Hacker Read top | best | new | newcomments | leaders | about | bookmarklet login
The Brilliance of Dwarf Fortress (2011) (www.nytimes.com) similar stories update story
334 points by 0wl3x | karma 236 | avg karma 3.63 2017-02-28 09:18:46 | hide | past | favorite | 234 comments



view as:

Dwarf Fortress is incredibly hard to learn.

Gnomoria and Rim World might be simpler games to learn. However not as deep or complex.


I don't think "incredibly hard" is a good term.

In more development terms, to me, "incredibly hard" is embedded assembly programming (im probably not even using the right term for this) ...

The level of hard i would put DF at is "learning VIM"... which I think is apt, because 1) its a steep but rather short learning curve to get around, 2) over a while you naturally get better at navigating/commanding/etc and 3) there are plugins and "distributions" just like vim, to make it easier for a beginner, etc.

Now that I write this out, im thinking VIM is the perfect analog for DF :)

[NOTE] I'm a DF n00b. I have less than 5 hours playing. I'd say it took me about 30 minutes to "get it" though, and now I'm actually playing what I would consider "normally"


It may be instructional that I think learning VIM and DF is harder than assembly programming?

=)


I think vim/emacs is an apt comparison.

I like your analogy, but what's missing is that the vim interface is much (much!) more powerful than the DF interface.

Any number of DF players who are also UI designers (hey, there have gotta be a lot of 'em) could propose a dozen minor interface changes that would make DF feel more consistent, and easier to learn without even changing any of the mechanics of the game. You couldn't say that about vim. (Or, if you could, I'd love to hear your interface change proposals for vim!)


Move the default movement keys to the home position -- shift them right by one. Right now I have to choose between shifting my hand off the home position or my index finger doing double duty. JKL; is very comfortable to move around with.

But that's totally possible with a slight change in .vimrc. I would argue that since you can change it in .vimrc your proposed rebinding is still part of the vim "interface".

This sounds right. And there's a reason more people don't use VIM, or play DF.

Indeed, the high barrier of entry, but for those of us who have entered, we enjoy a very privileged membership that just keeps on giving.

I'm a ten year veteran, and have helped many other "hardcore" gamers to get into it, but none of them have stuck with it, citing difficulty. It's a shame, as it has more to give than any other game.


Agreed but it's quite fair in the context of modern gaming. A lesser game might even have a tech tree and a progression acting as a tutorial. "You have reached a population of 20! You can now build cabinets and statues!"

Please do not get me wrong. Dwarf Fortress is a fantastic game.

The only thing I am saying is that the usability is complex with respect to what I am used to, in games.

It reminds me of using flight simulators. Many instruments and controls that you need to learn and understand.


Because I am old now and am not as easily entertained by being in fantasy worlds all by myself any more, I can't handle it. But games like this that are either MMOs or small scale persistent worlds where you friends can join (like minecraft) are still really fun for me.. Building an amazing Dwarf Fortress would be much more fun if your buddies are in the world too.

I disagree, respectfully, I think that the stop-start nature and complex workings of DF would mean that any sort of co-op would severely constrain the game, well at least with respect to fortress mode.

I suppose there could be an argument for taking adventurer mode and making it more like a MUD but I'm not really an AM person.


I would love to play a DF equivalent on a dyson ring. The cooperative elements would be for ring maintenance - keeping the station-keeping thrusters in good health, keeping the trams between territories running, whatever. The balance of the world play could certainly be enough for anyone even dividing the ring up into per-player planet-sized chunks.

You'll have to invest a couple of evenings reading the tutorials/installing the right stuff/setting up a playable world/getting lost in the awful menus/etc. before you get to have fun, but it is worth it!

Dwarf Fortress is an example of how a great idea can be held back by a horrible user experience. The UI is a nightmare, and the performance worsens as the game gets bigger.

I've recently started playing Rim World, which is essentially a Dwarf Fortress light. I'm enjoying it way more than I enjoyed Dwarf Fortress despite being a less complex (relatively speaking) game because it offers a FAR superior interface and presents it's mechanics in a friendlier way.


Perspective?

I find the DF UI, while minimal, to be great. Everything can be done on keyboard shortcuts... its more challenging initially, really easy as soon as it clicks... then the idea of doing it all with a mouse seems crazy


The problem is not keyboard vs mouse.

IIRC (foggy memory, several years since I last played) there are several ways to move through lists, depending on which list you are. Military was a chaos. People recommend an external application for better review of your dwarves, as the main interface is clunky.

It's a great game, but the UI could be much better.


Utilities is part of the game ecosystem though. Dwarf Therapist is great, but after 30 dwarfs it becomes better to enable autolabor. To be fair, I'd like to see RimWorld* manage 200+ pawns.

* Not to say I dislike RimWorld, in fact it is my favour dwarflike.


Oh, the agony of the military system. It has a learning curve all of its own.

RimWorld is great, but it just feels so much more sparse than DF - you only have a handful of folks. Graphics are nice (and the Prison Architect style is nice as well) and the way to assign orders is much needed (identical to a DF community add-on), but when it comes to Building Your Thang, it's just not a bustling community working against the odds.

It's interesting that so many things are based off what DF lacks (basic graphics, ui-friendliness) but don't implement the heart of DF (complexity), which is admittedly hard and was built up in DF over the course of years.


It's not either the traditional DF UI or a mouse-driven interface: other alternatives would be an improved keyboard-driven interface or an interface which uses both the keyboard and the mouse well (yes, I'm aware that technically DF does have some mouse interacton: http://dwarffortresswiki.org/index.php/DF2014:Mouse_control).

Dwarf Fortress is an amazing game with an incredible engine, but its UI is notoriously terrible, and IMHO deservedly so. But then, I've never managed to last very long, so perhaps there's something wrong with me.


The UX is a hurdle. There are inconsistencies and the last time I played, third-party utilities were pretty much mandatory in order to manage your dwarves effectively.

That said, the great thing about the UI is that once you've learned it you can discard it. If you know the commands you need, you can hide the 'commands' area to give you more information on the current gamestate.

FWIW, DF is absolutely fantastic at doing what it's designed to do. It's primarily a story generator not a game, which is partly why interactivity is almost an afterthought


I find the DF UI, while minimal, to be great.

Do you use Dwarf Therapist?


I think I need one.

It's not even the UI, or the graphics. It's the fact that the game exists in some sort of "hobby" or forum limbo. That is why it's not more popular, and simply exists in a perpetual status of some sort of "geeky" reverence.

You search for it, and the main page for Dwarf Fortress is on a "bay12games" domain rather than its own. No one knows "Bay12Games", and it's been 10+ years since Dwarf Fortress was first released. The guy running this should get over it, and admit that the only reason he's got some sort of fame is because of a single game. He should focus on that, not "Bay 12 Games". That company doesn't even have its own Wiki page.

Then the page itself. Looks like it's made in the early 2000's and hasn't been updated since. I just, I don't even know where to start with this one.

Then the actual download for the game: "df_43_05_win.zip". No installer, no MSI, nothing. I opened it just now, and couldn't find any sort of "Instructions" file or manual, either. Not to mention that all the "nicely packaged" versions of Dwarf Fortress exist in random forum threads.


> The guy running this should get over it, and admit that the only reason he's got some sort of fame is because of a single game. He should focus on that, not "Bay 12 Games".

AFAIK, that's all he's doing right now.

Also, Liberal Crime Squad is entertaining too and I know there's a few communities around it.


> No one knows "Bay12Games"

Really? You can find some of us here. Length > 0. http://www.bay12games.com/champions.html

> The guy running this should get over it, and admit that the only reason he's got some sort of fame is because of a single game.

And if you don't mind me being so blunt, what is your reason? I don't know of zo1.

> Then the page itself. Looks like it's made in the early 2000's and hasn't been updated since. I just, I don't even know where to start with this one.

Some of us remember the internet back when it looked like this. Some of us miss the internet when it looked like this. For this reason, many of us like the website to look _just like this_.

