Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 [2] 3

Author Topic: Long Term Automated Fortress (Planning Stages)  (Read 6019 times)

They Got Leader

  • Bay Watcher
  • Current Fortress:
    • View Profile
    • Steam Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #15 on: March 14, 2017, 01:05:20 pm »

Glass walls still obstruct view. You can achieve the same effect by placing a fortification in front of a window however. View will not be obstructed and the fortification will prevent monsters from destroying the window.

Yes, this would work. Requires having isolated a vampire and keeping them from accidentally stepping on the plate randomly.


Also, atom smashers do not deallocate memory, so that in itself poses a problem.
« Last Edit: March 15, 2017, 01:41:31 pm by They Got Leader »
Logged
Quote from: Urist McDwarfFortress
You do not understand the ways of Toady One. He is not a business, he's just a guy trying to make a fun game. He's invited people to come along and experience the journey with him (and help him test it out as he goes along). At the end of the day, I don't think his main goal is to sell Dwarf Fortress, its just to create the best game possible.

milo christiansen

  • Bay Watcher
  • Something generic here
    • View Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #16 on: March 14, 2017, 02:39:06 pm »

There is a x64 DFHack up, the DFHack thread title doesn't mention it, but it is available on their github releases page.
Logged
Rubble 8 - The most powerful modding suite in existence!
After all, coke is for furnaces, not for snorting.
You're not true dwarven royalty unless you own the complete 'Signature Collection' baby-bone bedroom set from NOKEAS

They Got Leader

  • Bay Watcher
  • Current Fortress:
    • View Profile
    • Steam Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #17 on: March 14, 2017, 02:52:36 pm »

There is a x64 DFHack up, the DFHack thread title doesn't mention it, but it is available on their github releases page.

I just saw it, thanks for the heads up.
Logged
Quote from: Urist McDwarfFortress
You do not understand the ways of Toady One. He is not a business, he's just a guy trying to make a fun game. He's invited people to come along and experience the journey with him (and help him test it out as he goes along). At the end of the day, I don't think his main goal is to sell Dwarf Fortress, its just to create the best game possible.

Squirrelloid

  • Bay Watcher
    • View Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #18 on: March 14, 2017, 04:39:48 pm »

As far as automating a cloth industry goes, llamas appear to be the most efficient, with stacks of 15 wool per shearing.  (~Double that of Sheep or Alpaca). 

You can also run a reasonable meat/bone/soap industry off them, as they give 18 bones and comparable amounts of meat, although I'm not sure how automatable butchery is.  But that's bigger than a cow.  (Something seems off about that, but :shrug:  Wiki gives the bone range as 12-18.  In my experience, as long as you've properly pastured them so they aren't starving, it's always 18.)

Running most of your dwarves as military will let them wear armor instead of clothes, but this won't work for miners (and probably not woodcutters).  Miners will dump all their armor every time they take a mining job, because they won't use the pick they use for mining with their military outfit.  (Or at least they kept doing that to me).
Logged

milo christiansen

  • Bay Watcher
  • Something generic here
    • View Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #19 on: March 14, 2017, 04:44:01 pm »

Miners and woodcutters not being able to work and wear armor at the same time has been an issue for years now. Also if their "military self" has a pick/axe assigned they will have two of them, one for work and one for war.
Logged
Rubble 8 - The most powerful modding suite in existence!
After all, coke is for furnaces, not for snorting.
You're not true dwarven royalty unless you own the complete 'Signature Collection' baby-bone bedroom set from NOKEAS

They Got Leader

  • Bay Watcher
  • Current Fortress:
    • View Profile
    • Steam Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #20 on: March 15, 2017, 01:54:09 pm »

Miners and woodcutters not being able to work and wear armor at the same time has been an issue for years now. Also if their "military self" has a pick/axe assigned they will have two of them, one for work and one for war.

Thank you for the heads up on this information. I assume it does not affect any job that does NOT require a tool (basically everything but the mentioned). This means that I can armor every civilian that I have. Can civilians also carry weapons when off duty?

When responding to this, I realize that I am starting to set up preliminary goals for this fortress: armor and arm as many dwarves simultaneously as possible...


As far as automating a cloth industry goes, llamas appear to be the most efficient, with stacks of 15 wool per shearing.  (~Double that of Sheep or Alpaca). 

But are llamas and other shearables more efficient than designating plots of farms of pig tails/rope reeds? I feel like it is easier to deal with plants than animals.

