Dwarf Fortress: A Case Study in Game Design

Dwarf Fortress. If you’ve heard of it, you’ve probably heard of how hard and unapproachable it is. What I find more interesting about it is how it has shed many of the things that people have come to expect a game to be. At the same time, it remains one of the most captivating game experiences ever created. For a long time now, I’ve wanted to write something about it to share some thoughts on why that might be.

The first thing that people notice about dwarf fortress is its graphics and UI, or, that is to say, its complete lack of graphics and UI design. Most people have come to expect games to be at the vanguard of user experience, and graphics and games seem to exist in a world where one cannot be without the other. It’s at this first barrier that many people are turned away from the game.

But the graphic and user experience aren’t the only (or even the most fundamental) assumptions about games that Dwarf Fortress forgoes. It does without things like creating “choices” for the player, designing mechanics to let the player feel powerful, or creating reward loops to manipulate the player into feeling good about spending their time playing.

Now you may be saying that Dwarf Fortress has all of these things, and it’s true that all of these things exist in the game. The fundamental difference is that these concepts aren’t placed there intentionally, but are instead emergent properties that come from a much more deeply important goal for the game. The most important thing, the thing that captures the attention of its fans, is the huge volume of narrative space to explore.

I’ve talked about narrative space in the past (I’ll do it again in the future), but for people who are confused by the term, here’s a brief rundown. A narrative space is the fundamental feature that defines a game and differentiates it from other narrative mediums. It is the collection of all possible narratives that can be created given a state, and a set of rules for interacting with that state. In other words, it’s the set of “world lines” through a game’s configuration space (if you’re into that kind of talk). In other other words, as you play a game, the sequence of events that happen in the game by interacting with the rules forms a narrative. The narrative space is all the different sequences that can come from playing a specific game.

In Dwarf Fortress, the number of different narratives to explore is ENORMOUS. It’s larger than any other game I’ve ever experienced, because the game creators have been specifically spending their time to make it that way.

This narrative space is where literally all of Dwarf Fortress’s value comes from. The game is given away for free, but there are players that find the game valuable enough to just donate money so development can continue. Remember, this is a game developed by two brothers, with no marketing budget, no graphics, and basically the same user interface as the vim text editor. Compare that to the many, many, many games on Kickstarter that fail to reach funding goals with the advantages of being pretty, and having a budget.

Dwarf Fortress shows us that many of the things that people believe must be designed into a game are really just emergent properties of a narrative space that has many compelling narratives inside it.

There is no system in Dwarf Fortress that was created to reward a player for doing specific tasks. Instead, the game models tasks and behaviors that are interesting to the players, and as players interact with the simulation, they are rewarded when things that are intrinsically interesting to them happen.

Many games assume that they must manipulate players into wanting to play more by exploiting reward circuits in players’ brains. In reality, players are motivated to play games because the process of voluntarily exploring the different stories that a game can produce, and finding the ones that appeal to the player is a feedback loop. The player chooses what to want on their own, then explores the narrative space to find it. When the player finds what they want, they feel so good that they go back to exploring the game. Repeat. The designers didn’t have to tell the players what to want, because the player just decided that something that was in the game was what they wanted.

The game doesn’t care about player choice, but in order to have a large narrative space, there must first be many different ways of interacting with the state. A large narrative space will necessarily give the player many different choices while playing. This is because, in order to create a larger narrative space, there must be more game interactions, that is, rules.

A metaphor for this is to imagine that a game is a two dimensional plane. The rules are the things that define the two axises, and specific game states are represented as points on the plane. Adding another rule to the game is akin to adding another axis to our spatial model. Now there are many different planes, and the number of different states was dramatically increased. The player has many more rules and interactions to play around with, narratives to choose between.


It’s commonly accepted that a game should be designed in such a way to let the player feel empowered and assertive. Often this means that the game will be easily approachable, will allow the player to quickly understand it and how to do things, and has “easy parts” that are created purely to help reinforce the sense of empowerment.

Dwarf Fortress does none of this. Its UI is great for players who already know all the keystrokes, but it’s horrible for helping new players feel like they know what they’re doing. The game will often randomly throw impossible challenges at a player, and there are hundreds of little ways that a fortress can go belly up without the player noticing until it’s too late. Yet the word that players have chosen to describe all of these barriers and things that can go wrong is “Fun” (with a capital ‘f’). They really mean it, because they just keep coming back for more. With every new attempt, something new is learned, some previous barrier overcome, and it’s EXTREMELY rewarding. This can only happen because the designers decided to focus on creating a large narrative space over “balance” and “player progression”. Compare that with bland industry games where you know what to do from the beginning, and the only thing that changes is the size of explosions on screen.

In the end, playing Dwarf Fortress is an extremely fresh experience that really stands in stark contrast to the shovelware that dominates the AAA games industry. It’s a dramatic case study into what really matters in a game, and how to achieve it. Luckily, with games like The Legend of Zelda: Breath of the Wild, it looks like some of the good ideas that inspired Dwarf Fortress are finding their way into the mainstream industry. Here’s hoping for a revolution in the way we understand games, and to the role that Dwarf Fortress helped play in bringing it about.

One thought on “Dwarf Fortress: A Case Study in Game Design

  1. Todd, great read. I’ve always been aware of this game but have never played it. The philosophy of this game I intrigues me to the point of downloading it. Maybe not playing it but at least downloading it so I can play another time. Thanks for putting this out there.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s