Bay 12 Games Forum

There are of course many things I could say about the UI, but it sounds like that's considered to be under control for the time being. (Not that I really believe that, but still.) I also remember Toady saying something about fixing aquifers, which is great to hear; and I won't repeat anything about the stress mechanics here because I already posted about that in the other thread.

As far as werebeasts go: These are definitely an issue the way they are right now. They're basically a perfect combination of threatening attributes: They're trapavoid, they're building destroyers, they're invisible until they're right on top of you, and even if they're defeated they can still leave your fort infected (and the only way to analyze your dwarves for infection is to search through the combat logs). I'm not sure the 'invisible enemies' (the way kobolds also are) mechanic is a good thing to have in the first place. But even if that remains, it's not logically clear why werebeasts would be trapavoid when regular goblins aren't (cage traps are probably overpowered as it is, I'm just saying the logic of what gets trapped and what doesn't seems pretty thin), and they just seem to be excessively strong in combat. Also, most players are going to assume that silver weapons will be good against werebeasts, but the actual randomized metal weakness is non-obvious and there's no way to ascertain which metal is effective other than by experiment. Additionally, there should be some easier way of quarantining dwarves (for instance, it should be possible to chain up or cage any dwarf, rather than just convicted criminals), and some easier way of keeping track of the phases of the Moon in order to know when the time of transformation has come (although this basically comes back to being a UI issue).

Werebeasts aside, some other thoughts on things that should be improved for noobs:

