Dwarf Fortress’ graphical upgrade provides a new way into a wildly wonky game
arstechnica.com
arstechnica.com
I enjoy ASCII games (Cataclysm: Dark Days Ahead [1] is surprisingly easy to get into in ASCII mode), but the obtuse UI of Dwarf Fortress was a little too much effort to learn. I hope the Adams brothers get some well-earned financial stability with this incredible endeavour. They are some old-school software craftsmen, devoted to the art of excellent game design for the sake of it, money be damned.
And I can't wait for the next Kruggsmash [2] series, now with actual, beautiful sprites to pair with his beautiful drawings.
EDIT: Kruggsmash is currently streaming his first look at this new version! https://youtu.be/sByxI7UX5-g
--
Yeah. I like RimWorld but it constantly throws ridiculously powered raids against my startup village no matter what difficulty settings I use. It's a story generator... but the stories aren't really that varied in my opinion.
Good and cheap defenses are intermittent walls (which have 75% cover).
The problem is that there are two ways to play RimWorld which will lead to very different results there.
A.) You do what "makes sense" to a normal human being dealing with a real situation.
B.) You know the algorithms and minmax.
If do A.) the game can be incredibly frustrating as you watch the colony you spent hours carefully building go down in flames despite the fact that you seemingly did nothing wrong.
In contrast, if you do B.) the game becomes so dead easy that you can't even impress other players anymore by starting alone, naked, as a tribal, on an ice sheet - with one leg!
Have you ever wondered why RimWorld likes to throw that "Imperial Deserter" quest at you right at the beginning of the game?
A.) people will think "This is silly, how could I possibly defeat an Imperial raid right at the start?"
Meanwhile B.) people are like "Sure, no problem. I can build a wodden trap corridor on day 1 which will kill the entire raid before they can even see a single of my pawns."
Also note that RimWorld raids are scaled according to colony wealth and number of pawns in the colony. Thus if you are an A.) player you are probably accidentally signing your own death sentence by e.g building fancy bed rooms for your pawns. Or givign your craftspeople chairs to sit on while they work..
I think the game needs to be rebalanced somehow to make it a less frutstating experience for A.) players. Cheesey tactics like kill boxes which exploit the stupidity of the AI, and seemingly non-sensical decisions like not letting your colonists have chairs, should only be required at the highest difficulty level IMO.
The raids compensate for their dumb AI with sheer size. A smarter AI could be more challenging but use fewer soldier. And maybe the AI could adapt its tactics...like once they learn there is a big wall outside your fortress, then maybe they should park themselves outside and call sappers for support or air droppers to arrive first. Maybe the game could introduce cannons to counter wall building. And to prevent cave dwelling, maybe there could be some crucial supplies that spoil such that it is essential to have some trade, and so enemy armies instead of doing a dumb assault could instead do proper siege warfare to conquer you by depriving you of crucial supplies until you die of starvation, thrist, disease, like was done for most of human warfare history.
Not only that, but the only defense against artillery is to build in a mountain. There isn't otherwise a defense against artillery except to go kill the artillery before it starts pounding you. And good luck doing that when you're outnumbered ten to one...
but ... that's dwarf fortress' motto too! (even if it's not 100% accurate)
You can have mods that warn you of incoming raids. Mods that add point defense laser turrets, suck it enemy artillery! Mods that add magic, so you can send your fire mage out to huck a fireball the enemy siege outpost and cook off their ammo, or call down a firestorm and watch the chaos. Mods that make tamed animals rideable so you can root out their artillerymen with a cavalry charge. Mods that add drop ships and bombing runs so you can counterbattery them before they even get set up, and so on and so forth.
The best defense is a good offense.
There are plenty of ways to solve each challenge the game throws at you.
There's... there's something already implemented to prevent cave dwelling. Some things, actually. Just keep living in the cave, you'll meet them eventually.
Yeah tell that to my 5 pawns with bows or maybe pistols looted from a previous raid when they're attacked by 50 raiders or a mech drop or whatever...
> Meanwhile B.) people are like "Sure, no problem. I can build a wodden trap corridor on day 1 which will kill the entire raid before they can even see a single of my pawns."
Been there, done that. Then my pawns hit their own damn traps and get raided while they're in the hospital minus a leg
> Also note that RimWorld raids are scaled according to colony wealth and number of pawns in the colony. Thus if you are an A.) player you are probably accidentally signing your own death sentence by e.g building fancy bed rooms for your pawns. Or givign your craftspeople chairs to sit on while they work..
Yup, exactly.
> I think the game needs to be rebalanced somehow to make it a less frutstating experience for A.) players.
Yes.
(Rivers with mountains are good to defend)
The StarCraft player in me wants to say: work on your micro!
