Hacker News new | past | comments | ask | show | jobs | submit login
The Brilliance of Dwarf Fortress (nytimes.com)
442 points by pcestrada on July 22, 2011 | hide | past | web | favorite | 144 comments

A few things.

First, as mentioned in the article, the Museum of Modern Art in NYC will be featuring Dwarf Fortress as part of its "Talk to Me" exhibit. The exhibit will be open from July 24th through November 7th. http://www.moma.org/visit/calendar/exhibitions/1080

Second, for those looking to get into the game, the two recommended video tutorials are SippyCup's (humorous) http://www.youtube.com/51ppycup, and Captain Duck's (down to business) http://www.youtube.com/user/captnduck#p/a/u/1/yn1iW1QN7_s.


Tarn has The Fisherman Parable philosophy and I have a tremendous amount of respect for that. Doing what you love, no matter the "sacrifice", above everything else. He's probably the biggest reason why I've been working on the side to try to create my own games, so I can get the heck out of finance and do what I wanted to since I was 6 years old. Until that works out for me, I send a monthly donation to ensure that at least someone can fulfill their dream.

I tried DF awhile ago and liked it, but realized it would take more time than I was willing to put in it. Reading this article makes me want to try it again, just out of respect for Tarn's humble devotion and his pursuit of the joy of "just building things". I'll definitely check out the MoMA exhibit next week.

I used to play a lot of Dwarf Fortress but like you I realized that it takes a lot time. I ended up quitting the addiction because I needed to get some real work done. One thing about DF: there is never a lack of things to do with your world. In my last game I built an engraved obsidian cast tower. In Dwarf Fortress terms this means I built a powered pump system to extract magma from the earth and water from a nearby river and used the water and magma to cast obsidian for a stone tower which I then hollowed out with rooms and storage areas as a dwelling place for my dwarves. Around the tower I built a magma moat for defense from the marauding goblins and zombies. The fortress has survived more than 15 years of attacks so far. I just don't open it up for fear of getting hooked on DF again. ;)

It's actually not that hard to get the basics if you're used to ASCII graphics or using a tileset. I was able to figure out how to play it (with the help of the wiki for laptop-friendly controsl) in only half an hour of poking around, mostly.

Very interesting: "Tarn sees his work in stridently ethical terms. He calls games like Angry Birds or Bejeweled, which ensnare players in addictive loops of frustration and gratification under the pretense that skill is required to win, 'abusive' -- a common diagnosis among those who get hooked on the games, but a surprising one from a game designer, ostensibly charged with doing the hooking. 'Many popular games tap into something in a person that is compulsive, like hoarding,' he said, 'the need to make progress with points or collect things. You sit there saying yeah-yeah-yeah and then you wake up and say, What the hell was I doing? You can call that kind of game fun, but only if you call compulsive gambling fun.' He added: 'I used to value the ability to turn the user into your slave. I don’t anymore.'"

Completely agree with Tarn. I would go further, and describe certain types of games as evil. Social games, in particular, I believe have evil game design.

Here's an interview with the somewhat controversial but always interesting game designer Jonathan Blow: http://www.pcgamer.com/2011/02/15/jonathan-blow-interview-so...

In the second half, he talks about social game design. I'm very much with him on this.

Hostile, thank you for posting this, Do you think that any other designers are like him?

I'm an indie game designer, and I find this is by no means an uncommon view among game designers.

It's a view I've often come across on indie game-related blogs.

Companies like Zynga help push the issue to the fore-front: They get continuous real time data about how changes in their games effect player behavior, and try to use that to maximize the addictive factor.

To put it another way, the makers of Angry Birds are crack dealers while Tarn prefers to expand minds by selling LSD.

I was going to say: I admire Tarn's exquisitely esoteric sense of taste, which informs his art and makes both him and others happy, but I try not to shame other people for being addicted to achievement. We are all addicted to achievement. Some of us are just much, much kinkier about it than others. ;)

I have yet to definitively locate the bright shining line between a great game and an evilly addictive game. I certainly believe the people who say that WoW, Minecraft, and now Dwarf Fortress are excellent games, yet I don't play them because I don't want to be addicted. And the fact that, say, bridge and chess are socially approved addictions, requiring much more practice to attain skill, doesn't make them harmless. People get addicted to Scrabble, ice hockey, knitting, model making, composing tiny essays on Hacker News, mathematics, and of course programming just as they get addicted to Angry Birds. Is it the fact that these hobbies are less accessible that makes them less "evil"?

The line is in the intent of the design. A certain class of game is designed to immerse you in addictive feedback loops purely for the sake of keeping you in place and (usually) draining your wallet.

It's the difference between the architectural design of a casino and a museum. They're both very technically beautiful and nuanced in the subtle intent, but one is about trapping you by exploiting your animal brain whereas the other is about something else, like walking you through a narrative.

Alternately, distinguish between writing a computer virus and any other kind of excellent code.

FWIW Bejeweled [Blitz] requires skill. I'm moderately good, second in my circle of friends and I go off the boil if I've not played it for a while, it takes time to get back into the groove. That said they do manipulate the scoring a lot and there is a lot of pressure to buy, buy, buy to get a higher score (that I've resisted, that's not my thing).

