was; the game did not change after “going gold,” nor was there any expectation that developers keep it fresh.

Nowadays, events (and more broadly, live operations, a.k.a. “Live Ops”) are critical to engaging players over prolonged periods of time. Seasonal updates, brand collaborations, new character packs, and so forth are all planned well in advance and released to players with specific goals of driving engagement, reactivating lapsed players, and encouraging additional spend.

Live Ops is perhaps the most overlooked aspect of fully onchain game development today. To my knowledge, only Pirate Nation has even attempted something like this to-date.

Once onchain games achieve greater scale and adoption, Live Ops may be another area ripe for unbundling and decentralization. Contributors to permissionless gaming ecosystems could feasibly create their own events, aggregate and arbitrage resources via sales or bundles, or contribute and monetize additional content that expands the game world in new and interesting ways.

As a simple hypothetical, consider the aforementioned Pirate Nation. Recently, the game underwent a spooky, Halloween-inspired transformation, complete with various themed cosmetic items for players’ customizable islands. In that same vein, what’s to stop passionate creators from contributing (or even selling) their own themed cosmetics for other occasions? Say, Carnivale-themed cosmetics from the Brazilian community, or Diwali-inspired decorations from Desi creators?

I’m sure that I’m glossing over technical implementation challenges here, but I think you get the idea: a properly motivated and enabled community can be encouraged and empowered to grow a game’s ecosystem in ways that the original development team may not ever have even imagined.

In theory, these sorts of contributions can even be professionalized into meaningful businesses. An interesting comparison here is Chess.com. Chess is fully decentralized — no one truly owns or updates the game of chess itself. Yet the developers of Chess.com have managed to build a robust business around the game — one that generates north of 

Lasky identifies “Player Investment” as opportunities for a game’s players to spend their time and/or money in building up skills and assets. This could take the form of in-game progression, where XP grind or currency sinks allow for characters and items to level up. It might also be something less tangible, such as investing time in becoming more skilled at a competitive game, resulting in higher ELO scores or player rankings.

In web3 gaming, player investment is most frequently associated with purchasing or earning NFT assets. Importantly, spending heavily on hefty gas fees likely doesn’t count, unless perhaps you’re “investing” in block priority and learning how to front-run competing transactions.

I’ve written previously about revenue streams and questions of what to sell in the On Business Models miniseries (most notably, in Part 4), so I won’t spend much time on it here. However, it’s worth highlighting something that Lasky himself calls out in Gamecraft, which is that player investment must accommodate different styles of play and pay.

Gamers engage with virtual worlds in different ways and choose to spend their money on different things. Not only does this vary from player to player, but it also changes for the same player based on where they are in their journey within a given title. New players will almost always purchase different items and invest their time in different parts of a game than elder players will. [3]

An important aspect of player investment that seems particularly relevant to web3 games is the endowment effect. Given the emphasis on ownership of digital assets, the idea that players may perceive their investments (of both time and money) in blockchain-enabled games to hold greater value than similar investments in traditional games makes a lot of sense. This is a big reason why I am bullish on free-to-play approaches to blockchain games business models, such as the “free-to-own” strategy popularized by Limit Break.

1. I’m sure that some forever games might also be said to foster unhealthy competition (looking at you, League of Legends), but Lasky makes no differentiation here.

2. If you are aware of any, please let me know! You can get in touch by emailing DarkTunnelsNewsletter@gmail.com.

3. For a deeper dive on this concept, I highly recommend the “Tower of Want,” a seminal GDC talk from Ethan Levy, former Executive Producer at N3TWORK. This may be my single most recommended games industry talk — I tell everyone about this video.

Though onchain games may face greater technical and operational hurdles (what else is new?) in excelling at Lasky’s “five attributes of long-term engagement,” I see no reason why they can’t keep players entertained for just as long as other so-called forever games.

Forever games don’t necessarily have to be persistent, MMO-like virtual worlds. In fact, as Lasky and Robbins discuss, session-based games seem to have overtaken MMOs in popularity and many have succeeded in persisting for decades.

In some ways, “forever game” is not necessarily a label for a specific title, but rather a type of play pattern. For example, FPS games vs. any specific Call of Duty; RTS games vs. StarCraft 2 or Command & Conquer: Red Alert.

If a similarly long-lived play pattern can be replicated through onchain gaming, it may be even better equipped than its web2 forebears to stand the test of time.

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 #9 of Dark Tunnels. This edition, along with the prior edition (#8: Finding Forever(Games)), are centered on the idea of the “forever” game and its place in the fully onchain games discourse. Enjoy!

In the podcast miniseries Gamecraft, legendary games investor Mitch Lasky provides a useful framework for evaluating what he and his Benchmark Capital colleague Blake Robbins refer to as “forever games.”

The whole nine-episode run of the podcast is well worth a listen, regardless of your level of familiarity with the topics presented. Mitch is an excellent storyteller and shares all sorts of fascinating anecdotes from his time as an operator and investor in the games industry.