> Then the actual download for the game: "df_43_05_win.zip". No installer, no MSI, nothing. I opened it just now, and couldn't find any sort of "Instructions" file or manual, either. Not to mention that all the "nicely packaged" versions of Dwarf Fortress exist in random forum threads.

Now, I'm going to do something rare and separate from the usual jovial friendliness associated with the community to give make a very stark statement: if you can't figure out how to run the game, you probably shouldn't play it.

Now, returning to a more friendly disposition, I think you need to chill out a bit, stop picking everything apart so much, and start enjoying things for what they are.


">> The guy running this should get over it, and admit that the only reason he's got some sort of fame is because of a single game. >And if you don't mind me being so blunt, what is your reason? I don't know of zo1."

I was a tad bit to harsh for polite conversation; but I have no idea why the creator couldn't let a community helper just fix things up a bit. Give it some love, so to speak, like it seems he does to the game itself.

>"Really? You can find some of us here. Length > 0. http://www.bay12games.com/champions.html*"

Sure, length of the amount of "champions" is approx 2100. Bay12Games is still a pretty unknown game company. But everyone knows Dwarf Fortress, so my point still stands. In case you glossed over it: It was a bit of hyperbole.

>"Now, I'm going to do something rare and separate from the usual jovial friendliness associated with the community to give make a very stark statement: if you can't figure out how to run the game, you probably shouldn't play it.*"

It's not that it's difficult to get into, it's that the whole presentation of the game and how it is presented to the user are completely unpalatable. Even before the user gets to see the game UI on his own machine.

Am I saying that your community needs to cater to me, make the game easier, etc? No.

I'm not trying to pick it apart, just adding my point about what sort of disservice the surrounding presentation of this game is doing to it.


And yet the guy makes a comfortable living on what he does have, despite the fact that it is 100% donation driven. For me, one of the most interesting things about DF is that virtually everything is done "wrong" and it is a massive success. How many fingers do I need to count the number of people making a living off their programming art project?

Could all these things be done better? Absolutely. However, he has found a way to prioritise the things that are required to make him successful. I look at that priority list and think it's insane, but you will note that I don't receive my salary from kind hearted people who appreciate my quirky way of doing things. I get paid to do as I'm told.

It's pretty easy to think "I could do it better". WRT to DF, there are so many places you can criticise, that it's not funny. But, TBH, I don't think I could do it better, no matter how crazy I think his way of doing things is.


Gonna call this for the elitist bullshit it is.

I play, know the bindings etc, but the UI has enormous problems with consistency.

The older interface like the jewlery encrusting and labor assignent work in a completely different way than the less old interfaces like military management and hospitals, which are yet again inconsistem with the latest interfaces like libraries and taverns

Some room get built on the room menu, oter in the zone menu, both can have either modern or old menus to manage assignment and roles and while the whole thing is workable is far from being great, good, easy or even likeable.


I agree in a lot of cases. It's actually really interesting to watch development because you can see him incorporate less obtuse patterns in the new interfaces as the game has progressed. Some of the most headache inducing parts are the oldest and (from how he's described it) more brittle.

I frankly don't blame the guy for wanting to keep it closed source.


Honestly I feel the opposite.

I agree Dwarf Fortress is quite hard to learn and the UI plays a large part of that as it's completely keyboard based and requires a lot of upfront effort to learn, but once you do you become much faster than you would otherwise (much like text editors).

People often complain about Dwarf Fortress's graphics in the same vain and breath as the UI, but I think these are parts of the charm and instead of being weaknesses they are leveraged as strengths.

For instance, the lack of fidelity of the game allows any new character to be added in 2 seconds, yet Rimworld needs a considerable amount of time and effort developing each texture. To an extent I think the 'horrible' user experience cannot be divorced from Dwarf Fortress. Losing is fun, after all.

I think what you consider a 'horrible' user experience cannot be divorced from what we know as Dwarf Fortress today. Losing is fun after all. I love Rimworld and Dwarf Fortress, but they occupy different spaces within a similar genre in my mind.


I agree about the UI, but the performance degradation is still a significant issue.

Kill cats before they adopt a dwarf. Clean up old clothes with dfhack, and stop making dams!

The other big one is making sure dwarves don't path to complicated places, or animals don't path at all. Needing to know about how the game is written to make it useable is not great, though.

I think it's part of the package.

Name me anything that is perfect without flaws and still useful? Water is great to drink yet it gets you wet and drowns you if you breath it. Fire is lovely and warm and cooks our food but burns and melts our flesh if we play in it. Lids keep felt tip pens wet, and choke our children. Fish are pretty and taste nice but occasionally swallow fisherdwarves. The internet is great and connects us but causes divide by mixing people who were better separated. Light is fantastic, but I can't sleep if I can see any. Tea is just perfect, but it mAkEs Me ShaKe toO mUcH aNd ThaT caN be a ProbleM.


We're not asking for perfection.

> Water is great to drink yet it gets you wet and drowns you if you breath it.

If we're going to carry this analogy to Dwarf Fortress, if water was DF, it would be great to drink unless you happened to drink it from an aqueduct that carried it more than 20 miles. It would also make you spontaneously explode if your cat also drank from the same container, and you would be able to demand that your landlord replace your front door with one made of water and would go berserk when he didn't comply.


But isn't figuring this out part of the !!FUN!! ?

Here's the thing; I like the !!FUN!! of Dwarf Fortress. I like knowing that some giant with poisonous spikes can come in, kill half my dwarves before succumbing to arrows, and let me carry on as usual... until I realize that cats are tracking poison into the kitchen, which leads to everyone dying. That's !!FUN!!.

Not fucking remembering how to engrave gemstones because the key combination is unlike anything else, or how to put down a butcher table, or how to to set up a training rotation for my military? That's just shitty design.


That's actually a good point - I like that I can figure more about the game by looking how it's made.

That's the point of it, and I think it attracts all sorts of interesting comments and exploration, like the massive study on the impacts of various types of arrows and bolts on subjects.


If you haven't played Dwarf Fortress recently, the performance seems to have improved nontrivially in recent releases. I was out of the game for a couple years and came back to a big jump in perf when I've got a lot of dwarves.

I'm reading a play through story posted elsewhere in this thread[0] and inbetween instalments by the OP there are plenty of comments by people talking about how their ability to play the game is constrained by fps.

This is from Feb 2016, are the updates you're talking about more recent?

[0]http://www.bay12forums.com/smf/index.php?topic=156319.0


I agree. I've played DF for probably about a decade now, and the UI isn't an issue. I know every keybind so I don't really need to go searching for it. Yes, to begin with, it is a bit of a struggle, but the wiki (http://dwarffortresswiki.org/) is probably the best game wiki out there.

Get past your dislike of the UI and you'll discover a game so rich it will outlast any other.


But thats just the thing. Its hard to get past the UI. You really have to be dedicated to get over the steep learning curve of both the game mechanics and the UI. For new players its often just too much effort for a game.

Ideally the game should be easy to learn the basics, but hard to master. The UI hinders this significantly.


Well, my game recommendation may seem out of place, but if you look for building&fighting strategy game that is very easy to learn and awfully hard to master, try GO (baduk/weiqi/igo).

And you play using a nice physical goban, the UX is really hard to beat.


People make tools that reads memory of DF to create better GUIs. Compare dwarf therapist to standard gui of DF to assigning jobs, it is just horrible.

I honestly kinda stopped playing after military gui is changed. The new system is unplayable. New conversation system is also very frustrating in adventure mode.

I just wish the game will be open source at some point and the things will be improved.


I want so badly for dwarf fortress to just be an API to the simulator and let the community build the user interface and view layer. That would make my life complete. And also completely unproductive.

That only solves half the problem, though. There's still the issue that it's a single-threaded game which allocates and accesses memory randomly on the heap. So there will always be a limit on how large and complex the game world can be.

What I'm still waiting for is the game that comes after Dwarf Fortress, the game which is appropriately engineered for performance and is thereby able to be even more ambitious. We've seen a ton of less ambitious DF clones, but nothing on the other end so far.


