Bay 12 Games Forum

Oh wow, is this back? This is some mighty nostalgia! Thanks for doing the DF talk again, everyone!

(goes to listen)

One thing that has been bugging me since forever...



...is that the cat is absolutely monstrous. As big as a dog or a pig! I know you have limited options, having to show everything in a same-sized square, but the cats never stop bothering me.

Other potential scale issues (albeit not as strong ones): the goose and the muskox.

But overall, I'm amazed with the great work and can't wait to play this!

Hey there. I played the 2D version, which was an approachable easy-to-play game. Then I spent quite a lot of time in the 40d days, when things were a bit more complex but also more fun, and I grew to love Dwarf Fortress. However, the last time I played DF properly was in 2012. I still follow the game religiously but I find myself unable to play it. After games like Rimworld or Oxygen Not Included, I just can't bring myself to endure Dwarf Fortress. It has become such a hassle to play. I do love the idea of Dwarf Fortress but I'm mostly just waiting for it to become playable.

In general, I think the game needs automation, streamlining and useful defaults. So much of the development has been adding complexity. Not only adding the ability to micromanage but also the need to do so. What it needs now is to reevaluate each feature and automate it so that it works by default, and micromanagement is just an optional choice for advanced players.

Specific examples of things that are noob unfriendly:

- ALARM! When the fortress is in danger, I want to have a single button I can press and everyone retreats to safety (with no need to set up burrows, alerts or anything). I don't care how exactly it works, there are many options. Imagine, for example, that after clicking ALARM, dwarves would automatically retreat to meeting halls. Even noobs have those, and for them the effect would be that the game just magically works. Burrows and alerts should be completely optional for those who want more granularity.

- MILITARY. This is such a huge deal. I want to click a dwarf, draft it to military, and be done with it. Anything more is just a hassle. This is how it used to work in the 2D days and it was glorious. Basically, things like uniforms, squad rosters, training schedules etc. should use automation and useful defaults in such a way that I should never have to open the dreadful military UI at all. I draft dwarves to be soldiers and it just works - everything is automated. The military UI should be completely optional for people who want to micromanage.

- STRESS has been mentioned many times, and is another example of this. The game should just work by default - stressed dwarves should be able to relieve their stress by themselves with no action on the player's part. This would result in them spending a lot of time off-duty, so the role of the player is to optimize the system (to get more work out of the dwarves), not to make the system work at all.

- JOB ASSIGNMENT, perhaps the worst culprit. It's overcomplicated to the point of being unusable. External tools are considered a must just to be able to play the game. No amount of UI wizardry will fix this - the system is just too complicated. You can't reasonably expect players to manually assign dozens of specific labours (a lot of them obscure and useless) to dozens of dwarves. This needs to be streamlined massively, to the point where players are assigning only the broad categories (All farming, all metalsmithing, all clothes making, all doctoring...), not specific labours. This goes hand in hand with redesigning the categories because right now you have to go deeper in the menus and use labour-lever granularity because just using the top-level categories isn't useful. Some categories are too broad (farming), other tasks that should logically be together are spread amongst different categories (clothes-related stuff). Better yet: Just let the game automate labour assignment completely (DFhack already does this and it's lovely), with no input on the player's part. And give powerplayers the optional ability to overwrite automation.

- DIFFERENT MODES OF SELECTING AND BUILDING. There needs to be just one of each, period. The different modes of building cause endless amount of frustration. Just have everything work the same way as hospitals and be done with it.

Etc. etc. The desired state is to have a game where the basics just work so that the player can then focus on improvement and optimization. Right now, the role of the player in DF is to struggle with the game just to make it work at all, which is just too frustrating. There has been a lot of talk about the UI* but talking about UI may obfuscate the fact that the game mechanisms themselves are set up in an overly complex fashion, and can't be saved by an UI designer. They need to be fixed at the programming/design level. For example - no UI designer can fix the different modes of selecting and building stuff.

I think what's most needed now is just take one year off, not adding any new features and just focusing on streamlining the game and doing quality-of-life improvements. Take each feature and think: „Can this be streamlined or automated? Can I set some useful defaults that will just work?“. Rethink and reorganize the menus. The game is more than a decade old now, it deserves a massive polishing pass. Do that every ten years, that sounds about right, no? The news about the Steam version and the associated plans to make it more fun to play made me excited about DF again. I know that for you, Toady and ThreeToe, this is a huge change in the way you approach the game and that you may not feel completely comfortable in your new role of game polishers. But I think that it's a great decision, and thank you for it! This sounds like the start of a new, more exciting era of Dwarf Fortress.

