 |
| |
|
|
Velociryx
|
 |
of Candle'Bre
Apr 1999
time: 05:20
|
|
LOL 'bout the comment on controlling the enemy's empire. Point taken.  I guess I mean I don't mind that demigod level of control in this game....with the inability to dictate to specific citizens, you don't have in CB the level of control you have in Civ, but I'll grant ya that Governance Points would further restrict what you're able to do.
It's an idea that I like in theory, and can see a variant game being born of it.
In fact, I firmly believe in keeping an open mind. If somebody can come up with a way to implement the Governance Point system in a way that it's more of a carrot than a stick, I'm not at all opposed to it! (I really like the Agent of the King, and letting the player choose what level of beuracracy he/she wants in the empire...less Beurocrats = less GP's and less direct control. If you're okay with that...coolio. If you want to micromanage, then provide a mechanism to allow for that....at the cost of a spiraling upkeep)....HOWEVER, despite the fact that the model loosely outlined above pays homage to realism, it also becomes less of a game and more of a simulation. The more power we give GP's, the more you basically just sit and watch the AI run your empire, making the odd decision here and there as you can. Fun factor heads south, and players get frustrated cos they can't exert the kind of control needed to really RUN the empire.
Having said all that tho...again, if we can find an implementation method that treats them as rewards for good play, I'll be quick to get behind it, cos I like the idea!
-=Vel=-
|
|
| |
|
|
|
JMarks
|
 |
formerly known as the artist
Mar 2000
time: 05:20
|
|
Vel,
Point taken about the fun and the simulation. The more 'real' we get, the less it'll probably interest the public. They need a pat on the head and told 'good boy'.
Still, if we 'role play' our populace, GPs can come in handy. Especially with a MP where more than one person can play in the same house. Power plays between the king and his nobles could become interesting. Oh well, not for the free release. 
Of course it would depend on what we have GPs for. Will the ruler still be a demi-god, with extensive powers in one way, but not in another? Could we have GPs only for economic and military, leaving trade (purchasing, selling), and populace for free reign (with perameters of course)? I'm not saying add them, though I'd like to see it, but it is doable.
But back to the society thingy. Populace can help determine government type, and government type gives power to a certain part of the population.
For instance, if you have no nobles (or higher clergy.  ) one can't really have a government that has aristocratic tendancies (fuedal system). If there is no king, you can't have monarchic tendancies. But these instances would be rare.
But for my second point, a fuedal sytem would give power to the aristocracy and the monarchy. So your nobles, and your higher clergy would hold some semblance of power. A pure democracy would invest the populace with power, (though you would still technicly control everthing 'cause you're a demigog) and less influence on the other two groups. Not that a pure democracy could, or ever would be achieved except mabey for really small states.
So, a specialist for the nobles could be a Duke.
The Duke cannot be trained, but he can be appointed after an extremely successful battle field performance.
The Duke benafits you in this way: Leads troops in province (or you can send him on campaign) giving them a moral bonus. He also can draft greater army numbers in times of crisis.
The position is hereditary, assuming the Duke actually has kids.
A Duke can only be promoted from a noble that has enogh 'experience' or 'training' points. The lower order of knights cannot become a Duke.
|
|
| |
|
|
|
moomin
|
 |
Moo Like In Moomin
Jul 2000
time: 05:20
|
|
quote:
Originally posted by Velociryx
The more power we give GP's, the more you basically just sit and watch the AI run your empire, making the odd decision here and there as you can. Fun factor heads south, and players get frustrated cos they can't exert the kind of control needed to really RUN the empire.
|
Then again, let's recall why GPs where first considered: every single 4x strat games ever designed has fun going south as the game progresses - the MM soon eats all the enjoyment. Any improvement on that is to be considered a godlike step forward. How many people basically give up games of Civ, Smac, Moo and Mom once they established themselves? Almost everyone, almost every time...
That said, let me offer a few ideas for GP alternatives that are, eh, carrotier rather than stickier:
A game like CB, being compartamentalized in provinces, lends itself to other forms of MM reduction methods as well. Consider a model where the player gets a couple of "flags" that he can pin on any province he owns - each flag means the player can directly MM that province _and_ is another build/production/goal queue that the player can assign to provinces that he can't MM directly. Let's say you start with tree flags - you can MM tree provinces yourself, and you get tree different queues that are essentially tree different "roles" that the rest can be assigned to. At any time, the player is free to change his flag to a province he'd rather MM now... well, once per turn and flag that is.
