Simulate your very own fantasy universe: Looking at a decade of Dwarf Fortress

Dwarf Fortress is one of the most ambitious and non-traditional games being developed in the modern era. It’s essentially the closest you can get to a 'free will' simulator when interacting with a computer. On the surface, it looks like a bunch of ASCII characters flashing rapidly around a screen with a few menus; charmingly dated visual nonsense at best. To the initiated, those colored characters mean so much more. Behind the primitive art-style and somewhat dated controls, the most detailed simulation of a fantasy world ever conceived is pulsating with life. Everything from your dwarves' moods and relationships to the mineral properties of the soil, temperature and fluid dynamics enjoy as faithful a simulation as is possible in ASCII graphics in Bay 12’s genre and trope-defying title. Minutiae as small as musculoskeletal interaction of combatants during fights, the density of the material your hammer is made from, and the interaction of the mucous tissues in eyelids on eyeballs enjoy mechanical consideration in the Dwarf Fortress universe.

Nearly everything in DF, from the history and storytelling to the appearance and personalities of your dwarves is procedurally generated, governed by behavioral and physical properties rather than scripted interaction or set rules. This gives the player and amount of freedom and choice that makes Gary’s mod look like Papers, Please. We’re not talking about choice in the narrative sense, either, where the player is made to feel like they’re making a difference when all of their actions will cumulatively result in an explosion that looks a bit different when they beat the game. We’re talking, raw, unadulterated, unbound decision making. Naturally, the more creative players have taken advantage of this freedom to weaponize, well, just about everything. Catapulting large animals into hordes of invading goblins? Check. Auto-firing super weapons that hurl dozens of spears downrange? No problem. The medieval equivalent of a high pressure firehose? You got it, boss. There are even players that start chemical and biological weapons programs that would make the Albert Wesker Blush. You want to build a monster trap using nothing but high quality furniture and some structural engineering? You bet you can. Ultraviolence not your thing? You can also engineer mechanical and structural marvels, or follow your Dwarves’ day-to- days and interpersonal relationships with a level of detail so granular you’d have to dumb it down an order of magnitude before it could be compared to the Sims.

The beauty of all of this is that none of it is planned. there’s no “craft elaborate, Rube-Goldberg death trap” button. It’s just things that the game allows through simulation of it’s own little reality. The player is in effect, god of his own pocket universe where the main intelligent form of life are squat alcoholic socipaths. As you can imagine, this kind of power attracts a vibrant and diverse cult following, still highly active to date despite a ten year plus beta dev cycle.

You may not have heard of Dwarf Fortress, but if any of this sounds even vaguely familiar, it’s because it’s been a powerful influence on the gaming industry during it’s quiet decade of development, spawning industry titans like Minecraft, and pioneering procedural game development as we know it today. Despite the massive scope of the game and long development cycle, The Adams Brothers (Tarn and Zach) responsible for both say the game is less than halfway done.

Last month marked the ten-year anniversary of full-time development on DF for Bay 12 Games, the Brothers’ dev house, of which they are the sole employees. We sat down with them to discuss what that decade has been like, and how they plan on handling the long road ahead. Here’s the meat of that conversation:

Bay Twelve Games is not a traditional game development studio, in much the same way as Dwarf fortress isn't a traditional game. The concept for DF has been around since just after the turn of the century and seen full development for just over a decade, which fans of the game celebrated at the Dwarf Moot event held last month. Tell us, what does a decade of development feel like, and did you envision the .43 release would have the scope that it does starting out?
Haha, no we didn't have any idea it would come this far. Dwarf Fortress started out as a mining game with a high score list, and now the dwarves are doing procedural dance steps. It's cool to still be going with the game on track and a great community helping us along. The Adams Brothers
How does your relationship as brothers affect the Development process?
Having a vision of the overall world simulation project which we've shared for decades now makes planning and design go smoothly. I don't think we've had a real disagreement since a yelling match over 20 years ago over whether a bizarre gladiation game we were working on should show striated or color-coded muscles when the skin is removed. The Adams Brothers
I imagine that the planets must have had to align to keep you going for ten years. Can you tell a little about what got you two working on the game full time?
Asking for money helped! The first six years the website was up, we were just releasing small strange games for free while we worked on an ill-fated fantasy RPG. Then somebody wanted to give me birthday money over PayPal, and we were like, okay, sounds cool. That was four months before the first Dwarf Fortress release, and we made just enough to scrape by, so I was able to quit my job back in 2007. Zach was also able to go full-time some years back. There have been some close calls. Going on Patreon last year has really helped. The Adams Brothers

