Population Model v. 0.1

[amjayee]

Joker and S.Kroeze: The population/age group/tile population system in the game is a hybrid of those two systems: We have regional population, age groups, and we store the amount of people in every tile, assuming that every tile has the same properties as the whole region.

So, what we can do with the diseases killing people: we find out the percentages of people killed in each age group. Then we simply take the tile population, subtract percentage*age group percentage*tile population from each age group, count all those figures together, and finally subtract the total amount of people killed from the tile population.

About nourisment: Isn't the demand figure for food just the _amount_ of food eaten? It does not tell how balanced diet that is. I think we could use the grade of the food to determine that; grade 1 is badly balanced, 3 is normal, 4 and 5 would have some bad effect since they would most propably be some heavy, fatty and non-healthy diets, so they are also badly balanced, but not as deadly as 1 and 2.

Your messages always contain lots of interesting data. I'm sure they are of help when creating a realistic model for population changes. We should make the model ready, and then balance it so that it works as much like reality as possible. The tech level system should perhaps be used for that.

About the starting year:

It hasn't been decided. Personally, I would like to start from somewhere around 6000 BC, when cities were started to be built more commonly. It might be interesting to go as far back as 10000 BC, when agriculture was invented. But that depends on can we make enough interesting things to do during a non-developed era like that. 5000 BC might be quite good year to start.

[The Joker]

S. Kroeze:

Thank you! It is really great to be presented with some actual numbers from reality. This is the only way we can figure out what mortalities should be at what tech levels.

I agree with you that tile population is more important than age group population, and I agree on the food effecting resistance to diseases thing.


Amjayee:

Unfortunately doing that will mean that changes in total population will occur on a province level, and not on a hex level. So it would mean that the pop would increase/decrease by the excact same amount in all hexes in a province, which is unfavorable, since e.g. city hexes (before modern times) constantly should have higher mortalities than rural ones.

So I have been thinking about maybe making the age groups a "pro forma" system, where we just find out how many people are killed in the province, and then have the tech level decide from what age groups they come from. It wont be entirely accurate, but it should work pretty much as good as handling age groups hex by hex, something that is just too powerconsuming.

Nutrition:
No, that is the whole concept! People don't eat tons of food just because they get more wealthy! In stead they eat better food. But in a grand scale strategy game we don't really care whether they eat McDonalds burgers or Italian sandwiches. What we do care about, is what type of food they eat. And the food grading system should show this, nothing more. So maybe we just need 3 grades of food: Grain products, meat products and vegestable products. And especcially between the two first there is a relation. To make 1 kg of meat you need 10 kg of grain. And this is primarily the relation I based the food demand thing on. When people become wealthier they want more meat. When they eat 10 food per turn they don't get any meat. So I thought that later on we could just replace the single food grade with 3, and then have the demand for each type depend on people's incomes.

Starting year:
I think it should be up to the player. Personally I would actually prefer a starting date a bit slower, llike 2000 BC, since the first 2000 years with 1 year turns could become a bit boring. But it should be up to the player.

------------------
Vote Gore. For the sake of people, not god.

GGS Website

[amjayee]

Joker: No, no. I was meaning, that also the "hex population" would be changed. We might decide that hex y is infected by a disease, and it kills people like this. From the tile we get how many people there are; that information we would use to kill the right amount of people from the region population; meanwhile we would keep track of the people killed, and finally subtract the right amount of people from the tile population.

About food grading, I don't think it is that black and white. People don't eat either grain or meat, but both, if they can. I think we should use 5 grades, as with other goods, and it tells the quality of the food.

[The Joker]

Hmm...

Yes, maybe you're right. Again I am too tired right now to think more about it.

I have been working on a way to implement diseases, using Heardie's disease model, in a way so they are ready to be put into a map and a pop model. There I have left out age groups, for now, since it will add a lot of complexity. But age groups could be added later on (I think) on a province level.

Food:
No, what I meant was that people would eat some grain, some meat and some vegestables. And a balanced diet would include the right ratio of all. What I think may be most important here is, that to create 1 unit of meat you need 10 units of grain. So it will take much more food to have a healthy population.

------------------
Vote Gore. For the sake of people, not god.

GGS Website

[amjayee]

About food: Hmm... you might have an idea here. So we would have many food trade goods? Perhaps we should add also some "luxury" food. But wouldn't that be a little copmlicated? I mean, much the same effect would be got simply by grading food. Grade 2 would need more land and more resources than grade 1. And this way all goods would work much the same way. But I don't know.