You gain flags by building special buildings that crank out flagcarriers... let the flags accumlate against a claim to the throne, or something... so that the one who has many invariably ends up the target of AIs and other players
Call flags and flagcarriers whatever fits. But something along these lines should be workable.
|
|
| |
|
|
|
JMarks
|
|
formerly known as the artist
Mar 2000
time: 05:20
|
|
quote:
Originally posted by moominparatrooper
A game like CB, being compartamentalized in provinces, lends itself to other forms of MM reduction methods as well. Consider a model where the player gets a couple of "flags" that he can pin on any province he owns - each flag means the player can directly MM that province _and_ is another build/production/goal queue that the player can assign to provinces that he can't MM directly. Let's say you start with tree flags - you can MM tree provinces yourself, and you get tree different queues that are essentially tree different "roles" that the rest can be assigned to. At any time, the player is free to change his flag to a province he'd rather MM now... well, once per turn and flag that is.
You gain flags by building special buildings that crank out flagcarriers... let the flags accumlate against a claim to the throne, or something... so that the one who has many invariably ends up the target of AIs and other players
Call flags and flagcarriers whatever fits. But something along these lines should be workable. |
Moomin, your a god send for PGs GPs and all the rest. The flag idea (while I'd rather have it being a person, ala agent of the king) is a great idea, and kinda what I was going for in my own schpiel on beaurocrats. All we need to do is make it more visible and 'rewarding' to move them around. I like it, reintroduces the whole idea, and even allows 'individualizing' of the agents. At least if I'm reading you properly. You are saying that some will be able to do economy MM and others Military MM, right? If not, thats ok too, just clarify yourself.
And we can use it in the societal model too! (I'm still going for beaurocrats and agents here) The agents and beaurocrats really piss off the nobility because they are burghers or peasants by birth and they get pretty high ranking stuff to do. Too many can piss off the nobility in a highly aristocratic house where the nobility actually controls a decent amount of power. Oh, I'm all excited again!
|
|
| |
|
|
|
Ramo
|
 |
Austin, Texas, USA
Oct 1999
time: 23:20
|
|
I've been implementing some of this stuff, so I've been looking over your posts, and I have a few more comments:
quote:
1 Mercantile Improvements
2 Industrial Improvements (relatively easier to keep happy, again, thanks to the large number of Industrial Improvements) |
What's the distinction between the two, exactly?
quote:
Clergy (Burghers/Nobles - Priests and their supporters)
Rogues (Vagabonds and Ruffians - Criminal Element - Base % represents crime bosses, with this archetype growing as unrest and unhappiness increases)
Rural Folk (Peasants - Farming Communities)
Cosmopolitan (Burghers - Businessmen/Bankers)
Merchants (Burghers - Tradesmen/Free Market Folk)
Military (Nobles - Military activists/Professional Soldiers)
Fringe (Burghers/Nobles - Artists, poets, and other “dangerous” folks)
Academics (Nobles/Burghers) |
Would you mind it if I made the following naming changes?:
Rural Folk = Farmers
Cosmopolitan = Bankers
Fringe = something else... artists, maybe?
It's just that the feel of you names you had don't seem right if you catch my drift...
quote:
Clergy - Religious Stronghold - Any troops trained in this province gain +10% Morale, due to strong religious convictions (which some may regard as bordering on fanaticism). All religious structures in this province produce double their normal outputs. Province is considered to have a level 2 fortification in it (non-upgradable). Taxes in this province, and revenues from economic activity drop by 20% due to the Church taking in increasingly more money from the population. Organization level of the Clergy Archetype is +1% per tile granted.
Rural Folk - Farming Collective - All members of the Rural Archetype get a + 20% productivity boost. Up to fifteen workers may now be assigned to work on farming/hunting tiles (+1 per tile beyond the first so granted to the Collective). Organization level of the Rural Archetype is +2% per tile granted.
Fringe - Artist Colony - .5% per four turns, per tile granted the group of generating a Great Artist.
Rogue - Syndicate - Provincial Income drops by an additional 20% beyond the loss caused by making the grant, but C&B is eliminated as a possibility from the province in question. For every tile beyond the first so granted, Mercenary hiring rates increase by +1 in the province.
Merchant - Trade Guild - Unleashes pure free market forces in the province. All future provincial improvements built in a province containing a Trade Guild cost 10% less gold to build. Further, depending upon the size of the grant to the Merchants (total grant size/3), one additional provincial improvement may be constructed in the province.
