Food Shortage

[GitiUsir]

隐藏: Bannerman Man said: you allow caravan trade, banditry, and other things to play a role in the health of a town

That's kind of already how the system works honestly.

1. Villagers supply the town with food by selling it to the market. Other sources of food income are trade caravans and passive bonuses.
2. A portion of that food is consumed/removed from the market and added to the food stocks on each daily tick.
3. When the food stocks are full the surplus food consumption is added to prosperity at a 10:1 ratio. A well fed town therefore grows faster.
4. If food is suddenly removed from the market (e.g. by an army), the negative balance is withdrawn from the settlement's food stocks on each daily tick.
5. Once food stocks hit 0 the negative food balance begins having an effect on everything else (garrisons starve, etc.).
6. Villagers and caravans eventually return to resupply the town with food and reverse the food supply crisis.

If a siege happens or a village is raided the food supply gets cut off and the stocks begin to be consumed and don't replenish until the cause of the disruption is removed.

The passive modifiers to food from bound villages and the orchards project are directly applied to the "Expected Change" number that flows into food stocks. This is effectively similar to villages selling a portion of their total goods directly into a town's food stock rather than the market. Armies buying food cannot and do not affect these passives, and they can only be removed by sieging and raiding.These passives make up anywhere from roughly 25% to 40% of the total food income of a town. Maybe more if you rush-build the orchards project while your prosperity is still low.

By not making the passives dominate the food supply you allow caravan trade, banditry, and other things to play a role in the health of a town. For instance, I can break into a siege with food in my inventory and feed a garrison that's in the midst of starving.

With food stock capacity moderately increased in 1.4.0 the issue of armies buying all the food out is really only a problem if a town was already stuggling to feed itself to begin with. If an otherwise healthy town has a full food stock and all food is purchased from its market, the town will be able to recover in a couple days before even half of the stock is wiped out, meaning minimal harm is done.

*NOTE*: It looks like something to do with orchards broke between 1.4.0 and 1.4.1. All orchards were removed from towns that had them before updating the patch, causing most towns to now be in a food crisis and in economic freefall. If you start a new game however they are still available.

First thanks for the explanation because I know it takes time to write it and I'm sorry if my english is weird sometimes.
My objection to this system is that surrounding villages should sell to local market and this would be the only real factor when considering city starvation. Food from other sources should be "luxury" goods, attracted by high prosperity and consumed by wealthy population. Cities should export only those goods not consumed by people or workshops, and they should consume luxuries only if prosperity is high enough. Food consumption is not really increased substantially by prosperity, it's just wealthy people tends to spend more in luxuries. Then prosperity as an abstract value increases with the amount of goods imported and exported. If the flow of caravans stop, prosperity decreases and imported food demand decreases because there is less money. Essential food demand remains constant for the most part.

It could be possible to sustain a population exclusively with imported food but only in heavily centralized empires or in highly merchantilized cities.

I apologize if my comments were too harsh. After reading this, maybe it's not so broken as I thought. I still think those passives should be higher. It gives too much randomness to bandits, caravans or armies. Not to mention players can play an economic war just buying and selling food massively.

I hope I explained myself. It's different to buy from local market than to buy from outer markets. Having the same stock for both worries me.

NOTE: I hope you are right and it's just a bug, but after reading drallim33 comment I'm appaled, I never took the time to watch how AI manages food

 

[Bannerman Man]

The food economy is broken at a fundamental level, it does not just need minor tweaking. Everything is starving everywhere all the time. Armies are walking around constantly starving because they don't buy enough food. They probably couldn't buy enough food even if they wanted because there isn't enough food to buy. If there is food to buy they probably can't afford it.

Just a ridiculous situation, absolute clown show. Started a new unmodded 1.4.1 game with no player interference, AI army starvation is in full force.

What day are you on in that campaign by the way?

Something is definitely wrong with 1.4.1. The widespread starvation and fiefs' prosperity spiralling to 0 has never been this bad before. I'm still optimistic that they can salvage the system though. It's hard to judge the mechanic by this patch alone because they've changed a lot of things all in a short timespan.

