Economics 101 Thread

Users who are viewing this thread

bonerstorm

Genghis Khan't
Sergeant
What I'd like to do is collect a number of suggestions to improve the economic simulation in BL, which I think a lot of players have been struggling with. Once it is fleshed out here, I'll x-post the best ones to Suggestions. It's also in the realm of imagination that I might make a mod to do some of this stuff myself - though I doubt the engine can handle all of what I want, since a lot of my suggestions are AI changes rather than economy changes.

I have a decent Econ background to start with, but I'm lacking in knowledge of the game's code and principles game design in general, so I'd like to summon: @SadShogun @Flesson19 @Bannerman Man @Apocal @Kentucky 『 HEIGUI 』 James, @Spyware and @Ananda_The_Destroyer.

Here's a summary of consistent economic problems I've seen in EA:
  1. Caravans and workshops are generally extremely weak investments
  2. Towns regularly fail to stockpile necessary materials, leading to unnecessary price shocks and economic collapses
  3. Bandit scaling at higher levels breaks the econ simulation and bankrupts the entire realm
  4. Pricing of certain goods fluctuate too much or too little - according to a semi-broken equation with maximum and minimum prices
  5. Denars magically appear out of nowhere and are hoarded/spent in massive quantities by lords, who also do not participate in the horse economy and bypass markets entirely when they sell loot
Here are some basic economic principles I'd like to see reflected in-game: (I'm sure some already are and I'm just not aware)
  1. Price elasticity of demand: Some goods (like food and firewood) should have near-constant demand which does not substantially decrease as price increases (because, no matter how much it costs, you still have to eat)
  2. Time value of money and opportunity cost: Money you have now is worth more than money somebody promises to give you a year from now. Anything you purchase now as an investment needs to be worth more when you sell it than if you'd invested it on something else.
  3. Price elasticity of supply: Greater profit opportunities should motivate traders to create more caravans and vice versa
  4. Substitution effect: Some goods should share a demand curve because one can be substituted for another (out of grain? eat fish!)
  5. Prestige/luxury pricing: Some retail goods should have demand specifically because there is limited supply (fish should be luxuries inland and dates should be luxuries in the north)
  6. Income effect and marginal propensity to consume: Higher income levels should mean higher spending in both increased consumption of basic goods and increased consumption of luxury goods
  7. Monetary Inflation vs. Supply-Crunch Inflation: Prices in general should reflect the supply of money as much as or more than the supply of goods
  8. Speculation vs. Investment: Market prices should take into account prevailing trends: for instance, a town with 3 silversmiths should reasonably speculate that demand will remain high for silver ore in the future and keep prices high even when current demand is met
Here are my own starting suggestions for fixing the economy:
  1. Every town should stockpile enough goods to supply them for the next year (half a year or 1 month at least), with demand slacking significantly ONLY after the supply chain is secured. I could also see a tiered discount of 5/10/20% per additional month up to a year. Also towns and villages should demand firewood depending on their climate and time of year.
  2. Prosperity should no longer simulate luxury demand or market decisions: notables/nobles should be practically the ONLY purchasers of luxury goods, with demand increasing based on their disposable income - with a daily/monthly spend that prioritizes consumables and then spills over to pure luxuries like velvet/jewelry. This would more accurately simulate the extreme price swings in luxury retail, which would be less about "how much stuff there is" and more about "how much the rich folk have to spend". Also, notables should respond to market conditions - if, for instance, their pottery workshop isn't making money because nobody's buying their goods... then they should invest in a caravan to get their goods to market.
  3. Goods should substitute when it makes sense: Peasants don't really demand "grain" - they demand "food". Once there is enough of at least 2 kinds of food to eat (grain + butter, fish + meat, dates + ale), then demand should drop to near-zero for everything else. So a settlement with a brewery that doesn't have enough grain or any other food should see the cost of grain skyrocket to near-equal the cost of ale, since "not starving" takes priority.
  4. Explanation of 1-3: A town has residents that eat 10 staple foods per day and 6 notables which demand 1 of each consumable per day. Minimum demand for "food" stands at 1320 (the minimum everyone needs to eat in a year). There are only 900 units of Grain in the settlement, which means the price of all food consumables will be high until 420 more units of food are produced/sold. Once base demand for food is met, then excess food demand is a function of notable/noble demand for greater food variety and their ability to pay for it. Let's say 1 out of the 6 notables can only afford basic food with their current profit, then that's all they buy. Let's say 4 out of the 6 have enough money to buy all different kinds of food, then they'll spend their "food budget" (a fraction of their income) on as many as they can get. Let's say the last notable has so much money that they can buy every kind of food... that notable buys a year's supply of them all and then spends the rest of their disposable "food" income on luxuries like jewelry/velvet. The order in which purchases are made should be richest-notable first, since those with more buying power should have more market muscle.
  5. EXTRA Explanation of 1-3: If the settlement has a production workshop like a brewery, then annual demand for grain should increase to factor in the increased need and annual demand for ale should lower to factor in increased supply. Even with the settlement starving for ANY food, the reward for delivering grain should be higher than the reward for delivering ale - because
  6. Weaponry, armor and horse prices should scale on troop replacement: Nuff said - if nobles are recruiting everyone in sight (especially high-tiers), then there should be more demand for armaments.
  7. Weaponry, armor and clothing should probably be aggregated into easily-tradable goods "packages": If part of the problem with the economy is that equipment prices need hard ceilings/floors because they're not traded between settlements, then this should change by allowing production to be easily transported from one town to another.
  8. MOST DANGEROUS SUGGESTION Denars should NOT be created ex nihilo - they should only be made at silversmiths: One of the problems with creating endless money out of nowhere is it screws pricing everywhere - especially when you dive headfirst into price ceilings and floors. The reason why silver historically has been such a good currency is that the production thereof has historically mapped well with the production of all goods in an economy and plagues/wars/collapses that affected the production of silver generally affected the production of everything else. This would give real wartime significance to the capture of silver mines and towns with silversmiths. It would also give rise to a more robust barter economy in regions without silver production. Also, ideally % tariffs should be paid by caravans selling goods regardless of whether the transactions are bartered - not a magic number appearing out of nowhere (at least tariffs should be earned as coinage but represented in the sim as a % of barter goods taken).
  9. EVEN MORE DANGEROUS Prices should only be high when there's enough money to pay them: Prosperity should drop as wealth transfers from peasants to the market and vice versa. If the peasantry cannot afford to buy food, then Prosperity (which should be a measure of the town's working-class and middle-class wealth) should drop as they starve and prices should reach a ceiling that matches their ability to pay. If Wine costs 100 denars and the peasants cannot afford to pay that much because they aren't earning enough, then the price shouldn't increase to 500 denars - because nobody cares if peasants "demand" wine if they just don't have the money to buy it. In math terms: total buying power divided by total demand should be maximum price increase created by peasant demand. This is a much more sensible "price ceiling" that doesn't have to be arbitrarily set by devs.
  10. SUPER DANGEROUS Crime should appropriate market goods to the peasants: When there is demand for goods that the peasants can't pay, then what goods there are should "fall off the back of a wagon" and transfer wealth from peasants to criminal notables at a discounted price, while cutting into the profits of merchants/artisans.