Cosmopolitan - Bards Guild - .5% per four turns, per tile granted the group of generating a Great Artist
Military - Military Academy - Counts as a second barracks for the province in question, effectively doubling the number of troops that can be trained in the province. For every tile beyond the first so granted, this number increases by +1.
Academic - Scholar's Forum - Doubles the output of all scientific improvements in this province. Additionally, the Scholar's Forum generates 1 research point in each tech tree per tile granted them, per turn. |
More renaming...
1. Syndicate = Knighting Rouges
2. Scholar's Forum = University
3. Religious Stronghold = Monastic Order
4. Bard's Guild = Banking Guild (why would Bankers like Bards?)
5. Artist Colony = Artist Patronage (wording sounds better IMO)
Also, the Farming Collective and Communal Enclave seem redundant...
What do you think?
|
|
| |
|
|
|
Velociryx
|
|
of Candle'Bre
Apr 1999
time: 05:20
|
|
Hey bud! Lemme see if I can whip out some quick answers on those:
1) Mercantile Improvements - Generally these guys will directly increase the income of a province.
Industrial Improvements - Generally, these guys will allow different types of stuff to be built.
There *is* some overlap in functionality, but this is mostly done so as not to hobble or paralyze players who find themselves dealing with a massive segment of population with an aversion to one type or another.
2) Those name changes sound coolio to me. As we proceed, we may tweak them further, but I'm game with those!
3) Again, those name changes are coolio, 'cept that you might wanna rethink university. IIRC, the core rules list an improvement bearing the name "University" which could cause some confusion there.
4) Farming Collective and Communal Enclave are very close, but the key difference is that one caters to the Farmer Archetype direct, while the other caters to the Peasantry (that part's done from memory, tho I will go back and recheck that, and edit here as needed). A slight difference, 'tis true, but IMO, the distinction allows for a shade more in the way of player options.
I've not forgotten the ol' societal model, btw, and am still hammering (well, chipping might be the better term) at creating the next version of this....'s slow goings, trying to frame everything in my head and make it all work the way I envision it to, but it's getting there....
-=Vel=-
|
|
| |
|
|
|
Ramo
|
|
Austin, Texas, USA
Oct 1999
time: 23:20
|
|
quote:
Mercantile Improvements - Generally these guys will directly increase the income of a province.
Industrial Improvements - Generally, these guys will allow different types of stuff to be built. |
I'm not sure if I understand the distinction. Doesn't one usually imply the other (i.e. economic growth implies diversification of industries and vice versa)?
quote:
4) Farming Collective and Communal Enclave are very close, but the key difference is that one caters to the Farmer Archetype direct, while the other caters to the Peasantry (that part's done from memory, tho I will go back and recheck that, and edit here as needed). A slight difference, 'tis true, but IMO, the distinction allows for a shade more in the way of player options. |
But aren't the vast majority of peasantry farmers, and all farmers peasants?
|
|
| |
|
|
|
Velociryx
|
|
of Candle'Bre
Apr 1999
time: 05:20
|
|
True, yes, but here's how I see it working in my head:
Mercantile improvements (ie Banks, Markets, etc.) improve the province by providing a direct cash boost (let's say that building and fully staffing a market gives the province a boost of 5gpt), while industrial improvements (fletcher's guild) make building specific KINDS of units cheaper (archers cost 2g less, each time you build one here).
As to the collectives....quite true...the vast majority of the peasantry falls into the "farmer" archetype, but there are some scattered exceptions to this, and so I included another sort of collective (hopefully) to give players a choice of how they wanted to arrange their grants (purely along class lines, or catering to a particular archetype).
-=Vel=-
EDIT: Industrial improvements would also include stuff like mines, which harvest resources that can be converted into various finished goods for trade or sale, and/or allow for other types of provincial improvements to be built. Mercantile improvements, on the other hand, deal primarily with the more amorphous aspects of trade (financial markets).
|
|
| |
|
|
|
JMarks
|
|
formerly known as the artist
Mar 2000
time: 05:20
|
|
quote:
Originally posted by Ramo
I'm not sure if I understand the distinction. Doesn't one usually imply the other (i.e. economic growth implies diversification of industries and vice versa)?
|
Exactly! But one is actual economic diversity, the other is the actual facilities. You can't make tons of weapons with a bank, you need a weapon's manufactury. The one provides the funds for the outfit, the other provides weapons for conquering, which will compliment the other. And you can't make tons of weapons unless you have nice mining outfits and later on big forges and stuff. Unless of course I'm missing the whole point myself.