Some possible factors of the economy breaking in 1.4.1:

• War is up. Factions are averaging around 3 or so wars at a time with the new voting system or whatever they did. This stifles trade, as caravans have fewer options for destinations now. More war also means more raiding, which negatively impacts hearth numbers in villages, which in turn lowers production rates of goods. Lower production rates means there's less food to go around in general.
• Faction rulers are keeping almost all of the newly conquered fiefs for themselves. This might cause issues with NPC lords' income sources, and might have other effects as well.
• Garrison numbers have apparently been increased (so food consumption is higher and armies take longer to siege). From loading a 1.4.0 game into 1.4.1 I've noticed lords are adding 80-90 more troops to a town's garrison over time as they visit.
• The garrison starvation rate is still artificially low from the temporary changes made in patch 1.3.0. This means the extra garrison numbers don't bleed off when minor food shortages happen and continue to consume food (anything less than -20 doesn't affect garrison).
• Lords are keeping garrisons topped off regardless of the food situation. I don't think this is necessarily new to 1.4.1, but they might be more aggressive about it now.
• Item pricing or workshop mechanics or both were altered in some way. The number of breweries around the map may have possibly increased too. I'm not quite sure what's actually going on here, but something is different.
• The new "Housing" daily default from 1.4.0 is nuts. It's a percentage based modifier to prosperity and causes prosperity to grow faster and faster out of control. The AI has no concept of how to manage this. When it's active it drowns out any negatives that are applied to prosperity. And as we know, artificially high prosperity causes major problems to the economy and eventual starvation. The new Housing probably has to go.
• They replaced the guaranteed +10 food from "Land around Settlements" with the "Orchards" project in 1.4.0, which gives a progressive bonus of +10, +20, or +30. However, sometimes NPC lords don't build orchards at all for some reason, so towns don't get any passive bonus to replace the previously guaranteed +10 from LaS. Jalmarys in your screenshot doesn't seem to have any built (as -19 + 8 = -11).
• The hearth threshold required for the passive food modifier to increase from +8 to +12 for bound villages was raised from 500 hearths to 600. And one of the towns affected by this change? Jalmarys (Dradios starts at 502 hearths). Not sure if the number of hearths a village starts with was altered at all, but it's possible. Also don't know if goods production from villages was changed in any way either.
• NPCs buy food from neutral settlements now. Not sure what kind of impact it has on food shortages if any at all. I also don't know if the amount of food lords are buying has increased at all either. Judging by your examples I'd say no.

The prosperity/food system definitely still needs work, and I've been raising a lot of the same points you've made here for weeks now. The devs had to put a hold on working on it while they fixed the bankrupt lords issue, but now that they've fixed that they've returned to the prosperity system. The fact that everything broke in 1.4.1 is a good sign to me, as it means they are working on it and they're not afraid of making big changes.

I'm not ready to deem the mechanic a failure yet. I'd expect the economy to be rocky for a few more patches at least as they work out some of the kinks.

Let's wait for a hotfix.

 

[Bannerman Man]

First thanks for the explanation because I know it takes time to write it and I'm sorry if my english is weird sometimes.
My objection to this system is that surrounding villages should sell to local market and this would be the only real factor when considering city starvation. Food from other sources should be "luxury" goods, attracted by high prosperity and consumed by wealthy population. Cities should export only those goods not consumed by people or workshops, and they should consume luxuries only if prosperity is high enough. Food consumption is not really increased substantially by prosperity, it's just wealthy people tends to spend more in luxuries. Then prosperity as an abstract value increases with the amount of goods imported and exported. If the flow of caravans stop, prosperity decreases and imported food demand decreases because there is less money. Essential food demand remains constant for the most part.

It could be possible to sustain a population exclusively with imported food but only in heavily centralized empires or in highly merchantilized cities.

I apologize if my comments were too harsh. After reading this, maybe it's not so broken as I thought. I still think those passives should be higher. It gives too much randomness to bandits, caravans or armies. Not to mention players can play an economic war just buying and selling food massively.

I hope I explained myself. It's different to buy from local market than to buy from outer markets. Having the same stock for both worries me.

NOTE: I hope you are right and it's just a bug, but after reading drallim33 comment I'm appaled, I never took the time to watch how AI manages food

Discussion is always welcome! Ive written a lot of words about the prosperity system over the last month and a half so it's getting easier now haha. If they can't figure out how to make this system function smoothly then a system like you suggest might be the better option. I agree though a 50/50 split with passives and market goods might be a good place to start.

 

[kweassa]

A town actually falling to 0 prosperity sort of sounds way too extreme to be treated as something normal, and IMO would probably require some analysis into how the "Great Depression" fell upon Calradia. I've seen in worst cases towns fall under 2k, but so long as peace is restored and some time to bound back, most towns usually keep somewhere between 4k ~ 8k. Those that are in a very good situation start soaring over 9k prosperity, well into 12~15k ranges.