On the flip side, how do you think DF's long development cycle affects the fanbase? Ten years actively playing a game, while not nearly as transformative as the same time developing one, is quite an achievement.
Yeah, it's amazing, though I guess I can also think of a few games I come back to over the years. It's the same way with many of the players -- they can take a few months or a few years off, and they know it'll be there waiting for them when they come back, with whatever new features have been added. People get into different mods and add-ons, or they hang out on the forums even when they aren't playing. It really is a community of people, with different subgroups and a diaspora-like group of players that are currently away from the game, or who have never played it but still like reading the stories. The Adams Brothers
Bay 12 Games works on a donation-based funding model, and yet you've succeeded in creating possibly the most complex and vast game experience of all time. We're seeing this more and more in entertainment with the advent of streaming media platforms and crowdfunding, but you've been doing it successfully since the early 2000's. What do you think most contributes to that success despite relying on monetary support from your fans?
Dwarf Fortress is unique in several ways, so if people like the game, they can become pretty dedicated fans. The sense of community helps, and we've tried to support it as best we can. The press exposure we've gotten has been positive. The game is not a runaway financial success, but it works well enough that we can keep going, and the independence from a publisher etc. has allowed us to continue to work on the sort of game we want to make, so we're happy with the situation. The Adams Brothers
Dwarf Fortress doesn't have a story in the traditional sense, and if you were to use buzzwords or category tags to describe the game, you might boil it down to something like 'Fantasy Physics Simulation/Sandbox.' The game, however, has a lot of narrative elements that make it feel uniquely handcrafted and lifelike relative to other procedurally-generated game worlds. These elements make it much more than the sum of its parts in the eyes of many fans. Can you tell us what elements of the creative process go into making those elements a reality?
When we started out, Zach and I (and then mostly Zach) wrote and analyzed short stories of potential playthroughs for the game, and a lot of what we were thinking about there involved small narrative building blocks. Simple things like giving each dwarf a name and different interests can go a long way, and trying to make as faithful a simulation as possible so that the interactions happening in front of the player make sense. You don't have to get all the way there -- people are pretty creative and will fill in the blanks where necessary as long as they aren't put off immediately. If you think to yourself, "what is it about playing this game that I'd tell a friend afterward?", you're in the right frame of mind. The Adams Brothers

Despite the long development cycle of the game, you still describe it as being in early beta, and the development road map often describes shorter-term goals for the project. Is this due more to to the more holistic organic approach you have to the creation of DF, or the sheer scale and complexity of the title? Do you have a 'master plan' for what the 1.0 version of DF should look like?
We used to have more complete public development notes, but it became a pain to keep them 100% up to date as circumstances changed. We still have a detailed outline of the 1.0 version -- the current version number starting with 0.43 refers to being 43% done with 2600 "points" worth of development on that list. We do try to be as public as possible, so we post a more detailed list of the nearer term goals, but sometimes that can still be uncomfortable since it becomes necessary or helpful to reprioritize, and we've had some things up there for years that haven't been worked on. The Adams Brothers
With the continuing growth of DF, updating the game must be getting increasingly difficult to manage. Have you ever considered open-sourcing the project and delegating development in a style similar to, say, the FreeBSD foundation?
Haha, that wouldn't make it easier to manage! A whole new can of worms, and not one we have the skill set for. It would quickly be the death of the project, in terms of our involvement, even if the forks would be interesting. The Adams Brothers
Moving on to a more personal thread, DF has been cited by many as a major influence in scads of sandbox, roguelike, fantasy, and procedurally generated titles. We'd like to know what your influences are in developing DF, and not just the ones that stem from the gaming industry.
A forum-goer says we work from the most banal possible sources, and there's a lot of truth to that, he he he. Cheesy movies and cartoons (Ray Harryhausen stuff, Ralph Bakshi, the Conan movie, that sort of thing), but also historical tidbits (lots of Assyrian translations have come up from Zach's history studies) and stuff like the Silmarillion and the Romance of the Three Kingdoms. From games, mainly the Ultima series, 80s roguelikes and Starflight. The Adams Brothers
This is the stoned gamer after all, so I'm most likely legally obligated to ask: the Dwarves in the DF universe are quite literally functional alcoholics. In the same vein, have any mind-altering substances been used as aids in the creative process in making DF?
Just caffeine! We drink sometimes, but having fun adding bugs to the code is just wasted time you have to spend cleaning them up afterward, so it's not worth it. Sleep deprivation has come up... our terrible game Star Zoo was written in 36 hours straight without a nap, back when I was younger and could pull that off. The Adams Brothers
DF's main focus is on the Fortress and Adventure modes, but there's a lot more to get out of the game than the stock-standard interpretation of those game modes. I have on several occasions, for example used the world generation to browse a unique history that I can use as source material for tabletop roleplaying campaigns. Do you see more modes coming into the game down the road?
Pretty soon, we're going to add creation myth generation to the game, and I suspect that'll give rise to a bit more fun on the world generation side, though I'm not sure it'd qualify as a mode. The same goes for editors for site maps and so on. Adventure mode just got cabin building, and that has the potential to morph into some sort of dwarf/adventure mode hybrid. I suspect we'll see more line-blurring like that -- fortress dwarves will be able to go off-map in the next release. Our larger plans had stricter ideas for different modes, playing dragons and deities and human towns and so forth, but I'm not sure how it'll play out, versus the natural evolution of existing modes to other styles of play. The Adams Brothers