quote:
But aren't the vast majority of peasantry farmers, and all farmers peasants? |
Depends on which period. To begin with, yes, most likely. But there are, and will be free farmers. Some pretty wealthy. Most will be peasantry, at least to begin with, and they'll have lords over them. But just because it starts out that way doesn't mean it should remain that way. Of course, we could flip scales and have most start out free, but then become peasants as wars ravish the land and force them to seek the shelter of a lord, or are pillaged so badly that they can't support their families, they get drafted leaving hired hands to do the work, etc. Just a thought. Hope I didn't miss the point here either.
|
|
| |
|
|
|
Ramo
|
|
Austin, Texas, USA
Oct 1999
time: 23:20
|
|
quote:
Industrial improvements would also include stuff like mines, which harvest resources that can be converted into various finished goods for trade or sale, and/or allow for other types of provincial improvements to be built. Mercantile improvements, on the other hand, deal primarily with the more amorphous aspects of trade (financial markets). |
Gotcha.
In that case, why don't we collapse bankers in with merchants, and invent a new archetype - industrialists to take over the characteristics of the bankers, since bankers would be more inclined to prefer improvements in financial markets rather than, say, inventions in textiles production.
quote:
As to the collectives....quite true...the vast majority of the peasantry falls into the "farmer" archetype, but there are some scattered exceptions to this, and so I included another sort of collective (hopefully) to give players a choice of how they wanted to arrange their grants (purely along class lines, or catering to a particular archetype). |
Hmm... point taken. But I prefer not to clutter the player and AI with unnecessary additions.
quote:
Depends on which period. To begin with, yes, most likely. But there are, and will be free farmers. Some pretty wealthy. Most will be peasantry, at least to begin with, and they'll have lords over them. But just because it starts out that way doesn't mean it should remain that way. Of course, we could flip scales and have most start out free, but then become peasants as wars ravish the land and force them to seek the shelter of a lord, or are pillaged so badly that they can't support their families, they get drafted leaving hired hands to do the work, etc. Just a thought. Hope I didn't miss the point here either. |
Free peasants are considered villeins and unfree peasants, serfs... The 'peasantry' generally refers to both serfs and villeins.
Keep in mind that armies are very small in this period. The amount of peasants actually taking up arms were very small, relative to the population in general. Also, the military archetype is intended to mean professional soldiers - knights and men at arms.
|
|
| |
|
|
|
JMarks
|
|
formerly known as the artist
Mar 2000
time: 05:20
|
|
Could I put an idea out there? Could we have each province have some ratings? EG: influence, honor, loyalty.
What I'm thinking is, at the VERY least, each prov has one set of these ratings. Preferably three sets (or more), one for the common people, one for the clurgey/wizards (whoever is dominant in the prov), and one for the nobles. Here's what the ratings mean.
Influence: Who has more power in the province. Its divided between you, and however many types of people we include. Naturally, its the nobles or the clurgey, or some balance therein. In republicized provs, its the people. This rating can have various effects including tax collection (the more influence you have, the more taxes you get, the more influence the nobles have, the more taxes they get, etc), may be important in making and dealing with vassals, and may limit what you can do in the province. This is NOT to be confused with loyalty. You cannot loose a province by having a very low influence as long as the loyalty is high, though the two should be connected.
Honor: How much they can be trusted. Might not need this one. Ideas? Could be used in espionage, and or orders to build (low honor means they'll go WAY over budget, 'cause they're spending half the money at the pub)
Loyalty: Each societal faction has a loyalty to someone, whether it be to you, themselves, another societal faction, or another major faction. Each faction's loyalty is measured in percentage (not to exceed a combined 100%). Anyone who has no (or negligable) influence is not displayed to prevent clutter whereever we put this for player's view. Used for various things, including the expanded espionage model, deciding probability of bandits and free companies occuring, etc, etc, etc. If only one of these three ideas get in, I'd prefer it be this one.
Ideas? Comments?
|
|
| |
|
|
|
Ramo
|
|
Austin, Texas, USA
Oct 1999
time: 23:20
|
|
First off, I completely agree that there should be an 'influence' rating, but I don't agree with what you its implications should be.
quote:
Who has more power in the province. Its divided between you, and however many types of people we include. Naturally, its the nobles or the clurgey, or some balance therein. In republicized provs, its the people. |
To be more accurate, the burghers. Peasants shouldn't have much power regardless of what kind of governmetn ti is.
quote:
This rating can have various effects including tax collection (the more influence you have, the more taxes you get, the more influence the nobles have, the more taxes they get, etc), |
Actually, the direct opposite was true. The more influence a social group had, the less it was taxed. Historically, nobles and clergy were lightly taxed, with burghers somewhere in the middle, and peasants harshly taxed.
