Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[ManaUser]

My fortresses tend to follow the central stairway design, a design I've seen criticized, and which has contributed to FUN on occasion. The thing is, I'm having a hard to thinking of any other way to build a fort... I mean anything that halfway makes sense. About all I can think of is to make greater use of horizontal space, or else simply replace the stair with ramps.. but I'm not sure that's much of an improvement.

[Shonai_Dweller]

Make your fortress however you like. It's your game,  Having other people get annoyed at your design is completely their problem.

Anyhow, a lot of the 'advice' thrown around here isn't about reducing Fun, but optimization of pathing and prolonging the time until fps death by a few days.

Fun avoiding advice: Don't build your staircase all the way down to the caverns. Put in a corridor of traps or two to slow the bad guys down before you're sure your caverns are secure.

[ManaUser]

It's not that I care what other people think of my fort design... as such, but if the central stair has detractors, there must at least be some good alternatives, at like I said I can't figure out what they are. I'm just trying to broaden my horizons a bit as it were.

And delaying FPS death is good too.

[Kogan Onulsodel]

I've recently enjoyed the idea of vertical compartments (which isn't really inconsistent with the central stairway, but it can decentralize things to some extent): I assign a horizontal area (minimum is obviously 3x3, but I often do 5x5) and an appropriate number of z-levels for production of something, with a level for bedrooms and dining rooms for workers, a level for input stockpiles, a level for the workshop, and a level for output stockpiles... and if there's a more involved production process, it chains for multiple workshops. So my textile industry might go, from top to bottom, bedrooms, pig tail (and similar things) stockpile, farmer's workshop, thread stockpile, loom, cloth stockpile, clothier, clothing stockpile. Of course for that particular one I need farms as well, but I'll usually put those in an adjacent 5x5 block which also hosts brewing (that one will go bedrooms, seed stockpile, farms, plant stockpile, still, drink stockpile), and since that's a shorter stack of workshops (there's something satisfying to me about having the stacks of workshops start and end together), and I don't usually need 25 tiles of farm plot just for brewing, I share the farms between the two workflows. I can also throw in dying wherever I want, but I don't always even bother with dye. The biggest downside is that it takes some extra planning to set up, and some extra micromanagement as you end up having lots of very specific stockpiles (stockpile only for brewable plants, stockpile only for things I can turn into thread at a farmer's workshop, etc.), but it does leave your workers walking less when there are jobs to be done, and it is consistent with decentralized designs (though I usually still use it with a central staircase... but I also tend to put all my defenses out away from my central staircase, and if an invasion actually gets all the way to my central staircase, I'm probably already dead, for all practical purposes).

Also, from an FPS point of view, I could have sworn that my game actually started running slower when there were two distinct routes between different parts of my fortress (there were essentially three different areas in that fortress, and there was a connection between each pair... as soon as I made the last of those connections, I'm pretty sure my FPS dropped, though I suppose it could be a coincidence with a migration wave or something). I think it cost a lot as the game was spending more time calculating the pathing costs associated with two very distinct routes. On the other hand, I hear you can get an extra FPS cost when dwarves start bumping into each other, but I tend to make my central staircase three tiles wide (certainly never a 1x1), so I don't think I have to worry about that too much (3x1 staircase means very few traffic jams, and with a lot of my workers staying in their particular area most of the time, there are even less, you only have to worry about dedicated haulers most of the time). If I go to a truly massive fortress (hundreds of dwarves), that probably starts to be less true.

[Tryss]

Personally, I like to make a spiraling 3 tiles wide ramp, around a 1x  stair

Code: [Select]

level n

#<<<###
#,,,#X#
#,,,###
#,,,>
#,,,>
#,,,>
#####

level n-1

    ###
    #X#
    ###>>>#
    <,,,,,#
    <,,,,,#
    <,,,,,#
    #######


level n-2
     ####
      >,,,#
      >,,,#
      >,,,#
    ###,,,#
    #X#,,,#
    ###<<<#

level n-3
#######
#,,,,,<
#,,,,,<
#,,,,,<
#>>>###
    #X#
    ###



[Xinael]

There are a lot of competing draws on your fortress design:

• aesthetics
• pathfinding efficiency
• hauling labour efficiency
• downtime minimisation
• defence

And this is far from an exhaustive list. The joy and terror of this game is that very few things are impossible or off-limits; they're just bad on one or more of these axes. So the process you're going through when choosing a fort design is deciding what balance of these priorities is most important to you. This is ultimately a personal choice and it's why the advice in these threads is usually "do what you like!".

You mentioned that you've seen the central stairway design criticised and have encountered problems with it yourself. So, my question to you is: why do you think a central stairway is bad, and what problems have you encountered with it? The real challenge we're trying to address is mitigating those problems.

This might not be a radical design change, but rather a small evolution to try to address those problems. Then you'll no doubt encounter more problems and evolve the design again. This is a much easier process than pulling something completely different out of thin air.