That being said, I myself am more of an "A.)" player. I got bored with RimWorld after I realized it's more of a puzzle game than a story game. It fooled me longer than others, but eventually, the stateless RNG nature of it becomes apparent.
Chairs for workers?? Everyone who’s familiar with Russia’s history knows that automatic cannons should go first (and they did a rather good job in my case). It is non-sensical to let workers have chairs unless they refuse to work without them. Or until all your neighbors are reduced to the status of Canada and Mexico.
It can! As the player you can fine tune just about every aspect of the storyteller you might want. You can disable events you may not like, like toxic fallout or sunspots. You can tweak raid sizes, disable certain types of raids, change the impact curves that time and/or wealth have on raid sizes, etc.
Yep. The problem is some people want to play like A but the game seems balanced around people playing like B.
Dwarf fortress lets you opt into a lot of this madness with their biomes. You still survive in a friendly biome if you're playing like A.
And no, hand tuning every aspect of the game isn't a good option. By that point you've lost most of your players. There should be an easy default option.
Francis John on Youtube basically did this recently. He talks a good bit about the math of the game, and he minmaxes his minmaxing. You could just spend all your time cheesing the game, or you can cheese it when it doesn't really affect whatever narrative you are trying for in your playthrough. There are materials in your environment that ding you even if you don't gather or process them. There are materials that ding you less once processed. Knowing what's what mean you don't mine a mineral formation until you need the material.
If his camp wasn't constantly attacked by giant furry creatures he would have probably been toast. The RNG kept throwing him meat and cold weather clothing materials. He had to kite them but then real hunters of megafauna kind of do the same thing. He fed his animals meat that the humans would not eat (namely, the flesh of dead raiders).
Anyway, I think if you want to understand Rimworld 'physics' more, that series is probably a decent starting place. But he also makes no bones about how frequently the game is one or two clicks away from war crimes.
I've mostly just gone back to Oxygen Not Included and make do without narrative and variation that I kind of wish were there.
I'm not surprised he's done so well with the minmaxing, he mentioned in one video his day job is a network engineer :)
I played RimWorld that way and I had a thoroughly fun experience learning the tech-tree. It's arguably not the way the game is meant to be played, but for the casual player, it works
I've done that. But it gets repetitive when every fight has three or four or ten red pawns for every one of mine.
Even read a DF book. No help.
I just bought it on Steam.
Probably I will never play it, but it's one of those purchases that are… aspirational. I'd like to be a person who plays DF. I know it's stupid.
But graphics and better onboarding are a real argument for me.
What tipped me finally over the edge towards buying was the background about the creator's health problems. I like that he's doing DF single-mindedly and I can afford the game without thinking twice.
Not stupid. The feeling of mastering a properly hard game is a good reason to try it.
I've used the same "strategy" for most technically challenging games: play in long stretches, trying to make any kind of progress. Have the wiki open and read it any time you are confused. Just keep trying to do _something_.
Eventually it will either click and you'll have a bunch of fun, or you'll tire out. If you tire out, just don't play the game again for a couple days, or whenever you regain your energy.
Once it clicks once, it will instantly start being fun. The interesting part is that the click happens multiple times with challenging games. I've been playing EVE for more than a year now and the combat system _still_ gives me huge revelations. Same with DF, once you learn how to get your fortress rolling you can very quickly learn to build a fully defensible mega castle. Then the next click is realizing that makes the game boring and allowing some danger by fulfilling self placed challenges.
edit: I just bought it and played it for a bit -- the updated graphics/UI are a great improvement! Obviously bit hard to judge but now there's proper icons indicating common actions (mine, cut trees), a tutorial.. give it a try!
I learned DF by setting aside two weeks and playing ~16 hours a day. Was a piece of cake.
It's not the kind of game you can pick up with an hour here or there.
Now that's commitment.
I still dig and irrigate farm chambers if there's a nearby river as though out of reflexive PTSD from that one time where the game required you to moisten soil before you can grow crops. (That hasn't been the case for a looong time, fortunately.)
For what it's worth, if someone told me they beat Morgoth in an interview I'd probably hire them.
That's a classic scenario with the gravity hounds!
I was a human paladin. Took me a while to find it, but here is my place on the Angband ladder:
The steam version has added tutorials in-game for the first time. In the past many people used "guides" or such, but unless you started with the exact same seed/location, you'd often have to figure out things as you went.
https://www.oreilly.com/library/view/getting-started-with/97... was a decently well done guide for it's time.
https://www.youtube.com/playlist?list=PLXX7Rp0iXj0nPwER_CwBD...
He's a fantastic artist, storyteller, and content creator.
Hopefully its a bit more stable now. Would love to lose many more hours to this game.