• Endless combat is a big problem. Right now, if your civilians encounter a hostile wild creature and don't get a chance to run away, they will start beating on it until one of them dies. If the civilians are unarmed and the creature is highly armored (but has weak attacks), this can take a very long time. This results in ridiculous situations where you have half your fort's workforce crowded around an unconscious alligator punching it repeatedly for months on end, falling unconscious when they exhaust themselves in combat only to wake back up and keep wailing on the alligator, with nobody able to deliver a lethal blow through the alligator's hide with just their fists. This is utterly unrealistic and can bring a fort's economy to a grinding halt for no good reason. Civilians should be more willing to break off combat (especially with unconscious creatures) in order to flee to safety or grab a weapon that can actually affect the hostile creature. (Aside from stress, this is the closest thing I can think of to a 'game-breaking' problem. I think I abandoned at least one fort as a result of this.)
• The mechanics of stairs and ramps are rather non-obvious. I've answered multiple questions on the DF subreddit by noobs wondering how to make stairs work. Some sort of ingame guide to stairs and ramps is warranted. Additionally, it would be convenient to have a 'smart stairs' designation, where you can designate a column of stairs and get only downward stairs on the top (if the top is a clear floor), only upward stairs on the bottom, and up/down stairs on all intermediate Z-levels. This would make it more straightforward for a noob to designate his first mining shaft without getting his dwarves stuck in a pit or wondering why no stairs are being dug on the surface.
• The problems with grazing animals are very non-obvious. The average noob is not going to expect their two starting yaks to starve in the midst of plenty just because they haven't been pastured. It makes no sense and it's an unnecessary barrier to understanding and utilizing the livestock industry. Grazers should graze wherever they please, and should only starve if they are not permitted to find groundcover at all (either because the local biome has none, or because they are pastured in a zone that has none, or because you have so many grazers that they have eaten it all). There should also be a UI indicator for whether a particular animal is a grazer or not.
• The irrigation mechanics are very non-obvious. When people think 'irrigation' they think of constantly bringing water to otherwise fertile soil in order to grow crops there, not spilling water once on bare rock in order to grow mushrooms. There are a variety of ways this could be done better. Farming is actually so easy right now (everywhere other than on bare rock floors) that it would be okay to make irrigation a little more complicated as long as it's intuitive. My suggestion would be: Require all farm plots to either have proximity to standing water or receive occasional watering (presumably with buckets, although automated systems with floodgates or minecarts would also be Fun), but allow farm plots to be built on rock floors, and even artificial floors, by conveying dirt, clay or sand there.
• The mechanics of wheelbarrows are non-obvious. The fact that 3 wheelbarrows means more hauling than 1 wheelbarrow but 0 wheelbarrows means more hauling than 3 wheelbarrows is going to fly over a lot of people's heads. There should at least be an ingame guide to this.
• Depletion of wildlife populations seems too severe. For a noob who finds fishing to be a convenient food source and plans his fort that way, running out of fish and realizing this cannot be reversed is a bit of a shock and kinda feels bad. In my view even pond turtles and cave fish should gradually regenerate from zero given a few ingame months, but more importantly, the populations of fish in lakes, streams and oceans that touch the edge of the map should be effectively inexhaustible. The idea of being able to completely deplete all the squid in the ocean just by fishing along the shore is ridiculous and counterintuitive.
• The fishing industry should be less broken. As it is, fisherdwarves will leave giant piles of fish on the shore to rot, and eschew all other jobs, including fish cleaning, as they seek to make the piles higher. Fishing should at least be a lower-priority job, and it would be good to allow the same dwarf to handle both fishing and fish cleaning depending on which looks more urgent.
• Groundcover blocking sand and clay access is kind of non-obvious and unnecessarily annoying. I get it that sand and clay are kind of overpowered as it is, but having them blocked by a bit of floor fungus doesn't make much sense. If your dwarves can clear the groundcover away by making a dirt road there without requiring any tools, they should be able to automatically clear a tile of groundcover within a sand or clay collecting zone in order to get at the sand/clay.
• Cooking all your plants and then running out of booze ingredients is a common pitfall. There should probably be some extra mechanic implemented to prevent this. For instance, instead of toggling 'cook' vs 'no cook' for each plant, there could be a limit so that the plants are only cooked above a certain threshold (e.g. if the limit for sweet pods is 20 then your dwarves only make dwarven rum as long as there are at least 21 sweet pods available). This way, the threshold could be set to some appropriate default value (like 10) at the start of each embark, and most noobs wouldn't have to worry as closely about what to do with their booze ingredients until they learn more about the game. (Similarly, cooking of booze should be disabled by default.)
• Constructions sequences could be improved. Right now, if you try to build a corner of wall at the edge of a cliff, what tends to happen is your dwarves build the two sides of the corner first, then suspend the actual corner piece because they've blocked their own path to that tile. This seems fairly obvious once you see it happen, but it's kind of annoying and silly. It doesn't seem like much of a stretch for the game to detect this and sequence the construction orders correctly. Additionally, dwarves will build constructions off the sides of bridges even when they have no support, causing instant cave-ins. Again, this is just silly and dwarves should have enough common sense not to do it.
• The fact that constructed floors block constructed walls, while natural floors don't, is very counterintuitive and makes construction of aboveground buildings more annoying than it needs to be.
• Fruit-picking dwarves getting stuck in trees is annoying and there should be some better way to handle it. (I don't do a lot of fruit-picking in my games, but last I heard this was still a problem.) Either dwarves should simply climb or jump down from the trees, or some other dwarf should come along with a fresh stepladder to get the stuck dwarf down. Having them sit up there dying of thirst is kind of silly and unrealistic.
• Dwarves not collecting corpses from outside by default is counterintuitive and the menu for finding this option is non-obvious. Usually players want to collect corpses so they can be entombed or disposed of (and at the very least, kept away from their dwarves' innocent eyes), so either this should happen by default or there should be a more straightforward way of recognizing that it needs to be turned on.
• Trees opening up holes in the ground when logged is a non-obvious way to completely compromise your fortress defense. I would suggest either removing this mechanic entirely, or having trees automatically fall when the tile below them is dug out (and preventing any new trees from growing directly above an excavated tile; saplings there would simply die like they do in 1-Z-level-high rooms).
• The inability to designate a bin for trading if it is currently being hauled by a dwarf doing a job is kind of annoying, and rather mystifying if you don't know what's going on when you see it.
• It should probably be harder to accidentally trade wood to elves. Sure, veteran players all like to piss off the elves and start a glorious war against them, but a player trading for the first time is just going to be frustrated when they try to trade the elves 30 perfectly good stone trinkets in a (wooden) bin and get slapped in the face. In particular, the fact that obsidian short swords qualify as wooden items is utterly non-obvious. One way or another, this mechanic can be made more player-friendly.
• 'Recover wounded' should be a much higher-priority task. Seeing a dwarf lying unconscious and bleeding out while the rest of the fort ignores him in order to make cheese faster is not only unrealistic, it also makes the player feel more helpless with regards to caring for injured citizens. Similarly, doctors should probably attempt to care for sick dwarves even without a dedicated hospital zone. Again, this is realistic (how many medieval doctors had purpose-built hospitals to work in vs attending patients at home?) and makes the player feel like something is at least being done to keep their dwarves from dying.

I'm not saying all of these are game-breaking or even close to it. But the compounding effect of so many counterintuitive mechanics can be more off-putting than any one of them is on its own. (When one or two things are counterintuitive, players can focus their attention on those; but when everything is counterintuitive, a player is more likely to just give up.) Wherever there's an easy fix to make something work in a more intuitive way, that should probably be done. Even if this leaves a few problems not attended to, those will come across as more manageable to players. Of course, the more integral a particular mechanic is to gameplay, the more important it is to get into good shape for new players. For instance, a problem with minecarts will usually be much less important than a problem with booze production. Debugging and mechanic-fixing efforts should prioritize the things that every player is going to have to do in their first few forts.