Bejeweled is a very well constructed Skinner Box that has been effectively optimised over time. TBH it should be part of a course on addiction psychology, it probably is.

Interesting comments nonetheless. Thanks for sharing.

Yeah right. Every game triggers OC tendencies and desires in people like me. I've regretted binging on both Minecraft and World of Warcraft.

The real reason for tension seems to be this: game designers see Zynga et al achieve the same level of player engagement without careful polish and production values (seemingly).

In other words, it's the age old plight of the artisan who invested time and money into details and craftsmanship, just to see it lose market share to cheaply-produced slop.

In that sense, old juggernauts like EA, who spend many millions in budgets on games, should be concerned too.

Some notes:

First, these guys are the real hackers. Not Carmack, Woz, Bill, or Zuck, or all the super-cool guys that made the bucks and then coasted (technically). These guys are living the life just because they love it. They're a much better representation of the inner hacker in all of us than those stories of riches and fame. I got the impression they would continue doing this no matter what their financial circumstances.

Second, they've done their time. Somebody should set them up with an annuity so they can a) continue, b) learn to live without worrying about money, and c) tell stories to kids 40 years from now about how it all came together

Third, this article left me gobsmacked. I'm left with one conclusion: this is art. If you're collecting these crayon receipts? Save them. They're going to be worth something one day.

Of course, just like the game plays out, it might all amount to nothing. If I had to bet about the value of their work, I'd say it drops off short-term (next 20 years) then becomes super valuable around 2030 or so. This seems to be the pattern with semi-famous labors of love with cult followings. Seriously. Save whatever you can get from these guys.

Very cool story.

I wish I could have ignored you saying that John Carmack is not a real hacker, but I couldn't.

Carmack has consistently pushed forward multiple industries on multiple fronts, and remains to this day a far better programmer than any I've ever worked with. He revolutionized computer gaming while living the life just because he loves it. He's not famous because he has Ferraris; his Ferraris are famous because they're his.

He doesn't need anyone to set him up with an annuity--maybe because his love was more mainstream than theirs, or maybe because his skill was far greater than theirs--but that in no way diminishes what he has accomplished or his motivations for accomplishing it.

These days he splits his non-Ferrari-driving-time between hacking on next gen. cross-platform game engines and hacking on rockets (at Armadillo aerospace). He may be a successful hacker, but he's most definitely a hacker right down to the core.

Started following him on twitter recently (http://twitter.com/#!/ID_AA_Carmack).

Definitely not coasting.

Sorry. I hear what you're saying and I think you are deeply wrong. (Oh no, someone is wrong on the Internet! :)

Carmack, Gates, Zuck and Woz are the real deal. Saying they aren't because others toil in obscurity is some hopeless romanticism about hacker culture. A heart warmer, maybe?, but not justified.

Just because financial outcomes are not matched to your hacker ideal doesn't invalidate either approach. The rich famous hackers were also good at business and publicity. Not everyone wants that or is good at all three, which is the case with Dwarf Fortress.

Money is not the scorecard. Read the article. That point is addressed.

To be fair, Gates was more business man than any sort of hacker. While Zuckerberg may have actually coded his first product, unlike Gates, making a web app, no matter how popular, is far cry from what Carmack and Wozniak have done.

Perhaps it's an illustration that it's often not fancy maths and algorithms that pay, but rather being in the right place at the right time with the right product. That said, there are instances like Google and Carmack where it really was the algorithms that made the difference.

"unlike Gates". Pardon? As Gates was co-author of Altair Basic, which was Microsoft's first product, I don't understand your comparison. Could you elaborate?

Gates wrote, by himself, the best BAISC interpreter in the industry in a matter of months. At a time where there was no internet to check out the latest papers, StackOverflow to help you out, etc.

Way more impressive than sticking together a few perlscripts for a social networking website. Zuckerberg wasn't the main programmer behind Facebook, btw.

Gates wrote the interpreter along with Paul Allen and Monte Davidoff... while he distinguished himself as a programmer, he's not quite that level of amazing. I get the impression that in a situation similar to (but not as lopsided as) Wozniak and Jobs' arrangement, Allen did a bit more of the technical heavy lifting and Gates planned and arranged the business side.

I'd recommend you read http://www.joelonsoftware.com/items/2006/06/16.html

wherein Joel Spolsky was reviewed by Bill Gates. It is pretty clear the man had serious technical chops.

I don't know where you are coming from saying Carmack 'coasted (technically)' after he made his bucks. He is about a hard core coder as they come. Its not like he took some CEO desk job after Doom came out. From the first commander keen all the way through probably Quake 3, the stuff he wrote was all cutting edge.

My point was, from a "how well-known are they" standpoint, these other guys earn a lot more points for just being rich and famous than they do for being great hackers.

No doubt many of them continue to do great technical work that constitutes "great hacking" in anybody's book. But that's not why (by and large) people buy the books and become fans.

If you are focusing on the technical skill of hackery then you are going down the wrong trail. I'm talking about living the life of hacking, as most hackers understand and would actually practice it, as opposed to being a famous hacker. Let's face it, odds are you'll never be a billionaire with your own rocket company. But you just might spend 20 years of your life working on a pet project. These guys are our guys. They're us.

