Print

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

[bill857]

    Being an economics major, one of the things that really exited me about DF when I started playing was all the complex interactions that could happen between so many different people.  One thing I thought would be cool was if prices for items (and maybe the reward for jobs too) could be determined dynamically based on supply and demand.  The problem is that trying to calculate this in a realistic/semi-rational way in this game seems to me to get very complicated very fast and not very easy to implement either, plus it probably would take huge system resources to constantly be doing the calculations.
    Then I had this idea: there is already a system set up where various items have a certain value in gold (or whatever money they use).  The numbers seem to reflect the value of the thing in terms of how much people like it in general.  What if we kept those values but multiply them by another set of numbers, which is the item's scarcity.  The scarcity can be determined by how many of that object exist in that particular fort.  The scarcity would therefore change over time while the old numbers stay constant.  By factoring in both the old prices already in the game and the scarcity we get the new prices that are used.

So if there is a golden salve barrel that costs 100 in the current version and there are 50 of them in the fort the high quantity would result in a lower actual price used for trading.  I'm not sure what actual formula should be used but think of the relationship this way:

old price / quantity in fort = new price used for transactions

maybe something like:

old price / (1 + .05Q) = new price

It might need to be a bit more complicated (perhaps to make falling prices decelerate) but the important part is that it would be relatively easy to keep track of the variables and calculate it and the same formula could be used for every item.

    So the changes in scarcity would effect the prices used in the fort.  Goods that were being mass produced would go down in price.  And if their was a sudden scarcity of iron thrones the price would go up, making dwarfs unlikely to buy it at that time.  (which could helpful to the player trying to run everything I might add)
    Where it gets really cool though is when different civilizations start trading with each other.  See if Civ #1 has tons of gold and Civ #2 has none, Civ #2 will naturally give it a very high price (using our formula).  But Civ #1 has a much lower price because they're swimming in the stuff.  So Civ #1 will be able to make what they consider to be a huge profit from trading with Civ #2.  This gives civs (whether NPC or player controlled) a good reason to set up trade routes and find resources other civs don't/can't have.  From a historical standpoint this makes sense too; the some first European ships that got to the far East apparently loaded up on spices and sold them back in Europe for ridiculous amounts of cash.  If civs still do negotiations like they do now where they are willing to pay 121% the regular price for miniforges this could still be factored in as well without causing problems.
    Factoring in scarcity not only makes trading more profitable and accurate in terms of the amount civs benefit, but scarcity can be used to a civ's strategic advantage.   If they can dominate the market, the relative scarcity of something like Silver in other civs makes them have a huge advantage.  Also civs might go to war to gain control of precious resources.  This might result in some Mercantilist strategies.  But if we didn't factor in the scarcity being different in different civs, civs would have little reason to invade and do other diplomatic things over resources like silver mines because there just wouldn't be much advantage in having control of a resource besides the direct value of them using it. 

Some potential problems in implementing this:

- First, there lots of types of items so there might be only one Microcline throne but twenty Birch chairs.  The Microcline one would be very expensive even though this did not really reflect how scarce thrones/chairs were.  A solution could be to calculate the scarcity factor of more general types of items like all chairs/thrones not specific types.  A more complicated option would be to partly factor in the large number of chairs but also factor in the number of the more specific type.  But since you already have the old prices factored in as well which would make the Microcline throne more expensive than the chairs anyway that not might not be necessary (K.I.S.S.).

- What if a fort has no lumber at the moment, but they are surrounded by a forest?  Whoever's running the fort shouldn't value that item especially highly when trading with other civs because they can easily get some even if they don't have it right now.  You could try to factor in the uncut trees in the forest as part of the tallies, like make each uncut tree worth half a felled tree (a bird in the hand is worth two in the bush) but frankly I think this is a slippery slope towards a complicated mess that should just be left alone.  Besides, maybe the militant elven empire nearby doesn't let them cut any down anyway.