[Guildmaster]

I actually agree that we don't really need agegroups to calculate death rates.
If we can get away with not having them it would take up a lot less computing power

------------------
How will you make it if you never even try?
-Macy Gray

[The Joker]

I don't think there should be "many" grades of food. Just the 3, and maybe, as you said, a 4th for luxury food. I think this system is more versatile, since all people eat the same food. Wealthy people just eat more of one grade and less of another. This will mean that you could easily rationalize food, if there is an emergency, without having to change the production of virtually all food producing hexes from high to low grades.

Tile pop versus age group pop:
Yeah, your system would work. But that will mean that age groups will be stored independantly in each hex, in stead of at a province level. I don't have anything against this, but I fear it might mean too many calculations (1.000.000 hexes, 10 or more age groups, and several diseases on each hex). If I am wrong about that then we should use your system.

------------------
"It would sharpen you up and make you ready for a bit of the old ultra-violence."
- A Clockwork Orange

GGS Website

[amjayee]

Age groups are needed for also other things than death rates, so I think we should keep them. And about hex population, no; in every hex, we would store _only_ the amount of people. All population statistics would be assumed to be the same as in the mother region. Generally I think we just have to accept that our game will require less computing power and memory; but of course we try to control the situation.

About food, that sounded like good reasoning. If we can make that work, let's use it! But would it be that there are grades, or would the grain, meat and vegetables be different goods? If grades, how would that work? Of course if we use floating point grade value, we could consider that as the "average" grade the people buy, and that would decide the "balancedness". Or what?

[The Joker]

I think that we should propably try to implement at least diseases without age groups first, and then we can add them later on. But if it is possible I think they should be there.

Food:
I think the difference between grades and goods would not really matter. It could be seen as different grades of one good, and it could be seen as different goods.

------------------
"It would sharpen you up and make you ready for a bit of the old ultra-violence."
- A Clockwork Orange

GGS Website

[VetLegion]

Population is my favorite topic

I think age groups are a good thing, they add depth to population.

I have some dilemma about other "population properties", I ll post it as soon as I can think clearly (it is night here)

I think the combination of population and characters/advisors (mentioned in another thread, "how this game will work" I think) has great potential - enough for a separate game. We ll see

[VetLegion]

I think that we should base our population model on social classes or social groups..

(Group in sociology: a collection of people with a set of common goals and rules of behaviour, and has an individual tied to it by feeling of belonging.)Individualism is not needed as a group variable. Group is by default opposed to individualism. If people have high individualism they cannot belong to group.

Now add a few more important group characteristics like job/wealth. There are some groups that need them to be modeled correctly, some need others.

The single most important action of a group is conflict with other groups, with you, revolutions etc.

So the basis of conflict are needed.

which is the most important characteristics of a group?:
location /old times, war times
ethnicity /old times mostly
religion /old times, new times (some countries)
nationality /modern times
job /all times
[ other maybe ]

I may sound like a marxist, but peoples jobs are what qualified them most of the history (in peaceful periods especially). Sometimes it was religion, Sometimes others were most important. Very early, it was ethnicity. So as groups are dynamic maybe it is good their properties have also dynamic importance -- not set in stone. People can switch groups they belong to.

Like this.
From the definition of group it is obvious that they must have a tie that links people in them together. All group charcateristics should have a percentage. In
order to have a group, more then 50% is needed for one property. That is the most important property then. If it falls group falls apart.

inforcement_strength is also a variable that could find its place in there. When it comes to fight, can a group resist? Controll over military best be represented in here rather then have military as a separate group.

How it works in game on Russia example:

1900-1917.
Job/wealth based group "Aristrocracy" has most power. Efforts to better living conditions in the country are ruthlessly crushed by them.

1917.
People are divided by what they do, religious and nationalistic groups are small and not powerful. Revolution comes because job based groups villagers and proleteriat had little wealth and power (hunger etc).

1917-1939.
Education level grows. Industrial workers grow in numbers and villages are becoming towns, villager groups fall in numbers. Job based group of services
grows but is small.

1939-1945.
War time. job division is military, worker, villager, but Nationality (soviet) is the strongest characteristic. Groups and group conflict therefore doesnt exist.

1945-1990.
Across the country in all cities, simmilar groups are formed. Development of industry and services. Groups divide among work again. Class of politicians is emerging.

1990-2000.
The dissipation of middle class. Job becomes secondary in group priority. People group around religion, ethnicity and nationalism again. In cities, the
groups of oligarch and merchants (mafia too) dont care about religion, they are pursuing to perserve their wealth only.

2000.
Which is most important now? It depends on area. In some areas religion (islam) is rising again as workers, ingeneers, politicians see it as the thing to fight for. In the areas where there is peace...like in all history people divide themselves among the wealth line.