What makes this specific episode (“Episode 004 - The Forever Games”) particularly salient to our discussion of onchain games is the introduction of Mitch’s “five attributes of long-term engagement.”

“I really do believe that there are attributes of these forever games…whether or not a game has the bone structure, essentially, to become a forever game”

Mitch Lasky, Gamecraft (Episode 004 - The Forever Games)

While successful forever games don’t need to be top-tier in every single category, they must (according to Lasky) be strong in at least one or two of the following areas:

• Balance

• Competition

• Community

• Eventfulness

• Player Investment

Given the aspiration that many onchain builders have expressed around creating truly immortal games, I thought it’d be useful to unpack this framework in greater detail. Let’s tease apart these defining characteristics and examine some of the considerations that go into each one:

Lasky uses “Balance” to describe games in which all players have a fair shot at winning the game. Interestingly, Lasky also makes a connection to the balance between players’ boredom and anxiety, otherwise known as “flow.”

Normally, balance in a multiplayer experience is achieved through a continual series of updates, patches, and stat changes to the core game, shifting the “meta” from one weapon, character, or team composition to another.

In single player games, this amounts to a vacillation between difficult challenges (tricky puzzles, overwhelming boss fights, etc.) and easier or slower-paced interludes (cutscenes, narrative exposition, exploration, and so on).

In all cases, balance is often specific to each individual player, requiring careful planning and extensive playtesting in order to deliver the optimal experience.

When applied to onchain games, a few challenges immediately spring to mind. For one, the idea of continuous balance updates and game patches is fundamentally at odds with immutable smart contracts.

“If any entity can change the rules of an on-chain game, it becomes a player so powerful that no mechanics other than the mechanics of changing the rules really matter.”

Bohdan Melnychuk, Founder - Mithraeum

Of course, a developer’s first game need not be their forever game. Pivots and iteration are inevitable. However, while it is possible to create updated contracts and then migrate your game’s community from the old version of the game to a new one, this is likely easier said than done.

Another challenge to creating a balanced forever game comes with the addition of bots. How can any game ensure balance when humans and bots operate in the same arena?

One potential solution to this is to make the game inherently social, leaning heavily into the most “human” elements of gaming: activities like negotiation, social deduction, cooperation, and the like. Yet even these areas may be susceptible to competitive imbalance over the long run. Early experiments with Meta’s CICERO AI and turn-based strategy game Diplomacy give us an idea of just how convincing non-human players may soon become.

Note also that balance will always be a moving target in a game or virtual world that is composable and extensible. Even if the base set of rules is perfectly balanced, some players may still gain an edge through the use of bots or other automations, more efficient UIs, information marketplaces (e.g. Nightmarket in Dark Forest), and so on.

Competition encompasses more than just player-versus-player games. Yes, competing against others is a key element, but there are many other components that go into fostering healthy competition. [1]

Matchmaking, lobby composition, player ranking systems, leaderboards, and other features such as these must all be taken into consideration. Some of this overlaps with the previous discussion around balance, but other aspects exist, too.

For example, is the game’s competition direct (i.e. one player’s performance impacts that of another) or indirect (e.g. high score leaderboards, time trials, community goals, etc.)? Where do players encounter competitors? How is competition rewarded at various skill levels? What are the stakes of the competition?

In both onchain and traditional games, competition can take many forms. The way competition functions in-game is likely similar in either scenario, but the means by which it is achieved is likely more challenging in a fully onchain environment.

For example, I’m not aware of any games-specific matchmaking algorithms that operate entirely onchain, but that seems like a nontrivial technical challenge. [2] There is no trusted moderator or centralized service pairing opponents of similar skill levels together. As such, skilled players will almost always seek out weaker or more vulnerable opponents, as opposed to equally skilled competitors. When surveying the current crop of onchain games available today, many have opted for a gigantic shared “world map” or global view as a sort of proxy for matchmaking, as this format lends itself well to transparency and ease of understanding.

Given the inherent and inevitable financialization of onchain games, it is likely that competition will often be realized in the form of high stakes, skill-based gaming. I’m speculating here, but my assumption is that this form of competition will allow players to self-select based on risk tolerance and personal liquidity, rather than forcing matches based on skill level or tenure. I offer no judgment on the validity of this approach (I can see both pros and cons to this form of competition), but would simply point out that it is fundamentally different from traditional gaming and should be planned for accordingly by both players and developers.

This one is pretty straightforward: the ability to form relationships with other people who play the game. We’ve all heard of or experienced our own tales of meeting people, building friendships, or even falling in love in the context of virtual worlds. This is frequently referenced as a positive emergent phenomenon of gaming’s continued proliferation and is easy to understand in the context of forever games.

Social bonds are extremely sticky and can help to keep players retained and engaged far longer than they might otherwise be. Personally, I probably would have (rage)quit playing fantasy football or Call of Duty years ago were it not for the group of friends I play with. I’m sure many readers will have similar examples from their own lives.

In blockchain gaming more broadly, community-building seems to happen largely via established social media channels (Twitter, Discord, Telegram, etc.), rather than through in-game features. That said, there are definitely products out there seeking to bring this behavior onchain, such as Summon.xyz.