Regarding tutorials: Personally, I don't like tutorials very much, and I don't think Dwarf Fortress is very suited to them. I've seen them done badly many times, and I often see them used as an excuse for not fixing bad gameplay or a bad UI. Having intuitive game mechanics and an accessible UI does way more for a game than having a tutorial. Aside from obvious UI improvements and ingame guides, the one thing I would recommend in place of a tutorial is some sort of 'noob embark finder' that automatically searches the world map for really good beginner embark locations (warm, forested, non-evil, non-savage, away from goblins and necromancers, have water access, etc). Even a 'noob world generator' (makes small worlds with no werebeasts, vampires, necromancers or evil regions, and plenty of islands away from the goblins) might be warranted.

I posted a response to this on Reddit that some people said was good and that I should post here, so here's a copy+paste of it:

[start copy+paste]

I would agree that dwarven stress is too serious of a problem right now, in a way that feels like it constrains gameplay. Some points that come to mind are the following:

• The effect of rain on dwarven mental health is blown way out of proportion. Dwarves should not be going insane after spending years haunted by the recurring memory of getting rained on that one time when they went out to collect a log and put it in a stockpile. It's not realistic and it's not fun. I get it that dwarves are supposed to prefer living underground, but this is a bad way to incorporate that into the game.
• The psychological effect of seeing corpses is probably also blown out of proportion (especially when each tooth and bone counts as an entire corpse). I get it that in real life people do get pretty disturbed by seeing corpses, particularly those of people they know, and in that sense this is somewhat realistic. However, we have to remember that this is a high fantasy game and combat is a big part of it. There are a lot more opportunities for dwarves to see corpses in the game than there are for people to see corpses in real life. Trying to give dwarves realistic reactions to this completely unrealistic level of violence is just not good balance for a game. (Additionally, dwarves lack any tools for extracting items from underwater. If a dwarf falls into water and drowns right next to a busy area, their corpse lies there just under the surface causing traumatic thoughts to every other dwarf who walks by, and the process of trying to pump out the water just to retrieve that one corpse is much slower and more complicated than is either realistic or good for gameplay.)
• Dwarves respond very negatively to the corpse of any sapient creature, even an enemy or something that doesn't resemble dwarves at all. This isn't very realistic. Dwarves should not be that upset from seeing the corpse of a goblin, troll or plump helmet man. Or at least, seeing dead enemies should give them happy thoughts about victory and security to counter their feelings of disgust.
• A lot of dwarves just have too many needs to take care of. Some dwarves might get unhappy because they didn't socialize enough, or because they didn't get enough intellectual stimulation, or because they didn't get to pray to their favorite deity enough, or because they haven't been able to let off steam in a fight, or because they haven't done anything creative, etc. Again, that may be quite realistic, but the problem here is that players don't want to micromanage every dwarf's work/life balance. Either the seriousness of this mechanic should be turned down, or some sort of additional automation should be implemented to allow dwarves schedules for doing all the various things they need to do in order to have fulfilling lives (the way they already take time off for sleeping, eating, drinking, breaks, and strange moods).
• The amount of damage dwarves cause during tantrums is pretty ridiculous, and bad for gameplay. Rebuilding random structures that get knocked down by tantruming dwarves is annoying at best, and dangerous at worst (when they decide to knock down the one drawbridge protecting you from the goblin hordes outside). In general, it's way too easy for dwarves to kill other dwarves or knock down buildings when they get upset. (When's the last time you saw somebody in real life dismantle a stone drawbridge in a fit of rage?) The escalation of tantrums should be more gradual, with more, and more effective, methods of intervention available along that route.
• There aren't really any ways for dwarves to alleviate their stress, other than complaining to somebody in charge, and that doesn't seem to be very effective. Dwarves should be able to get similar positive effects by talking over their troubles with family members, priests, philosophers, etc. There could even be dedicated dwarven psychiatrists the way there are already dedicated surgeons, that is, unless we're just assuming that priests fill that role in the medieval sort of world the dwarves live in.
• Another way of balancing mental illness is to replace most of the tantrums with suicides or suicide attempts. In real life, there are more suicides than murders; the average person is more likely to be killed by themselves than by someone else (at least in a direct sense). Giving dwarves a tendency to commit suicide rather than punching everything around them would help the mental illness component of the game feel more manageable; the loss of one unhappy dwarf, while unfortunate and sobering for the player, does not interfere much with the proper functioning of a fort the way wanton violence and demolition do. Moreover, if dwarves can fail their suicide attempts (again, highly realistic), logging these attempts in the announcements screen would help the player to identify unhappy dwarves and get them help.
• Cave adaptation, while perhaps not a major contributor to stress the way rain and corpses are, is honestly just not a fun component of the game. I'm not sure why it's there. Nobody gets any enjoyment out of dealing with it, least of all new players who don't know as much about how to arrange fort layouts. The game would probably just be straight better without it. (And with the myth/magic release coming up, there'll be a lot more opportunity for new ailments that are actually interesting to play around.)