So to conclude:
I propose we make a set of dynamic groups. Each time something happenes we check should the action form a group or should an existing group respond. For example if you raise taxes on priests they may try to use their enforcement strength against you... or in a newly occupied region people may gather around nationalism and try to break away, etc...

do you think it can be done like this? There are alternatives...not to do classes, but to calculate percentages of people in a different way... Anyway, shoot. I d like to get closer to a solution, because I could speak all year about possibilities, but some compromises have to be made .

[Guildmaster]

Something has occured to me:

Age groups DO affect birth rates. We can do the pro-forma thing and base it on the tech level to determine death rates, that would be fine. BUT:
What about birth rates? What about social changes?
The easiest example is war, it kills young men depending on what tech level between the ages of 16 and 30 or thereabouts. War doesn't kill too many men older than that and unless you're talking about genocide or infantacide it doesn't kill as many younger than that either. So there's a death factor taken from a very specific age group.

Childhood diseases like polio or chickenpox, which mainly affect young children, you might see a vast number of adults with strong immunities while the deathrate would be raised substantially for a very small part of the age group.

All sorts of things.

I never really said anything about age groups before but this is my beef on it:
If I look at the thing from a practical viewpoint, that is in terms of what we can use in the game, this is what I see:
1) Children that are too young to do anything no matter how oppressive the government is
2) Children that are too young to work or fight but could be forced to do so if the rest of society is desperate enough
3) Technically old enough to work or fight or marry but wouldn't do so in modern society (under 18)
4) Full-fledged adulthood in any age
5) Too old to work high-octane jobs or fight in the trenches, but still plenty of vitality
6) Old enough to retire, but still can work
7) Too old to work

You as a dictator could, for example, force all but 1 and 7 to work in concentration camps if you felt like it, but you would incur great penalties. No player under any circumstances force either 1 or 7 to do anything no matter what, the individual would simply be killed in the process-try to imagine a 92 y/o man working in an anvil factory. If 2 and 6 are forced to work or fight, you will incur severe penalties both social and productive (eg, a 7 y/o or a 60 y/o won't produce as much in a factory or put up as much of a fight as a 22 y/o, doesn't have the strength) no matter what age or tech level. Depending on your tech level and also the tech level of your competitors, you might incur a social penalties for forcing them to work. 3 and 5 have more to do with tech level and social attitudes than anything else, and there are no production penalties for employing these age groups. It can be assumed that in the modern age a vast majority of the work force is age 4 while in an ancient society it will be 3 and 4.

It all, of course, depends on the player and what he wants to do and how desperate he is.

------------------
How will you make it if you never even try?
-Macy Gray

[The Joker]

Guildmaster:

Yes, this was the reason why we should have age groups.

I still see some rather large problems with implementing them without spending loads of computer ressources while doing it, though.

For instance, like you said, most adults should have resistance to a lot of diseases, while kids will not. But since diseases are done at a hex level we will have to have age groups at a hex level, too, or things wont work.


Vet:

Cool chat we had sunday night!

Anyway...

We are going to have social classes. Check the econ thread for information.

And we will also have nationalities and religious groups.

generally I like your ideas. They wont be too dificult to make. Just have certain things determine whether peopel care about work or nationalitity etc.

------------------
"It would sharpen you up and make you ready for a bit of the old ultra-violence."
- A Clockwork Orange

GGS Website

[DarkCloud]

About the immigration model; just take the happiness of the country
to which they are immigrating, then rank it against all other countries
x= country ranked against
y,z,a,b,c,d= other countries

Happy Rates
y- 55%
z- 43%
a- 22%
b- 87%
c- 44%
d- 44%

x- 51%

If it is above x country's rating then they only 1-5% of the population may immigrate to country x.
If it is below x country's rating then there will be fluctuating amounts wishing to immigrate to x.

Then you take the technology and taxes rates. Depending on where country x ranks in those
then you will have (x) number of immigrants. Now you will calculate just as you did for the
"happy rates"

For the age groups:

I believe these will be sufficient:
0-5
6-12
13-19
20-25
26-35
36-41
42-49
50-55
56-60
61-65
66-70
71-76
77-85
86-90
91-100
over 100

[The Joker]

Migration:
Thanks, that is a bit like what we had in mind. I don't think tax rates would matter, however. In stead the real income of the people would matter, and of cause tax rates would effect this.

And the distance would have some effect, too.

Age groups:


Is there any system in these groups? 42-49?

We are propably going to have simple 10 year groups, so 0-10, 10-20, 20-30 etc.

This will be easy to see and work with. And there will be fewer groups than with your system.

------------------
"The future is that mountain."
- Bret Easton Ellis

GGS Website