You can also run a reasonable meat/bone/soap industry off them, as they give 18 bones and comparable amounts of meat, although I'm not sure how automatable butchery is.  But that's bigger than a cow.  (Something seems off about that, but :shrug:  Wiki gives the bone range as 12-18.  In my experience, as long as you've properly pastured them so they aren't starving, it's always 18.)

The problem with running a meat industry is that this requires animal slaughter, which as we saw before, can cause FPS death over the course of several in-game years. Likewise, I can not automate slaughtering. I believe that a fully vegetarian fortress may be required.


A successful FPS-preservation strategy will also probably include healthy amounts of dfhack to clean contaminants, delete items (though I've seen claims that atom smashing fully deallocates the item), and remove records of dead units. (The deceased unit history page by itself can kill a game.)

Does DFHack remove ALL records of deceased? Does this affect ingame stories; i.e. if I kill a bronze colossus and wipe the deceased page, do the dwarves no longer remember fighting the colossus?
Logged
Quote from: Urist McDwarfFortress
You do not understand the ways of Toady One. He is not a business, he's just a guy trying to make a fun game. He's invited people to come along and experience the journey with him (and help him test it out as he goes along). At the end of the day, I don't think his main goal is to sell Dwarf Fortress, its just to create the best game possible.

mikekchar

  • Bay Watcher
    • View Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #21 on: March 15, 2017, 08:48:16 pm »

No.  Even a llama is nowhere near as productive for cloth as a single rope reed tile.  A llama only produces 1.5 wool per month.  Even considering meat and milk production, crops are still going to be more productive.  However, I think you're going to have to slaughter *something* occasionally just to get bones for moods.   I can't really imagine a way to run a completely automated fortress...
Logged

Mostali

  • Bay Watcher
    • View Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #22 on: March 15, 2017, 09:05:03 pm »

It's Miners, Woodcutters, and Hunters that have the uniform switch.  You might notice that if you try to assign a second of these jobs to a single dwarf it will disable the first.

If your squad is assigned to wear the uniform while off-duty, then any weapon that's part of that uniform will also be carried.

Just some behavioral notes on this:
- You can still assign miners and woodcutters to your military, even with a uniform.  If you're going to be going very long periods without designating anything to mine or cut, you shouldn't notice any trouble at all.  It's when they change roles that causes problems, so the moment a miner takes a mining job, it drops its entire uniform on the spot and goes to grab its mining uniform - which is clothes and a pick.  It's literally the dropping of the uniform that causes all of the chaos associated with miners in the military.
- If your squads that are off-duty are assigned to a barracks then they will typically spend all of their free time in training.  Personal preference can override this behavior, but I'm convinced it's default since so many dwarves do it.  This might even be what you want, but it will take away from time socializing, praying and reading.  If you're trying to run generations in a fort like this I don't think it's nearly as likely to happen since they won't be making friends or lovers while training. 
Logged

Fearless Son

  • Bay Watcher
    • View Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #23 on: March 16, 2017, 04:05:33 pm »

No.  Even a llama is nowhere near as productive for cloth as a single rope reed tile.  A llama only produces 1.5 wool per month.  Even considering meat and milk production, crops are still going to be more productive.  However, I think you're going to have to slaughter *something* occasionally just to get bones for moods.   I can't really imagine a way to run a completely automated fortress...
Hunters will automatically hunt and bring in their kills to be butchered, so long as they have some crossbow bolts to fire and wild animals wander into your territory.  It may not be as predictable as ranching for your meat, bones, and leather, but it can be made automatic. 
Logged

bloop_bleep

  • Bay Watcher
    • View Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #24 on: March 16, 2017, 06:25:52 pm »

Glass walls still obstruct view. You can achieve the same effect by placing a fortification in front of a window however. View will not be obstructed and the fortification will prevent monsters from destroying the window.

Yes, this would work. Requires having isolated a vampire and keeping them from accidentally stepping on the plate randomly.

Actually, I think a pastured civilian animal would work. Do pastured animals ignore pasture restrictions when they get scared? My experience says they do.