[Salmeuk]

I had this problem after playing for a long time. It felt like I had tried everything at some point or another, and even with a new idea I had no motivation to finish. Part of what helped me was taking a break from the game. When I came back after a year or so, I had been reading a lot of fantasy and browing images of monasteries and castles and things. I then built the awesomest surface fort ever, and actually completed the walls for once.

So, spend some time rethinking why you enjoy the game in the first place, and seek out inspiration either from real life, the forums, or places like the DF map archive. Try to forget about things like efficiency of design or perfectionism (unless, of course, that's the only reason  you play then it's totally cool) and focus more on roleplaying your design. What would your dwarves do or build, if they didn't know they were part of a computer simulation? A lot of fortress design is constrained by quirks in the game engine, but if you don't mind a little inneficiency you can make some really awesome things.

And lastly, if your current fortress is feeling stale just abandon and start anew. Find that awesome set of cavern lakes, or aboveground waterfalls, or whatever inspires you to build a fortress. There is no reason to settle for an "alright" embark when they are literally limitless.

[Fleeting Frames]

Decentralizing certain work areas, perhaps aided by minecarts, can be more efficient than keeping all the piles, bedrooms and dining near central stairwell with some industries:

The classic example is building magma workshops near magma. Non-magma variants will require digging/chopping, hauling and making fuel for potentially across the map, while magma variants can just have track for items out (or at least be easier to haul). Albeit magma is movable....

When working with large stone quarries, it is often better to use temporary workshops as the product is probably going to be much lighter than the boulders, and it often is possible to haul several items at once as well by some means of storage.

For safe fishing, you might want your fisherdwarves to be shielded either by brook tiles or by grates that an invisible weremammoth cannot path to from above. The second one's micromanagement alone is easily reduced by using separate living complex for the fishers where they can send fishes out from.

Soil floors and walls are rather difficult to create in native game, aboveground plants are biome-limited and growing orchards is not very controllable nor fast. If you want bayberry wine you will likely want to be picking existing bayberries, and perhaps copy that for red sand and fire clay for pottery and glassmaking.

Underground gathering/cutting areas are more easily created, but if you have no near-fort infinite water and easily sealable caverns you might take a page from house wulfenbach, cave in a few drop walls and harvest the preexisting flora and fish.

If you're prefer pointing military at targets instead of waiting for them to fall in traps, avoiding cave adaptation would require frequent patrols or skylight (bit easier near the surface). Making their need areas also on the pathway into fortress helps ensure any visitors will move past them instead of them all sitting deep in earth, praying to be not dumped in magma.

If you like your marksdwarves to patrol on minecarts it is easier to ensure only they can access the push-off tile instead of outfitting everyone with a quiver, crossbow and bolts.

...

Now, most of the above projects and probably some I missed can always have a fortress built right next to them, but when you want two of them at separate locations travel distance might become significant consideration. *pushes minecart full of fish across 4 embark tiles*
Thinking about it, ramps seem like they'd benefit somewhat more horizontal/separate designs more than submerged skyscraper design. Consider the following:
z1:
Stairway
SSSX===
SSSX===
SSSX===

Rampway
SSSW===
SSS^===
SSS. ===

z0
Stairway
<===
<===
<===

Rampway
.
W===
^===
  ===
z2
Stairway
X===
X===
X===

Rampway
  ===
^===
. ===
W

For stairway, from smelter to the center of 3 respective stockpiles there is 5-4-5 long path. For rampway, z-level traversal is replaced with diagonal movement, thus the path is 4½-4-4½ with flat 4 squares traveled as far as pathfinding calculation goes.

{Though, note that when speaking of actual stockpiles, both ramps and stairs are eligible tiles for placing them, meaning small enough (possibly quantum, even) piles will benefit more from extra walkable space around workshops with stairways.}

Otherwise, I suppose for security design ramp+hatch cover might beat straight tunnel and door, depending on other pathways and who has ranged fire advantage.

I've recently enjoyed the idea of vertical compartments

Note that depending on your production amounts it might be more efficient to ditch stockpiles for non-food items like thread and just leave them in the workshop. Clutter for small items can take a while to start (75 items max iirc?), and this way you don't create extra job for the thread, making it inaccessible until hauler who is perhaps coming from several tiles away can move it 1 square.

[Kogan Onulsodel]

Note that depending on your production amounts it might be more efficient to ditch stockpiles for non-food items like thread and just leave them in the workshop. Clutter for small items can take a while to start (75 items max iirc?), and this way you don't create extra job for the thread, making it inaccessible until hauler who is perhaps coming from several tiles away can move it 1 square.

That's a fantastic point. Stockpiles are really important for anything that can rot, but I could do away with some of my output stockpiles. On the other hand, for a few things (e.g., weapons and armor, which are all manufactured way down by the great magma sea, but are used near the surface, where the gobbos show up), I'll use stockpiles simply to get things kind of close to where I want them... for another example, I might make some trap components near the magma sea, and as I'm planning out a meat grinder I think, "I need mechanisms and trap components," start manufacturing a bunch of things before I'm ready to put them in, and my haulers have them available right near the entrance by the time I actually want to put them there. On the other hand, it would probably be best to just not have a furniture stockpile for the first year or so, so I can just say, "I want 7 beds and 7 doors for bedrooms," and I can start placing them as soon as my carpenters and masons make them.