> Probably I will never play it
Even if you never end up playing it, it's worth buying it to support DF. We need more works of art that are a product of passion and vision and fewer that are just a clumsy, accidental byproduct of profit-seeking by way of mindless, manipulative psychological tactics.
Can't agree more--look forward to the day when something better replaces capitalism.
For me games are an escape from reality and the moment I have to context switch and go look up something in an FAQ or read a guide just to be able to play a damn game I completely lose interest. I hate the idea that I'm having to study or work for the purposes of entertainment. Zero interest in memorizing inconsequential fictional information that only applies to a game. If I'm going to study it's going to be some thing that's applicable to the real world around me, Not some bull crap crafting recipes.
Games should always aspire to be self contained with the appropriate levels of guided discovery within the system itself.
There are people who either don't work full time, don't work at all, or the games are their paid job who are completely fine to put as much effort in as a regular job. And there are people who want to watch these people play the game to a level of ability difficult for most to achieve.
Instead, I watch TV or play games to relax. Dwarf Fortress is one of those games. It may sound counterintuitive but immersing myself in something like that, or a puzzle game, clears my mind and helps me wind down.
This is not necessarily about the richness of the game, it's about how you present information, and not merely accepting that a player will have to have an external wiki open or memorized and dwarf therapist running at all times to play the game.
except I want to enjoy it... its sounds like something i will enjoy... also the new graphics are while simple... very pretty.
No, I said that ONI crosses a threshold for me where I feel like I'm investing too much time into learning a simulation of a real-world trade to be worth it, it feels a bit hollow after a while that I'm devoting so much time to it and not learning the real thing.
I guess that Guitar Hero and Farm Simulator wouldn't be games for me either, I feel some guilt of not learning the actual thing after a while.
Edit: But to the credit of ONI, that's after ten hours of fun, and I go back to it once a year or so. My bases won't extend to the hundreds of hours and explore all the game mechanics, but I've still gotten a lot of enjoyment from the game.
That's what gets me ultimately. I get a good colony going, but end up trying to do something just a little bit ambitious, but there's something that starts breaking and requires so much effort to fix, after a while I just give up.
Adding one new station machine can result in you having to redesign the whole plan of your power distribution network.
Want to make those cool atmo suits? The noob trap is refining metal with the crusher and running out of easily accessible copper and iron.
The more long-term sustainable way is using the electric refinery.
It wasn't the only piece of infrastructure that led to it, but in my last base I had to redesign the whole power grid to have segregated electric networks of high-voltage wire leading to various power regulators that acted as substations delivering power to separate lower-voltage networks.
It was a cute puzzle but I didn't feel like I accomplished something from all that work, damn. I guess it's an analogue as to how it works in the real world, but I don't know enough about how it works in the real world as to tell how simplified and fake the game version is, what did I learn really?
And now you have to deal with all the heat dissipation from all this, time to learn the mock version of another infrastructure-related profession.
That said I got many, many hours of fun mastering the simulation up to that point, so it’s still a strong recommend for me.
A few examples to make that clear:
- to get petroleum, you have to refine crude oil. There is a building that does that, but you only get 50% of the mass out. On the other hand, if you boil it yourself (raise its temperature to 402°C), you get 100% of the mass. Cue furious designing of an efficient boiler. Then you gotta figure how to efficiently burn it to get water out to produce more crude oil.
- if you think that's cute, you can instead continue the boiling to 510°C and get Sour Gas. Then you freeze it to -162°C (not an easy task) and get Methane. Heat it up again and you get Natural Gas that you can burn for power. That comes out to 6X the power of the previous petroleum boiling.
- you have some salt water provided by a geyser (an infinite source at a fixed output). You can use a building to get water out. But that means Dupes have to operate it. What if instead you boil that salt water to get steam, then cool that steam back to get water in a fully automated setup? Turns out it can be more power-efficient too!
Oxygen Not Included is filled with unexpected behaviors, way more than meet the eye (there is a dedicated wiki page, which is lengthy but still far from complete) that enables to make all sort of crazy contraptions.
I prefer to consider Oxygen Not Included as 2 separate games : the survival game you get at the beginning, and the engineering game you get once you overcome the survival issues. Unfortunately, many people drop off before reaching the engineering stage, or don't expect it and drop off because they only wanted the survival part.
The Spaced Out DLC also has smaller asteroids, it changed the game so it now requires more "inter-planetary" automation. I highly recommend trying the game with the DLC, but I'm not sure if Steam allows refunds for DLCs.
Life also doesn't come with an instruction manual nor any handy mouse-overs, but in life I can use my existing years of experience and non-2D view of the world to deduce what's wrong. In a game, I only have access to the information the game chooses to share with me
I tried to get a refund, but Steam was all "nope"
The community is overall very welcoming, and it's an old game by now, with most things figured out.