BTW, I don't know if this has been brought up yet, but I think we should have varying tax rates for each of the social classes (and the clergy should be considered distinct for this purpose).
quote:
may be important in making and dealing with vassals, and may limit what you can do in the province. |
Can you elaborate?
quote:
This is NOT to be confused with loyalty. You cannot loose a province by having a very low influence as long as the loyalty is high, though the two should be connected. |
I'm not sure what you mean by this. By your description, influence is a property of a social group, not a principality. How can you have low influence? Do you mean some sort of centralization level?
___________
That said, this is what I think influence should mean:
scrap out the part of the "organizational" level that refers to forming cities, and that's a social group's influence (since there should be no coorelation between being urban and having power). The summation over all social groups of the products between a social group's population and influence is considered happiness. A province with a happiness level that is below a certain minimum may try to revolt.
quote:
Each societal faction has a loyalty to someone, whether it be to you, themselves, another societal faction, or another major faction. Each faction's loyalty is measured in percentage (not to exceed a combined 100%). Anyone who has no (or negligable) influence is not displayed to prevent clutter whereever we put this for player's view. Used for various things, including the expanded espionage model, deciding probability of bandits and free companies occuring, etc, etc, etc. If only one of these three ideas get in, I'd prefer it be this one. |
By what you've written of 'loyalty,' I assume you mean loyalty to a faction, not loyalty to government policies (some sort of nationalism). If that's the case, I don't think it should cause bandits to necessarily spring up - that probably should be more tied to happiness.
I think loyalty should determine which power a province revolts to; if it's primarily to some other faction, it could revolt to that faction; if it's loyalty is primarily to you, it could become independent.
Also, it could be connected to war wariness with certain factions which a province is loyal to.
quote:
Honor: How much they can be trusted. Might not need this one. Ideas? Could be used in espionage, and or orders to build (low honor means they'll go WAY over budget, 'cause they're spending half the money at the pub) |
Seems extraneous...
|
|
| |
|
|
|
JMarks
|
|
formerly known as the artist
Mar 2000
time: 05:20
|
|
quote:
Originally posted by Ramo
Actually, the direct opposite was true. The more influence a social group had, the less it was taxed. Historically, nobles and clergy were lightly taxed, with burghers somewhere in the middle, and peasants harshly taxed.
I'm not sure what you mean by this. By your description, influence is a property of a social group, not a principality. How can you have low influence? Do you mean some sort of centralization level? |
Yes, influence is what social group has it. What I meant is if you have a high influence, you can collect taxes easier. If the influence of the nobles or burghers or clurgey is higher (not only would they pay even less in taxes) they would end up siphoning taxes from the lower classes for thier own purposes. Influence is basically who, in any given province, has the ability to do things. While you'll always be able to build and draft, etc, etc in a province, if the nobles or burghers have a very high influence, they can really end up being a pain in the arse by siphoning taxes, increasing costs, etc. So it would be connected to how centralized a province is. For example, noble influence would increase costs for military stuff (or would it be non-military stuff), burghers woudl increase cost for economic stuff (or would it be non-economic stuff), clurgey would increase cost for religeous stuff (or would it be non-religious stuff?)
quote:
By what you've written of 'loyalty,' I assume you mean loyalty to a faction, not loyalty to government policies (some sort of nationalism). If that's the case, I don't think it should cause bandits to necessarily spring up - that probably should be more tied to happiness.
I think loyalty should determine which power a province revolts to; if it's primarily to some other faction, it could revolt to that faction; if it's loyalty is primarily to you, it could become independent.
Also, it could be connected to war wariness with certain factions which a province is loyal to.
|
I think thats what I meant for loyalty. If not, then I like what you said better. It would be a good thing for espionage/diplomacy.
"Hand over the woods of Arden, they are Furies by nature, Furies by their wishes, and thus should be Furies. Hand them over or we shall anihilate you into utter nothingness."
|
|
| |
|
|
|
DeepO
|
 |
supporting Candle'Bre
Jan 2002
time: 06:20
|
|
Guys, for the past few weeks I've been trying to read up on the Candle'Bre project, and so far I've loved what I see. The information is huge, so excuse me if it takes a while to catch up, the leads in the column were great, but man, there are pages and pages of information 
I don't know whether this has been discussed so far, and I don't think I know enough yet of the societal model to comment in full, but I noticed this post below on page 1, and want to add an idea to it. I don't know if this has been suggested already, and don't know if the idea was forgotten, so please forgive me if I'm not up to date. I'm continuing to read the thread (only at page 1 now, but proceeding), but if I don't comment now, I'll forget what I want to say.
quote:
Originally posted by Velociryx on 14 June
WOW! Rich, that parental pairing notion is top notch! You're right...in my head, I went with a static framework (whereby the archetype percentages were constant, and each time new people were born in the province, they simply got "plugged into" the various archetypes according to the baseline percentages...but I'm LOVING the notion of an evolutionary population!!