Whilst I get where you're coming from, to me you are falling into the 'mainstream can't be cool' point of view. Just because someone is successful and rich doesn't mean they can't be judged on what they do anymore.

In some respects, I think that continuing to do what you love even after you have 'fuck you money' is one of the strongest indicators of a persons passions. Carmack could have stopped programming games years ago, yet he continues to innovate and push boundaries. For him, the rocket company IS his pet project and is just as brilliant an example of hacking as anything else. Both Gates and Jobs could have retired to a private island (or even private archipelago) years ago, instead both continued to work at their passions long past necessary. And now Gates has a new pet project, it just happens to have a lot more money and publicity than Dwarf Fortress.

The notion of hacker-points is just so tiresome and not the kind of discussion that I want to have. We don't need to constantly one-up people, even if we're comparing this person to that person.

Put another way: it's possible to praise his behavior without pissing on someone else.

Woz isn't a real hacker?

Disagree. Read up on him.

Second your disagreement.

Woz is a real hacker, if not one of the best ever.

Long excerpt from his amazing Founders at Work interview.

"This was two weeks away and Mike Markkula said yes. So that was my motivation. I always had these little fictitious motivations that motivated me and got me to do such great work. So I sat down and designed the floppy disk, and Randy Wigginton (he was the guy just out of high school) and I came in every single day including Christmas and New Years for 2 weeks. I came in every single day leading up to, I think it was January 3 or 5, when we went off to Las Vegas. I almost had this floppy disk done.

I got it to where it was writing data on a track, reading the data on a track. Then I got it to where it was reading the data in the right byte positions. Then I got it to work with shifting tracks, and we wanted a simple program where we would say "run checkbook" or "run color math" and it would run the programs that were stored on the floppy disk. So we went off to Las Vegas, and Randy and I worked all night and we got it done to where it was working. At the very end, it was 6:00 a.m. and I said, 'We have to back up this floppy disk." We had one good disk that we prepared with the data hand-massaged to get it just right. So I stuck it in the floppy and wrote a little program, and I typed in some data and I said "read track 0," stuck in the other floppy and said "write track 0, read track 1, write track 1." There were 36 tracks—I had to switch floppies back and forth.

When I got done, I'm looking at these 2 floppies that look just the same. And I decided that I might have written onto the good one from the bad, and I did. So I had lost it all. I went back to my hotel room. I slept for a while. I got up about 10:00 a.m. or so. I sat down and, out of my head and my listings, recreated everything, got it working again, and we showed it at the show. It was a huge hit. Everybody was saying, "Oh my God, Apple has a floppy!" It just looked beautiful, plugged into a slot on our computer. We were able to say "run color math," and it just runs instantly. It was a change in time.

But the real eureka moment for me was the very first time I ever read data back. I wrote it on the floppy, which was easy, but read it back, got it right. I just died."

The whole interview (and book) are inspiring:


What's so hackish about this? Seems like pretty typical low-level programming.

Please read the whole interview. If not, we'll agree to disagree.

"My whole life was basically trying to optimize things. You don't just save parts, but every time you save parts you save on complexity and reliability, the amount of time it takes to understand something. And how good you can build it without errors and bugs and flaws.

Livingston: You were designing all of these different types of computers during high school at home, for fun?

Wozniak: Yes, because I could never build one. Not only that, but I would design one and design it over and over and over—each one of the computers—because new chips would come out. I would take the new chips and redesign some computer I'd done before because I'd come up with a clever idea about how I could save 2 more chips. "I'll do it in 42 chips instead of 44 chips."

Steve Wozniak recreated a disk controller in assembly on a motorola 68k chip in less than a day, and it worked pretty much perfectly first time, and you are complaining that he's not much of a hacker?

What do you do in your spare time, if I may ask politely?

6502, not 68k

Oops, thanks!

From his head and from his listings? How much was from the listings? Do you know what listings are? (I know Woz is a godly Hacker, I'm just playing devil's advocate)

Heh. For what it's worth (and at the risk of karma annihilation), you're probably correct. A lot of the seemingly-godly things people have done are from 'simple tricks' that they've learned. People just don't like to admit it.

Woz was as brilliant as technical brilliance could possibly ever be, though.

You need to find out more about Carmack and Woz. Agree with the rest.

I've never heard of other games spawning threads like these:





I've never seen other games where people use that as a reason to study geology, farming, beekeeping or whatnot. One person, not satisfied with the material properties of Saguaro wood and unable to find good data online went so far as to track down a sample and measure them.

One person even mentioned that they neglected to study for their geology test and played Dwarf Fortress instead. They got an extremely high score because so many of the questions were relevant to DF, such as asking for the names of common iron ores. Any good DF player can list at least magnetite, limonite and hematite without any trouble.

I know of several people who have gotten history degrees because of an interest sparked by the Europa Universalis series (usually either EU2 or EU3), and of course some people have become fighter pilots after playing flight sims. But the sheer breadth of DF is different from anything I've seen.