- Also, up until now I have mentioned the interactions of forts and civs as if the words were interchangeable but a civ could have multiple forts/towns/towers trading.  So let's say one of the forts in a civ starts mining huge amounts of gold.  What if another fort from the same civ trades with them.  In a sense it's a different place but they're still on the same "team" so it seems like maybe they should get less of a markup.  This would also reflect that its easier to move resources around within a smaller geographical area than a larger one.  I suppose we could adjust the amount scarcity is a factor in the price based on the distance between the two trading parties but now my bain is stating to hurt so i never said that.

Please be aware that I know the real economy does not work this way at all.  That's not what I'm trying for.  The goal is to get something rudimentary in place that adds a new level of depth to this amazing game.  Please let me know what you think.

Sorry if this post was too long, just remember this haiku:

Make trading cooler:
Divide existing prices
By fort's quantity

[profit]

I think this should really only be applied to traders.. I know that the fort wealth computations are a lil on the flaky side.. however its the traders where the problems become most glaring.

I think a better formula would be for traders and have it be slightly more complicated.

I am thinking:

Price = Default / ( 1 + (Items Sold - Demand) * Price Factor)

(Note: if items sold is less than demand, the price will actually increase by design)

This would allow a lot more variability in the economy and allow for raw file tweaking of it.

For example. if you wanted the price of Microline Mugs to fall faster and rise sharper you could increase the "Price Factor" variable in a raw.

If you wanted to have a high demand item like dwarven wine have a higher default price unless a lot was sold, you could set the "Demand" in the raws to a higher value.

I think this would add some needed flexibility and would prevent low value but numerous items from collapsing too fast and high value low demand items from having only trivial price reductions if done right.

* demands would likely be procedurally generated in the future by each civilization, based on their production capacities, the wares price, "desperation levels", and stockpile levels.

Umm don't have a fancy haiku... But I remember the thing to memorize the color bands on resistors...

"Bad Beer Rots Our Young Guts But Vodka Goes Down Well"

(Black Brown Red Orange Yellow Green Blue Violet White)

Also the other version "Bad Boys rape our young girls, but violet gives willingly.("for silver and gold" sometimes added)" 

[The Architect]

I didn't read every word. It's just too long. However I wanted to tell you that the prices already change based on scarcity. As everything belongs to the government and every dwarf works for the government, the nobles make decisions about what the prices will be. These decisions are heavily influenced by the available supply and the number of dwarves.

Edit: the wording was a little off.

[Lord Shonus]

Also the other version "Bad Boys rape our young girls, but violet gives willingly.("for silver and gold" sometimes added)" 

When I learned it, it ended with "God She's Nasty" for the tolerance band (Gold, Silver, and None)

[dakenho]

well the only problem with this i for-see would be a reduction of total production why would I make 5 GCS silk socks when 1 *being the only one* is worth just as much?

[The Architect]

Because you want your dwarves to have socks. The value of everything in the fort is still counted normally during economy; changing the price dwarves pay doesn't alter actual wealth.

[Xveers]

One thing that occured to me is a way that might make money AKA coinage a bit more useful and sane (and non-gamebreaking than before) is to make coinage somewhat abstracted instead of having them as items that dwarves can hold and hoard.

Dwarves internally still have a counter that tracks their money/debt, but you have three new global variables, Optimal Money Supply, Actual Money Supply and Money Liquidity. The first is calculated via some mechanic that takes into account population and relative "earning" based on their assigned profession, perhaps. The second is the actual total amount dwarfbucks actually minted and in the economy, and the third is a variable that compares the amount of coins available versus the total money supply.

The ratio between the first and the second gives you an inflation/deflationary adjustment to prices of your fort (you can't just mint more coins!). The third number (liquidity) offers an additional adjustment based on how much currency is actually useful IE if you only mint adamantine coins, then everything is going to be insanely expensive since that's all that someone can pay with (gives reasons to mint lower value coins like copper or lead).

The coins themselves are stored in a new stockpile tagged as a "Vault" (new target for kobold thieves?) with their actual movement of "currency" being handled invisibly without the actual movement of coins...

Just a thought I had.

[NW_Kohaku]

I think this should really only be applied to traders.. I know that the fort wealth computations are a lil on the flaky side.. however its the traders where the problems become most glaring.

I think a better formula would be for traders and have it be slightly more complicated.

I am thinking:

Price = Default / ( 1 + (Items Sold - Demand) * Price Factor)

(Note: if items sold is less than demand, the price will actually increase by design)

This would allow a lot more variability in the economy and allow for raw file tweaking of it.

For example. if you wanted the price of Microline Mugs to fall faster and rise sharper you could increase the "Price Factor" variable in a raw.

If you wanted to have a high demand item like dwarven wine have a higher default price unless a lot was sold, you could set the "Demand" in the raws to a higher value.

I think this would add some needed flexibility and would prevent low value but numerous items from collapsing too fast and high value low demand items from having only trivial price reductions if done right.

* demands would likely be procedurally generated in the future by each civilization, based on their production capacities, the wares price, "desperation levels", and stockpile levels.

What about the difference between advanced and third-world nations?

Generally speaking, what happens when we currently trade, we behave much like an advanced industrial nation trading with a third-world nation - we take in raw materials we don't have enough of, ourselves, and then sell them back the finished products we have produced out of the raw materials we purchased.

This means that, for example, if we buy up wood/charcoal/coal and iron/iron ore, and sell something made of steel, under a system like this, we are likely to wind up making wood and iron more expensive, while devaluing steel. 

While, granted, increasing the global steel supply should make it a less precious metal, the problem comes if you don't really bound this right - what if iron becomes more "valuable" than steel because it's suddenly more rare (assuming for a second we don't get limitless supplies of goblinite).

[Norseman]

While, granted, increasing the global steel supply should make it a less precious metal, the problem comes if you don't really bound this right - what if iron becomes more "valuable" than steel because it's suddenly more rare (assuming for a second we don't get limitless supplies of goblinite).

This is a very good point. However, I think we could get around this problem by determining the rarity of something based on how many other things are equal or better. Previously, we were considering the rarity of iron based on how much iron is in the world. However, since steel is better than iron, and adamantine is better than steel, the rarity of iron should be considered the rarity of adamantine+steel+iron. The rarity of steel would just be steel+adamantine. The rarity of adamantine would include only adamantine since the only thing better than or equal to adamantine is adamantine.

[lucusLoC]

who sets up those "equal or better" relations? how do you calculate the difference between usefulness and simple desirability?

say i have a wood sword, a silver sword, a bronze on, an iron one and a steel one. wood<silver<bronze<iron<steel

now i have a wood statue, a silver one, a bronze one, an iron one and a steel one
iron<wood<steel<bronze<silver (subjective of course, which is another issue entirely)

now i have the same set of materials, but in doors. we cannot even assign a value scale to that because we do not know what the doors are to be used for.

if it is for a vault then: wood<silver<bronze<iron<steel
if it is for a throne room: wood<iron<steel<bronze<silver

i am also sure you could find scenarios where wood is the most valued, or any other combination.

i think the point i am trying to make is "equal or better" is rather arbitrary, and will vary depending on what needs to be done.

lets take a look at a specific example of that.

i have a king and a mayor. the king wants a single chair for his office, and the mayor wants 200 chairs for a meeting room. the mayor decides he needs 200 cheep wood chairs, and the king decides he wants a single expensive gold throne. under the current calculations the run on cheep wood chairs would cause a spike in *all* chairs, because the category of "chair" is depleted. but this should not really be the case, for a few reasons:

1. the demand for 200 chairs is not a global demand, but rather a 1 time shortage caused by a single event. while it is understandable that this might cause prices to slightly rise for a brief amount of time, the overall net effect should be unnoticeable.