That being said, it could be something about 1.4.1 itself, where people are now reporting all the kingdoms are locked in multiple wars -- which I suspect has something to do with how the declare war/make peace system changed, and probably some parameter governing it got zonked. My own experiences with 1.4.1 match the reports as well, all of the kingdoms are in a war with at least 2 major factions, with a slew of minor factions added in as well.

There might be something with 1.4.1 that makes it very difficult for kingdoms to stop fighting, which engulfs the entire continent into wars and just pouring down economic ruin everywhere.

TEST

 

[drallim33]

@Bannerman Those screens are from about day 600.

• The garrison starvation rate is still artificially low from the temporary changes made in patch 1.3.0. This means the extra garrison numbers don't bleed off when minor food shortages happen and continue to consume food (anything less than -20 doesn't affect garrison).

Yep it was a band-aid fix that didn't actually solve the garrison starving problem (used to be -8 I think?), and now is causing other problems. Point is the underlying system is the problem. Tweaking numbers like this won't fix it.

• Lords are keeping garrisons topped off regardless of the food situation. I don't think this is necessarily new to 1.4.1, but they might be more aggressive about it now.

Yeah they definitely are, although this SHOULD be a good thing. We want castles to be hard to take. As it stands it can easily bankrupt a lord (because castles do not earn enough to support a decent garrison), if it doesn't starve first.

• The new "Housing" daily default from 1.4.0 is nuts. It's a percentage based modifier to prosperity and causes prosperity to grow faster and faster out of control. The AI has no concept of how to manage this. When it's active it drowns out any negatives that are applied to prosperity. And as we know, artificially high prosperity causes major problems to the economy and eventual starvation. The new Housing probably has to go.

It's based on construction now. So if you pay money to get the bonus construction it goes even faster, and also gets the boost from the "workshops" improvement. It would actually be fine if the underlying system had sensible limits, I don't see anything wrong with boosting a town up a bit for a while to recover from a war or something. Again the problem is the food situation is completely ignored causing it to grow out of control. If there were caps in place based on food then all housing would do is let the town grow up to it's equilibrium point faster.

• They replaced the guaranteed +10 food from "Land around Settlements" with the "Orchards" project in 1.4.0, which gives a progressive bonus of +10, +20, or +30. However, sometimes NPC lords don't build orchards at all for some reason, so towns don't get any passive bonus to replace the previously guaranteed +10 from LaS. Jalmarys in your screenshot doesn't seem to have any built (as -19 + 8 = -11).

Good point about losing the "Land around Settlements" bonus, I didn't notice that

• The hearth threshold required for the passive food modifier to increase from +8 to +12 for bound villages was raised from 500 hearths to 600. And one of the towns affected by this change? Jalmarys (Dradios starts at 502 hearths). Not sure if the number of hearths a village starts with was altered at all, but it's possible. Also don't know if goods production from villages was changed in any way either.

In my game that town got raided multiple times and is down to 150 hearths. It doesn't matter though, it still provides more grain than the town needs, since it is missing everything else. Caravans still come to Jalmarys, but the don't buy or sell anything. They just arrive, wait for a moment, and then leave. If there's any grain there they buy it. So even if the village was supplying 800 food at a time it would all get bought by caravans since it's so cheap.

There are really two incompatible things here. They really have to pick one or the other.

The idea of artificially increasing grain prices in a town with a brewery, or increasing prices based on prosperity makes some abstract sense. Because demand for grain would be higher than if there was no brewery, or low prosperity. This is the sort of thing you might have in a board game or grand strategy where all the economy is abstracted.

But they have gone and mixed that with an actual economy simulation, where individual units of goods are produced, bought, sold and priced.

So you have two layers of economy mechanics, one that wants to do free market supply and demand simulation, and another that is arbitrarily setting prices based on board game style abstractions. And you end up with a complete mess where nothing makes any sense.

They need to pick one, either go full simulation and remove the prosperity and workshop-based price modifiers, so that prices are entirely supply and demand based. If stocks are high, prices are low, and vice versa, end of story. You lose direct control over things doing this since the system will auto adjust to even things out.