Don't miss the chance to read through the "Boatmurdered" Dwarf Fortress succession campaign:

Perhaps most fascinating are the stories that fans share online, recounting their dwarven travails in detailed and sometimes illustrated narratives. In a 2006 saga, called Boatmurdered, fans passed around a single fortress — one player would save a game, send the file to another player and so on, relay-race style — while documenting its colorful descent into oblivion. (After a vicious elephant attack: “A single untrained marksdwarf stands ready to defend the crossing, but I doubt he’ll be enough.”) Boatmurdered spread across gaming sites and made the front page of MetaFilter, a popular blog. “That did a lot to make people aware we existed,” Tarn says.


In addition, might I also suggest the illustrations done by Tim Denee.

* Bronzemurder: http://www.timdenee.com/bronzemurder.html

* Oilfurnace: http://www.timdenee.com/oilfurnace.html

Bravemule (http://www.bravemule.com/matulremrit/) is the best Dwarf Fortress-related writeup I have ever read. It is fantastically illustrated, and captures the black humor and horror of Dwarf Fortress better than anything else I have read. Check it out.

Seconded. Even if you don't know anything about DF, Bravemule is a wonderful read.

My teenage son - who, raised right, grew up on Nethack - absolutely loves this game, and will occasionally stay up all night and be found still feverishly working on some elaborate construction in the morning. From what I have seen, it is ASCII crack.

edit: I guess it's modified code-page-437 crack:


edit again: This article really is a very interesting story behind the development. Even if you don't play, it's worth reading.

freakishly, the game actually uses opengl to display the grid of character sprites...

It has half a dozen rendering modes too, due to various driver bugs and GPU performance profiles. There are massive threads on the forum of people trying different settings to squeeze out a few more FPS, especially on lower end hw.

haha! that's awesome.

there's an ncurses mode in the linux version. i use it to play over ssh, as my laptop is weak. the opengl is mainly for tileset support.

Oh god there is a Linux version now? Which means I can play it under screen? or from my iPad?

Thanks, there goes all my free time.

ipad play is probably wonky as df is heavily keyboard-based, but let me know how that goes.

If you're ever worried that a feature will be too complex to implement, take a look at the ridiculous number of things Dwarf Fortress keeps track of and simulates some time. The article briefly touches on it, but like any look at Dwarf Fortress, it bared dips its toes in the water. Just a quick list off the top of my head...

In world gen, it simulates geology and erosion, climate/biomes, growing civilizations, politics, war, trade, attacks by various "megabeast" creatures and their battles (in which injuries down to losing a tooth will be kept track of), migration and refuges as a result of war, (I believe) deforestation as a result of logging, and probably a lot more that I'm forgetting. That's just in world generation, before you're actually playing it.

Remember, what one person can program when they're truly dedicated is a pretty incredible feat.

Dwarf Fortress has a level of micro-complexity that I wish Civilization had.

I think there is a niche -- albeit probably a very small one -- for more super-hardcore strategy games in a variety of settings and genres. Folks like me would love to play a game like Civilization and have incredibly rich, complicated diplomacy and trade and other issues that simulate, as closely as possible, real-world geopolitical issues. (In some ways, for instance, I'd love to be able to fight a cold war with an AI opponent, rather than an actual war).

I can vouch for Crusader Kings, Hearts of Iron, and Victora (all from the same developer as Europa Universalis: Paradox Interactive) as having the kind of beyond-Civ levels of depth and complexity you're looking for. Each is tailored to it's own slice of history in a way that Civ by nature can't be.

While I have never played it, I have heard good things about the complexity of the Europa Universalis series. IF author Adam Cadre wrote a mini-review of it a few years ago:


Have also heard good things. Given that I keep hearing them, I think it's time I got off my butt and checked it out!

Remember, what one person can program when they're truly dedicated is a pretty incredible feat.

And I wonder if the level of dedication he has demonstrated requires Asperger's. He has all of the classic symptoms, down to the the involuntary rocking. I've worked hard, and I've sacrificed a lot of my time to my work. But I'm not willing to sacrifice as much of my time as he has. And it's quite likely his life's work will be both larger in total and have more impact than mine.

That all leads me to my question, which is if his level of dedication requires Asperger's. Before people bring up people like Einstein and Feynman, consider that they had personal lives. They pursued personal lives.

> And I wonder if the level of dedication he has demonstrated requires Asperger's.

People always jump to Asperger's the moment they hear this was all put together by one guy. It is as if they refuse to believe a normal person can have willpower enough to accomplish so much. It is simple: take a single step every day towards a single goal for several years.

Listen to his podcasts[1]. He is about as chill/normal/well-adjusted as people come.

[1] http://www.bay12games.com/dwarves/df_talk.html

It's entirely possible, but the big three cues to me where his extreme social awkwardness, the rocking, and finally, his dedication to technical interests. The final one alone wouldn't make me think Asperger's. Even the first and the third wouldn't make me confident, but unconscious rocking is a symptom of autism in general.

I'm a big proponent of the idea that genius is more about raw effort over extended periods of time rather than innate intelligence. But perhaps some people are wired for that kind of dedication moreso than others.

The only thing he's missing out on is romance and sex. His occupation and lifestyle appears to satisfy all the other typical human needs. Taking a DSM perspective is not the most enlightening perspective.

The sex thing is a big problem.

In the past, great artists were able to find companionship and sex despite poverty and oddness. Why is it that hackers have such difficulty? If Richard Stallman's medium was anything other than computers he would have to fight off the women, but somehow hackers are girl-repellent despite being devoted artistic geniuses?