2. the wood chairs and the gold throne are really in two separate categories, serving separate functions. the mayor wants chair purely for their function and chose the cheapest product that filled that need, while the king wants his not only for its function, but also for the status it can convey. he chose the most glorious and expensive chair available.

we also need to take into account that the mayor is specifically looking for *cheap* chairs. if the "demand" for wood chairs pushes their price above that of iron somehow, then the mayor will simply stop demanding wood chairs and switch to iron. this would mean the the demand for wood chairs evaporates. this means that their prices would plummet, and the mayor would them demand them again. there needs to be a way to account for the fact that pricing sometimes drives demand on its own, outside of other factors.

on top of that you need to model in the motivation of the suppliers side into equation. i may have just produce 500 socks, but just because no one wants them does not mean that i will just give them away. i may not be willing to let them go for less than y. y may also fluctuate depending on my circumstances.

i guess the whole point is that so much of supply and demand is subjective that is gets really difficult to model in any way that would make sense.

as i have posted elsewhere i think df needs its own, less volatile economic model. please read that post before you comment futher, as it explains my thoughts on the matter in detail (the rest of the thread is good too ;-).

[praguepride]

I like the idea of eliminating coins as an actual, carryable item in the game and instead just making it invisible and having dwarves carry dwarfbux in their pockets/bodycavities

[Lmaoboat]

My economy is the one where the laborers slaughter the nobles and redistribute the wealth equally.

[praguepride]

My economy is the one where the laborers slaughter the nobles and redistribute the wealth equally.

Ahhh, so Communist Revolutionary

[Firehound]

I like it so far, but yeah, a temporary large demand should not cause a insane price hike for until the demand stops.

Perhaps let the dwarves that own the items choose how much they sell the for, randomly choosing to do the job 'evaluate prices' and adjust accordingly, which should stop prices from randomly shooting to 10,000,000db a chair because you want to put 200 wood chairs in a dining room. Nobles, including the king, should change the prices of unownned items fairly frequently then, considering the lack of things for them to do, other then party. This might cause a sudden spike in cost of this or that, or a sudden drop, while private businesses carry a more reasonable price.

Also would be interesting if dwarves started out owning items they created, sans valuable coins which would instead raise the nobles accounts and the metalsmiths maybe get about 50 db's go to the guy and maybe an icrease in a inflation modifier that doesn't decrease to make this not so good of an idea, to pay for placing/trading items. Of course, one could seize it, but it would instead of magically creating money, and nobles had accounts that was usedcause unhappy thoughts in the owner. Would also make the tax collector much more important, since he would pretty much be the only one to be able to reverse poor nobility, other then wild minting getting inflation up by making everyone have more then enough money. Of course, this wouldn't be retroactive, and it would be interesting to see a noble have to buy twelve crossbows so your fortress can shoot with crossbows. Also, Jobs that are simply buildings or designated(such as digging or engraving) should be payed directly for from the nobility.

Coins should be brought on caravans, which should just simply increase the account of the trader, who can get taxed, and probably should be a very wealthy man pretty quickly considering the mayor/expedition leader/count/baron/king have to buy traded things from him once the economy starts.

[NW_Kohaku]

I kind of cut my last point short, but...

When you have something like a resource-rich third-world nation and a first-world nation as trading partners, what you tend to wind up with is a third-world nation hungry for finished goods, but which can only trade with the (inexpensive) raw materials.  It winds up with the third-world nation doing things like selling the first-world the cloth and the dye, and then buying back the clothing.

I suppose the best way of saying this is that this model only looks at supply, but demand is variable, as well.

If we want a "realistic" (HA!) economy, then we need buyers who are motivated by what they can afford, a desire to buy what they need, first (cheapest food they can survive on, low-quality weapons), then climbing up the ladder to toys and instruments or anything masterpiece quality or other luxury goods, depending on the wealth and stability of the trading partner.

If humans are getting trounced by goblins, they should be willing to keep buying more and more steel weapons, regardless of price, just to try to stay alive...

If humans are just plain floating in wealth, they'll ignore low-quality stuff, but buy anything masterpiece or with neat decorations you can haul to the trade depot.

That sort of thing.