Summary:
The current system is simultaneously too complicated and too simplistic, because I understand it relies on a bunch of assumptions that don't really mesh and then a few stopgap rules to stop it from going out of control.

In the past, the system would break when a town would - for instance - produce too many horses (increasing Prosperity), increased Prosperity would increase food consumption and then the whole town would starve, wiping out half of the garrison/militia.

BTW: Here's a great - if somewhat outdated - breakdown of how pricing works currently.

My ideas may seem kind of complicated, but they'd actually streamline the game by turning it into an internally-consistent system that doesn't cause runaway-stupid economic breakdowns. It would simultaneously gear the economy to first-and-foremost prevent starvation (90% of any medieval economy), allow massive profit potential for selling luxuries to prosperous areas and also fix the denar-sink issue.

What do you think? Do you have more suggestions?
 
Last edited:

Ananda_The_Destroyer

Master Knight
1 Make parties and garrisons dessert🍨 immediately when wage is not payed. Could be a threshold depending on how short the wage is, but there needs to be an immediate martial penalty for the the AI (or the player...) not DOING THINGS to have ENOUGH MONEY for their troops! Waiting for moral loss is not enough and may never happen since so many NPC lord have leadership + moral anyways.

2 No free horse/warhorse upgrades for AI NPC, don't add extra either, make the AI have to DO WITHOUT upgrades if they looses to many parties and there aren't enough horses for everyone. If the raid all the ****ing horse villages, Oh too bad ***** you don't get any horses for a long time! Now you're slower and you're gonna lose like you SHOULD because you ****ed up!