I don't think his devotion will last long or stay fresh unless he starts fulfilling sex and romance needs. Exposure like this NYT article should help, and also a rising income would help a lot. Then he still has to make time for women and he might find the intimacy too threatening.

Anyway for the sake of Dwarf Fortress and his art ambitions I hope he figures out a way to fulfill sex needs. Other than that I don't see anything unsustainable or obviously unhealthy about his lifestyle.

You've got it backwards. Women aren't scared of hackers, hackers are scared of women. This is a gross generalization, but it fits for the examples you mention. Yes, some nerds are completely unattractive and girl-repellent. But it doesn't take that much effort to become attractive, and a lot of hardcore computer guys are attracted to computers precisely because working with computers allows you to lead a solitary lifestyle with minimal social interaction.

I'm certain that there were people before that were as scared of social interaction as the biggest hardcore nerds today, but they weren't working in a technical profession. Working with technical things 50 years ago required interacting with other humans much more than computers do, because the machinery in question was physical and more distributed.

You're right. The issue is that for a woman to pick up one of these nerds she would have to be unusually aggressive in pursuing the nerd-genius and yet also unusually submissive since the nerd-genius (like any great artist) would have strange and demanding needs to which she would need to conform and satisfy.

She'd have to be Courney Love, and there aren't many of those. And shed have to somehow meet the nerd despite his reclusively.

The fear of girls is definitely what makes hacker geniuses more like monks than painters. I admire the crap out of great hackers but my sex needs come before my art.

  my sex needs come before my art
That's a good way to put it, that's how I feel as well ;) The monk analogy is really good; I'm guessing that monks were attracted to their lifestyle for the same reason the most introverted nerds are attracted to computers today.

Find a partner with equally strong passions.

I.e. he/she understands that you need your time to geek out on your stuff because he/she needs the same thing.

>The sex thing is a big problem.

Seriously? Not everyone in the world is a sex addict. He's living his own life. He'll look for what he needs when he needs it.

Everyone in the world has needs for sex and romance. Some people have their needs unfulfilled or repressed, and they suffer for it.

Sex is one of the most basic biological needs, and romantic intimacy (which is a part of good sex) is crucial for happiness, especially for a deeply social species like humans.

Don't you think you're simplifying it a bit? Some people get along fine without having sex, you seem to suggest that anyone who doesn't have sex is bound to be unhappy and lead a horrible life.

While for most people sex is important, I think you need to take a step back and look at how different people's needs can be.


NB. There's of course a gradient between complete asexuality and total 24/7 horniness. People are different.

Errr, have you considered the possibility he doesn't have sex needs, or that his sex needs are not as much as a "normal" human, or that his needs can't be fulfilled by "normal" humans?

You are making it sound it he CAN'T get girls. I m sure he can if he puts in the time. It is just not his priority right now. Time is limited. And he chose to devote his time to the game.

I wonder how he manages to balance all these elements. Even in much simpler games balancing the strengths and weaknesses of all the actors can be very difficult. With so many interlocking factors it seems it would next to impossible to avoid inadvertently making one tactic or actor too powerful.

For example, in the early Civ games the optimal strategy was to just build up to chariot technology and then go 100% on the offensive.

There are a lot of releases, and fairly often; I've had versions go "out of date" on me within a week of its release. The forums are very active, and when they say something is unbalanced or broken, it tends to get fixed. Some things are not serious game-breakers, even if they are mildly broken (see: elephants in the versions that Boatmurdered used), and those tend to stay in until it becomes a pain to deal with.

Basically, it's just trial-and-error. But it makes for an interesting game, because you're never quite sure if everything will work as expected.

The early civ games were susceptible to the mongol hord method, which the designer never considered.

I wonder what techniques Tarn uses to manage that complexity. Does he write extremely clean code? Use test-driven development?

From what I've heard, not really. He tests a lot (he posts about it on the devlog), but I've heard he doesn't use version control even.

Supposedly his code is pretty rough. That would explain why development is slow and the game pretty bug-ridden.

For those of you who are seeing this for the first time and have any love for games like Minecraft, Civilization, etc... please take the time to break through the very steep learning curve to give this game a fair shake. It is truly a work of art.

Please note that the Wiki helps make the game playable: http://df.magmawiki.com/

please take the time to break through the very steep learning curve to give this game a fair shake

I don't know, I don't think we have the same attitude towards games. I already spend hours a week sitting in front of a computer doing really difficult stuff with screenfuls of ugly text for non-obvious rewards. It's called "work", and it pays me in real-world money and real-world glory. I don't have much desire to do the same sort of thing in my spare time.

Nope, give me a beautiful, overproduced and shallow game any day.

Reminds me of the experience of sitting around a table with some brilliant people arguing about what our role-playing-game characters should do next. Hey, we're in a heated, passionate, frustrating meeting, and we're not even getting paid!

We should start a company or something...

You might benefit from having played Nethack first. At least the culture shock will be a smaller.

I still can't get a grip on Nethack. I think I need to print out the list of key bindings in order to actually play.