OR

Go fully abstract so that the system is more predictable. In this case you have to cheat things and do unrealistic things (like resetting inventories/monies, or giving certain things infinite supply) to keep the system afloat. So you lose the organic nature of the simulation but you get predictability and stability.

A town actually falling to 0 prosperity sort of sounds way too extreme to be treated as something normal, and IMO would probably require some analysis into how the "Great Depression" fell upon Calradia. I've seen in worst cases towns fall under 2k, but so long as peace is restored and some time to bound back, most towns usually keep somewhere between 4k ~ 8k. Those that are in a very good situation start soaring over 9k prosperity, well into 12~15k ranges.

Yeah the towns had a lot of war, Southern Empire and Aserai were fighting over Danustica for for basically the entire game, until Khuzait stepped in and brought peace and order to the galaxy. It's been stable since then but is only going downhill.

There is a certain threshold when prosperity drops below a certain level, caravans won't sell anything anymore, because the prices are too low. At that point the town is relying entirely on it's own villages for food, but because most towns only get one or two food types from the villages they don't have what they need and the spiral begins.

Just look what Danustica is offering for food. If you were a caravan, there's no way you would sell here. You would buy everything they had instead, leaving them with even less food.

This means the town is doomed, it will never get the food it needs.

Situation is similar in Argoron, only slightly less bad. There is enough food thanks to two fish villages, but the town is not growing, it is stuck and will probably remain like this for a very long time if not forever. Caravans do not sell here. It should have a huge food bonus to prosperity because of the full granary and extremely cheap fish but it does not, it only gets +0.2 prosperity bonus. So it won't go to zero but it won't get any bigger either.

 

[Badcritter]

brb, going to Argoron to buy 376 fish so I can make bank while the townsfolk starve.

artificially increasing grain prices in a town with a brewery

Is this done via an arbitrary modifier IN ADDITION to the literal demand element that takes place? ie. workshops have defined inputs which they buy from their local market (these can be edited in a text file), and I assumed that was the extent of their influence on that commodity's price: consumption reduces stock and the price rises as a consequence.

If the mere presence of the workshop has an additional layer of impact on the price of its inputs, that would indeed be a bad thing for any viable economic simulation.

 

[Bannerman Man]

Yeah, I think the extra wars are the primary factor in the screwed up economy at the moment. Constant sieges and raiding cause the prosperity of highly contested settlements to rapidly fall. Then as you pointed out, eventually a point of no return is reached and prosperity falls to 0 with no real way to bounce back. This also happened back in (I think) 1.2.1 when the revolving door sieges on settlements were decimating prosperity levels. I remember seeing Sibir hit 0 in one campaign back then too.

I think a safeguard is needed for low prosperity towns to prevent them from being unable to climb back out of the hole due to the feedback loop that drives prices down. Maybe something as simple as a flat boost to prosperity if it's <1000 or so would work , like: "Rebuilding from the Ashes: +50."

The economy only really breaks down at the extremes of prosperity, but when a town either grows to extremely high prosperity or falls to zero it has an effect on the entire world's economy because the price extremes break caravan behavior. Town's like that just can't be allowed to exist.

Yep it was a band-aid fix that didn't actually solve the garrison starving problem (used to be -8 I think?), and now is causing other problems. Point is the underlying system is the problem. Tweaking numbers like this won't fix it.

It was never meant to solve the underlying problem though. It was mostly meant to appease players complaining about their garrisons starving while they were working on siege mechanics and the economic system. No, it hasn't totally prevented garrison loss, but it definitely has had an effect on how rapidly garrisons starve. I believe mexxico stated that it'll be reversed in the long run, once everything else is figured out.

Yeah they definitely are, although this SHOULD be a good thing. We want castles to be hard to take. As it stands it can easily bankrupt a lord (because castles do not earn enough to support a decent garrison), if it doesn't starve first.

I agree. Towns will need strong garrisons so warfare and sieges actually involve starving out the defenders rather than just an immediate battle after the barricades go up.

Again, the garrison system mostly only breaks down at the extremes, when the negative food modifiers of prosperity and garrison plus the positives from passive and item modifiers sum up to between -20 and 0, meaning food can't rise and garrisons can't starve.

It's based on construction now.

Oh good catch, I hadn't noticed that. It mostly ends up being a distinction without a difference for NPCs though, because a) they don't appear to use the boost from the money reserves system, and b) most of the NPCs have lousy engineering skills which contribute nothing to the construction stat.