3 Remove telepathic instant free event for AI NPC such as marriage, mecr recruitment and lord recruitment, make the AI waste it's time (and money)doing all these things exactly like the player does, and mercs should be paid on hiring and leave if payment is missed, make mercs get paid first so parties and garrisons get disbanded for running out of money. It's not okay to have poor ruined factions having mecrs all the time, it's fake and cheaty and you know it. You wanna play you gotta pay, can't pay go AWAY! If the AI king can't spend 300k-1m to get lord like the player they should never ever get 1.
 

ldgarrett

Regular
I guess I'm violating the Never Volunteer rule...
While I don't have detailed code analysis, lots of the folks you've already called in do.
What I do have is a ton of observational analysis of how the Strategic Layer (towns, trade, lord parties, clan finances, some political choices) works, current to an e1.7.0 campaign ongoing.
Please feel free to contact me here, in private, or on STEAM (same user name).

You're up against a scalar problem that may not be addressable without a complete redesign of the costs and values in the game, where entire towns operate on 1,000's of denarii a day and consume 10's of units of materials, yet a single war party of quality has a couple thousand in upkeep and millions of denarii in gear and horses that can be lost and need replacing in days. The game barely considers this, when you look at the definitions (and AI behavior determined by) Clan Wealth. It's why the single most devastating advantage a player can give a faction is to run around to every clan leader and hand them 200,000 denarii or so; Their garrisons and war parties and recovery rate of recruits will all more than double compared to the usual "our clans are Poor or Average" situation. And yet... that less-than-2-million denarii which is a thing that a wise mid-game (post game day 300 or so) player can and should be able to afford and recoup quickly... won't buy a fraction of the value of what those war parties will put in the field. Right now, the AI still mostly thinks in 1,000's of denarii at a time. One huge improvement would be getting the clans and factions to think in 100,000's of denarii (or do that massive rescaling job mentioned above).

You are also up against an inherently unaware Town and Castle Management AI, and almost anything that encouraged the AI factions run their settlements better (or have an obvious excuse like a Greedy Lord Governor that changed away from that) would be a huge improvement. Here's one tool that might be available for that, from the purely economic side: Currently, settlement caravans are formed by Merchant Notables and Players, and only base in Towns. They work a series of concentric "go to these markets" radii seeking beneficial trades and some specific demands (roughly 40% of destinations are "close", 30% "moderate distance", 20 far 10 go to forever far away :razz: ). Things that make these AI caravans stay closer and specifically fulfill local needs would be great, but the Merchant is quite reasonably sending them to make money not be a public service. That they are dumb about it is another part of the problem... Instead, I would strongly suggest that Artisans and Governors (including Castle governors) originate specific, one-time, caravans any time an obviously threatening situation to the viability of the settlement occurs. It doesn't need to be a free success; in fact those would be prime war and bandit targets. But a settlement should not be obviously dumb about the basic stuff it needs to keep its military granary full, and townsmen of the commercial class (small that it was) were not entirely unaware of basic feedstock and sustenance flows on a larger scale than "what's in the market?", even in the era the game draws inspiration from.

I hope those help, and here's to others helping here too.
 

bonerstorm

Genghis Khan't
Sergeant
YES! The AI may be too stupid to compete with real humans, but there's no way humans can compete with deity-level omniscience and cross-continental telepathy. The AI really needs to stop getting all these freebies, without even the courtesy of having to schlep miles to do it like we do.

The Diplomacy mod's messenger system is a really helpful stopgap solution to the problem, but it still REALLY burned my guts the last time I played to kingdom-tier and saw my favorite mercenary get picked up by the enemy AI DURING WARTIME seconds after the contract timed out.

This is also part of the broader BL problem that everything in the game feels sterile compared to Warband and Viking Conquest because the AI never socializes... hell, it never meets up at all except to join an army.
It's not okay to have poor ruined factions having mecrs all the time, it's fake and cheaty and you know it. You wanna play you gotta pay, can't pay go AWAY! If the AI king can't spend 300k-1m to get lord like the player they should never ever get 1.
The endgame is especially painful in Bannerlord because of this issue. I mean, I'd hardly know because I always got too bored to follow through.
THIS is exactly the kind of feedback I was hoping for! Def let's chat. Steam invite just sent.

Reacts:
1) The distribution of notable wealth really is kind of a mess and - for me - I believe the majority of that is that prices are completely nonsensical and so is the mechanism by which nobles get money.