I would try playing Dungeon Crawl: Stone Soup instead. It's another roguelike that is heavily inspired by nethack but is more modern and more beginner friendly. I love nethack (I've ascended twice) but it's definitely showing it's age (not to mention that the game is pretty broken in some regards). Crawl has a better interface and tries to remove all the tedious parts of roguelikes, though there's still a learning curve.

What's broken in it?

A lot. Much of it isn't really relevant to beginners, but here's a quick list off the top of my head. Maybe broken wasn't the right word, these are just general problems with the game

1. The endgame is too easy and boring.

2. Elbereth and #pray make it too easy to get out of tough spots

3. Most classes end up playing similarly after the midgame (either wizard or melee typically)

4. Wands of wishing, magic markers, and polypiling are too powerful

5. The game is nearly unbeatable without spoilers

6. Sokoban is boring after you've played through it a few times

7. There's a bunch of cheap tactics like pudding farming

8. It's too generic and boring to put together an ascension kit. There are very few trade-offs you need to make in choosing what armor and weapons to use

9. The religion aspect is too simplistic

10. Combat gets tedious in the endgame. Most of the time you're just plowing through monsters. This results in tons of "battle spam" with you getting several messages like "You hit the minotaur!" every turn, forcing you to not read them and possibly miss important information

11. The game requires several patches to be playable (menucolors, sortloot, hpmon, etc.)

12. The default options are pretty bad

13. Traveling and exploring can be boring

Wow, that was a lot longer than I thought it would be. But I only write so much because I care :) I had a lot of fun beating nethack and and it's one of my favorite games of all time. I'm possibly prouder of beating nethack than of my MS degree :)

This is all true. However.. how many years of gameplay did you have to put in to learn it?

Speaking for myself, I've been playing nethack since the 80's. How many other games can you name that will keep an audience interested for 20 or 30 years?

I'd count that as a fantastic success in game design.

Despite that, I do agree that the game has its weaknesses and could stand to be improved quite a bit. One of its greatest weaknesses is its glacial pace of development. The last release was in 2003.

Eh--I had my first ascension (Val) about four months after I started. I've done so another three times since. (Wiz, Arc, Rgr).

I think that there's a possibility that it's not as difficult, in relative terms, for people around my age who grew up with similar games, and fairly hard ones at that.

If you want to stay with Nethack, but want those flaws addressed: Try Sporkhack, a fork of Nethack. The author tries to make the game harder in the endgame, but not harder and possibly easier for beginners.

Not sure if this is what johnswamps was referring to, but the two big things that come to mind are the Sokoban levels (there are only two fixed variants which can only guarantee you one of two things, and a lot of the time if people don't get the thing they want (usually the bag of holding), they end up restarting since the odds of getting it anywhere else in the game are pretty low after that) and the fact that you can do things like pudding farming to obtain items/corpses easily. A lot of actions in Nethack are formulaic and based on tricks, whereas DC:SS encourages things like unknown item use more. In the end, I love Nethack and it was my first roguelike, but DC:SS is just a lot more fun.

Also, here's a post comparing Dungeon Crawl and Nethack that I just remembered: http://nethack.wikia.com/wiki/Linleys_Dungeon_Crawl

I actually prefer getting the amulet of reflection from Sokoban, and bags of holding get randomly generated relatively often. But start scumming is damnable anyway.

POWDER is worth looking at as a starter - it feels like a scaled-down, more accessible NetHack with much of the the identification fun and unexpected combinations of interactions, and it's just as playable with a control pad, due to its GBA origins.


Key bindings is the least of your problems. If you want to play Nethack well (ie with any chance whatsoever of ascending), you'll need to read hundreds of pages of FAQs to understand exactly what the hell is going on.

And once you've read all those, most of the surprise is gone, since you'll know that kicking a sink has a 1/20 chance of producing a black pudding and so forth.

I liked nethack, angband, and rogue.

Then I found ADOM -- (www.adom.de), which is great fun, but turned into a "second system" problem with JADE (development slowed so he could rewrite entirely in Java...). Still, amazingly fun.

I never got into dwarf fortress or minecraft, but will check them out more after 25 August :)

Or if you cherish your time, you keep away from it. It's quite addictive and you can have a lot of Fun.

Having never been really involved with the Dwarf Fortress community/cult, but having played the game quite a bit, this exposé on the creator is extremely interesting. I always thought it must have taken a certain level of cleverness to build such an impressive simulation, but I honestly didn't think he'd be a Stanford Ph.D. Also, I'm honestly surprised they get that much in donations. It's very well deserved indeed and I'm happy to hear that they do.

What's interesting is that they could likely be making notch-like levels of money if they went the minecraft route; some public server code and subscriptions would yield a huge amount of revenue. Honestly, I don't think they want it. If they were interested in doing that, they'd have found a publisher, and started segmenting up the game for sale.

I don't play it, but Tarn really deserves more support as it sounds like he and his brother skate by on very little.

He makes all of the donation information public, so you can see for yourself:


June: $2374.03

May: $3010.58

April: $3018.42

March: $4220.20

February: $2113.11

January: $2158.91

I added January from the previous post, just to get the whole year in there.

I have long assumed that it will eventually be unethical to turn off dwarf fortress.

Based on the type of world being built, it might eventually be equally unethical to turn it on.

Seriously, though, bits don't rot.