--

* Ages ago I made some suggestions for interface overhaul. Most of them still seem relevant. There's even a non-graphics version, and a revamp of the military UI.

This green sash looks nice. I just think that when you keep adding badges, pips and crowns for nobles etc., you end up hiding so much of the original sprite that the base clothing isn't really visible any more. At this point you may as well draw a unique noble sprite.

And with this, I think I've said everything I have to say on this topic.

Whatever the end result, I have confidence that Mike and Meph will do something gorgeous. Can't wait! 

Hi everyone, great work, you're making me very excited about the game. Some humble feedback:

In general, I think you have the tendency to make small things too oversized. Specific examples:
- Weapons lying on the ground. (Compare them to the size of wielded weapons.)
- Tiny items (Meph's crafts posted recently.)
- Meph's tree trunks (they are huuuge... the size of a bed.)
- The cats!

It seems like you always try to use the whole tile. It allows you to use more pixels, which allows you to make fancier graphics but I'm afraid it's a bit counterproductive. Let's take Meph's crafts or the weapon items. I'd say you're adding too much visual details to the game - while the crown graphics are fancy indeed, the actual in-game item is very unimportant and doesn't deserve so much focus. It really shouldn't be the size of a dwarf! You need to realise that by making unimportant things big you're stealing visual focus from more important things (creatures) which harms clarity. Items should be small and easy to ignore, almost merge to the background. If a dwarf walks through a stockpile, he/she should be clearly visible even in the corner of your eye. I'm afraid that with these huge items, dwarves would be drown in visual bloat and unnecessarily hard to spot.

Item sprites should only be as large as you need to in order to make them distinctive, and not a single pixel larger. Take the weapon items. You are already able to make weapons distinctive enough when wielded in hand, and there really isn't any reason to make the item sprites any larger than that just for adding more detail – that's just a visual distraction.

Making small things oversized also makes you want to make big things even bigger - see the elephant debate. I understand that if a cat is almost the same size as an elephant, you want to differentiate them a bit. Instead of making the elephant bigger, make the cat (much) smaller.

===

As for the clothes debate:

While theoretically, showing the actual items that the dwarves wear seems cool, you have to realise that the in-game economy / clothing system is very primitive. IIRC, dwarves currently have only 3 different dyes, and everybody wears the same set of clothing. Showing realistic items would only have one effect: everybody would look the same, it would be a visual mess, and you would lose any ability to tell different dwarves apart. You can already test this in Stonesense - all dwarves are interchangeable, you can't see individuals.

Telling dwarves apart is critical! You need that for narrative purposes, to be able to create stories in your head. Anything that makes all dwarves look the same is the exact opposite of what you want to achieve.

Even just using profession colors gives you much more variety that using the 3 dye colors that the game has. But forget about professions! Much more important are the unique profession sprites, like mayors, nobles, priest, guard captains. You really want these to stand out from the crows, not to merge with everyone else.

When, and only when, the game actually makes nobles wear fancier clothing, makes blacksmiths wear blacksmith tools, guard captains wear different-colour uniforms, makes priests wear priestly robes... only at this point it becomes useful to show the actual items worn. Until that time it would be actively harmful because it would just create a visual mess of non-distinctive individuals.

===

It all comes down to this: the purpose of graphics is to add clarity. Not to add fanciness.

===

And please allow me one wish: different tiles for different tree species! This would probably mean different bark and different leave sprites. Trees are so omnipresent that it would be a shame to have them all look the same.

Keep up the good work, I can't wait to play this!

I've asked this at the beginning of the thread but let me ask it again now that the work is done.

Could this "technology" be used to prepare a completely new interface for PC/Linux/Mac and play the game through it?

Will you do that? 

I've just rediscovered this mod, and thanks for doing this. I never liked mods that add stuff because I believe DF is already so complex that we can indefinitely improve it just by expanding the already present mechanisms - Toady doesn't really utilise his own features to 100%.

Anyway, a random thought:

Hey Button,

I know you're probably super busy preparing a release for this version, but I thought I'd ask any way if it would be at all possible to make a fix that makes at least cage traps only buildable with metal components. The things are OP to the point where it would be pointless to use anything else. At least this way cage traps would take more resources than with the near infinite supply of surface wood, and make traps more of a mid-game option.

Also, the idea a Dragon can be contained in a wood cage just boggles my brain meat.

I used to solve the overpowered cage traps by adding TRAPAVOID quite liberally to all sorts of creatures - giants, dragons and other stuff that's simply too big for cages, and to all intelligent and humanoid creatures, even goblins, so cages are no longer the ultimate solution to everything but merely a way to catch normal animals.

Perhaps a suggestion for including into ModestMod?

That's an understatement.