Some people talk about hating Shenzen I/O and TIS-100 for being "too close to my day job," but for me having to troll through wikis, forums, and (heavenforbid) chat channels in order to debug a computer problem caused by someone not caring about my experience is too close to my day job. I got an adrenaline jolt just reading your sentence because my experience with those things is so incredibly enraging
I just chalk it up to "yet another game that's not for me," similar to my prior contact with Dwarf Fortress. I had actually considered buying the graphical version if it was a "reasonable price" but I'm not paying $25 for masochism
It's blananced to a point it always feels incredibly fair, and there are sliders and fine-tuned adjustments you can do if you wan to make it "just a little more chaotic", or "a slower start", etc.
It's brilliant.
Dwarf Fortress in all its complexity, embraces fuzziness in how things work, and focuses on a bit of a higher-level focus (fortresses reach many more dwarves than a Rimworld colony). However, this complexity makes the game a lot less accessible too. I hope the UI changes help bring ease of use and clarity!
Perhaps it's gotten better in the few intervening years since I last tried it.
Dwarf Fortress ;)
DF is also not as 'hard' as most people think, if you start your fort in an easy location you can quickly build a self sufficient colony inside a hill or underground and make it inaccessible to most threats once you understand the basics of production in the game. And with the UI upgrades managing your dwarfs' moods is a lot easier too.
The only thing that has been preventing me from being an annoying df evangelist for years has been the horrible UI :D
https://www.youtube.com/watch?v=cIkNXqlIIas
And here is the list of mods for Rimworld which creates this world:
https://steamcommunity.com/sharedfiles/filedetails/?id=27809...
Rhadamant shows a ton of different Rimworld scenarios in his Youtube channel, and does marathon Twitch streams of the game as well. Many of them heavily modded to do all sorts of fantasy and sci-fi scenarios. You can get many ideas for your own scenarios from them.
To bring it back slightly on-topic, I strongly suspect that while Dwarf Fortress influenced Rimworld, Rimworld had some influence on the Steam version of DF in terms of the graphical user interface.
It's been a couple of years since I last played using this mod, so it's probably even better now. The only down side at the time was that it would desync sometimes, so you'd have to wait a bit a resync, but it's still so worth it because the experience is just so amazing! I would probably pay the price of the game again just to have an official version of multiplayer.
This is co-op multiplayer, at least how I played it.
Highly recommended to anyone who likes RimWorld and has at least one friend!
It's frustrating because it's the system which really drives the whole game, and it's so simple, and so arbitrary. Without Dwarf Fortress' depth of simulation, it needs some stronger glue to tie the game together.
I repeatedly tried DF and despite having spent hours in similar games, this one eluded me to this day due to it's borderline unusable UI (for me at least) and art style.
I'll be happy to give it another go since the mechanics seem to be supremely refined.
And I say that as one of the 5% who still played and loved it.
It needed various external tools to alleviate the painstaking manual labor that went into managing the colony, like macro generators for floor plans and the memory hacking of Dwarf Therapist to have high-level management of the colony, and all that with wacky menus.
I really love DF but at some point, I know that I can surrender my fortress with walls and only let a small entrance with a labyrinth of traps that the pathfinding will lead enemies to.
In Rimworld such strategy is not very effective since some enemies will dig into your base walls.
That is why meta players always leave one door open. The AI will brainlessly walk in .. right into your kill box.
The open-door killbox is still super strong, but you can't survive with only that defense.
In Dwarf Fortress, there are just too many details to look at them all, and the reward isn’t so great anyway. IMO it is much more enjoyable to just embrace the fact that there will always be too many details and, as you say, it is better to just focus on the higher level. The details aren’t really there to look at all of them or play with all of them, they are there to interact with each other and with the things you do, to result in emergent phenomena that (within the limitations of the simulation) “make sense.”
Or at least when you are picking over the rubble of your base, after one of those phenomena really emerged all over the place, you can piece together where it all went wrong.
Rimworld in my limited play time, seemed like I’d really want to optimize everything fairly well, and like that was actually do-able. So it while the games are pretty similar mechanically, the actual feel of playing is pretty different, and I can’t really compare them.
And that, from a game that looks like it's still using stock, placeholder graphics and interface elements.
I played DF a lot around 09-11. I knew all the commands by heart, experienced so many fun situations that it absolutely holds a special place in my heart.
Then life happened, I didn't play it for a long time and after that trying to get back into it just felt like such a time sink. Played rimworld more during this time.
But now it's back and I am not disappointed. I would have payed 70 euro for this no problem.
Also I'm so happy Toady and Three are getting some money out of all this. At least I hope they are.