If we could pause the world, would it be unethical to do so? It's not the same as destroying the world. I'm just responding to the implicit reference to pausing-ai-being-unethical. pause != destroy.

If I had to choose a single game to play for the rest of my life, Dwarf Fortress would be an easy choice. The ability to build massively complex systems in a massively complex world appeals to the part of my brain that made me an engineer. There is so much potential in this game. So many things... hidden. So many stories to share.

Just watch out for the Elephants. Or Carp. Or Badgers. Or the random, world-swallowing bugs.

that is a pretty strong recommendation. Must find time to try it out.

Here's the game's homepage - http://www.bay12games.com/dwarves/

This looks very cool, and there are versions for Linux/OSX/Win

You're better off with your first experience being the mayday set that uses tiled graphics:


Here's the wiki, which has some quickstart guides:


If you're going about it for the first time, I'd recommend the Lazy Newb Pack:


Because of the somewhat eccentric nature of the DF menus, this contains a number of utilities like Dwarf Therapist that are, IMO, indispensable for running a fortress of more than about twenty dwarves. It also has a custom tileset built in and an isometric visualization tool.

DF does have very impressive depth and game mechanics. Unfortunately, its interface is one of the worst of any game I've ever played.

It's a pity that DF isn't open source either, as then its interface problems would have long since been fixed. But, as it is, its lead (and only) developer doesn't seem to care enough to fix it himself.

The interface for DF is actually not bad at all. Once you learn the keystrokes the interface disappears, and you don't even need to have the side menu open at all except on context menus where you need to scroll up or down. I find it to be vert fast and efficient and would hate the game it I had to use point and click or some other slower method of manipulating the world.

I have learned the keystrokes, and have spent countless hours playing it. The interface is still absolutely awful. It's so inconsistent and needlessly convoluted.

Just to be clear, I'm not wishing that DF had a point-and-click interface. I'm happy that DF is keyboard-controlled. It's just that this keyboard control could have been made a lot less painful with a tiny bit of consideration for consistency and efficiency.

As far as roguelikes go, nethack has a pretty nice interface (except for the stupid # commands), though even there there's some room for improvement. DF should have been modeled on that, or at least with some input from players on how the interface could have been improved.

At the absolute least, the author could have let the same actions on different screens of the game be activated with the same keystrokes, rather than requiring completely different keystrokes for some perverse and unfathomable reason.

Another quite workable solution would be to let the keystrokes be user-configurable. That would go a long way towards lessening the pain.

The keystrokes are user-configurable. Both in-game and in a human-readable text format.

Ah.. I didn't know that. Next time I give the game a try I'll have to look in to that.

> It's a pity that DF isn't open source either

I don't get that. It's totally donation-supported. Is it a pride thing? A control thing? I have already tried it out if it were in apt.

From what I remember reading when I was frequenting the bay12 forums, it's a mixture of pride (the code is very rough, and it would probably take a huge amount of effort to get it into a form where contributions from others could be useful, or even where others could contribute) and control (he doesn't want it to get out of his control, and as it's his only source of income he doesn't want to deal with donations and community being fractured between several versions). The fanbase both doesn't care (based on donations) and sees it as a challenge (there have been some interesting analyses posted of what the game is doing, and there's a thriving modding community - both in the form of modifications to the game's files, which is supported and encouraged, and modifications to the game through memory editing, leading to the creation of invaluable tools like Dwarf Therapist).

If your only reason for not trying it is that it's not in apt, just go to the website and download it. It doesn't take any installation; just download it, extract the contents, and run the game.

Being a Nethack fan who has also lost countless hours managing civilizations, space fleets, cities, insect colonies, zombie apocalypse survivors' camps, I have three words to say:

Must... not... play...

Dwarf Fortress sounds to me like nethack in its difficulty but unlike nethack in its endlessness. You can play all you want, but there will be no end!

Dwarf Fortress is, in many ways, the anti-nethack.

In Nethack, you can parse the source code to figure out nearly any situation, that is, it's a closed, nearly solvable system. (See http://sporkhack.com/design_concept.html for some commentary on nethack being 'solved'.)

Dwarf Fortress, on the other hand, is totally emergent. Toady logs in and plays to check if his coding worked well, that is he goes looking for the situations he is referring to in his code and observes them.

I love them both, but I'm enjoying dwarf fortress more right now.

a fusion between nethack and minecraft (and the sims) is a good description I feel.

I have always had an inkling that video games should be considered as applied math. According to Tarn, the author of Dwarf Fortress, who also has a Ph.D in math, making games "scratches all the same itches" as math. That sounds just right to me.


Programming is one of the most difficult branches of applied mathematics;

Great article. Funnily enough, the Metafilter post referenced is actually the one I put up in 2007:


For those of you worried about losing your life to another simulation game, I present:

How Dwarf Fortress cured me of my Dwarf Fortress addiction once and for all - even though I've never played it:


Nice to see such a great game receiving some well deserved exposure.

Just as an aside, a game like this with deep algorithmic complexity but trivial graphics seems like a natural fit for a functional language like Scala or Haskell. I wonder why Tarn chose C++ instead. Was it just out of familiarity? With his background in math I'd imagine he'd be very comfortable in an FP language.