[Carpecanum]

It's amazing how different each fort can be and still be viable.  I use the vertical compartment mostly for nobles with multiple rooms, my farming levels and I love having one stair going from tavern to temple to library. 
For pre planning I base EVERYTHING on 7x7 rooms.  That holds 4 workshops or 4connected rooms makes a large pile.  3x3 bedrooms fit 2 up facing a 3 square wide passage.  This way I know my rooms all line up along z levels and my OCD is kept at bay.  Tavern etc are 4 7x7 connected rooms so far (I haven't seen a need to go bigger).
 The central stairway is wonderful but after you hit solid stone follow the wise mans advice and make a secondary stair.  Always dig 5 or 10 levels deeper with the secondary stair so you can seal it quickly when you hit caverns.  If you know it's safe to dig extend your primary stair. 
 My buddy,however, makes huge one level forts(sometimes building more levels above ground).  He clears out an entire z level and rebuilds all the walls with stone blocks. It looks pretty

[nimbus25]

I've always done the 3x3 vertical staircase design, with (almost) always the following design (and yes I don't play with aquifers):

Spoiler (click to show/hide)

• Above Ground Entrance
• Skip first underground layer to avoid pools
• Military/Hospital
• Graveyard
• Skip any extra soil
• Farms/Animal Breeding/Cage Storage on last soil layer
• Workshops (one wing of stone/wood/cloth production, one wing of food)
• Stockpiles
• Inn/Temple/Library
• Noble Quarter Floor
• Bedrooms as needed
• Branch off main stairwell to 1x1 cavern stairwell
• Magma Forges/Bar Stockpile down by magma sea

And while I've always enjoyed it, I started trying to build a castle this latest fort with the end goal of a completely self-sufficient above ground fort (minus below-ground farms/maybe non-grazing animal breeding), and while it is EXTREMELY slow, I am enjoying it and looking forward to the final product, if I ever reach that point.

[Daris]

I use lots of ramps, almost no stairs, and almost zero stockpiles.  I quantum stockpile inputs like stone and wood, which are more efficient to move in large quantities, but not inputs like sand, which can be generated quickly when needed.  I don't use output stockpiles at all, except for food and booze.  Made stuff stays in the workshop where it was made until it is used, traded, or melted down.  This completely solves the tied-up-container problem for virtually all resources that aren't on the food stockpile list, and it ensures that trade goods are moved to the depot ONCE, and then stay there until traded and are never moved again.  SO MUCH HAULING never happens anymore.

So, for instance, the workflow for clear glass involves 1 quantum stockpile for the wood, and 1 normal stockpile for the sand.  Ash moves from the furnace straight to the ashery, and from there straight to the kiln, and from the kiln straight to the magma glassworks, without touching a stockpile in between.  If these workshops are aligned in a row, very little dwarf movement is necessary, and NO hauling of the intermediate products.

[Urist McShire]

I use lots of ramps, almost no stairs, and almost zero stockpiles.  I quantum stockpile inputs like stone and wood, which are more efficient to move in large quantities, but not inputs like sand, which can be generated quickly when needed.  I don't use output stockpiles at all, except for food and booze.  Made stuff stays in the workshop where it was made until it is used, traded, or melted down.  This completely solves the tied-up-container problem for virtually all resources that aren't on the food stockpile list, and it ensures that trade goods are moved to the depot ONCE, and then stay there until traded and are never moved again.  SO MUCH HAULING never happens anymore.

So, for instance, the workflow for clear glass involves 1 quantum stockpile for the wood, and 1 normal stockpile for the sand.  Ash moves from the furnace straight to the ashery, and from there straight to the kiln, and from the kiln straight to the magma glassworks, without touching a stockpile in between.  If these workshops are aligned in a row, very little dwarf movement is necessary, and NO hauling of the intermediate products.

Dangit. I should just do that instead of quantum stockpiling everything! It would save me so much time on hauling everything. The only thing I may actually stockpile besides food/booze would be ammo near my archery towers/ranges, but that's about it.

[Fleeting Frames]

Well, while QSPs save almost no distance (oh look, a track stop moved them a single tile) and stockpiles generally at most break even, don't forget the efficiency of having your legendaries work on materials and others haul instead of legendaries doing both - Daris mentions wood and stone for a reason after all, heh. Other options might be controlling what your jewellers encrust or hauling in dead bodies from outside - but only when there aren't hostiles there.

Can't really save any distance with the thread/loom example, but it is something to not forget. Another use would be to, say, have a statue stockpile 1 tile from their planned location for visual indicator when you can start building them - if making lot of them this would also save on time as heavy statues clutter workshops more easily.