You'll be able to see the interior of buildings and design the buildings as you please, save them as blueprints, and be able to plop down duplicates.
You can find progress updates on: https://www.reddit.com/r/Archapolis/
Or a few video logs at https://www.youtube.com/channel/UCUYLstskSLNcSkFg1giTBtQ
I had had a normal sleep schedule before that bout, and was otherwise a functional and responsible adult.
I’ve been waiting for this release for ages, but I do not think it’s going to do my mental health any favours Hah
“Better UI” was the #1 DF feature for as long as I played it; it will be interesting to see how many more players they get in this incarnation.
https://github.com/bottlesdevs/wine/releases/tag/caffe-7.20
Then you can drop that into the default folder in Lutris and choose it as the runner. Lutris also integrates with Steam so it'll see your entire steam library and you can choose a different wine runner and wine options on a per-game basis.
I was personally able to download on Linux (Pop!OS) and run it without enabling proton experimental or installing anything else. Worldbuilding didn't seem noticably longer than classic.
[1] https://www.gamingonlinux.com/2022/11/dwarf-fortress-release...
It's still a lot of fun to play around with for a while. I would love to see what could be done with a larger development team working on it, but I suppose it would be hard to keep the concept focused.
And I get what you mean, it's one thing to add new lighting effects, but they also swapped out textures which changes the feeling a lot more, and not necessarily for the better.
Instead of the silent protagonist they have a Marvel movie-esque protagonist who keeps doing "well THAT just happened" lines after every joke so none of them actually land.
The beauty of DF is the subjective standards you refer. DF would have just been yet-another-world-generated-game.
https://steamcommunity.com/app/975370/discussions/0/37093075...
I don't know, maybe it's my fault for expecting too much from it, but I was legit hoping for a UI and UX experience equivalent or at least close to that of RimWorld or Amazing Cultivation Simulator.
I hope that It'll improve as time passes or that the community will develop something amazing.
If you choose to disregard this warning: At least buy it when you definitely have the free time, and don't mind losing a lot of sleep. ;-)
And the new game that I just begin (it's yet early summer) and I got a siege of zombie elfs. and I don't have any defense... funny times!
You can see the passion on their faces. That excites me. That deserves reward.
In RimWorld, you can install some slightly (read: extremely) overpowered turret mods, plus a moderate amount of save scumming, I can make my colony last pretty much forever (until I am bored hundreds of hours later). Are there techniques enabling similar playstyles with Dwarf Fortress?
- Fill a tunnel with traps.
- Embark in a location with no enemies. But maybe instead with undead.
- Set up a lava chamber above your entrance hall, and a remotely triggered floodgate…
- Close the doors on your enemies, and have archers on walls above them.
- Create a “danger room”, make an OP army.
There’s no shortage of ways to survive, once you dig into a few of the tools.
I think you just have to set your own goals. Maybe you want to get a population of 1000 before death, or maybe you want a fortress to last 100 years.
If you're not very good, then your goals can be very small, and get larger.
To mix some metaphors, the house always wins, but can you make par for the course, or "justify your seed" (to use the tennis term I picked up from Infinite Jest)?
You can pick extremely peaceful biomes to settle in.
You can learn how to use traps and drawbridges (completely overpowered).
You can learn how to make perfectly self-sufficient farms and trading crafts.
Once those are in place I've never lost a fortress to anything but boredom.
The great thing about DF is that after the initial difficulty of figuring out the complex interactions, it is actually an easy game in many respects. If you look at the ratio of the amount of work you have to dedicate for necessities in comparison to the total work your fortress is capable of producing, well this ratio is actually the lowest for any city builder I am aware of.
Even if your fortress reaches 200 dwarves (the usual maximum), you can still feed it with 2-3 farmers, 1-2 cooks and 1-2 brewers once you get them all experienced. You can have one experienced dwarf working on trade goods that will produce enough wealth to get you anything you wish to buy from the caravans. One clothier can make all the clothes your fortress needs.
You do not have to fight invaders or monsters if you really do not want to. There is a way to seal off your fortress and make it impenetrable ... and then you just wait for the invaders to leave. But it is not that difficult to create a well armed, armored and trained military to head off any invading force.
There is a complex stress system, but really it is not that difficult to keep your dwarves happy and free of stress.
So once you have figured out all the possible pitfalls and how to avoid them you end up with this fortress that has massive amounts of spare production capacity.
So you start doing fun stuff: building megaprojects, or trying to make the most beautiful and expensive objects and pampering your dwarves with every possible luxury, or making a powerful military and seeking out and fighting the monsters, etc.