A lot of the draw of DF is the painstaking attention to detail when it comes to simulation. It creates a living breathing world that feels uniquely alive. What would you say is the smallest detail you’ve explicitly accounted for in the game?
Eyelids cleaning eyes when they are dirty?


I'm not even sure it works anymore, since tears seem to stay in the eyes. It's funny, similar code for cats cleaning themselves was what was causing all this alcohol poisoning in the tavern release, since they'd lick it off their paws and there was a bug with how much it affected them.

The Adams Brothers
Part of what makes dwarf fortress so popular is the amount and depth of emergent gameplay, you get a feeling of free will that runs deeper than most other games can manage because the consequences of your actions actually have more or less logical real world effects. So many things are simulated to achieve this that DF sometimes runs at a slideshow's pace on modern enthusiast PC hardware. The game thrives on undefined behavior because of the amount of faithful simulation going on as a result. Do you think we'll see this approach to creating choice more in traditional gaming as computing power increases?
I don't think computing power is the limiter on this, and it hasn't been for many years. DF's speed problems have more to do with my lack of skill -- more talented programmers can and do make more optimized code than I do. On the flipside, the limits on other games taking our approach have more to do with what risks larger developers are willing to take, and how practical it is to write a longer-term project. Throwing more resources at a game doesn't help with design, except around the edges with some tools perhaps, and a game like DF would have many interlocking systems that need to be carefully managed for the experience of playing the game to be enjoyable. It's possible this situation could change somewhat as additional tools are created (as with the physics engines and third-party procedural content generators), but it's not a simple problem of resources or processing power. The Adams Brothers

DF is one of the most modded games in existence. There are countless management and tweak applications that improve quality of life in the incredibly complex fortress management mode, mods that update the ASCII graphics with 2D tile based textures, mechanical and game feature extensions, full retheming and reworks of the game, and even plugins that allow isometric 3D graphics and high-resolution 3D renders of gameplay snapshots. You can even find massive bundles with GUIs that make managing all of these easy for the average Joe. All of this obviously wouldn't be possible had you taken a less liberal approach to distribution and end-user rights. As the driving force behind this vibrant and active modding community, how do you feel about the trend of locking down traditionally modded franchises by large development houses? (Snapmap in Doom, The Autumn Leaves mod theft controversy, and server centralization to enforce DLC purchasing over modding, for example.)?
I don't know very much about what's going on with mod horror stories in the larger industry, but more generally I think it's cool when creative people are allowed to play around with stuff. That said, DF is closed-source, so I can't say we're not also being restrictive. There's a balance people have to strike, and what happens can be forced by practical needs or perceptions of what's practical, anyway. That said, a lot of what I hear about seems Draconian, reactive and/or bizarre, but that's partially bias, based on what I read or what's likely to make the news. The Adams Brothers
The nature of DF's world generation engine makes it such that each player experiences their own canon, but the shared experiences of the community and 'Threetoe's Stories' (Narratives written by Zach or 'Threetoe' that guide the development of each game release) often guides the current of the collective 'head canon' have events like 'Something Awful's BoatMurdered,' or the meme of deadly sponges and ambulatory carp guided your design decisions in unexpected ways?
I'm sure this happens a lot, though it's hard to think of a really striking example. One thing we could bring to mind was the players attempting to summit mountains. The game didn't care at all, so we added an announcement and a historical event in the legends so people wouldn't be disappointed again. Mostly it's just people surprising us with their weaponization of game mechanics, like intentionally pouring vampire blood in a well in one section of the fortress to create a race of specialized soldiers separate from the rest of the population. The more creative fun they have, the more it encourages us to add in similar interactions and pay attention to possible connections in the code. The Adams Brothers


Ten years of development on the most complex game in existence, working on donations, and it’s about two-fifths done in the eyes of the creator. Whether you appreciate the game itself or the guiding influence it’s had on the industry, you have to wonder what the 57% of the game still in the minds of Zach and Tarn looks like. I for one look forward to revisiting Dwarf Fortress in 2030 for its 1.0 release, and all the extra fun that will entail.

Published in /Gaming

This email address is being protected from spambots. You need JavaScript enabled to view it.


SGL Players Online

We have 4441 guests and one member online

Latest Articles