Print

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

[GavJ]

No, in terms of dwarf fortress reactions, they are just ingredients with the tag [PRESERVE_REAGENT] or whatever it is. I.e. the ingredient doesn't get used up in the reaction.

Pets and war dogs are just final end user products, creating happiness or utility and can be treated precisely equally to furniture and swords.

The point here is that you should only need one algorithm that traces recipes, much more elegant and easy to debug and tightly coded instead of 500 exception cases.



In fact, I strongly suspect that this is ALREADY how they are treated in game code (as a preserved ingredient) for shearing and stuff -- since it's something you order at a workshop and they lead the animals back to the workshop to have it done.

[Dirst]

Just to piggyback a bit on what GavJ said, an input/output matrix won't be confused by an input that can be used several different ways (otherwise heaven help us when it comes to dolomite boulders).  Things can get hairy if a good can be produced using different sets of inputs (for example how some alloys can be made from bars or ores), and I'll have to look into that.

Most economic value can be derived from the raws or the pseudo-raws from hard-coded stuff.  The reason we want to give the AI some strong hints about weapon-grade materials is that there is no straightforward way to determine their usefulness from the raws.  Developing a system to evaluate arbitrary modded materials would require a huge investment in !!SCIENCE!! for little incremental benefit.

By the way, in any simple implementation the AI will believe the weapon material matrix over its own experience.  You could tell the Elves that wood will slice through adamantine like butter, and they would show up with wooden weapons even if you sold them steel.

[heydude6]

Ok, if we are going to be working on making prices dynamically change based on the ever changing supply and demand, we should figure out what things Dwarf fortress will do to actually cause supply and demand to be ever changing. After all, Urist isn't more likely going to buy a wooden sword just because his grandma died.

Let us start with demand first
-a war will increase demand for weapons, armour, weapons grade metal and maybe other stuff
-expanding the city limits will create a demand for building materials
-a rise in the percentage of upper class will increase the demand for luxuries
- a change in the population percentage of citizens who have preferences for certain objects( more people with preferences for teacups, higher demand for teacups)

that's all I can think of for now, I believe the majority of the dynamic economy in dwarf fortress will be caused by supply changing(this is where banditry, natural disasters and unexpectantly good harvest go into play)

anyone have more suggestions for what changes demand in dwarf fortress? Or should we move on to supply next?

[GavJ]

Let us start with demand first
-a war will increase demand for weapons, armour, weapons grade metal and maybe other stuff
-expanding the city limits will create a demand for building materials
-a rise in the percentage of upper class will increase the demand for luxuries
- a change in the population percentage of citizens who have preferences for certain objects( more people with preferences for teacups, higher demand for teacups)

- Different species have dramatically different demands, so if one takes over another's site, the market will change.
- Lots of small villages nearby will tend to demand more staples. As cities grow in your area, more diverse and high end goods are relevant.
- Megabeasts dying = less demand for weapons and armor
- There could be climate change. Ice ages historically did sometimes swoop in in a matter of a few years, during the timeline of a fort game. Needs other changes first, like dwarves not ignoring weather until their fat melts, for starters...
- Overfishing/overgrazing/overforestry will encourage more of a natural colonial setup.
- You could code in fads and trends. Maybe whatever the current monarch likes is fashionable, so when they change around, the civ as a whole gets significantly different demands. And year to year clothing fashions, etc. (mostly high end, nobility wear)
- Fires and natural disasters require rebuilding.
- Droughts and famines change demand for foodstuffs.
- Disease changes supply for labor-intensive goods (too many people die, they can't make as many of these competitively) and alchemical products.

[Dirst]

I'm a bit worried about the game moving from a "happiness" metric to one or more "stress/stability" metric(s) because it's less obvious how the latter would map into utility.  The great thing about utility is that it's basically your "score" and a really good rubric for economic decisionmaking is to do what maximizes your "score."

Individual utility and site profit roll up into a site-level welfare "score" very easily (though we need to tack on site-level concerns like security and justice), so the broker's guideline is to make trades that maximize the site's welfare "score."