When they say losing is fun, losing happens in a couple of stages. First, when you are newbie you just kind of hit all the possible pitfalls that the game has for you. Ware-wolves, (or ware-tapirs), vampires, goblins, dwarves in strange moods going mad, goblins, underground monsters, necromancers, ghosts, drunken guests in your tavern, etc.
And then you learn about them and how to deal with them. So then you get to the second stage of losing is fun. This is when you have a safe and prosperous fortress and get kind of bored so you embark on a grand and ambitious project, that turns out a little too ambitious and your entire fortress ends up flooded with lava.
Or you get a military so powerful that you decide in hubris that you can face off any challenge and go to a place where really no mortal dwarf has any business going.
But this second stage of fun losing is entirely optional. You can have a happy and prosperous fortress for as long as you want, once you have figured out the game.
During world gen, you can set the number of sites, (which includes goblin camps and the like), the natural savagery, and the number of beasts to very low, and set mineral abundance as high as possible
As a beginner, you want the threat to be as low as possible (serene or calm), the climate to be warm, NO AQUIFER, and make sure that there's iron-bearing ore and coal. In such a situation, the worst you need worry about are some of the late game !!FUN!! sources, which are only really relevant if you delve too greedily and too deep.
This new release has added some additional difficulty settings that makes embarking on a peaceful adventure much easier. In particular, there are explicit difficulty settings that can be changed while in-game, and you can do things like disable different kinds of attacks.
It is possible to turn off all attacks on your fort.
In DF random things will happen (sieges, monsters, etc) but they don't cheat to get to you, with one glaring exception. If you have a path from your stuff (people or objects) to an enemy the enemy immediately knows it, scouting out the most-hidden paths instantly. To be locked down you need to have a raised drawbridge that actually blocks access.
Turn on seasonal saves and replay when you lose, until you think you've lost in a cool and epic enough way that you'd rather remember it that way than restoring. But play out the fails for a while - often it's easier to recover than you'd think.
Which kinda looks like it's facing right ... (depending on the font)
I’m really hoping this graphical update will lower the barrier to entry for me.
The "original" DF of course supports both.
https://www.routledge.com/Procedural-Generation-in-Game-Desi...
DF is too complex in terms of world simulation (dwarfes have actual organs/limbs etc, liquids have different behavior)
RimWorld simulates organs in a more complete way than Dwarf Fortress does, and lets you surgically remove and replace those organs. That’s pretty cool, and a lot fun. Each part has hit points, and each can be lost in combat. A character without body parts suffers from infirmities or death, depending on how critical the part is. A character who is missing a leg can still function, but moves around more slowly. RimWorld tracks similar information about the body parts of animals and robots, which makes combat against them a lot of fun. Herbivores run away from pain, and robots still have organs like cameras and chemical reactors that can be destroyed to incapacitate them.
Dwarf Fortress simulates body parts in a more complete way than RimWorld does. Each body part (limb, torso, head, etc), has hair, skin, fat, muscle, and bone layers each with their own properties. These parts also have sensory and motor nerves as well as blood vessels. DF keeps track of the color of the skin and hair, the depth of the fat and muscle (and those are related to the Dwarf’s stats), and what type of damage each layer has taken (bruises, cuts, tears, etc). Damage to some layers can heal naturally, be surgically treated, get infected, leave a scar, etc. Others, like nerves, often won’t heal. This sometimes leaves characters unable to use body parts that they haven’t completely lost. It keeps track of some organs, such as the brain and the gonads, but not others like stomachs, hearts, lungs, etc. Surgical options are much more limited, as befits the setting. You can clean a wound (with or without soap), set bones, suture cuts and tears, and use casts or traction tables to immobilize broken limbs. A character who is missing a leg cannot walk normally, but the game will try its hardest to allow them some mobility using other parts, or you can give them crutches. Like RimWorld, Dwarf Fortress tracks the overall body plan for all species in the game (thousands upon thousands of them), as well as the unique information about each individual. Dwarf Fortress gains many of the same combat and storytelling benefits as RimWorld from its in–depth simulation.
Dwarf Fortress goes a step beyond RimWorld though, when it comes to Forgotten Beasts and the Undead.
Forgotten Beasts are horrific monsters left over from the chaotic creation of the world. They have _randomized_ body plans. Sometimes they will be a combination of two or more creatures, like a giant feathered iguana with bat wings and a fiery breath. Other times they will be a standard body plan (such as a colossal troll) made from an unusual substance (such as quartz, or fire, or iron). And it knows the material properties of quartz and fire and iron, and it can simulate how your weapons will do against them!