I have to think that anything more ambitious than DF would either take a crazy genius with lots of free time or a large team. Possibly both.

I don't think that the criticism is that easily dismissible.

It's not so much that it's keyboard controlled (that's great for the reasons you mention) or that it's hard (that's ultimately part of its charm). It's that the interface is inconsistent with itself. A great example (from when I last spent any comsiderable time with it) is that different menus use different controls and mechanisms for selecting a menu item. Some have you type a single character, while others have you scroll through them with varying pairs of keys for no obvious reason. It's a natural result of organic growth and IMO outweighed by a great game, but the cognitive load of using the interface definitely isn't its charm for me.


The criticism is not being dismissed. It is simply an opposing and simultaneous viewpoint.

The jeweller's workshop UI is a weird disaster. You need to specify specific a task to cut a specific type of gem. You can make it repeat, but the task will erase if that gem type is unavailable.

Then you need to create a task with a specific type of gem to encrust, again the task will delete if that gem type is unavailable.

Finally the encruster will select an item in your fortress at random to encrust with that type of gem.


There should really be an option to handle gemstones by value rather than mineral type, which would require (and exercise) the appraisal skill.

So if you want to cut semiprecious stones, queue a task for cut gemstone, worth ¤1 to ¤5. It would also be nice if you could specify the type of cut, since the game engine already randomly produces cabochons and baguettes and cushions and such.

The encrusting randomness can be controlled by locking the jeweler in a room with only one encrustable item, but I shouldn't have to. If I want a masterwork silver hammer encrusted with small jade cabochons, I want to be able to specify that.

Also, building multi-part instruments is just insane.


dfhack has an automatic jeweler script now.

Once you have a manager though, you can set up jobs that are highly configurable and repeatable. Learning the job system really took my DF productivity to the next level.

> For instance, the lack of fidelity of the game allows any new character to be added in 2 seconds, yet Rimworld needs a considerable amount of time and effort developing each texture.

This is only true for tiny developers. If you have dedicated artists then you aren't gated in this way as you work in parallel.


If you pay artists, development is expensive. There is pressure to sell and to break even. Is there any game which is developed continuously for years and pays artists?

Certainly there are a number of AAA games that fit that description.

At this point we are conflating too many topics to really talk clearly. Simplest response would be "you should plan to spend years developing if you want to and if you do you need a budget and a way to fill that budget". See early release games as one example, expansions are an older one.


The problem for me is... I love simulation games (love Rimworld and Prison Architect for example) but the Dwarf Fortress learning curve is too much for me. I've attempted a number of times and learning it just was not fun for me at all. I want to enjoy good simulation games, but I don't have enough free time to force myself to play something that I don't enjoy until the point where I've learned it enough to enjoy it.

For instance, the lack of fidelity of the game allows any new character to be added in 2 seconds

I get this and actually completely agree. I loved MUD's back in the day because they could be so much more complex and detailed because they didn't have to worry about graphics and whatnot. Rimworld and such also can do a lot more than fancy 3D games for the exact same reasons, so there's definitely a spectrum.

But having said that, I still find the DF UX to be almost unapproachable. It could be low fidelity for the reasons you mention without being quite so hard to learn and internalise, but I get the impression that its just not a priority for the dev at all. Oh well, its their choice, but I feel a bit sad that I'm missing out on an otherwise amazing game.


> I want to enjoy good simulation games, but I don't have enough free time to force myself to play something that I don't enjoy until the point where I've learned it enough to enjoy it.

I am the same way. If I wanted to struggle with difficult tasks, I would work. I realize this puts me more in the "casual gamer" category.


You might enjoy Stardew Valley then.

"In the same vein"

> I agree Dwarf Fortress is quite hard to learn and the UI plays a large part of that as it's completely keyboard based and requires a lot of upfront effort to learn, but once you do you become much faster than you would otherwise (much like text editors).

Is there a VI interface?


I've wanted to love Dwarf Fortress, but the fans sometimes come across like Stockholm Syndrome victims when you mention the UI and UX.

I mean I came at the game from a long history of playing Roguelikes, so the ASCII graphics weren't an immediate turnoff, but figuring out how to do anything is just plain daunting.

Also, there is a big difference between most Roguelikes and DF on the graphics front. In Rogue you start in a small room with just yourself and maybe a treasure or a monster. New monster icons are introduced slowly so you can learn them at a reasonable pace. In Dwarf Fortress you are apparently expected to learn dozens of symbols right from the start in addition to figuring out what keybinds do what and trying to figure out what you are supposed to be doing or if the fact that there are dozens of different kinds of rocks is important yet or not.

If there was ever a game crying out for a hand holding tutorial it is Dwarf Fortress.


This is one of the reasons why Youtube video tutorials of DF are hugely popular. I think there are probably even a lot of people who really enjoy watching them for the game play, but don't want to invest the effort in learning how to play for themselves.

Dwarf fortress is a great idea made possible by the winnowing effect of its UI.

I've found that this particular game has found its niche because of what it is.

The point of the creator is this: the game isnt even past pre-alpha. Lots of things will change, which a UI will only bog down.

So for example rimworld is still currently 2d. (will possibly change), but that a big change

DF plans to change the full world gen and design in the next arc - its a reasonable view.


This. The UI is the final step. Right now it's about experimentation and determining the ideal game mechanics. The obtuseness means that the most obsessive and hard core are your play testers. This is a bad UI as part of good game design for what the current goals for dwarf fortress are.

As far as I understood, the UI is there to stay. The creator wrote his own game engine that is optimized for printing ascii to a screen-- it's not architected in a way that can plug-and-play with a different renderer.

Cogmind is an interesting example of pure ASCII holding itself to a very high UI standard.

Here is a quick animated gif with a few examples of interactions:

http://www.gridsagegames.com/blog/gsg-content/uploads/2014/1...


I would absolutely love if DF looked that good at some point! You're definitely right that just because it's ASCII doesn't mean it can't be visually appealing

it took an hour here just to get a half-ways readable font. after a while the game was completely destroyed by constant notifications. so I couldn't agree more with the horrible user experience.

I have played both, and while I got bored of RW as soon as I got to the siege stage, I had to force myself to stop playing DF even though I still had about a dozen ideas for lava traps. YMMV.

Edit: I still have to stop myself from re-downloading dwarf fortress about once per week because there is so much left to explore. I just barely scratched the surface of power stations and automated rail systems. You could spend thousands of hours just designing traps to kill goblins. The fun is in the complexity. Perhaps you just did not get deep enough into the game. Now I have to talk myself out of downloading DF again. The sheer maniacal glee of knocking a dozen goblins off a narrow path (use a rail car attached to a pressure plate) into a pit and filling it with lava after spending hours setting it up just can't be described.


To be fair, RimWorld is a much newer game than DF. It hasn't had nearly as much time to add features.

I think I agree. I'm up for complex keyboard interfaces and simple, ASCII-ish graphics - I play a lot of roguelikes - but something about Dwarf Fortress's interface still rubs me the wrong way. I do want graphics to be pretty, even if they're simple, and I do want controls to be elegant, even if they're keyboard-based.

The default tiles are ugly - unnecessarily ugly - and the keyboard controls don't seem great. (Like a sibling to this comment says, consistency has a lot to do with it, and not having to move your hands far around the keyboard is nice.) Then again, I admit I haven't spent that much time with it.

So I do find the interface offputting, but it's not because I want a flashy, dumbed-down one. And it's a bit frustrating whenever I see criticism of the Dwarf Fortress interface dismissed in those terms.


I don't want to disparage Rimworld, it's great.

But the game pales in comparison to the beauty and replayability of Dwarf Fortress. I've been playing DF for 5 years and I still feel like a novice. Toss in modding and the game goes even further.

I understand your point of view, but I'm sad that amazing and well-designed games like Nethack and Dwarf Fortress don't get a fair shake because they're fixed width tile games. DF is a simulation game of stunning proportions.


Agreed. And at its current development I wouldn't be surprised if Rim World would reach just as much complexity as DF, of course always lacking a few years behind (more than a decade to begin with). But as you said, due to a better interface it's probably in the leading position.

Got to meet Tarn Adams a few years ago when I was into Dwarf Fortress and the community. Really cool guy, you can tell he loves what he does.


Wow, the creator of Dwarf Fortress doesn't use version control? I don't even know what to say about that. It's like when I heard the creator of KeePass doesn't use version control either... why are people so resistant to VCS?

I've been trying to understand how the author of DF does development. My impression is that he doesn't really add features the way most of us add features. Instead he'll rewrite large portions of the game, or he'll write some stand alone test application and then integrate it into the mainline code. Apart from safety (I really hope he has backups), there probably isn't that much use in VCS because he intends to simply rewrite everything every time. He also does not like to have anyone touch his code. He allowed a few people to improve the performance of his display code once and has since sworn off it. While the performance improved as promised, he says that he can't understand any of the code and now never touches it again.

Not the way I would do my work, but when you are your own boss you get to do whatever you want ;-)


I heard once that he is burning the code on discs once in a while.

I'm glad for the repost. Going to try this game.

I understand the keyboard only controls, but man, seeing what is going on is a challenge. Yeah all the PC power is used for the emulation of the world and stuff, but how much power does it take to do 2d sprites these days?

Oh well, here I go losing again!


As someone who has wasted far too many hours in games, i believe that DF to be simply the best game ever made. The UI is rather horrific but something you can get used to. Admitedly it took so many efforts to figure out the game but that becomes half the 'fun'. Anyone wanting to play, find a lets play video (for the current version) and just follow along. Sounds silly, but thats how i finally got it.

Semi-related: Don't Starve is a great indie game with a different theme but IMO similarly complex world simulation. [1] [2]

[1] https://en.wikipedia.org/wiki/Don't_Starve

[2] http://store.steampowered.com/app/322330/


I found DS to be too hands on. I like to be more elevated, with my pawns having some autonomy.

How are they pawns if they have autonomy ;-)

I love Don't Starve, but it doesn't even approach the complexity of Dwarf Fortress. I think the similarity is mostly thematic, both being pretty dark and having a techno-fantasy element.

I have enjoyed many, many hours with don's starve. The modern randomized, living world games are a lot of fun.

Dwarf Fortress' complexity is far, far superior. It's not comparable.

More relevant, their latest game: https://www.kleientertainment.com/games/oxygen-not-included

System-driven space colony simulation, with a particular emphasis on maintaining breathable air.


>space colony simulation, with a particular emphasis on maintaining breathable air.

Seems only logical...


Dwarf Fortress was one of the first computer games that I really ever got into. I was a little 15 year old that had a shitty desktop computer that wasn't always connected to the internet and DF has provided me with many tens of hours of entertainment. I remember the depth and realism included in game caused me to want to learn to program myself. I wish the UI was better, but honestly I think the engagement with the keyboard in DF is it's strongest selling point. Every game action can literally be performed with the keyboard, and this means that I don't need to wait for the text on menus to mentally register, move a mouse, click. I can essentially type the commands at speed.

TLDR: DF is a great game that shaped my childhood and motivated me to become a programmer.


> Every game action can literally be performed with the keyboard, and this means that I don't need to wait for the text on menus to mentally register, move a mouse, click. I can essentially type the commands at speed.

This is a sentiment that crops up again and again. In particular as long time users of a terminal ui (not necessarily CLI but using ASCII/ANSI to draw the interface, and keyboard for input) when they transition to a GUI replacement.

The thing about keyboards is that they are designed to be used without looking. Special keys have special shapes to make them stand out. Every key is in a fixed location. And certain ones are marked but a bump or similar to make them easy to find by touch and thus use as navigational aides.

Yes, mouse and touch screens make it easier and faster to localize and develop a UI. But we humans are quite capable of doing repetitive tasks with our hands without looking.

I think that in RTS and MMORPG circles there is a certain degree of disdain for "clickers". People that operate the game mostly via mouse. This because it is slower and thus limits how quickly they can react to in-game events.


> And certain ones are marked but a bump or similar to make them easy to find by touch and thus use as navigational aides.

Completely off-topic, but there are two keyboard changes over time that absolutely drive me bonkers.

First, caps lock replacing control on the left, but that's not what prompted this post, but I'm still grumpy at Sun and Apple for surrendering to the idiot PC world on that.

Second, moving the bumps from "d" and "k" to "f" and "h". Pretty sure Apple also had that right originally.

When the bumps are supposed on the 2nd finger, you always feel them, even if your fingers are offset by 1. Feeling the bump on the wrong finger is great feedback to fix your hand position.

When the bumps are now in the 1st finger, if your fingers are offset by 1 you don't feel it at all, and it's much harder to notice the absence of something than it is to notice the presence of the bump on the wrong finger.

Bad designers, bad bad.


I think you mean F and J.

edit:

As for the caps lock location, i think you can thank typewriters for that.

As best i recall, they did caps lock by having a mechanical locking lever/key attached to the left side shift key arm. Thus to do caps lock you pushed down the shift key and then locked it in place with that key. Once done you popped the key loose with a press and moved on.

That locking key is pretty much in the same location as the caps lock key is today.


Bumps on my MBP late '13 are now F & J for what it's worth.

Meant f/j. d/k is better for the reasons I outlined.

I agree, I wonder what it would be like if you spent you're entire computer upbringing with a brail keyboard, regardless of your sighted-ness...

Interesting about the bumps. Lately I've been rehashing some typing tutors (I learned how to type at a very young age). One of the pieces of advice I picked up was to retain your index finger on the home row when you press shift. This is actually impossible on some keyboards, but where it isn't impossible, I was surprised at how much better my typing was.

I learned on an Apple and I wonder now if having the bump on the second key has led to a few problems. I find there is a tendency for me to rotate my hand so that the fingers are no longer on the home row. One of the nice things about changing is that when I'm trying to code on the bus or train, avoiding rotating my hands keeps my elbows in and makes it possible for me to work (yes, I like working on the train :-P).

Not entirely sure what the best answer is for this.


I recently bought my first desktop keyboard and pulled all of the keys off and created my first proper Dvorak keyboard, but in the process created a keyboard without proper bumps in the home row. What I did was cut a double thick strip of electrical tape and stuck it in the correct position. This subtle bump of tape adheres extremely well and feels just as good as the button keys do.

>This is a sentiment that crops up again and again. In particular as long time users of a terminal ui (not necessarily CLI but using ASCII/ANSI to draw the interface, and keyboard for input) when they transition to a GUI replacement.

I used to play Bard's Tale when a kid and I had memorizedthe keyboard movements from town to various dungeons on the map, and would navigate the map without even looking as it was muscle-memory for where things were... I always loved that.


One of the central questions of our time may become whether or not it's ethical to turn off dwarf fortress.

It is probably ethically no worse than any killing simulation (e.g. first-person shooters). Creatures in such games arguably do not have a rich and lasting experience of processing information as humans and many animals do, so it only has a small ethical importance. My moral intuition on this matter is that the real ethical questions begin with systems that perceive, learn and can experience pain/reward: http://petrl.org

You may have missed the joke (such things are hard to judge by the medium of text). The parent was implying that the complexity of Dwarf Fortress will grow to the point of sentience of the dwarfs themselves.

Indeed, it is all about sentience. If we assume it's unconscious, which we probably can, then it's morally irrelevant in my opinion.

It is probably, in a limited sense, ethical to suspend (save) the game, but not ethical to destroy the saved game file. When the game is off, the dwarves do not experience an interruption of continuity to their time. So on the face of it, there is no harm done to the dwarves if there is an existential interruption from the perspective of some external context.

However, the abstraction of the dwarven universe from the universe in which it is being simulated is hardly complete.

By suspending the game and keeping it in storage for several days, for instance, you are placing the dwarves at greater and greater risk that some external-universe course of events will prevent their universe from ever resuming. Furthermore, as the cycles of time which can occur in the universe ultimately face a limit with respect to the rate of the processor and the cycles of time available in the external universe, so in a very real sense you are causing a sort of existential damage to their universe.

This, in fact, raises the question of ethics with respect to the resources devoted to the simulation. Is it ethical to run dwarf fortress on a consumer-grade general purpose computing device? Should not the dwarven reality be given the strongest underpinnings we can provide it, with equivalent respect given between an internal universe subjective & sentient existence an external universe subjective & sentient existence? In fact, given our moral debt as the creators of the dwarven universe, do we not actually owe them more than we owe each other?


Heh, falling into the void of never executing again is a much better outcome than normally occurs when I'm playing. Such as building stairs and ramps in the wrong place.

I think I'd enjoy philosophy a lot more if it was always presented in dwarf fortress terms.

The dwarves may simply wonder why Armok chooses to prolong their suffering.

How did Dwarf Fortress become popular?

I first saw it because someone made a calculator within their fortress. That interested me, but I don't know of any particular thing that made it popular with "the masses".

A lot of it had to do with creatively told stories about the events in a playthrough, particularly with some succession games - games passed along to others after a set time elapses. The first one I'm aware of is Boatmurdered:

http://dwarffortresswiki.org/index.php/Bloodline:Boatmurdere...


The complexity allows a lot of emergent gameplay.

For instance in one game a dragon attacked my fort. NBD. Everyone got inside and I was planning to catch it in a cage trap. Then use it as a decoration.

However there was a group of Human traders at my Trade Depot. They decided to run over and fight the dragon. Now the dragon is running around breathing fire. Which started a forest fire. Which destroyed my wooden bridges and doors.

Completely unscripted fun, just a result of the game simulation.

Also the game manages to make you hate goblins and elves through your interactions with them.

In evil regions eyeballs grow out of the ground, blinking at passing dwarves. However cattle will happily graze on them.

If you "delve too greedily and too deep" you'll unleash horrors of the underworld.

There's just a lot of neat stuff there.


I never understood the appeal of Dwarf Fortress until a good few hours into my first game I had just finished following a detailed tutorial and was starting to get a hang of the interface. I decided to start work on my first big construction project - a deep pit to keep the regularly attacking trolls away and a gargantuan drawbridge to lead over it. It's probably child's play by any non-noob player's standards, but I was so proud of it. All of the noble dwarves came to the grand opening of the drawbridge - the mayor, the sheriff, the captain of the guard and all of the highest ranking military dwarves in their bright shining armor. They all stood at the base to witness the first lowering of the grand drawbridge. It was lowered and with a mighty crash, flattened the gathered nobles and military elite...Because I had placed it facing the wrong way.

The fortress didn't last too long after that. I tried my best to clean things up, making a nice new dining hall and brewery, but the remaining dwarves took it pretty hard - one of them ended up losing his mind while drunk and slaughtered most of the rest in the great dining hall, the rest died of disease because nobody wanted to clean up the mess. I was so proud of my towering (and first!) fortress, and I lost it all because of one stupid mistake, but it's definitely the most fun I've ever had losing a game.


To anyone that's a fan of these kinds of emergent stories, I'd recommend giving crusader kings a go too.

I played NetHack for a long time and wanted something more. NetHack becomes pretty repetitious after a while. I tried various roguelikes and hit upon DF.

Also, I see Dwarf Fortress as the natural progression of programming games. As programmers, we always look to abstract away busywork. DF takes away the busywork of dungeon-building and hands it off to little applets.

Basically it became popular by word of mouth. I think the popularity of Minecraft led more people to hear about it, since Minecraft was said to be influenced by it. But as far as I know, it's always been a slow build, never went viral or became an overnight success.


One of the things that I find interesting about Dwarf Fortress is that (to use programming jargon) it's sort of a declarative game rather than an imperative game. In e.g. StarCraft you select an individual unit and demand that it move to a specific point on the map; in Dwarf Fortress you configure which dwarves are allowed to perform certain tasks, then you place a task in a queue, and some dwarf somewhere will (eventually (hopefully)) take care of it (until they get distracted by a party, or decide to go fishing, or get hungry and wander off to the dining hall, or fall asleep in a stockpile, or drop anything they're carrying and run screaming from the forgotten beast hurtling down the hallway at them). It's a fascinating difference in paradigm, and I wish more games would explore the idea of actors in the world being chaotic/free agents which will only somewhat prioritize your wishes.

I remember a strategy game called Majesty that worked in a "declarative" way as well. You didn't order your units around, you could just place bounties on things and your soldiers would go out and kill/destroy them

I owned that game, but seems to have misplaced the discs it came on...


There's a HD version on Steam.

Okay

yeah, i loved Majesty. You couldn't force anybody to do anything. That way it was more chaotic.

Many times you could have won quickly if only the units obeyed. But they didn't ;)


Yeah, your level 5 wizard would try to take on a vampire on his own, and quickly expire, instead of leveling up on rats like he ought to. Loved that game, and loved the voices. Don't buy the iOS version though.

What do consider lacking on iOS?

Assuming it's the same as the Android version, it's superficially polished and internally hollow compared to the original. I posted more here: https://news.ycombinator.com/item?id=13760412

It is available for iOS and Android I believe. Works great on mobile.

The mobile version seemed like it would be fun if it was your first experience with Majesty. It doesn't hold a candle to the original, though. Fewer heroes, monsters, missions, and worst of all, dramatically simpler and less character-ful AI, with few apparent behavioral differences between classes.

Also the art was entirely redone in the generic cartoony style of every other mobile RTS. I suppose it's more approachable than the original's 2000-era pixeliness, but it didn't seem like an improvement to me.

It's still better than 99% of all mobile games, but that's not saying much. The original is on Steam for $10 including the expansion and various bugfixes; I played it through again last year and I think it still holds up wonderfully. http://store.steampowered.com/app/25990/


I LOVED Majesty, getting high level wizards was my favorite thing, and could honestly be quite tricky considering how blatantly suicidal they often were. XD

The saddest part was encountering enemy units who had magic reflection.


Dungeon Keeper used to play quite a lot like this as well. It was an interesting aspect that set it apart from other strategy games that were around at the time.

Yep. You needed imps to dig out rooms, reinforce walls, and carry corpses (turned into zombies or something in the graveyard room) and loot (used to pay of your other monsters and cast spells, iirc) back to your dungeon.

You could do things like pick up imps to make them prioritize a task, or slap them to make the work faster (took a slice of health out of them though).


Evil Genius is similar in spirit to Dungeon Keeper and a fantastic game as well.

Oh dear deity i had forgotten about that one.

I seem to recall now that one of the ways you could torture a captured agent was by having him being squeezed by moving bookshelves or something...


Your comment made me think of neural nets. My knowledge is admittedly quite surface level but there seems to be an analogy to the emergent behavior of trained network. Also, I'm thinking about the recent (latest?) SimCity with it's much discussed ability to track each individual actor, which apparently became a computational crutch that limited the size of the cities you could build.

It seems like there are some interesting game mechanics possible (as you suggest) that could also benefit from the recent GPU enhancements for neural nets.


Interesting point but it's not exclusive to Dwarf Fortress there is a whole category of this kind of (management, simulation) games from SimAnt to Football Manager.

Heh, we used to call some of these games "managing a spreadsheet" -- Specifically the "Masters of Orion" series

Aurora 4X is almost literally that, but a very interesting take on the epic space 4X nonetheless.

The first one is pretty approachable. The hardware constraints of its time made the game very focused in comparison to other 4X games.

Have you ever played the game "Knights and Merchants" (1997) -- it was such a fun game, as you had your little serfs who had to do all the work, but you also had your little armies that you could fight another player with.

The AI was flawed, the serfs, once the population got to around 100 or so, would starve to death frequently as they couldnt figure out how to get to the inn to get food and drink...

It was especially fun and funny when playing against a friend in the same room, you've built your village to the point where youre able to raise an army to smite your foe - but the AI goes bonkers, the serfs start to die off and your economy is ruined to the point that even two soldiers can show up and raze your entire town...