Because of the interoperability of tokenized assets and the composability of smart contracts, it is also possible that social systems or community platforms traditionally contained within a game client — features like clans, friends lists, group chats, and so on — will never need to be developed from scratch by onchain game studios. Rather, products and protocols like the aforementioned Summon, Farcaster, Lens, or perhaps even gaming guilds might one day fulfill those needs.

Eventfulness represents perhaps the largest departure from the traditional model of gaming. In the pre-Internet premium games era of $60 boxes, whatever developers put on the game disc or cartridge was all there was; the game did not change after “going gold,” nor was there any expectation that developers keep it fresh.

Nowadays, events (and more broadly, live operations, a.k.a. “Live Ops”) are critical to engaging players over prolonged periods of time. Seasonal updates, brand collaborations, new character packs, and so forth are all planned well in advance and released to players with specific goals of driving engagement, reactivating lapsed players, and encouraging additional spend.

Live Ops is perhaps the most overlooked aspect of fully onchain game development today. To my knowledge, only Pirate Nation has even attempted something like this to-date.

Once onchain games achieve greater scale and adoption, Live Ops may be another area ripe for unbundling and decentralization. Contributors to permissionless gaming ecosystems could feasibly create their own events, aggregate and arbitrage resources via sales or bundles, or contribute and monetize additional content that expands the game world in new and interesting ways.

As a simple hypothetical, consider the aforementioned Pirate Nation. Recently, the game underwent a spooky, Halloween-inspired transformation, complete with various themed cosmetic items for players’ customizable islands. In that same vein, what’s to stop passionate creators from contributing (or even selling) their own themed cosmetics for other occasions? Say, Carnivale-themed cosmetics from the Brazilian community, or Diwali-inspired decorations from Desi creators?

I’m sure that I’m glossing over technical implementation challenges here, but I think you get the idea: a properly motivated and enabled community can be encouraged and empowered to grow a game’s ecosystem in ways that the original development team may not ever have even imagined.

In theory, these sorts of contributions can even be professionalized into meaningful businesses. An interesting comparison here is Chess.com. Chess is fully decentralized — no one truly owns or updates the game of chess itself. Yet the developers of Chess.com have managed to build a robust business around the game — one that generates north of 

Lasky identifies “Player Investment” as opportunities for a game’s players to spend their time and/or money in building up skills and assets. This could take the form of in-game progression, where XP grind or currency sinks allow for characters and items to level up. It might also be something less tangible, such as investing time in becoming more skilled at a competitive game, resulting in higher ELO scores or player rankings.

In web3 gaming, player investment is most frequently associated with purchasing or earning NFT assets. Importantly, spending heavily on hefty gas fees likely doesn’t count, unless perhaps you’re “investing” in block priority and learning how to front-run competing transactions.

I’ve written previously about revenue streams and questions of what to sell in the On Business Models miniseries (most notably, in Part 4), so I won’t spend much time on it here. However, it’s worth highlighting something that Lasky himself calls out in Gamecraft, which is that player investment must accommodate different styles of play and pay.

Gamers engage with virtual worlds in different ways and choose to spend their money on different things. Not only does this vary from player to player, but it also changes for the same player based on where they are in their journey within a given title. New players will almost always purchase different items and invest their time in different parts of a game than elder players will. [3]

An important aspect of player investment that seems particularly relevant to web3 games is the endowment effect. Given the emphasis on ownership of digital assets, the idea that players may perceive their investments (of both time and money) in blockchain-enabled games to hold greater value than similar investments in traditional games makes a lot of sense. This is a big reason why I am bullish on free-to-play approaches to blockchain games business models, such as the “free-to-own” strategy popularized by Limit Break.

1. I’m sure that some forever games might also be said to foster unhealthy competition (looking at you, League of Legends), but Lasky makes no differentiation here.

2. If you are aware of any, please let me know! You can get in touch by emailing DarkTunnelsNewsletter@gmail.com.

3. For a deeper dive on this concept, I highly recommend the “Tower of Want,” a seminal GDC talk from Ethan Levy, former Executive Producer at N3TWORK. This may be my single most recommended games industry talk — I tell everyone about this video.

Though onchain games may face greater technical and operational hurdles (what else is new?) in excelling at Lasky’s “five attributes of long-term engagement,” I see no reason why they can’t keep players entertained for just as long as other so-called forever games.

Forever games don’t necessarily have to be persistent, MMO-like virtual worlds. In fact, as Lasky and Robbins discuss, session-based games seem to have overtaken MMOs in popularity and many have succeeded in persisting for decades.

In some ways, “forever game” is not necessarily a label for a specific title, but rather a type of play pattern. For example, FPS games vs. any specific Call of Duty; RTS games vs. StarCraft 2 or Command & Conquer: Red Alert.

If a similarly long-lived play pattern can be replicated through onchain gaming, it may be even better equipped than its web2 forebears to stand the test of time.

Thanks for reading.