The 'Base' number of construction is just 1% of prosperity, so without the flat boost from Reserves and Engineering the construction rate is purely based on prosperity. The 'Workshops' project only ends up multiplying the base rate that's in turn determined by prosperity alone.

That means at 0 prosperity the construction rate for NPC towns is also effectively 0 and they can't build anything or use any daily defaults. If they haven't built 'Orchards' by that point they never will. Even at low prosperity the construction rate is so low that an individual project takes ages to complete. Now might be the right time to give NPCs a boost to engineering.

For the player though the reserves boost ends up being very powerful because the money is not consumed by daily defaults, meaning potentially just 1 denar is enough to boost the daily defaults percentage by +50 construction points forever (not that money is an issue right now anyway).

I still think Housing in it's current form has to go though, because I don't really like the idea of there being a hard cap to prosperity, and the runaway effect it has on prosperity is too damaging to the overall economy.

There are really two incompatible things here. They really have to pick one or the other.

I don't really know what's going on with the price model right now, but I have a feeling different departments pushed out different changes at the same time without fully coordinating and the cumulative effects broke the economy somewhat. It's hard to totally account for all of the effects a change will have before it goes live, and it's not like they can't continue tweaking things to bring things back into order. These things are expected to happen on the beta branch (not to patronize anyone).

That said, I still think the current system of linking prices, prosperity, and food together can work with a little tweaking.

My Suggestions:
Maybe they should consider factoring in a settlement's food stock levels into the pricing model somehow. Having high or full food stocks and high food surplus will put downward pressure on the price of food items, while having low or no food stocks and a food deficit will drive prices up to entice caravans to visit them, and not buy out all the food when they do. Basically, the lower the food stocks go, the higher the prices on food items go.

Or, another way of doing it could be to increase prices on food items in the market and decrease prosperity progressively for each day that the food stocks are not full. A high food deficit or empty food stocks could make the system progress faster to draw more caravans in at a faster pace. The price change from quantity alone does not seems to be enough to create the urgency for caravans to visit a starving town.

Those are just suggestions. I'm sure there are flaws with those methods as well. And I still think the special treatment for grain is hindering the prosperity system too.

Edit: Here's Jalmarys at 430 days in my game too. Same thing. It's definitely from the numerous wars causing villages to be continuously raided. Towns off the beaten path seem to be mostly faring well prosperity wise. Reminds me of 1.2.1.

 

[drallim33]

brb, going to Argoron to buy 376 fish so I can make bank while the townsfolk starve.


Is this done via an arbitrary modifier IN ADDITION to the literal demand element that takes place? ie. workshops have defined inputs which they buy from their local market (these can be edited in a text file), and I assumed that was the extent of their influence on that commodity's price: consumption reduces stock and the price rises as a consequence.

If the mere presence of the workshop has an additional layer of impact on the price of its inputs, that would indeed be a bad thing for any viable economic simulation.

Prosperity definitely does that and is the main problem here as far as I'm concerned, it seems to be the biggest factor on prices.

For the workshops I was only guessing, I tested with a brewery, and this doesn't happen from what I can tell, breweries don't affect the prices unnaturally. I didn't test other workshops though.

But SOMETHING else besides prosperity does, I'm not sure what, maybe it is the production of the villages, not sure. Look at these two towns, pretty much the same prosperity, same stock levels, but different prices for many things. Hides and leather much more expensive in the tannery town. In the pottery town, pottery is cheaper and clay more expensive. Flax half price in the town with the flax village.

So there are definitely some arbitrary modifiers, prosperity is one, and causes the food death spiral. There are some others that I'm not sure about.

 

[Badcritter]

The way Tanneries and Leather work is stupid. From what I've seen the only thing that consumes Leather is Tanneries, and Tanneries are of course also what makes Leather. This probably contributes to why Leather price is always in the red and Tanneries have been the most consistently profitable workshop for some time now, they have the Leather market cornered.

 

[drallim33]

Yes but in addition to that, there is some arbitrary modifier, look at the stock levels, they are the same, the price should be the same but it's not.

 

[Badcritter]

Assuming everything has exactly 20 stock because you bought and sold to make it that way, the reason for a difference in price at the same stock level is that you probably had to buy or sell different amounts in different places to get to that specific arbitrary stock level.