Probably, it'd require at least one other attribute on the buildings in the Economodel (call it "imprinting" or something, lending to its ability to slow stamp itself on the minds of the population, making it more likely that their offspring will drift toward the archetype that a specific buildig imprints.
So...the basic system could be something like this:
Mom and Dad give birth to a new CB citizen. That CB citizen is 50% likely to go with mom's archetype, and 50% likely to go with dad's archetype.
If there are buildings in the province with 'imprinting' toward a particular archetype, then this becomes a third force (with percentages being pulled evenly from mom and dad's pool to create the third choice).
Example....let's say that scientific builds come with a 2% academic imprinting, and the province in question has 5 Scientific buildings in it.
Mom's a farm gal, dad works in the church. Junior would normally be 50/50 split between farming and church work, but because of the 10% Academic imprinting, he now has a 45% chance to become a farmer, 45% chance to become clergy, and a 10% chance to become academic.
Would be REALLY interesting too, if we gave the imprinting value a time-driven increasing effect....say that every x turns, it's effect increased by an additional 2%, to a maximum of 20% or something.
In the above example, if mom was an academic and dad worked in the church, then the imprinting chance would be added to mom's chances of influencing junion (50 base, plus 10 due to the pull of the facilities, makes it 60% likely that Junior will drift into the Academic Archetype, and only 40% likely that he'll join the church).
I LOVE this idea!
|
I like the evolving population a lot, even if I should do some calculations to see whether the data needed is possible to calculate in a decent time. You're talking about storing a maximum of 100,000s of citizens, each with several attributes, that might be a big database to go through each turn. But apart from that, if you want a truely evolving population, the parental model above would be a start, and with some modifications would become a Genetic algorithm, thus really evolving.
The point is to make the chance of a child becoming one of the parents archetypes dependent on how happy the parent is. In the case of a academic / peasant couple, the academic would be happier if the focus in the province / empire is on science, where the peasant is happier if he can work his farm (If I understand this correctly). Say, as example, that the academic is 80% happy, where the peasant is 40% happy. Thus, a child of the two would be in 2/3 of cases an academic, 1/3 of cases a peasant. This is in genetic algorithm terms the crossing of genes.
Further, a random component willl change the child in e.g. 1% of the cases to something else, totally randomly choosen between the different archetypes. This would make it possible that an academic-peasant couple produces a merchant, even if the chance of this is low. The weighted chances to get a different archetype can depend on many things, but the current division is probably a good one: if your province consists of 60% peasants, the mutation will lead to a peasant in 60% of the cases.
With these two algorithms, a population can really grow and evolve, as long as the happiness algorithm is balanced. Therefore it is needed to have negative factors as well: a peasant that is not working a field is an unhappy peasant. I think that so far you have a lot of these already, so probably only some tweaking is needed to get the results wanted.
Further, a technical note: it is not needed to treat the population as individuals for this, you can calculate it per province with some statistical tools: If you know that there so many academics and so many peasants, you know how many mixed couples there will be. As you also know the average happiness factors, you can calculate what offspring that will give. Do this for every combination in each province, and you are done, you don't need to pair all citizens to someone else, and keep a record of who is married to whom.
Keep up the good work, I hope you like the idea,
DeepO
|
|
| |
|
| |
|
|
|
Ramo
|
|
Austin, Texas, USA
Oct 1999
time: 23:20
|
|
A few thoughts:
If we are going this route, there needs to be constraints on the pairing we have. For instance, no academic is going to marry a dirty, unwashed peasant farmer. A person MUST marry within the same social caste, and a large modifier should be given for marriage within the same archetype.
But the net effect of such algorithms is that the divisions would stay approximately the same. So I suggest dropping the idea.
Instead, archetype changes could be handled exclusively through a market model, which, IMO, is much more realistic.