Apart from age, there's performance - even written in c++ and running on a modern CPU, the framerate can get pretty slow.

Dwarf fortress was started in 2002. Scala didn't even exist then did it?

I didn't realize DF had been around so long. Scala wasn't introduced until 2003 and Haskell certainly wasn't nearly as mature then.

I seriously discourage you from checking out this game if you're on a deadline, building a startup or have any project you'd like to take on.

On the other hand it's a must have if you have a nuclear shelter in your backyard.

I haven't played it much, but I've always felt it would make an interesting iPad game. Seems like it wouldn't be too complex to put an objective-c wrapper around the code.

The problem is simple: The iPad lacks the horsepower to run the game at approaching a playable rate. Now if you were running it on another machine and playing by remote desktop.... Maybe. But to me this game simply demands a keyboard.

I have not played Dwarf Fortress, but is the game really so processor intensive that a 1 GHz dual-core iPad 2 (with GPU) could not keep up?

My 3.2 GhZ Dual Core system chokes at times with it, so yes. The biggest catch is that the core of the game is not multi-threaded (though graphics may be depending on config). The throughput of a single processor core is the most important factor in the simulation's performance. The iPad, while nice for the rather passive computations it's expected to perform, just doesn't have the muscle.

Now assuming the game could be ported to iOS (wrapper/emulator or whatnot), 1 GHz would allow you to generate a small world, and maybe play for a year or two at a low FPS. (I used to play the older version on my 1.2 GHz AMD "netbook," so it's certainly possible.) But in fortress mode, when you've got a great many dwarfs doing pathfinding (say, 80+) and a whole bunch of other stuff being tracked by the game, the simulation can be extremely taxing on system resources. Ultimately, most forts die from what is called "FPS Death," when the simulation frames completed per second reaches an unplayable level.

I wonder how much of the simulation code could be appropriately ported to some sort of GPGPU (CUDA,OpenCL, etc) code. There'd be something nicely ironic about running an ascii-frontend game which required a serious GPU to play :)

That's been proposed in the bay12 forums several times. While I don't think that Tarn ever officially responded, the reason that's not going to happen seems to boil down to it requiring a complete rewrite of all the code behind the game. Which isn't going to happen any time soon - I recall that, a while back, he was using implementing new features as a reward to get him to do bugfixes; putting off the fun things until after he'd done the unfun things. Rewriting all of the code would be extremely unfun, take forever, and prevent implementing new features.

You would probably want to run it on something like a POWER7 for best possible performance then, which have fewer but faster cores than comparable x64s, if there is an AIX port.

Nice article.

Reminds me a little of the two lead characters of "Makers", the Cory Doctorow book. But this is real, and mildly depressing. You want guys like this to be successful, and to have the choice to live the spartan lifestyle if they choose, or one with a better choice of food, beverage, sleeping, etc. options.

Sad to read that relationships are not important though :-(

Why? There is some sort of an assumption that a romantic relationship is something desirable by default, and someone's lack of interest is somehow "sad".

I agree that sometimes the reasons for staying out of the relationships business may be sad - or, more precisely, constitute sad stories. But assuming that someone can't achieve happiness in life without pursuing a relationship seems presumptuous, especially when faced with someone who clearly defines their objectives and needs differently.

Well you are making the assumption I am referring to a romantic relationship.

The article presents (and this may be the reporters bias) the picture of someone with minimal social contact. Whilst I accept different people have a different level of need for social contact, this certainly feels more than a little unbalanced.

I realise whilst writing this reply that "sad" can be construed by some people in a derogatory sense. I didn't mean it in that way. I genuinely feel upset that, for whatever reasons, the lead developer feels that almost anything would be a distraction from his life's work, especially a partner and children. Intimate human relations - romantic or not - offer things no personal achievement can.

OK, right, I did, though that doesn't really change that much.

Well, let's start by saying that intimate human relations are achievements as well. Also, computer programming, climbing and drinking soft drinks offer things no intimate human relations can. It sounds like sarcasm, but it is, actually, true.

I was, too, upset at first. I just noticed my upset has no real reason. It was about someone choosing a radically different way of life than mine, based on their own needs — that are radically different from mine. He's not even telling "no intimate human relationships." It's more as if he's satisfied by the relatively small group of close humans — his relationship with his brother seems quite close.

To sum up: I don't think his choice is sad. It's actually our limited view of possible life choices what's sad.

"Dwarf Fortress unfolds as a series of staggeringly elaborate challenges and devastating setbacks that lead, no matter how well one plays, to eventual ruin."

Soooo.... One can play this game, or one can be in charge of a real country's economic policies. Same outcome. ;)

I was terribly happy to see this article all over my Twitter feed. This game is an inspiration for budding programmers, even if the game is terribly difficult to get into comfortably. I've been playing it for quite some time now. Very happy to see it.

Wow. I thought Realm of the Mad god (http://www.realmofthemadgod.com/) had low-fi appeal...

Cool piece.

Dwarf Fortress is an outstanding game with features not available anywhere else, and I would highly recommend it. The background music, which the creator also composed and performed, is very soothing, as well.

Mostly I play it because I like chaining tame hydras in front of the bridge that leads into my trading post.

Registration is open for Startup School 2019. Classes start July 22nd.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact