Bay 12 Games Forum

They get attacked as soon as they enter the map.

There are probably too many undead on the surface of your map, as caravan guards should be able to punch their way through smaller common undead wildlife. You should build weapon traps in a trap hallway or other defense that will mulch the undead for you as they attempt to reach the majority of your citizens. Automating the process is advised since military dwarves should be reserved for fighting trap-avoiding monsters or overwhelmingly large sieges.

The ideal solution would be to force the wagons to spawn at a location of my choice close to my fort's entrance. My attempts to do this have failed. Is there a workable solution?

As stated on the wiki page for the Trade Depot, any location where the depot accessibility indicator shows wagon access, the game can spawn a wagon. What the wiki does not say is that any location which is "not wagon accessible" but also not completely blocked, is still capable of spawning pack animal traders. This results in a scenario where the wagon can spawn only at the available spot but the pack animals spawn at their intended location; walking the entire map surface with slow animals. If the task is to absolutely block all trader spawning anywhere you do not want them to be, there are very few options available:

The most reliable method and the one you should attempt in blocking the map edge, is with placement of the thinnest possible raising bridges on the map edge, raising towards the map edge. Once raised, bridges form a wall that completely blocks pathing, sight and spawning. Just keep in mind that by blocking the map edge, you're now forcing all ground-based spawns to appear on the remaining unblocked tiles; using your military may be required to clear any undead prior to the caravan arriving.

If the goal though is only to block the wagons from spawning, you can very quickly dig a simple pattern of trenches to block them from spawning where you don't want them:
..........
V..V..V..V
 .. .. ..
---------- UNDIGGABLE MAP EDGEPeriod is ground, V is down ramp, space is rampless ground 1z-level below.
Keep in mind that the reason this works in blocking wagons is the rampless 1z-level cliff the trench creates. You can absolutely dig an entire map-spanning continuous trench (remember, that has no ramps in it on the map edge side) that forces all spawns to walk around to the nearest bridge; but it could take a great deal of time depending on the number of miners you have. (and more dwarves on the surface means more risk for attack by undead... or more dwarves lost to an "fog of evil" event) By putting spaces in the design you allow full pathing for non-wagon units and make the job much faster. You can even revisit the trench later and extend it to a full border trench if things get too dangerous.

In evil biomes people usually put Trade Depot in caves. Evil biomes are only on surface?

Evil biomes extend all the way down into the caverns, which includes any kind of undead-raising biome effect. Trade caravans only spawn on the surface, but can also be allowed to exit the map through the caverns; surface entries are required, surface exits are not.

Having played Dwarf Fortress during the 2D era due to reading the Boatmurdered community game on the Something Awful forums, I've gone through the entire scope of learning how to play the game in 2D and then through the changes to 3D. There really wasn't the amount of community help, wiki space, youtube videos and individual knowledge about the game back then as there is now... But I managed, in part due to prior videogame genre experience and not necessarily because of available tutorials. I may have forgot the specific hang-ups I suffered in my first playthroughs, but ultimately I continued to play Dwarf Fortress because it was something I wanted to play and wanted to learn to play... So, I can safely say that: Someone who has no patience for learning will be a lost cause when introduced to Dwarf Fortress.

Reading through Boatmurdered again: New players will absolutely take choices that will result in "bad" outcomes. These outcomes may become "unfun" after a series of defeats to the point where a player may begin to suspect Dwarf Fortress is nothing more than a simulator for destroying hope and crushing souls. If the player is actually going to learn how to play the game, they have to be held back from making and taking the absolutely suicidal choices of figuring out how the game rules work by trial and error. The best tutorials prevent these kinds of suicidal tendencies from happening.

Tutorials should be:

• Advised to be required for best experience but ultimately optional so the advanced player isn't forced through remedial classes
• Able to load near-instantly so the player doesn't spend an eternity generating a new world to play in
• Replayable, in any order so they can redo any section until they feel competent
• Searchable for specific subjects so they will be able to jump immediately to what they want to know
• Summarized as text instructions so they don't have to actually play through the tutorial to be reminded what's going on
• Able to prevent the player from committing suicide in every way possible. except in the case where they need to die in order to demonstrate being undead ("The tutorial where you die.")
• Exposing how the game works at a fundamental level, for EVERY aspect of the game so that no stone is left unturned and no one bugs Toady with bug reports that an NPC "lied" to them about the location of an artifact (when being lied to is actually a story feature)

The last point is the most important: If the player does not know the fundamental workings of a system, how can they be expected to master it?
How does a dwarf die from starvation; better yet, how does hunger work? How does farming work? How does building construction work? How does skill assignment work? How does menu navigation work? -- How do hotkeys work and why should the player be using the keyboard 9 times out of 10 compared to the mouse? And how do inconsistencies in hotkeys change how the user experience (UX) feels when they press a key they think should work but doesn't?
Each of these questions appear seperated from one another, but anyone who's played Dwarf Fortress long enough knows I've just described an actual flow chart of actions that do happen in the regular gameplay loop, even if the player isn't always aware of it. When a player asks a question "Why are my dwarves starving?" they've actually asked several questions at once. It could be that they've physically blocked their dwarves off from their food with a locked door, or their dwarves are crippled and can't reach it, or their food is forbidden due to dumping, or their food stockpile has rotted because the stockpile wasn't actually storing food because, or they are producing food but not enough, or because, because, because...

• Start with hotkeys and basic concepts behind menu navigation and keep the controls consistent so that they're not switching constantly between 'umhk' or '2468' or '-+/*' or 'hjkl'. Tell the player if the mouse works on those screens!
• Identification of symbols or tileset images so that the player will know at least the appearance of a dwarf, a floor tile and a wall tile.
• Comprehension of game space and dimensions; 2D and 3D spaces being able to allow the player to visualize in their mind's eye what is actually above and below their character
• More symbol ID and more meaningful game menu navigation (inventory) being able to distinguish why that dwarf is blue and why that dwarf is yellow and how to put a box inside a bag and then wield the backpack the box is inside of and equip on a sock from inside the bag
• Building player knowledge and skills outwards from basic concepts so that the player will master adventure mode well enough that they can understand the world when they enter fortress mode, and have some mental fortitude when the nobles are shouting at them to not export leather caps and the sheriff is beating people about it (which should all probably get its own set of tutorials)

If I was personally tasked with creating a Dwarf Fortress tutorial, I would honestly avoid stepping into Fortress Mode until the player was fully familiar with the symbol ID, world space, path finding, and other very essential behaviors before... setting a werebeast upon a starving unarmed fortress of 20.

Lastly:
Test test test test test the tutorials WITH OTHER PEOPLE. Long-time players of Dwarf Fortress will never get you the authentic "noob" experience random people on the street or in a library could offer. (just ask for permission and don't be annoying)

Do you have any tips on making them more distinct?

Play around with some different angles on the facing instead of just straight forward.
Addition of colors always helps immensely where monochrome relies entirely on forms.
To help identify more critical icons, the white background on the bubble can be dropped for something else.
Not everything needs a face, sometimes abstract really is best.

20x20? Challenge accepted.

So I'm fairly new to this, but what's the best way to avoid overheating?

The obvious temporary solution is to put all heat generators on one side of the base, and all heat-sensitive objects on the other. Don't have 10 small batteries and a coal generator blasting heat into the air, directly inside your mealwood farm. It's obvious but i've seen it happen before.

Keep a wider margin between your base and the neighboring high temperature biomes. Remember that mining a tile destroys half its mass. 1800 kilograms of sandstone takes much longer to heat up than 2kg of oxygen, and the 900kg of sandstone that's now on the floor only really interacts with the floor's temperature, not so much the air that's around it. Realise that raw metal will conduct heat faster than oxygen can, so having a large copper deposit touching the hot caustic jungle is a bad idea, and it would be preferable to mine it out immediately. The ideal place to put your heat generators is at the top of the starting biome, and the farms beneath the pools of water or near the center of the biome.

Heat rises and cold sinks. Don't dig too far down until you can get insulated tiles between your living and farming spaces, and the rest of the asteroid beneath you. Keep in mind that every single tile will perform heat calculations based on its neighbor, with a bias towards rising or falling based on its surrounding relative temperatures. If there's cold above hot, the materials adjust each others temperatures much faster than if it was hot above cold. Similarly, a series of tiles that look like "Oxygen - Insulated Tile - Oxygen" is only slowing down the heat transfer by 1 insulated tile math calculation; 2 insulated tiles are much better because of this. (a width of 3 tiles of any kind prevent liquid from breaking through due to over-pressure; 2 insulated and 1 regular tile surrouding a 10,000kg water tank won't ever rupture and will be very thermally insulated.)

If you absolutely can not research insulated wall tiles in time (due to finding an active steam geyser point of interest) build a vacuum wall to keep the heat at bay, which is done by removing an orthogonally surrounded tile through exploiting diagonal deconstruction:

edit - The ultimate solution to heating problems, however, is to research automation, liquid storage, and the aquatuner, then setup a cold-water cycling pipe in your base to keep your dupes comfortable.
As a side note however: Never ever use the special thermal-adjustment clothing, as they don't work as you would expect. Sweaters increase insulation which means they take MORE time to change temperature (even standing in liquids!) but they do still change temperature. Cool vests remove insulation on a dupe which means they'll immediately become hypothermic or heat-stroked if they step in a puddle for too long or stand near an active space heater.

I suspect that the problem is not so much running out of oxygen as succumbing to carbon dioxide accumulation.

Displacement of oxygen is a real thing and is a real hazard. It doesn't matter how well you can filter out CO2 in this case, because if there's no oxygen in the atmosphere you WILL suffocate.
Considering many parts of a fortress often rely on U-bend shaped areas (at least, if they want to make sure nothing can crawl up through a grate) and various low-laying areas beneath populated rooms, any gas which is denser than air would accumulate there over the course of the many days a dwarf doesn't pass through to disturb the atmosphere.

If gasses were properly modeled by Dwarf Fortress, most fortresses would slowly fill with water-like CO2 until everyone drowns in it.

A recent favorite of mine has been recreating the designs found in older DF versions:
23a (aka '2D') Since this ancient version doesn't have diagonal movement (orthogonal only!) the generated DF forts are very simple, almost a basic roguelike in layout and style. The "staircases" are really just doors that will rotate your view when you pass through them and the map is effectively "flat" Though the most striking feature however is the smooth walls on the cliff face entrance to the fort, which is exactly one tile more than the view distance a dwarf has (25 tiles from the character's position) that creates a nice effect with the end pieces. (visible because continuous walls are programmed to be visible at the edge of sight distance) I have never actually seen a 2D generated fortress reach the cavern river, likely because the river starts at 48+ tiles from the mountain face... Additionally, knowing that the game maps are partitioned into region tiles of 48x48 regular tiles, this fortress design is basically a nano-fortress fitting within 1 embark tile in modern DF. (Thus why i never saw a generated fort reach the river.) Fun fact: the total embark area of a 2D player fortress map is roughly 10x10 when fully explored (ie: to the adamantine)

40d One of the much more rigid generation styles, fortresses obey strict templates that are all 16x16 tiles, or 1/9th the area (1/3rd the width/length) of an embark tile. The fortress is sectioned into only a couple such templates, which are in the spoilers below:

Basic main hallway layout, straight piece with side hallways... Which they basically always do, unless it's a corner or edge of the fortress map or would intersect a map feature (though map gen "mistakes" do happen and parts can breach features, but very rarely)

The same main hallway design as a dead-end, but also with a ramp on it. Fun fact: I've only ever seen ramps be placed next to the left or the right pillar pairs, never the north or south pairs. (The x's mark the alternate ramp position, which would have a wall between the two pillars most adjacent, so the ramps would work) They also always face towards the center of the template; this is so there's always open space above them and can also be walked around in case of the infamous 180-turn on the next z-level down.