To answer ThreeToe's specific question:

I want to know what kind of play style is causing people to quit in frustration.

My conjecture here is: Basically, treating the dwarves as workers rather than people, and not bothering to micromanage them.

Toady and ThreeToe presumably approach the game from a sort of D&D perspective because their background is in that sort of stuff. The D&D paradigm teaches you to care about every party member, to think of them as individuals and treat them accordingly. Also, Toady and ThreeToe have the time to devote to doing this ingame because the game is literally their job. The average newcomer to the game, and the average person in the Steam audience pool they want to appeal to, is not coming at it from this perspective at all. They're coming at it from a standard gaming perspective where individuality and characterization are not emphasized, so they see the dwarves first and foremost as a workforce to carry out their commands. The villagers in Age of Empires don't get PTSD from walking past the corpse of a slain scout cavalry; the SCVs in StarCraft don't get bored because they're always mining minerals instead of vespene gas; and so on. Most players see their dwarves as having the same basic role as an Age of Empires villager or a StarCraft SCV. When you tell them how the dwarves all have personalities and react to seeing things around them, they go 'Oh, cool!', but they don't internalize this as meaning that they should treat the dwarves in a fundamentally different way. Besides which, they simply don't have the time for all that micromanagement. If you have 150 dwarves in your fort, and you spend just two minutes on each dwarf to assign him the jobs most suitable to his personality and fill his bedroom with furniture made of materials he likes, you've just spent five whole hours focusing on nothing but your dwarves' psychological needs. Most people just don't have the patience for that. They want to achieve something more in the span of five hours of gameplay than merely keeping their dwarves from going berzerk out of sheer depression.

(Also, it doesn't help that every new player wants to start their first embark where wood is plentiful, which is to say, where rain is also plentiful.)

[end copy+paste]

I also want to say that I've skimmed through this thread and seen a number of really great replies here already. I hope the devs are reading this stuff carefully. The general sentiment about stress being way too persistent and about the creativity being sapped out of the game by the need to orient everything around keeping dwarves from becoming miserable is very on-point.

Moreover, I would recommend that the devs find someone who hasn't played the game before and watch them play it. Only paying attention to gameplay and testimony from experts often gives a skewed idea of what's going wrong. Carefully watching a newcomer playing a game can be extremely informative. It lets developers cut through their built-up subconscious assumptions about what playing the game looks like. This is known to be a highly effective practice in software development generally.

To sum up concisely the fixes I would be inclined to implement right now, roughly in descending order of priority:

• Have negative thoughts, particularly minor ones, fade much more quickly over time. Getting rained on once for five minutes shouldn't be a source of stress a year later.
• Automate more of dwarves' psychological self-maintenance. The player shouldn't be toggling them between idleness and karoshi every time they turn a job on or off.
• Make tantrums less common and/or less destructive. (Possibly introduce suicide attempts as a less destructive alternative to tantrums, unless of course suicide is considered too touchy a topic even for this game.)
• Implement a better UI for detecting and managing dwarven stress. (The DFHack advanced labors screen is good, but something that shows specific sources of stress would be even better.)
• Ditch the entire cave adaptation mechanic.
• Make corpse disposal a higher-priority job.
• Give dwarves more outlets for their frustration: Talking over their troubles with parents, siblings, friends, priests, philosophers, drinking buddies, or even dedicated psychiatrists, rather than having to cry on the mayor (who may not be available) every time.
• Reduce the effects of seeing dead enemies or dead creatures of other races, and/or give dwarves positive thoughts of victory and security to offset the negative effects of seeing corpses after sieges.
• Give dwarves ways of praying that aren't just in temples (much less deity-specific temples). For instance, let them carry around a figurine of a deity for some religious comfort, or have a personal shrine in their bedroom.
• Give dwarves ways of getting intellectual stimulation that aren't just from books. Having an intellectual conversation with a friend, scholar, visitor, or master of a shared craft should have a similar effect as reading a good book.
• Give dwarves positive thoughts just for eating high-quality food, rather than requiring (often very rare) ingredients in their food. (A list of 'food dislikes' would probably be more realistic than 'food preferences', and easier to balance.)
• Let dwarves who miss their family members outside the fort alleviate their stress by sending/receiving messages with caravans, diplomats, or even trained messenger birds.