I thought I should add my two cents before heading off.
|
|
| |
|
|
|
DeepO
|
|
supporting Candle'Bre
Jan 2002
time: 06:20
|
|
Vel, I'm on page 3 of this thread now, and read a lot of the comments to the original idea. But, as a former AI software engineer, I don't see any practical problem with this evolving, as long as you do it right. The trick is to bundle it in groups, and I think that that is perfectly possible. As long as you know the breakdown in groups, and calculate their happiness levels, the rest is some fast code that will be harder to think out than to run. You will have a harder time with the path finding in game.
The training level of citizens is something else. If you want to adjust these individually, you will run into problems, even if the individual calculations will be small. 100,000 citizens means 100,000 loops... ouch. Even if you could devise an algorithm to do it with queries and use a high-end database engine this will take a long time to perform.
It might be an idea to abstract this, and have a training level for very group of citizen in a province. You could keep the rest of the model, but adjust it to this. E.g. if 80 out of your 160 peasants are working on farms, the average goes up by 1%/turn (IIRC 2%/turn was the individual level). Further, some adjustments are needed for the farmers dying and leaving their archetype, and for child peasants starting to work. And, calculations based on the training level (e.g. whether they can advance to mining) have to become a chance: with a population of 160, there are x peasants chance that will have a level above 20% if the average training is y%. It will make the calculations more difficult and abstract, but will also mean the idea is still possible.
But then again, maybe the 100,000 loops is not that bad... it depends on how much memory access there will be, and how many calculations have to be done. After all, by the time CB is a fully functional game, you can set a minimum requirement of a 500MHz, 128MB computer, if not higher. 500MHz means that in 1 second time, you can spend 5000 cpu cycles on one citizen, and still have all 100,000 of them done. 5000 cycles is not nothing... but everything that can be handled on a higher level (like the population growth I proposed) should be done there.
DeepO
|
|
| |
|
|
|
DeepO
|
|
supporting Candle'Bre
Jan 2002
time: 06:20
|
|
Ramo, I'm not sure about the constraints... true love has no boundaries I agree that irl there were and are lots of constraints, but this will only make it more difficult, so a simplification is possible. After all, it is still a game, and not a psychology experiment to accurately model who would marry whom why.
And even then, if needed this could also be worked in the calculations...
About your second comment: No way the divisions will stay the same, the population will evolve, and with the happyness provided can evolve quite rapidly! If you are focussing on science exclusively, the peasants are jealous, so every couple of a peasant and an academic will produce an academic child. After all, both parents wants what is best for their child, and if they see what makes them happy in their own situation, the child is easily lead into a certain profession. The mutation aspect will provide for those children that do not follow their parents advice, which will also lead to some randomness.
And this will balance again: at a certain time, a lot of the scientists have nothing to do, have to work the land, so are unhappy. While the peasants love their job, and even if they are jealous they are happier than the academics. So, children will have a higher chance of turning into peasants again.
Further, another balancing issue is that even when no peasant is happy, peasant-peasant couples will normally produce peasants as childs. So you won't run out of peasants in 10 turns time because you temporarily favor the academics more.
The only problem I see is how happiness gets calculated, and it maybe will be needed to somehow correct this value before using it as a cost function. For instance, adjust the happiness relative to the highest and lowest possible values present in the game. But this shouldn't be an major hurdle to take.
DeepO
|
|
| |
|
|
|
Ramo
|
|
Austin, Texas, USA
Oct 1999
time: 23:20
|
|
quote:
Ramo, I'm not sure about the constraints... true love has no boundaries I agree that irl there were and are lots of constraints, but this will only make it more difficult, so a simplification is possible. After all, it is still a game, and not a psychology experiment to accurately model who would marry whom why. |
I think the game should be reasonably accurate. I don't want weird stuff to happen for no reason, but due to clear socio-economic trends. I don't want half of the population to suddenly become academics because of "true love."
If we do go this route, constraints are absolutely needed.
quote:
About your second comment: No way the divisions will stay the same, |
Of course not, if you allow crazy pairs. We can't have a society where serfs and many poor villeins exist at the same time as frequent inter-marriage between social classes. It just shouldn't happen. I suppose I'm just a conservative that way. 
quote:
And this will balance again: at a certain time, a lot of the scientists have nothing to do, have to work the land, so are unhappy. While the peasants love their job, and even if they are jealous they are happier than the academics. So, children will have a higher chance of turning into peasants again. |
I'd prefer to use a simple supply and demand market model (which can be affected by decisions by the state - the player) exclusively to determine job changes, rather than modeling their parents occupation, etc., etc. I think we get the most simple and realistic population model that way.
|
|
| |
|
|
|
DeepO
|
|
supporting Candle'Bre
Jan 2002
time: 06:20
|
|
I finally reached the end of this thread, great read!