Equipment prices, in particular, are way too high. We have a pretty good idea how much a denarius was worth IRL and there's no squaring that with equipment cost in-game (commodity prices are mostly OK though). 100 denarii is on track for a cow and 10 denarii per day salary is aight for an elite soldier but 50,000 denarii for top-shelf armor is kinda nuts (especially the SAME top-shelf armor that magically appears on a T7 unit for pennies on the dollar).

Back to the point: it would make sense for sovereigns to NOT want to give their vassals money and power if the vassals might switch sides then use that money against them... but side-switching disloyalty isn't really implemented in BL the way it was in WB. You're right that the AI doesn't really know what game it's playing.

One thing that would make a lot of sense: high-rank clans with tons of money but less than their maximum number of parties should hire wanderers of their culture. And there should be an option for richer, more powerful clans to absorb noble men into their households through marriage rather than give their daughters away - robbing them of a crucial party leader. The one getting the new household member should pay the cost appropriate to their value. Also, orphaned nobles of adults-all-dead households should be adopted by surviving clans.

2) You're right about the stupidity of the caravan situation. I'd like as well for the merchants to be able to call in pinch-hitter caravans to save their towns or businesses from bankruptcy/starvation/disaster.

Though the root of the problem remains: If you dump a load of cheap production material (i.e. flax or silver ore) into a town with a relevant production facility, then immediately that settlement will be swarmed with caravans buying that cheap crap to sell it in other towns which do not have a relevant production facility.

Why do towns without silversmiths demand silver ore? Why do towns without potteries demand clay? IT'S LITERALLY JUST STICKY DIRT.

It looks like the current pricing equation - which does not factor in substitution effects etc. - scales demand with Prosperity and increased supply decreases the price asymptotically approaching the minimum price.

And - this is ****ing ridiculous - it does not appear that workshop input demand is factored into the demand curve. @Bannerman Man am I right here? If that's the case, then it's no wonder that workshops keep failing to profit because they're not giving off a market signal that they need input goods. That would explain why caravans keep carting away input goods.
 

SadShogun

A Furtherer of the Calradic Cause
Developer
Hey @bonerstorm thank you for your feedback and analyses.

I agree with some of the issues you point out, while we're working on a fix regarding caravan and workshop investments there are still a lot further balancing and upgrades required for economy to feel better to the player.

I believe that the several aims the developer has when creating a sandbox game which is not an economic simulation (yet the economy takes an important role) are the following
  • A system which has understandable and predictable consequences to certain inputs
    • It's behavior can be learned and exploited by the player.
    • It should not allow exploits which are very easy to pull-off, extremely immersion breaking or "necessary" in the sense that you have to play that certain way to gain meaningful rewards.
  • A system which the developers can control and maintain, meaning easy enough to debug, add new content or tweak existing ones.
  • A system which allow developer intervention; to introduce missions, effects other "hand-crafted" scenarios. I believe this is the hardest part to achieve while still trying to achieve others.
  • A system which allows unexpected things to happen. This might seem contrary to first point I am making, but what I mean here is that while similar behavior should produce similar outcomes. The state space of the system should be large enough that there can be many combinations which can give rise to different situations without developers explicitly programming to do so.
  • And contrary to prior one, a system which does not spiral out of normal values and reduce game unplayable because of external effects for extended amounts of time. A good system imho (while it might take some time) should try to correct itself to return somewhere near the "normal" values.

In our current system in Bannerlord, I can see that the game achieves a lot of the statements I listed above. The amount it achieves is of course open to discussion, but the general economics simulation we have such as the luxury demand, prosperity being stable etc. are all ways achieve this goal. I do like the ideas you present here, however I also believe that they are already (to some degree) handled by our current systems.

For example,
The notables creating luxury demand instead of high prosperity.
This is a very nice idea, however, our high prosperity also simulates this, the aim of this system is to diversify poorer and richer towns, and also make them destinations to sell luxury goods.
or
Every town should stockpile enough goods to supply them for the next year
This also seems to make sense, however I can see two problems with this one
If there is a supply (which is not available to the markets and cannot be bought) this is already simulated by the moving average calculations the game makes when calculating the supply of a good.
The "supplies" are already not updated immediately and they instead slowly follow the actual supply in the market. In a sense this acts a store, and protects against the shocks created by looting, caravans/player buying a lot of goods in a short time etc.
The second one would be the sieges being too long (should every siege really last several months?) but this is a rather minor one which can be tweaked by other means.

Also for the substitute foods idea (which we already have, but not show to the player) it uses certain pairings of goods to create a substitute demand for them. (For example a city buys fish if there is not enough grain, though note the other ways (the system has its limitations))
This is a good feature we can expand to make it more similar to what you describe in your suggestion. So I took its note.


Replacing currently working systems takes very long time on design, development, balancing and testing. So it is a hard decision to make and requires good arguments to have "go ahead" with the new system.
However, the feedback and new ideas we receive from you is always welcome and can be used to form arguments to change existing systems so as always thank your for your feedback and ideas.
 

bonerstorm

Genghis Khan't
Sergeant
Thank you so much for taking the time to respond in detail!

Since I first posted the OP, I've been doing some digging into the market price equation and I'm getting a better understanding of how it works (or at least how it worked last year, the last time @Bannerman Man looked at it).

I'd be happy to give more detailed and helpful feedback rather than generalized critique, if you like. I love doing bug-hunting and feature-optimization. Did a bit of that here IIRC when Mexxico was working on the snowballing problem.

I don't want to make promises before I look at the code, but I'm pretty sure I could design a food-substitution price equation that wouldn't take much effort on your end to implement.

What I'd do is keep the Luxury Demand intact for every staple food item (fish/grain/butter/oil/dates/meat) but use the same Base Demand for all of them. The tricky thing about this is that purchases of one good would affect the value of linked goods, but it should be workable if the system can handle more than one price changing at a time. It might, admittedly, cause issues with the caravan AI though.

I also have some thoughts about the Prosperity and Food stats - which seem to be the source of some economic breakdowns - that I need to look into more.

The main thing that's been bothering me, though, is the tendency of caravans to go to a town that's suffering a chronic shortage or recovering from a disaster like a siege, then buy up crucially-needed raw materials. The worst example is when caravans buy up grain from a starving town after a siege, but I also notice that caravans have a worrying habit of buying up tons of silver ore from Lycaron even when I have two silversmiths there.

I'm not clear on if the demand equation is still purely driven by B & L variables, or if daily workshop consumption is added to the demand number.

What do you think? What's the best way I can be helpful to you? I'm going to be bug-hunting anyway. I'd much prefer to do it helpfully than as a way to fuel my complaining.
 

froggyluv

Grandmaster Knight
NW
Man you changed your tune rapidly -from like "id rather kill myself than endure this game unmodded" to " Great Mr Dev - thanks so much and glad to help!!!"

People still get "celebrity shock" when a dev responds - better to stay even keeled level headed.
 

bonerstorm

Genghis Khan't
Sergeant
Man you changed your tune rapidly -from like "id rather kill myself than endure this game unmodded" to " Great Mr Dev - thanks so much and glad to help!!!"

People still get "celebrity shock" when a dev responds - better to stay even keeled level headed.
My keel is even and my head is level :smile:

This is the first straight answer I've received from a dev since Mexxico left. If homeboy is on board with fixing the game even a little and I can make that job easier, you bet your ass I'll change my tone from angsty to grateful in no time flat.
 

Ananda_The_Destroyer

Master Knight
I think for me, the economy needs to have a much stronger, blatant effect on the power of the AI versus the player, or else it really doesn't matter.
As is, it's mostly just obnoxious things that only effect the player (because only the player "cares") and little petty differences in the AI that don't actually change the game enough.

The AI can either raid your village right away with new party or it can't. It can make new army and siege your fief or it can't. It doesn't matter one bit if they have a few more or less t4+ units or what not, if I have to walk over and fight them it might as well be the same thing. So all these little effects of them having more or less money, it just doesn't do anything.

What's also missing is the AI spending campaign time like the player does. They finance everything from passive income and incidental earnings from winning battles and don't ever have to *DO anything* to get more money, they just zombie on and on, rich or poor. It really doesn't seem like a economic simulation at all if it's only the player participating. there really should be times when an AI lord says "No I can't raid or join an army, I MUST hunt weak parties or do tournaments for money because my clan is just *****ed and if I wander around in an army or get caught raiding again we're finished, I will do OTHER activity until I have 20K again, then I can join army or raid village."

The AI doesn't care one bit if it gets food penalty and looses a few random garrison troops before a caravan (or 10) bring more food. Meanwhile the player will STOP progression and go cart in food so there precious collection or rare and exquisite troops doesn't poof into thin air. The AI doesn't care one bit about -security from hideouts, the player is compelled to STOP and go clear them out! The AI spawns with all it's gear, so does it's kids! The player and their crew.... The player (sometimes) must do actual trading to to build trade skill, the AI starts with cheated on trade skill and also gets it somehow from loot-to-gold (not sure?).

And this isn't strictly an economy thing, but the AI doesn't understand policies! Raggy and Derthert (and their vassals) love to invade 4 factions at once and have fiefs in a every land, but they always have ZERO support for the +loyalty policies that are an important boon for a expanding faction with miss-matched owner culture all over!
 

ldgarrett

Regular
re: Ananda's "...and also gets it somehow from loot-to-gold (not sure?)."

Any AI party disposes of loot items at the arrival at a settlement at a fixed coin return independent of the local market's supply of coin (although you can artificially create a situation where the game does decrement settlement coin in very uncommon cases). This is why one of the ways to create wealth in the game was/is to gift or barter high value items to an AI Lord with a party (even your own clan's) as they resell them irrespective of local coin availability at roughly a 1:0.4 return. If they are in their own clan, this enriches the clan; if they are in yours, then the party "excess money" will be transferred back to you the Clan Leader as part of the daily readjustment of clan party purses.
 

Ananda_The_Destroyer

Master Knight
re: Ananda's "...and also gets it somehow from loot-to-gold (not sure?)."

Any AI party disposes of loot items at the arrival at a settlement at a fixed coin return independent of the local market's supply of coin (although you can artificially create a situation where the game does decrement settlement coin in very uncommon cases). This is why one of the ways to create wealth in the game was/is to gift or barter high value items to an AI Lord with a party (even your own clan's) as they resell them irrespective of local coin availability at roughly a 1:0.4 return. If they are in their own clan, this enriches the clan; if they are in yours, then the party "excess money" will be transferred back to you the Clan Leader as part of the daily readjustment of clan party purses.
Do they have to go into a fief now? I know in the past they got a daily tick of loot-to gold, or are there two things that happen, 1 for if they go to a town with loot and one for if they are in the field during a tick?
 

ldgarrett

Regular
No code access for me, so I can only observe by Clan parties' finances and transfers to the player Clan Leader, but all I see are coin transfers spiking after the AI party tags a settlement, not starting immediately after the gift.
 

ldgarrett

Regular
@SadShogun First, yes, thank you for joining in.

I would repeat a thing said in a private discussion for you to hear and hopefully either already know or are willing to test for yourself:

The absolute test of the macro economy's fundamental basis would be one that takes the established sourcing rate of all base goods as it is and sees if the ideal state of Calradia is viable at all. To do that, you would: 1 ) Re-unite the Empire as a single faction (giving all the clans to one faction; any, but I suggest the WEmp.) and console commanding a state of temporary peace for all factions; 2 ) Let the game run, with whatever wars break out, as all AI. No player role at all; 3 ) Examine regular intervals of 300 or 500 days of world progress, with the eventual goal of looking at the emergent state post game day 2,500.

I think you will still find some incapacity of the game model to reach a high development ideal, from the lack of continued expansion of basic goods sourcing beyond that which villages plateau at in the 600+ Hearths size.

I think you will find that the AI inability (other than by unreliable random "solving of an issue") to address local disorder and the banditry of some regions obliterating village Market Parties compounds the problem to the point of failure.

and, very likely I think, you will see that a mismatch happens of the number of workshops viably producing certain goods v. the demand of settlement economies in total across Calradia for those goods, in emergent play over time. Which matters only if you want the economy to be responsive enough to reach an ideal state, but should at least be a consideration.

I ran a version of that model early in the e1.5.x series of versions, and it failed utterly. Since then, all the changes to "villages produce some of the basic goods other than their main product" were made. I know that Mexxico stated about making the change that he'd seen the problem, and that change was to address it in-part.

I believe you'd be well served to do such a test again, current version. Game play is arguably best served by a Calradia that functions economically in the ideal state, and then all practical states emergent are from the horrors of war and (if intentional to the overall "story" of the game) mismanagement of fiefs by the greedy... and that it is explained to the player that this is what is happening.

Thanks for listening.
 

Antaeus

Sergeant at Arms
I'm less concerned about the overall economy... it functions well enough to be plausible.

I think it is important to address bandit inflation at higher character levels, and therefore improve caravan usefulness. Better caravan survival rates has a flow on effect for the player that might counterbalance some of the existing economy plateaus.
 