The game passively consumes commodities at define rates, and passively processes things too (eg. turns livestock into hides). Workshops increase the rate at which this happens for specific things. I don't remember whether Tanneries increase the rate of Hide production, judging from the prices in your screen shot they probably don't. They take the Hides input to make Leather (greater demand drives the price up) then they turn the Leather into armours, which consumes Leather apparently at a greater rate than the Tanneries produces the Leather, thus the price of Leather is actually higher in towns with a Tannery.

Those processing values can actually be tweaked in a text file. If I get time I'll find it again and fiddle some values to see if I can "fix Tanneries" (tm) lol

 

[ileboii]

Long ass ****ing story about a hungry Aserai army lead by bipolar commanders.

I laughed all the way reading through that lol

 

[drallim33]

It's almost like there's some code in the AI like, "if starving then raid tevea."

Rhagaea's army was besieging Vostrum, but began starving. So she went to raid Tevea, all the way from Vostrum, while starving. That Aserai army was besieging Morenia Castle, but also ran out of food and is heading to Poros. So those armies are at war but too hungry to fight.

Servic's army began starving while besieging Setadeim Castle, and did the only logical thing.

None of them actually raided Tevea for more than about five seconds.

 

[drallim33]

Monchug's 2000-man doomstack was starving while assaulting Zeonica. Victory was imminent but then the army lost cohesion. About one third of them went to Tevea.

A few minutes later Tulag turns up but is already starving before the siege even begins, and quickly turns around. They couldn't resist stopping in Tevea for 5 seconds on the way back for food.

 

[Badcritter]

Isn't Tevea the village used in the tutorial?

<insert any number of so many possible jokes about the AI's behaviour combined with this fact>

 

[Bannerman Man]

@drallim33 As Badcritter said, the player buying and selling goods will alter the price depending on how many goods were bought or sold.

However, price changes are not instantaneous; there is also a long term price trend that changes on the daily tick towards a stable value. You can read about it in heu3becteh's post here. If you only checked the prices for a single day, then the long term price probably did not have a chance to settle on a stable value yet.

As an example to illustrate this, here's Marunath and its inventory through the natural course of the game (in a new 1.4.1 campaign):

Marunath Natural Grain Stock​

And here is what happens when you hold the inventory of grain to a constant quantity of 100 over the course of a few days:

Marunath 100 Grain on Beginning Day

Marunath Prices after 1 Day

Marunath Prices after 2 Days

Marunath Prices after 3 Days​

So as you can see, it is steadily increasing towards a stable price each day for its given prosperity and quantity. According to @heu3becteh's post, the long term price trend rises slowly and linearly when the total quantity of a good is low, but if vast quantities are sold to a town's market, then the price shifts a great deal toward a stable value in only one day. That's why price drops so rapidly when you flood a market with goods.

As for the NPC lords constantly raiding Tevea, that looks like a bug, possibly introduced when the AI behavior was changed to allow NPCs to buy from neutral settlements, or perhaps during one of the other behavior modifications that have happened in the last few patches. Have you filed a bug report? Edit: See my post two down from here.

 

[ileboii]

A **** ton of pictures and Tevea.

You should be appointed as the Official Minister of Food&Supplies of Tevea. Good find.

 

[Bannerman Man]

There's nothing special about Tevea. Edit 2: I take that back, maybe there is somthing about Tevea that causes the AI to want to raid it. There's a bug of some kind happening here:

Tevea is owned by the Khuzaits at the time of this picture. It was at a later time than from the header posted below.


Original post:
Here is Tevea in my game, happy and healthy as ever:

I believe what is happening is that NPC lords can override their food seeking behavior if an easy target is nearby, even when they are starving. This seems to be because of the changes made that allow an NPC to finish out their sieges even when they are nearing starvation, so as not to call off a siege with only a handful of defenders left. This behavior, whether intentional or not, will allow them to take a hostile action on a nearby target if one is close enough rather than seeking out food. Eventually the food seeking behavior overrides the hostile behavior again when the action takes too long.

Drallim, my guess is that Tevea is poor in your game because of the constant warfare, and has only a handful of militia at any one time, making them a soft target.

It's a bug, guys. Someone should report it.

Edit: Here's a post I made about militia that is semi-relevant to this discussion.

 

[drallim33]

There's an economy forum? lol this place has too many subforums.

Ok so there is a double effect based on stock levels. One is immediate, the other a short term average. So, the product of a workshop will pretty much always be cheaper in that town, because the average stock level will be higher. So while there is no arbitrary modifier, the effect in the end is the same.