Ill never forget the laughter we had at one's serf's death knell "HNUUGGGHHHH" as they die, and the giddy feeling you had if you could take them out with just a few soldiers.

http://www.knights.de/


There is a remake with an improved AI and online multiplayer! If you're interested: http://www.kamremake.com/ it's free, but you need to have the original game to play the remake (which you can buy on steam, if I recall correctly).

i now have a mac........... i think you get my point

> It's a fascinating difference in paradigm, and I wish more games would explore the idea of actors in the world being chaotic/free agents which will only somewhat prioritize your wishes.

Check out Banished.


I really wanted to love banished, and I did love it for a while, but I found that once you learn how to create a stable town, you can apply almost the exact same strategies in every single game played no matter what the land looks like, as well as every time you branch off to expand an existing town into a new piece of land.

The mods didn't really do anything for me either, I felt like they just increased complexity and unbalance without really adding challenge or fun.

I haven't ever actually played dwarf fortress, but from what I've read, the inherent unpredictability of the world, your dwarves, and other creatures seems like nothing that any other game can even dream of.


Plus the AI in Banished liked to screw you over, and not because it was trying to help. When one villager is dying from cold and just needs a single piece of firewood, another villager will take the last 10 from the stockpile and hide it in their house.

Never trust your neighbors.


Several of Peter Molyneaux's [1] games have a similar mechanic where you set conditions for independent agents to (hopefully) do what you need done (My experience was playing Populous II Trials of the Gods [2]). His most recent creation, Godus [3], also uses this scheme.

[1] https://en.wikipedia.org/wiki/Peter_Molyneux

[2] https://en.wikipedia.org/wiki/Populous_II:_Trials_of_the_Oly...

[3] https://en.wikipedia.org/wiki/Godus


The first Populous game did that too. It's very simplistic in Populous, though - it's mostly about flattening terrain so they can build or calling them to battle and helping them by smiting your opponents in various ways, and their actions are similarly basic. Then again given the machines it ran on, it's not unsurprising.

Black and White stands as one of my favorite games of all time for this reason. I'd argue it was the pinnacle of this behavior as far as Molyneux's games go because of the creature. It had a sophisticated AI that enabled it to actually be taught what was good and what was bad through experience. If it ate a villager you smacked it around and it'd learn it shouldn't do that. If it helped water your crops then you'd rub its belly and it'd know to do more of that.

It was especially interesting because in order to train the creature what behavior was or wasn't desired you had to catch them in the act doing that thing. Since the creature was an independent actor who you could only influence, you had to rely on the randomness of the AI to do certain things so you could teach them what was good and what was bad. E.g. you can't teach it eating villagers is bad unless you catch them in the act and smack them around. Such a shame they almost completely did away with that teaching-through-doing part of creature-training in favor of simplifying down the creatures behaviors to basically sliders like "good vs bad" and "friendly vs aggressive" in BW2. Still waiting on a proper spiritual successor to BW1... what a brilliant game.


No surprise the guy who made it went on to create this little company called DeepMind...

Was going to mention Dungeon Keeper in this thread, another one of Molyneux's masterpieces, and probably the most relevant comparison to Dwarf Fortress IMO.

Similarly to DF, you just allocate tiles to be excavated, and can paint allocations on rooms, but never issue direct orders to minions. (Though you can possess them and take a first-person perspective, which is an interesting twist).


I'm missing Dungeon Keeper 1&2 on that list :)

The attempt to manage a dungeon and direct all your minions at a certain goal, compined with the idea to automate your defence (through traps and alarms) somehow evoked a similar joy to programming for me. It's especally ironic on how many misbehaviors of the creature agents I tried by locking (ha!) doors in front of them.

Also, the general game style (aesthetics, soundtrack and the narrator) remains unmatched to this day.


I think it's kind of a spectrum of emergent gameplay. There are games with more than that than others, depending on how detached your commands are from the actual action that is taken.

There is a great and popular game called Rimworld[http://store.steampowered.com/app/294100/] that does this. It's a colony management game where the default mode is colonists acting according to priorities and rules you set. You can override if you want but you quickly realize this is only for rare situations. It's really a fun game and a much easier learning curve than Dwarf Fortress.

Most games have a smaller learning curve than dwarf fortress.

That's also the point.

Dwarf fortress is actually a good study on community structuring via challenge based gating. Before the many tutorials, if you could hack df or aurora, you were a particular type of gamer, and the challenge itself was the fun.

I really wish people could know themselves through the game the way I got to.


I play both, and the communities are wonderful places to have mainstream-unbiased discussion about games and game mechanic.

I'm not playing that much these days but I still frequent bay12 other games section, it's the best source for opinionated discussions about games


Yes. Over Christmas period I played 88 hours of RimWorld, says Steam. It is a degradingly addictive game. Edit: and I'm not a gamer.

now that is an expensive 2d game. holy shit.

I downvoted you.

£23 for a game actively developed, with great replayability, tons of mods, a passionate fan base and community? Holy shit, that's a bargain.


$30 for a game that's early access, may never be finished.

I think GP is also potentially balking at the $300 DLC pack (what could be worth $300 from a game that isn't finished?)


That's interesting, I've never seen a pre-release DLC that expensive. I wonder if they are going for a kickstarter-style 'fund us' approach with that pack.

Or maybe they typo'd the amounts...


Look up 'Star Citizen' and marvel/despair.

Those DLC packs are actually more like Kickstarter rewards. The text of the Pirate King DLC is this:

> This DLC gives you the right to enter a name and character backstory into the game, with skills, appearance, and special work requirements. In addition, your character will appear as the leader of another faction!

and follows with a note that says it does not affect game play, only gives you the right to add your desired content to the game.


Sure would be nice if people would look at what the DLC actually is before complaining about it....

I'd like to note that I wasn't complaining about it, just noting that at a glance it makes the game look expensive.

$30 for a game that's early access, may never be finished.

$30 for a game that already has more content than many AAA games I've played, is still actively developed, is in many ways more advanced than most AAA games (in terms of the simulation and emergent gameplay). If it were abandoned RIGHT NOW, nobody would ever notice that it wasn't "completed".

I've personally got more value out of playing it than I get from most $60 games and I haven't even played the last two updates yet (which were HUGE).

The expensive packs are expensive because they let you add custom stuff to the game (e.g. Characters named after you) and not meant for the average player.


It's more finished than many "properly released" games. There are many games released to early and quite expensively as EA, but Rimworld really isn't in that category.

Early access is a descriptor that isn't too useful except to say the developer doesn't consider the product finished for a 'first' release. Some developers will label a game as completed when it has serious issues that prevent most from having fun. Other developers will label a game incomplete even though there are dozens of hours of enjoyment to be found. A better judgment is to ignore what is promised, and instead ask if the game as it currently exists justifies the price. In the case of Rimworld, that was a yes for me. It may not be a yes for others. That's fine; we all have different preferences.

Take a game with a lot of free updates, such as Terraria. Consider the game the day it was released. What if it had been labeled 'Early Access' and incomplete because it doesn't have all the features that were given in post-release updates? That wouldn't make it worth any less on the day it was released.


The problem, as I see it, is that the nuance you're talking about isn't visible from the Steam store page. When I see a game that's Early Access, there's not a lot to indicate to me:

a) How far through the QC process the game is

b) Whether the game is going to change dramatically in the next 6 months

c) Whether the game is about to be a cancelled project.

d) Whether there are features/portions of the game that will just _not_ work with <insert OS version/hardware/network configuration/...>

Basically, Early Access is too big a tent to be meaningful, except to take as a caution of "you might be paying for vapor."


I upvoted him.

$555 of DLC for an early access game?


To be fair, "name a character in the game" isn't really "DLC", that's just the interface Steam has to give devs more money ;)

eh, i was talking about "buy the bundle" offer of 500€...