Also, with another very, very dwarfy contraption, you could create an automated livestock breeding and butchering process. You would capture a bunch of wild animals, and *without* training them, place them in an enclosure so they can breed. When the population rises above a certain number, a killing mechanism that only kills a few animals at a time would activate, and would open an airlock that would allow a butcher to carry the corpses into the Butcher's Shop.
Logged
Quote from: KittyTac
The closest thing Bay12 has to a flamewar is an argument over philosophy that slowly transitioned to an argument about quantum mechanics.
Quote from: thefriendlyhacker
The trick is to only make predictions semi-seriously.  That way, I don't have a 98% failure rate. I have a 98% sarcasm rate.

gchristopher

  • Bay Watcher
    • View Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #25 on: March 16, 2017, 06:27:31 pm »

What about just stockpiling a huge quantity of bones and then count on the artifact count limit to keep you from needing more?

Or stay under 20 dwarves to prevent moods, but that might not be an indefinitely viable breeding population.

I can't imagine a military defense being any component of an automated fortress plan. If nothing else, you'd be required to go pick up their corpses for burial, which will eventually end the fort, through accumulated ghosts or by feeding everyone to a steel titan.

You'd have to either turtle completely, or maybe design elaborate airlock systems. (I've considered using minecarts to jump dwarves through fortifications as a one-way method of bringing migrants in.)





Logged

They Got Leader

  • Bay Watcher
  • Current Fortress:
    • View Profile
    • Steam Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #26 on: March 16, 2017, 09:52:18 pm »

What about just stockpiling a huge quantity of bones and then count on the artifact count limit to keep you from needing more?

This is actually a piece of my plan, I think. I tend to have this as a general rule of thumb in my other fortresses.

I can't imagine a military defense being any component of an automated fortress plan. If nothing else, you'd be required to go pick up their corpses for burial, which will eventually end the fort, through accumulated ghosts or by feeding everyone to a steel titan.

The plan is to actually have no invasions for a while. I will make an elaborate entrance, with as much !!FUN!! involved as possible.


I am going to update the main page with information as to my basic ideas of how I will run/setup the fortress.
Logged
Quote from: Urist McDwarfFortress
You do not understand the ways of Toady One. He is not a business, he's just a guy trying to make a fun game. He's invited people to come along and experience the journey with him (and help him test it out as he goes along). At the end of the day, I don't think his main goal is to sell Dwarf Fortress, its just to create the best game possible.

bloop_bleep

  • Bay Watcher
    • View Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #27 on: March 17, 2017, 10:17:48 pm »

What about just stockpiling a huge quantity of bones and then count on the artifact count limit to keep you from needing more?

Are you an elf?  :P

Anyway, I'm starting a fortress so I can do my first idea, and perhaps I could do my second idea afterward.
Logged
Quote from: KittyTac
The closest thing Bay12 has to a flamewar is an argument over philosophy that slowly transitioned to an argument about quantum mechanics.
Quote from: thefriendlyhacker
The trick is to only make predictions semi-seriously.  That way, I don't have a 98% failure rate. I have a 98% sarcasm rate.

Squirrelloid

  • Bay Watcher
    • View Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #28 on: March 17, 2017, 11:07:55 pm »

Well, if the plan is to stockpile massive amounts of bones, you are going to need cats at least.  (Otherwise vermin will eat your bones).  Admittedly, you can geld the toms now, so catsplosion risk is controllable, but cats only live like 15 years?  So you'll need some method of either controlling cat population or importing cats when needed.
Logged

peridot

  • Bay Watcher
    • View Profile
Re: Long Term Automated Fortress (Planning Stages)
« Reply #29 on: March 18, 2017, 10:55:00 am »

If you're comfortable with DFHack and the workflow plugin, you can achieve quite a lot of automation, for example:

  • Clothes are no problem: tell workflow to keep (say) 10 shirts on hand, and as dwarves throw out their old ones and put new ones on, the supply drops and the clothes makers start making more.
  • Chains of production are mostly no problem - just build all the workshops needed to make the items leading to soap, ask workflow to keep a couple items of each intermediate product on hand, and enough soap on hand at the end, and the whole chain will operate as needed to replace soap as it's used.
  • Tell the charcoal burner to keep 10 charcoal on hand, the smelter to keep 10 iron on hand, and you'll always have iron ready to make things from as long as you have wood and iron ore stockpiled.
  • Keeping manufactured furniture on hand means you can just build when you want (up to a point) and let the dwarved handle restocking.
  • Shearing/milking/cheese making are automatable if you're not concerned about efficiency: just tell workflow to keep milk on hand and every time it runs short the dwarves will try to find an animal to milk.

Workflow's format - "keep N on hand" - also generally makes it possible to size stockpiles appropriately. The tricky bits are the places where workflow isn't smart enough to know what the output of a process is - making powder can be tricky, since workflow can't really tell sand from flour. So I just don't ever make flour.