What does a main hallway "corner" look like? Take the dead-end, paste another on top after rotating it 90 degrees, subtract walls that intersect with hallway floor space.

Basic side hallway layout, but with all four rooms added:

"Corner" side hallway, with three rooms:
... And really, that's all the templates there are. No furniture or shops in these fortresses either, despite the rooms obviously being the perfect size for them, or for workshops and surrounding stockpiles. The primary entrance to the fort is really nothing to talk about; it's literally very often as if someone designated a 46x46 up/down staircase pit in the middle of an embark tile (so if it was a nano fortress, it would be THE ENTIRE NANO-SIZED EMBARK) and then placed the ramp-hallway in the middle of it once it was all dug out, with a ring of up/down staircases along the perimeter. Sometimes the entry pit is smaller (1/3rd as wide; a rectangle instead of a square) but not usually; only if some other feature forced it to do so.

The actual fortress generation rules are really very simple, and can be copied very easilly by a human:
- Make the entrance as described above; First template placed is always a dead-end ramp that leads to this entry.
- Pick another space some non-zero distance away in 16x16 chunks, and draw a "line" of hallway templates towards that point, and put a dead-end downward ramp template there if there's still more z-levels to go through; otherwise just dead-end the hallway.
-- Never ever place a downward template and upward template on the same 16x16 chunk, the shortest hallway possible is an up ramp template and a down ramp template.
- Place down the 4-rooms template out to a distance of 2-3 from the starting, ending, or every hallway segment. Turn the outside pieces into corners or 2-rooms if the site edge is reached.
- Try really hard not to put any templates intersecting map features like chasms
- Fail to avoid intersecting the templates and chasms, and cause the fortress to flood with water or get murdered by inhabitants that spawn around the chasm.

It's a very nice and straight-forward method of fortress building, and thanks to the 4-wide main hallway is actually time efficient by giving dwarves sidestep space (even though the 2-wides are often shortcut through rooms) Building supports between the stone pillars, and statues or windows between the pillars and walls (and one of the two doors of a room) and shrinking the rooms to a 4x4 so the walls will have engravings on the proper side (the room side!) are really all that needs to be done to "fix" problems with the design.

Modern Embark over an existing dwarf site and don't worry about it.

The planar/magical landform rewrite code is likely going to lead to a change to how caverns work as well; might as well clean up all the quibbles there if the underlying structures are going to be roasted.

What sort of quibbles and how far does the notion of a "quibble" or underlying structure extend to? Strictly map-gen related or just "in general" or super secret !!fun!! things you want to do in caverns/underground again?

- Geology-wise: What displeases you most about caves/underground at the present? Absence of features? Difficulty in access or travelling through? Lack of variety in flora/fauna? Bugs and errors, or oddities that are just annoyingly inconsistent?

- Equally, even if you are displeased by something currently, how likely would those things be to actually be DIRECTLY hit by the planar/magical landform re-write? I can definitely imagine something being done like each cavern layer being its own little plane of existence...

- BUT before i go any further with that, i should admit that last paragraph really just begins to sound like a suggestion list of features... It's very easy to get carried away with this. So: How do you intend to say "Ok, that's enough, for now."

- Lastly, about the "rewrite" itself ... If the events in the myth generator specifically influence the map generation at a fundamental "underlying structures" level, how intense of a rewrite would that actually be, without dealing with these "quibbles" at all? Not asking for an ETA, just how many times you'd have to press "delete" or "//" on the code. (Like, a % that would just be "gone" by the end of it.)