Back to the evolving population constraints: What if you add soft constraints that burghers only marry burghers, nobles only marry nobles, in e.g. 80% of the times? This will keep the option open, keep the more or less accurate societal aspects, and will keep the evolving population idea alive. If you forget about that, and just make it a supply and demand model, surely that is not accurate either. After all, it is not because the king thought that more academics are needed that the population shifted towards that, no, he can indirectly influence the grand situation, and hope that his population follow him. So favor the academics, and hopefully more academics will emerge...
DeepO
|
|
| |
|
|
|
Ramo
|
|
Austin, Texas, USA
Oct 1999
time: 23:20
|
|
quote:
If you forget about that, and just make it a supply and demand model, surely that is not accurate either. After all, it is not because the king thought that more academics are needed that the population shifted towards that, no, he can indirectly influence the grand situation, and hope that his population follow him. |
Right, that's exactly how a supply-demand model works... There's a certain natural demand based on the current education and technological sophistication of the populace and then there's a extra, public funded demand, based on state sponsership of education. Then, the number of academics would shift towards the equilibrium number based on the demands of the populace and the state.
quote:
What if you add soft constraints that burghers only marry burghers, nobles only marry nobles, in e.g. 80% of the times? |
You *could* do that I suppose, but that complicates the model even more. The number was likely so close to 100%, this element wouldn't be worth modelling.
|
|
| |
|
|
|
Ramo
|
|
Austin, Texas, USA
Oct 1999
time: 23:20
|
|
quote:
As to the supply-demand model, I don't see how you'd accurately get the same kind of emergence and indirectness with a simpler model, without grossely simplifying the effects. Would you care to explain what you had in mind? |
Like I said with the academics example, you have a function based on the level of technology of a populace, the current education of the populace, and whatever state sponsorship of education there exists to create a demand figure for academics.
Based on various variables for the other archetypes, you get demands for these archetypes.
Then you take the difference between the demand and the current supply of each archetype, and you get a sort of potential difference accelerating the growth of an archetype in one direction or the other (I'm a physics student, not an economist, I have no idea what the correct term for this is ).
Now we get into special cases. Based on some modifier for each archetype representing social mobility (for instance, the modifier for the farmer archetype would be based on the level of serfdom), these "potential differences" are scaled by these factors.
Then, new archetype figures are calculated based on these "scaled potential differences." Based on an assumption of exponential growth with a certain growth rate, the new population is calculated.
And the proportion of each archetype relative to each other stays constant, but is scaled based on the new population figure.
This solution is very simple and very realistic, IMO.
quote:
true, marriages might be bound to class, but I doubt that (temporary) couples were... |
So we would have a certain number of men having multiple sexual partners?
|
|
| |
|
|
|
DeepO
|
|
supporting Candle'Bre
Jan 2002
time: 06:20
|
|
Ramo, what if I call your potential difference happiness, and use the scaled version of it as the cost function of a simple GA? Wouldn't that make the two approaches close to each other what effects are concerned? Plus, you already calculate the happiness of a group in the game, so that would rule out another calculation, and there are other factors included as well (not all of them might be good, I agree).
For instance, when you don't have enough food, peasants will starve, are very unhappy, and will not want to have their childs in the same profession but try to get them into clergy. This did happen irl too... poor but bright boys were send to cloisters (sp?) by their parents, and in certain cases even were given a compensation by the monks. Or girls were 'given' to the nobles as maids, just to make sure they were fed properly. If I understand the societal model well enough, maids of nobles fall in the same nobles class in CB. A count wouldn't marry a peasant, but a maid could...
As to the multiple sexual partners of men: If you'd constrain this to man, a lot of girls would never marry. Plus, I think it was one of the cases where there wasn't a big difference between sexes... but I wouldn't model marriage in this kind of detail, I think that would be too much for the computer, keeping track of all the marriages of individual citizens. If you do it at a higher level, you imply that people marry and remain loyal, or you imply that there were free sexual encounters for all, it doesn't change the overall effect (provided that there are enough citizens around).
DeepO
|
|
| |
|
| |
All times are GMT. The time now is 05:20.
Apolyton Time is 00:20. |
top of page
|
|
|
Forum Rules:
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
|
HTML code is ON
vB code is ON
Smilies are ON
[IMG] code is ON
|
|
|
|
|
|
04-07-2002 00:42
|