CptMuppet

Sergeant
the AI starts with cheated on trade skill and also gets it somehow from loot-to-gold (not sure?).
I haven't look into the code, but I have tested it thoroughly in game on my clan members some time ago. AI party leaders get the same amount of Trading xp per denar from selling items like the player, but there is a key difference - the player gets xp only for gained denars from selling that make positive difference between buy and sell price of an item, however AI party leaders don't care about that and get xp for for every denar they get from selling items.
At least it works like that for items AI party leaders get from the loot and barters. I don't know whether they also trade like caravans or not, and if so, I also don't know how Trading xp is calculated for them in such case.

Edit: I have missed here one very important thing - actually AI party leaders don't get xp for jest getting denars from selling items, but for giving these denars to their clan leaders.
I am not sure how AI clan leaders get their Trading xp, but the fact that they indeed do get some xp in Traiding skill from time to time indicates that there is some other game mechanic present, and my guess is that it might be exactly the same trading mechanic like for the player.
 
Last edited:
1 Make parties and garrisons dessert🍨 immediately when wage is not payed. Could be a threshold depending on how short the wage is, but there needs to be an immediate martial penalty for the the AI (or the player...) not DOING THINGS to have ENOUGH MONEY for their troops! Waiting for moral loss is not enough and may never happen since so many NPC lord have leadership + moral anyways.

2 No free horse/warhorse upgrades for AI NPC, don't add extra either, make the AI have to DO WITHOUT upgrades if they looses to many parties and there aren't enough horses for everyone. If the raid all the ****ing horse villages, Oh too bad ***** you don't get any horses for a long time! Now you're slower and you're gonna lose like you SHOULD because you ****ed up!

3 Remove telepathic instant free event for AI NPC such as marriage, mecr recruitment and lord recruitment, make the AI waste it's time (and money)doing all these things exactly like the player does, and mercs should be paid on hiring and leave if payment is missed, make mercs get paid first so parties and garrisons get disbanded for running out of money. It's not okay to have poor ruined factions having mecrs all the time, it's fake and cheaty and you know it. You wanna play you gotta pay, can't pay go AWAY! If the AI king can't spend 300k-1m to get lord like the player they should never ever get 1.
I just do not observe these problems.

In my current game it is precisely day 500. The nobles of the world are basically broke. Even the pillar of safety (Aserai) have paper thin garrisons.
- Their castles have an average garrison of 60 men ranging between 31 and 91.
- Their towns have an average garrison of 136 men ranging between 93 and 174.

These are numbers you can role over without even stopping to take a breath.

In contrast my own nobles have much larger garrisons thanks to the sidepayments I need to pay for them to join me. The AI does not need to be punished; on the contrary they really need professional help.
 

bonerstorm

Genghis Khan't
Sergeant
What's also missing is the AI spending campaign time like the player does. They finance everything from passive income and incidental earnings from winning battles and don't ever have to *DO anything* to get more money, they just zombie on and on, rich or poor. It really doesn't seem like a economic simulation at all if it's only the player participating. there really should be times when an AI lord says "No I can't raid or join an army, I MUST hunt weak parties or do tournaments for money because my clan is just *****ed and if I wander around in an army or get caught raiding again we're finished, I will do OTHER activity until I have 20K again, then I can join army or raid village."
YES. The economy is really just window dressing if it doesn't affect gameplay in any other way than inconveniencing players.
The absolute test of the macro economy's fundamental basis would be one that takes the established sourcing rate of all base goods as it is and sees if the ideal state of Calradia is viable at all. To do that, you would: 1 ) Re-unite the Empire as a single faction (giving all the clans to one faction; any, but I suggest the WEmp.) and console commanding a state of temporary peace for all factions; 2 ) Let the game run, with whatever wars break out, as all AI. No player role at all; 3 ) Examine regular intervals of 300 or 500 days of world progress, with the eventual goal of looking at the emergent state post game day 2,500.
THIS!
I believe you'd be well served to do such a test again, current version. Game play is arguably best served by a Calradia that functions economically in the ideal state, and then all practical states emergent are from the horrors of war and (if intentional to the overall "story" of the game) mismanagement of fiefs by the greedy... and that it is explained to the player that this is what is happening.
+1