You did catch that the DLC is more like Kickstarter rewards right? The $300 DLC is getting your name and a backstory you create into the game.

no, i didn't. i did scroll down on the store page to see what the DLCs were about but didn't see any comparison. makes sense in hindsight considering that they don't actually change anything about the game.

RimWorld is kindof a cross between a sci-fi variant of Dwarf Fortress and Prison Architect, whose art-style it ripped off wholesale, with the permission of Introversion. I figure Prison Architect is surely worthy of a mention in this thread. (http://www.introversion.co.uk/prisonarchitect/pc.html)

Like DF and Rimworld, PA has that 'imperative' style, where your prison staff are given instructions, but may have problems carrying them out. Also, with the real-life setting, it might help stimulate some thinking about the nature of the penal system and the prison-industrial complex, while, at the end of it, still being a fun game.


Rimworld's play style is actually very little like PA beyond the general indirect management conceit. Even the way building constructions work is different enough to affect your gameplay (PA does all-in-one foundations, RW is more piecemeal).

I'll share an anecdote from my gamer bro in his early twenties. He likes Rimworld, but his complaint is that it's way too shallow compared to Dwarf Fortress. Here was his example:

"I decided to play a game of DwarfFortress in an evil biome. Only I didn't realize the weather was also evil. Within 30 minutes all of my Dwarves were dead from an acid rainfall, cause of death: drowning in their own puss."


Pus, I hope.

Could be a kiwi.

Dungeon Keeper was like this as well, except, IIRC where you could cast a spell and go into first person mode with a specific character for some time.

The Close Combat series is entirely based on this idea. You give commands to squads, but the individual soldiers respond to the conditions on the battlefield, the orders they have being but one part of that.

Yeah i remember that one. You could order a squad to move somewhere only to find them hunkering down behind a wall or something because they were afraid that some sniper was out there.

I think the map was even "deformable" in that if you ordered a tank to move somewhere it could break down walls and such to get there, walls that soldiers could have used as cover while moving around etc.

Note though that the graphics was top down 2D, so it was "cheap" to replace a standing wall tile with a broken one.


Close Combat is one of my favorite series ever. The morale and fatigue mechanics were some of the best I've ever seen. And the graphics were fantastic for the time, really excellent aerial photography-inspired hand-made maps and sprites.

There was nothing like setting up a textbook ambush, and then having your inexperienced gunner on an MG42 panic and run off when you gave the order to fire...


I remember buying Close Combat a Bridge too far at Virgin Records many many years ago. All those bright red and blue tracer lines suppressing my units and launching fog. I remember how underwhelming the tanks felt sometimes

It's been a couple of years now, but I still load up A Bridge Too Far every now and then! It's the only 20-year old game I know that's still fun.

It's getting hard to run on current machines, though, with its hard-coded screen sizes and a scroll speed that's unusably fast.


It's one of my favorites as well. I haven't run it in several years, but last I tried it you could get custom screen sizes by manually editing the configuration file.

The tanks definitely felt underwhelming sometimes, but overwhelming other times. It depended a lot on the environment, the type of tank, and the enemy units, which seems pretty realistic. It was interesting and terrifying how invulnerable some of the better German tanks could be.


Why not both? For example, the game of Risk where each battle is a game of Starcraft, and then Call of Duty at an even lower level. Maybe different players play the role of general or soldier etc.

With eve online to glue everything together.

Empire Earth 3 tried that(grand strategy-like overworld, rts-like combat), but poorly executed.

The total war series is a better example, but the games range in quality


The original Tropico game from the series somewhat felt like this, especially when it came to construction. Unless you were cheating, you could only declare where a building would be built - you had no power to instantly "plop" a building into existence - you had to then wait for your builder units to eat, sleep, get to the site and build.

It made you careful in where you place things. Placing an airport far from home? Good luck building that!


> Placing an airport far from home? Good luck building that!

Would you have to first build a road out there with public transit service stops, and then bootstrap an increasingly-large set of airport hotels for the work crew to stay at, before attempting the airport itself? If so, that sounds like my kind of game.


More or less that, yes.

To make large buildings in distant places you would build roads, then work crew related buildings, then whatever you wanted, including the needed housing for the new city area you ended creating in the process.

Another interesting game is knights and merchants, armies are under your direct control but require supplies, that need to be physically carted off to them, so you needed to either use hit and run tactics, or build infrastructure near the frontlines. Also traffic in that game can literally kill you, as your badly designed city becomes so inefficient that people die of starvation.


Settlers is like that too! Oh the memories.

I recently went on a hunt to find a game similar to the old Settler series, and found a game called Banished, where you have manage a small town against the elements. I put 10 hours into it so far and really enjoy it, check it out if you haven't already.

Of course you can still buy the old Settlers games.

Or check out Widelands, a free Settlers-like game.


I believe that we will see more games that have AI-ish systems. Randomization in games like Don't Starve is one of the mechanics that keeps the game going.

Well, I certainly think the Sims did that, to tremendous success.

Actually in the year 2017 there are loads of games with this indirectness, even quite a few new ones. Google it and you'll find quite a few interesting ones. E.g. gnomoria, or prison architect.

This. There are plenty of games based on Dwarf Fortress and indirectness, though many of them seem to suffer from lack of depth.

DF is nearly endless in terms of new content and events; once you figure out a pattern for setting up your fortress and finally feel like you're winning, something new comes along that you didn't expect. Once that feeling of "anything could happen" is lost in a game, it's tough to continue.

Factorio is a good example of this. It satisfies the desire for challenge and survival like DF, but the challenge drops off drastically once you have a good setup.

It's tough to match the content of Dwarf Fortress for a game that hasn't been around for 15 years though.


Yes, this lack of depth is really hard to take as a fan. But of course most of them are professional works not life long works of art.

I had hoped that this is what Spore was going to be like. I guess the good news is that I discovered Dwarf Fortress and thus have spent the rest of my gaming life quite happy. Remember kids losing is fun [0]!

0. http://dwarffortresswiki.org/images/4/40/FunComic.png


> It's a fascinating difference in paradigm, and I wish more games would explore the idea of actors in the world being chaotic/free agents which will only somewhat prioritize your wishes.

Sounds like current config management software


Sticking to your StarCraft example, the Total War series is a good example of a "declarative" game. You can initially start ordering your troops to engage costly battles, but as their morale falls, they'll be less committed to an engagement and may rout (fall back) without being told to do so.

The main reason I've pointedly avoided trying Dwarf Fortress is that it feels like a game where you can get noticeably better results through micromanagement than through high-level management. (For instance, "something bad would happen to that dwarf, but I can avoid it by reassigning that dwarf to different tasks".)

Thats an interesting point, I think DF is more macro than you seems to think, Micro is limited at dwarf profession (you can set profesions fixes, or by skill level)

Another great game was Banished, i tried when it just came out, It is still updated by developers?


Kingdom is a beautiful game which is declarative. You run around on your (asthmatic) horse marking where you want buildings and your peasant workforce struggles to build them. It works well, but the rules are almost too simple. "Winning" the game is easy, but if you make one single wrong decision e.g. marking a wall for rebuilding at the wrong time of day your entire fortress will collapse. I believe some issues were fixed in Kingdom: New Lands, but it was a bit of a stain on an otherwise lovely game. At least in DF, usually it's your fault when things get Fun.

For example, in Kingdom, you get more minions buy throwing gold to beggars. There are spawn points on either side of the map. Your fortress is symmetric and you can also pay for tools like bow/arrows or hammers. There's a tool collection point on either side of the centre. One rule you figure out in Kingdom is that minions pick up the first available tool they come across. That means that if you want archers, but you have a hammer waiting pickup (and your new recruits come from the wrong side of the map), you're going to get an engineer.

Watchtowers are automatically restaffed by your soldiers. Watchtowers cannot be destroyed. If you overextend and build towers too far away from your castle, then you'll enter a vicious circle where your new archers will immediately go to man the towers and die when the hoarde arrives. As the towers are always there, and you can't mark them for demolition, it's possible to get into a situation where you can't resupply faster than your men run to their deaths.