The Undead are a similar kind of fun. These are ordinary creatures (such as your own fallen dwarves) who have been reanimated by arcane forces. The Undead of course do not feel pain, do not need blood or intact nerves to use their body parts, and ignore all damage. And remember how a crippled character will still try to crawl if they are missing a leg? You don’t know fear until your dwarves are being chased by an undead skull that is pushing itself along by its eyebrows, biting anything that comes near. Your old nightmares will fade into irrelevance once one of your dwarves butchers a cow and the bloody skin and hair are reanimated to terrorize the innocent.
What I'm talking about here is best exemplified by looking at this image: https://cdn.arstechnica.net/wp-content/uploads/2022/12/df4-1... . There are two types of snow tile (full snow, and patchy snow) here, surrounded by grass tiles. The full-snow tiles are connected, RPG-Maker "9-cell brush-texture"-wise, as if they're little dug-out ponds full of snow. The patchy snow tiles are all individual, with no ability to blend into one-another, and so end up looking like a bunch of individual objects on a clear grid — like ore deposits in a rock face in Runescape.
Neither of these (square-grid connected runs, or non-tesselatable one-tile blocks of varying intensity) is a good choice for depicting snow on grass. A graphical designer who wanted to portray the scene of "a field patchy with snow" in a tile-based game engine, would not create these particular visual elements as the primitives to depict said scene. These primitives are instead only used here because they are the ones that can have a 1:1 mapping with particular underlying DF TUI tiles.
I feel like it would be quite practical, if you were building something like this from the ground up, to write an algorithm to extract from DF's raw tile data here, an "elevation map" of snowiness (heck, maybe it's already a pass available internally in the engine, before it gets quantized out as tiles!); to do some smoothing over said snowiness-map; and then to have grass/dirt/sand/etc tiles consume said "snowiness map" as a shader pass over their textures.
(I know that all this terrain is deformable. So what? Regenerating a 2D snowiness-map over a slightly-past-the-viewport crop of the tile data [i.e. a very low-res sample grid] would take less time than pressing the key to dig a hole. It could be done every frame, even.)
And you could do all sorts of other things like that — e.g. have the lava in https://cdn.arstechnica.net/wp-content/uploads/2022/12/df1-1... be a flowing mass of liquid rather than a bunch of particular static grid tiles.
I feel like the only reason the developers of these graphical adaptations of TUI games don't do this, is because they are trying — at least partially — to serve the part of the existing player-base who prefers the TUI version of the game; rather than focusing solely on the part of the player-base who doesn't like TUIs but plays the game despite that, and the untapped market of people who aren't playing the game because of how ugly and gridlike it is.
(And if you're wondering, this sort of higher-level adaptation doesn't have to be to the detriment of gameplay. "The grid" just becomes a UX feature, an extra lens and maybe an always-on fade around the player and/or the cursor, rather than something evident in the design of the levels. Compare e.g. Sim City 2000 [evident grid] with SC3000 or SC4 [UX-layer grid] for what I mean here.)
Yes. Your snow example is a good one.
My problem with reworked TUIs is that the graphics emphasize one element of the game and tend to ignore the others. Dwarf Fortress actually supports diagonal paths. With the graphics pack they look like a series of unconnected open rectangles even though a dwarf can walk down them without problems. With a "better" tileset you can see actual diagonal walls. (This screenshot doesn't have a diagonal 1-wide passage, but does have diagonal walls at the edge of rooms. https://i.imgur.com/tryTOUA.png)
The graphics are made to attract people to the game not to best represent the fiddly details of the mechanics. I prefer the text because while it's not a perfect representation either at least it isn't actively wrong.
> consume said "snowiness map" as a shader pass over their textures.
Yeah, to get good looking snow you'd have to do something like this - without the discrete blockiness. In the case of snow this would be easy because DF doesn't have actual game mechanics connected to snow depth or anything it so it would just be a graphics tweak.
> have the lava be a flowing mass of liquid rather than a bunch of particular static grid tiles.
In the case of lava and other liquids though, the simulation itself is tile based. If you made the surface continuous it would make it harder to tell what was going to happen.
> serve the part of the existing player-base who prefers the TUI version of the game; rather than focusing solely on the part of the player-base who doesn't like TUIs but plays the game despite that
I think DF is mostly the second category. Not hardcode roguelike fans but fans of the DF simulation which happens to be tile based.
> and the untapped market of people who aren't playing the game because of how ugly and gridlike it is.
I wonder how big this market is, for DF. If you want a game that looks good why go for graphical DF when you can go straight to Horizon or other graphics-first game. If you aren't interested enough in the mechanic to interact with it, will some 90s-era graphical skins make the difference?
> "The grid" just becomes a UX feature, an extra lens and maybe an always-on fade around the player and/or the cursor, rather than something evident in the design of the levels.
It would be interesting to maybe anchor building items into an easily understood and simulated grid, but where the grid was arbitrary in relation to the map and other grid-based elements. So you could point your fort in any arbitrary direction in relation to the terrain but the walls would still function properly.