On top of that there is the double effect of prosperity - the arbitrary modifier, and also prosperity controls consumption rate of goods which effects stock levels.

FOUR separate supply/demand based modifiers, one completely arbitrary and the rest simulationist. I think you only need one, not four. But whatever, this is what we've got.

First let's talk about the prosperity-based pricing. It makes some sense in an abstract way, seems like they were trying to get caravans to sell more to towns that need it, and buy more from those that don't. In theory, with enough supply, caravans would keep selling to the high prosperity town until the price went down to near the average price. So the end result would be similar prices to low-prosperity towns, but higher stock levels. But it doesn't work like that because there is no check on consumption, towns will just consume every last available resource no matter the cost, causing them to grow even more and jacking the prices even higher.

At the other end of the scale, in the low prosperity towns, the immediate stock level pricing does not have enough effect. Stock levels close to zero should cause extreme prices but they do not, so caravans just buy everything and leave the town with nothing.

So all the goods get consumed by a town, and any leftovers get funnelled to higher prosperity towns, where they are consumed, and the result is a global shortage of everything. Towns grow until demand exceeds supply, and everything is super expensive or starving.

One problem here is that prosperity seems to apply an offset or multiplier to prices. Currently it (probably) looks something like this:

Even with 0 stock, the price in a small town can still be lower than in a well-stocked, high prosperity town.

It should look more like this:

So that high-prosperity towns would still maintain a higher stock level, but lower prosperity towns wouldn't get completely bought out. A shortage would still result in high prices.

I believe this would solve the death spiral issue, because poor towns wouldn't be selling all their goods below cost and going bankrupt, and caravans would actually be willing to sell there because the prices wouldn't be constantly too low.

To solve the life spiral issue, there needs to be controls on the prosperity growth.

Prosperity growth should be linked directly to the market. Currently there are only some small weak modifiers, these are not enough. There needs to be stronger modifiers and hard caps. I propose the following controls on town prosperity growth and consumption:

• For towns to grow, the granary must be full. If growth > 0 and granary < full, growth = 0
• Towns should consume as much food and other goods as they need, if the food is available. They should not over-consume like they do now, and only buy as much as they need.
• If there are not enough goods to meet the town's demand, prosperity should be capped at a negative value. So that it cannot be overridden by other bonuses. -1 for each missing unit. If the town got everything it needed except linen, needed 4 linen but could only get 2, then growth should be set to maximum of -2 for that day.
• If all the above is satisfied, the town has bought enough of all goods and has a full granary, AND it still has some funds remaining in its budget, it should use the remaining budget to buy some surplus goods. By surplus I mean goods that are currently below their base price. The value of these goods should then be added to that day's growth. (maybe base price/100, e.g. 5 wool = +2)

Doing this would solve all the problems. Towns would not consume all of their stock under normal conditions. There would be no long-term global shortages, since towns would shrink and lower demand if there was. Towns will turn surplus goods into growth, so you wouldn't end up with giant stockpiles of underused goods. Towns would not grow beyond their means to feed the population. They would still be vulnerable to market disruption, and reliant on caravans for growth. Towns would grow when in surplus, but not before. Large towns would not consume all goods from smaller towns like vampires.

Stock levels and volatility would vary between towns. A large army might be able to get all the food it needs from a high prosperity town but not necessarily from a low prosperity one. It would be easier to find extreme prices at smaller towns, but the prices would move towards the base prices with lower volume trades. Whereas prices at large towns would be more stable but it would take more volume to move them.

 

[Badcritter]

I like your model, it looks robust. I'm confused about this point:

If there are not enough goods to meet the town's demand, prosperity should be capped at a negative value. So that it cannot be overridden by other bonuses. -1 for each missing unit. If the town got everything it needed except linen, needed 4 linen but could only get 2, then growth should be set to maximum of -2 for that day.

My intuitive expectation is that total prosperity movement for the day should cap at zero if a demanded item isn't available, rather than at a negative number. Lack of any one thing can stifle growth, but shouldn't cause loss unless the actual net prosperity modifier is negative after taking into account all influences.

In fact the average player would be annoyed even with stagnation because their town isn't getting linen but is getting everything else it wants. "Buy something made of wool today instead you fussy bastards!". I'm not able to concentrate enough right now to run the thought experiment on whether the integrity of the system could be maintained if very limited growth was still possible in that scenario, but I expect that simply capping the growth at 0 (rather than at a negative value) should be sufficient for sustainable integrity.