Winning go and poker ? Easy, I'll believe in IA when it will be able to handle dwarf fortress !

Here's a question about Dwarf Fortress: I play a lot of roguelikes, which are a genre of similar games, insofar as they're relatively complex, keyboard-and-ASCII oriented games made by geeks for geeks. Like Dwarf Fortress, these tend to have accumulated a lot of developer-hours, and like Dwarf Fortress, these developer-hours tend to get channeled into adding complexity to the game rather than superficial polish, like graphics and interface.

But they vary in terms of their approach this complexity: some seem to always want to add more, seeing more complicatedness as always better, and end up feeling like they contain everything but the kitchen sink - complexity for complexity's sake. (Nethack, I'm looking at you.) Others add it only where it's justified by producing interesting gameplay decisions. (Brogue and Sil are rigorous about stripping out unneeded complexity and getting the maximum amount of subtlety and nuance from a stripped-back set of mechanics. Dungeon Crawl Stone Soup is more complex, but seems aware of the trade-offs around complexity, and is known for removing features as often as it adds them.)

Which of these camps does Dwarf Fortress fall into? There's a lot of complexity, features and mechanics there. Is it all justified, in terms of adding interest to gameplay? Or is just for complexity's sake?


> Is it all justified, in terms of adding interest to gameplay? Or is just for complexity's sake?

I stopped paying attention to DF development around the time Toady One started focusing in depth on history and cultural generation - apparently nowadays, you can have temples, bards, all sorts of stuff that seems like it only impacts the Adventurer Mode, and will have some impact on Dwarf preferences, history, etc.

None of this strikes me as anything I need to know about when playing the Fortress mode of the game, and seems like just something that Toady enjoys working on.

But for all I know, there's someone out there who's going to love all of this stuff, and it's going to make gameplay much better for them.


I think Dwarf Fortress falls squarely on the Nethack side of things. The motto of Dwarf Fortress is 'Losing is !FUN!' (exclamation points on either side of an item indicating it is currently on fire, a common occurrence in DF.) The fun and the !FUN! in Dwarf Fortress is because of complexity, not in spite of it.

> His expenses are low — $860 a month in rent, $750 a month to Zach for his help and a few hundred dollars for utilities and food

Uh. The US... sorry, but if you spend like $1600 before food that is not "low expenses". If you can get below $1300 WITH food then I'd say that's low. Some people have to live with much less than $1000/month altogether.


$750 of that is an employee.

I cannot see why a PhD developing one of the most ambitious and complex video games in history for $20K a year is objectionable to you.

This winter a player has documented seemingly the greatest game ever to be played out on Dwarf Fortess. A lyrically narrated 320 year Saga, culminating in the construction of a glass fortress... in Hell.

Archcrystal - 320 years in a fortress (w/spoilers read 37592 times)

http://www.bay12forums.com/smf/index.php?topic=156319.0


Oh lord well there goes a good hour of my time.

I'm a sucker for a good DF story... :P


Just yesterday I was looking up how to recommend someone for a MacArthur grant specifically for the Adams brothers. Sadly, the nomination committee is anonymous and does not accept recommendations. It's a shame because this seems like the ideal case for their grant.

I've never played Dwarf Fortress (something I hope to rectify in the nearish future). However, friends have told me that it's somewhat similar to Factorio, which is a game that I highly recommend programmers to play.

I finally cracked open Factorio again over this weekend and sunk way too many hours in. It's quite literally Engineering: The Game.

I made the mistake of introducing some friends and team members to the game. Their productivity has gone to shit.

Of course, it's only a corollary that they now spend too much time playing Factorio. Their productivity was liable to go to shit anyway; I simply gave a purpose to it haha


How hard DF is compared to Nethack? I've ascended on Nethack and it is always a very tricky business... this game seems on a whole new level.

You can't win. That hard.

Well for one, there's no victory condition. At best, its do so well that you try to see how well you can survive is the face of stupid decisions/goals

A very touching story about the two brothers. I especially enjoyed the idea that they send drawings and stories to donors. I hope they're still doing well.

I dl'd it and started it up. I'm now wondering what I'm looking at. Not being a gamer, I don't have the visual vocabulary or expectations, so I'm glad there's a wiki; the in-game manual doesn't fully work. EDIT: My reading comprehension is not fully functional, the manual works.

> “Water’s not doing it for me these days,” he said. “I know it’s bad, but the sugar goes right into programming the game. If I don’t drink soda now, I get a headache and can’t do any work.”

I feel bad that he's sacrificing his health for our pleasure.

> He’d enrolled at the University of Washington, ... Tarn moved into a string of “dingy one-bedrooms” with “bad moisture problems” — in one, he discovered a shelf fungus growing behind his couch.

I probably lived in one of those when I was at UW. It was a "World's Fair" apartment building, garden level, the only window facing north and looking up an outside stairwell into the alley. Google maps shows me that much of all that has been scraped and replaced.


Very strange timing -- I spent about an hour last night looking in to DF and other games since I like that type of 'emergent' or 'declarative' (as another comment here put it) gameplay.

Other games I have played of a similar vein: Rimworld (great), Prison Architect (great), Banished (great, needs mods to add more content), Planetbase (EA, good, light on content after 2~ hours).

Others I found were: Stonehearth, Gnomoria, Town, Kingdoms and Castles (not out yet), Dungeon Keeper 1 and 2 (kinda) / War for the Overworld (fan remake of DK, essentially).


Rimworld and Prison Architect are practically training for DF. Now that you know you like the style of gameplay you will probably love DF

Only in DF you can talk about turtle reproduction....

here's a bug related to turtle pond going extinct :

http://www.bay12games.com/dwarves/mantisbt/view.php?id=2780

"Since turtles were my only source of shells, which are ever so important for moods, I am keenly feeling their absence after a few years. I went ahead and modded my world so that hoofs and horns can be used also for shells, (which is actually very cool btw, its awesome to have an artifact decorated with Elk Bird horn.)

However, my ponds were filled with turtles early on, and I accumulated many shells that unfortunately rotted away. It seems like fish populations are regenerating, but turtles are not among those, and since I found that odd I decided to report it."

"Pond turtles lay eggs, which might contribute why they will only breed in off site statistics if that's any relevance.

DF structures used for dfhack only implies that eggs that are hatched go into certain classes, and entity ID is one of them for intelligent species (crow men eggs layed and hatched on site belong to you etc., underground egg laying races have been in constant decline since they were added because of this until recently)

Egg layers have always had a hard time repopulating due to dependency on a object to breed which limited and stagnated them in world gen (made easier by spontaneous population regeneration in world-gen recently). I cannot recollect if its possible to offer pet pond turtles nestboxes to use by pitting or pasturing enough of them in a contained area as a alternative or even if they have additional orientation/marriage barriers to overcome we are not aware of.

All gendered vermin breed (or apparently breed, they have the prequesite animal tags but its uncertain whether they become pregnant whilst in the game world before leaving for the site population tally and being replaced with a new generated creature in their momentary existances, or even if they do become pregnant at all with child/adult born states) hermaphrodital or non-gender typical vermin are usually accounted for by being virtually innumerable to compensate for no breeding on site. Technically if the female pond turtle could get off the map by dissapearing and being replaced whilst pregnant it could spawn additional turtles slowly.

Fish are very visible with ASCII symbols and can be seen in murky ponds and rivers for periods of time if a example is needed, if they are close together they are in the capacity to breed and keep the numbers up. Drop a sizable amount of caught vermin fish into a empty pool and it should sustain the fish in theory as they repopulate with compatible mates in that area.

Similar designs have been used with isolated cave spider rooms with wild vermin which appear to be self sustainable and harvested with burrowed animal trappers & web collectors. "



The DF adventure mod have a very very potencial, Imagine skyrim in 2d, multiply it by 1000 in terms of gameplay/content

Anyone can write "code", but to design, write and maintain code that is large and complex is a specialty. This game has interesting emergent behavior that arises from complex systems that it's built from. It's impressive just from that standpoint.

Legal | privacy