The best way to design/test the economy is to start from a point of comfortably-buffered equilibrium and work backwards from there. Then you establish which goods are subject to substitution effect and tweak them to partially share demand curves. Then figure out price elasticity of demand for each good. THEN you make sure growth in the money supply scales with economic output. VOILA: functioning economy.
I haven't look into the code, but I have tested it thoroughly in game on my clan members some time ago. AI party leaders get the same amount of Trading xp per denar from selling items like the player, but there is a key difference - the player gets xp only for gained denars from selling that make positive difference between buy and sell price of an item, however AI party leaders don't care about that and get xp for for every denar they get from selling items.
At least it works like that for items AI party leaders get from the loot and barters. I don't know whether they also trade like caravans or not, and if so, I also don't know how Trading xp is calculated for them in such case.

Edit: I have missed here one very important thing - actually AI party leaders don't get xp for jest getting denars from selling items, but for giving these denars to their clan leaders.
I am not sure how AI clan leaders get their Trading xp, but the fact that they indeed do get some xp in Traiding skill from time to time indicates that there is some other game mechanic present, and my guess is that it might be exactly the same trading mechanic like for the player.
As someone who spends the majority of their time in BL grinding the Trade skill, this is maddening.

My workshop income should also increase my trade skill. Just sayin.
 

CptMuppet

Sergeant
As someone who spends the majority of their time in BL grinding the Trade skill, this is maddening.
Agree, getting a high level of Trade is painful if you care about roleplaying and you do this the intended way. However, if you don't care about merchant gameplay and just want to get it levelled fast, there is nice exploit for it, check out this madman:
 

Ananda_The_Destroyer

Master Knight
I just do not observe these problems.
Oh, you observe them needing to buy and consume horses to upgrade? You observe the AI walking across the map to talk to a clan leader to get married? Do your poor ruined faction not hog all the merc clans?

No you're saying the old "the AI sucks so it's okay if they cheat" excuse.
In my current game it is precisely day 500. The nobles of the world are basically broke. Even the pillar of safety (Aserai) have paper thin garrisons.
- Their castles have an average garrison of 60 men ranging between 31 and 91.
- Their towns have an average garrison of 136 men ranging between 93 and 174.

These are numbers you can role over without even stopping to take a breath.
^This has nothing to do with making the AI participate in the economy and be more then numbers moving on the map. Plus the number you listed for garrison are very optimal, that's what I do myself, you don't need hundreds of troops eating all the food: if you can't stop the enemy in the field you're goin to lose the fiefs because they won't siege if they can't win. Garrison is all about having enough security to have enough loyalty without running out of food. Now, I'm not saying the AI understands that, but them having modest garrison sizes isn't an indication of them being weak or compromised.
The AI does not need to be punished; on the contrary they really need professional help.
They need different behavior and to help themselves and do other activities in the world besides raid and siege. If they do run out of money they do need to have a strong punishment(now it's a stern talking to), but they should also DO more to prevent this, including taking breaks from war to safely rebuild finances via quests and such.
 
Oh, you observe them needing to buy and consume horses to upgrade? You observe the AI walking across the map to talk to a clan leader to get married? Do your poor ruined faction not hog all the merc clans?

No you're saying the old "the AI sucks so it's okay if they cheat" excuse.

^This has nothing to do with making the AI participate in the economy and be more then numbers moving on the map. Plus the number you listed for garrison are very optimal, that's what I do myself, you don't need hundreds of troops eating all the food: if you can't stop the enemy in the field you're goin to lose the fiefs because they won't siege if they can't win. Garrison is all about having enough security to have enough loyalty without running out of food. Now, I'm not saying the AI understands that, but them having modest garrison sizes isn't an indication of them being weak or compromised.

They need different behavior and to help themselves and do other activities in the world besides raid and siege. If they do run out of money they do need to have a strong punishment(now it's a stern talking to), but they should also DO more to prevent this, including taking breaks from war to safely rebuild finances via quests and such.
The problem is that they do run out of money. They are not choosing those garrison levels because they are optimal, it happens because they run out of money. It is a very direct indicator that they have exhausted their money supply.

In fact, my best guess is that they are particularly likely to do so when they are not challenged/at peace. When at war some of their people are captives and thus cannot muster armies, attrition leads to lower quality/cost armies and assumably they do make something from loot. War is probably the only thing that actually keeps them afloat (feels familiar).

So, you need to give them something to do, besides war, to fill their warchest, before you start punishing them. The poor buggers are already in a tough spot as it is.
 
Top Bottom