If you're worried about having to keep mining manually, many things can be made from green glass, which requires no consumables. In particular, you can make green glass "gems" and encrust everything in your fortress with them, just to keep jewelers busy. For the metal ores, you can dig out large areas and just leave the ores lying on the floor; when spaces open in your stockpiles dwarves will come refill them (with wheelbarrows, optionally).

Wood is annoying, particularly if you have to go outside to get it; I think there's an autochop plugin that sends dwarves to cut more wood when it runs short, but I don't really use much wood on a regular basis, so I just have big stockpiles I refill as needed. Underground tree farms take forever to build and get producing but do solve the danger problem nicely; in principle you can wall off a piece of the caverns but I haven't needed to do this. Also every time I let my dwarves into the caverns they all take off to clean up every bone, spiderweb, or lost sock anywhere in there, so mostly I keep them shut off.

You can very efficiently train up metalworkers by telling them to keep N gauntlets (say) on hand, setting up a stockpile for all non-masterwork gauntlets, and telling the dwarves to automatically melt anything in that stockpile. Depending on which item you choose, this can require only a small amount of metal on each make/melt cycle, or even generate metal (through a bug). If you build such a training forge, it can make sense to let only non-legendary dwarves work there. The main annoyance is when a military dwarf comes and equips some non-masterwork item otherwise destined for melting.

Autobutcher is also very handy - just tell it to keep (say) 5 female and 3 male cats, and you'll have a comfortable number of cats to keep your vermin down, and a steady supply of cat meat and bones. Using up the bones can actually become a problem; I decorate random things with bone, but it's hard to keep the numbers in balance so that the decorators don't stop because they're out of bones and the bones don't accumulate. You can also tell workflow to keep a huge stock of bone trade goods on hand (they're small), but eventually you'll need to get rid of some of them to get the job restarted.

Autonestbox is handy, though you could do the same thing by hand: assign every female egg-layer to a 1x1 zone on top of a nestbox. They'll produce a constant stream of eggs until they die of old age.

Seasonal plants are annoying; pig tails only grow half the year, so unless I can arrange to have a sufficiently massive stockpile the dwarves tend to run short of fiber-making materials when they're out of season. If I can get a hold of non-seasonal surface plants I build a greenhouse ("outdoor" garden roofed in green glass) and solve that problem.

Dyes are kind of annoying too; at least one of the jobs runs into the ambiguous powder problem, and dyeing fabric is also not a job workflow can really cope with.

Animal training can be largely automatic, too. Usually if I catch something mildly interesting but dangerous in the caverns, I set my dwarves to train it in its cage. No risk, occasional warning span when it goes wild, but eventually my civilization gets good at taming whatever it is. Numerous critters like crundles I tame and then butcher. Valuable creatures like giant jaguars need a little more attention to set up a proper breeding program, but as long as your animal trainers aren't all too occupied you can just rely on them keeping the animals tame until young can be born and then permanently trained. Egg-layers like jabberers are particularly annoying to breed, since the eggs need to be kept away from hungry dwarves but the parents still need to be kept from going wild. It can be done but it's not easily automated.

Military training can be automated; others have much cleverer setups, but I just make sure every entrance passes through a barracks where a squad is permanently on active/training. The dwarves get a little grumpy about never getting breaks, but soon enough they hit legendary and stop minding. It helps to put a few beds, chairs, and tables, and have stockpiles of food and drink that pull from the main one.

My fortresses usually run in a basically automatic fashion: every time I unpause after the seasonal autosave I go around and check the non-automatic things: have the decorators run out of input and need to be told to start up again? Is it time to cycle the obsidian caster again? Does the forbidden beast silk farm need its target captive replaced? Otherwise the fortress runs basically unattended apart from whatever project I'm working on. My game mostly stops for sieges caravans and moods, which I prefer to handle interactively - though to be honest once the fort is up and running and has a supply of basically everything on hand, moods generally go fine without intervention. Sieges I usually have reasonably strong static defenses, but it always takes some military dwarves to clean up the stragglers, and then everybody pitches in to clean up the mess. (I'm still working on the perfect entry defenses that can swallow a siege and extract the iron without ever exposing a civilian to the sight of a dead body. Obsidian casting is a good place to start.) Caravans are where I get weird items that are annoying to make - wood, seeds, beverages, pets, metals not found on my map and not brought by invaders.
Logged
Pages: 1 [2] 3