DF is a tunnel digging and wall building game where they have to contain fluids or block enemies and it wouldn't be fun debugging tunnels that almost met or had flow-blocking extrusions.
Oh, to be clear, I didn't mean with my suggestions that you'd have a non-quantized underlying geometry. (It is interesting to think about, but it would be pretty impossible to accomplish if your goal is to sort of do an "enhanced emulation wrapper" that just reads out a TUI state from a black-box underlying game binary, and renders it graphically.) What I'm suggesting here would still be "a GUI as an interpretation of an underlying grid-structured scene data structure." It's just that things could have greebles that stick out of or into their tiles, and that the textures for tiles would be dynamically composited depending on the contents of their neighbours, and so forth.
My example of Sim City 4 was meant to be instructive. Look at this screenshot, for example (https://images.gog-statics.com/2bfe9edb18ccfdecc47bf1ddf132d...).
• The underlying physics operate — in almost all passes — on a 2D grid of tile data. There's a height map, but it's very quantized, and only applies (IIRC) to the movement of cars and pollution. But graphically, you've got smoothly transitioning elevation, and even things like houses where the length of the driveway of the house changes in order to meet a sidewalk of a street that is closer or further away depending on where the house is relative to the slope of the street.
• Likewise, look at the boulevards in the image. Physics-wise, each street side of each boulevard is "one tile" wide, with a "one tile" median, making it a "three tile wide" roadway. (Thus the 3x3-tile intersection.) But if you look at what the game is doing when it draws all of this, none of it is snapped to tile boundaries. The roadways themselves are slightly wider than one tile, forcing the median to be slightly narrower than one tile; and the existence of the roadways causes sidewalks and lamp posts to spring into existence within the boundaries of other neighbouring tiles, where those tiles, and not the roadway tiles themselves, then handle the visual edge transitions around the sides and corners of the roadway (where you can think of the tiles that meet the inner corners of the crossing of the intersection, as having "the responsibility" to draw a little bit of the roadway inside them, to fill in the gap left by the curve of the curb. Though more likely what's actually happening is that the JIT-composited tile data for those tiles leaves that area transparent, and the roadway overdraws on a previous pass, which then gets mostly covered over by those tiles.)
• However, despite these render-time tweaks to how things look, the game is still "on" a 2D grid. Your city is fundamentally a grid of tile data; things can only be placed on grid positions, and there is only ever exactly one thing "in" each grid position, simulation-wise. If you have "grid visualization" on, then you get this: https://i.imgur.com/AiSuFKX.jpeg. And even when you have it off, a grid "spotlight" still displays in the immediate area around the cursor — so that you don't start drawing things on the edge/corner of a tile and then get mad that you selected the wrong tile. Even if you draw a diagonal road, that road still exists "on" a particular set of grid tiles.
He chose to skip all the graphics entirely, use ASCII to render the simulation, and focus on the "story-telling simulation" instead.
And though he added support for tile sets and some rendering, that's mainly how it worked since. It's taken about 2+ years for this Steam release to happen, and that's with a lot of support; during those years DF has basically been "stalled" and not much new has happened in the simulation. This has to be a bit disappointing to Tarn, as that's what he enjoys most.
Minecraft already took a huge chunk out of the industry. Which they then tried to solve by encoperating this and selling additions.
Now another game appears, that offers endless replay ability with fascinating stuff. If the onslought would be kept up, all those monetization schemes could collapse over time, by virtue of simply being boring compared to what the "indy" dev scene has to offer.
One starts to wonder, what could happen if a concentrated efforst started churning out ever better world generators and "deep" games, allowing others non "deep" players to join the world just as "action" heroes or adventurers.
Its really not that clear of a distinctions as it used to be.
Do you have an example of this happening short of blatant asset flips?
The biggest issue I have with the indie scene is that so many of them are solid base engines or aesthetics with a general confusion of what to do after that, and end up with something fairly mediocre — which could be fixed with fairly minor changes. But there’s a very limited appreciation for modding, and even less for open source, so they die in irrelevance.
Where are you finding all these open source games?
Also I think you're confused as to what open-source means...
Now dwarf fortress is a more interesting beast to me. It belongs to a niche where the ratio of players to developers is very small. Those games are in perpetual betas. The biggest ones are financed by patreon and by side jobs. The meat of those games is found in the complex systems that the players can interact with. The graphics are often dry and unappealing to the mainstream audience.
I recommend the grid sage games blog as an example of the dedication that those people put in their craft: www.gridsagegames.com/blog/
I don't think the big publishers are interested in this niche. Mostly designers will look at those game to study their innovations and try to replicate that in more polished games.