The Return Of The City

[amjayee]

All here is interesting; I'm sorry I can't answer right now. I have a problem with my home computer, I can't connect to the internet. I'm writing this in my school, but can't stay here longer right now. Tomorrow I try to write more. The same applies to my e-mail; I can read it only with my home computer, so I'll not be able to answer it for a while, a couple of days I think. In situations like this I really see how dangerously dependent some of us start to be of the internet... I'm dealing with most of my businesses on the net, like bank, library and school things etc... I'm severely crippled right now. Perhaps I should start creating a backup plan for more severe situations?

[amjayee]

Still problems with my computer, but I think I should get it fixed tonight. Anyway: I really liked S.Kroezes posts. Thanks again for your efforts - it must have taken long to gatjer all that information and to write it down. I don't have the knowledge to debate the facts you provided, but they sound like they could be near the reality. That package of information will be a great help when we start to balance the game system, which, as Leland said, is unfortunately quite far ahead of us. But at least we can base the plans for the game system on hard facts of real history.

I liked most the things of primitive people and dealing with them. I think those things will be the source of most conflicts and challenges for the players, at least in the early part of the game. It will be interesting to look for more resources when you are in need of them, and to try to subdue the primitive people living in the area. And I'm glad that VetLegion finally agrees that we need them. Usually the would just be people living on the tiles, but I think later it might be quite easy to create simple chiefdoms and things like that.

About randomness in the game, I think it would be good. But of course we would make it an option; that would be quite easy to do. But I think if we balance the system properly, it would become a real source of suspense and challenge for the player, without reducing the enjoyment of the game too much if it strikes you unguarded.

Also about the rise and fall of civilizations, I think that should be quite an essential part of the game. Would it be stupid if you had to start over with some other civ when your earlier one is conqured? Also I think the game objectives should be something else than conquering the whole world; that is quite an impossible task. US has quite strong military, but they could not even hope to conquer the world with it; I believe they would have huge problems conquering even single more developed nation, like Britain or China (nukes would of course be disallowed). Same thing for every earlier empire. If we include even a little realism, victory by conquer would be an impossibility. Perhaps only if the majority of the world would be allied against the rest, it would succeed.

The main objective of the game should be to guide your people, build your empire, fight your neighbors and to try and do your best. The success would be measured with a point system. When your empire eventually falls, you can do something else in the meantime, and perhaps the old empire rises again later with you back in charge.

[VetLegion]

Yes, most of this is purely philosophical

[S. Kroeze]

Amjayee:

I agree that civ2 system was very playable, working and fun. We just need to refine it a little. So, let's say that borders are not drawn by the player; that wouldn't work very well, and would be lots of work. Instead, they are decided by computer based on area control, transportation and communication etc. Usually the first player to get hold of some tile will have the right to keep it, unless other player conquers it. Units don't conquer terrain automatically, player has to ask them to do it. This might give us a way to allow the player to affect a little the region borders; he could conquer single tiles of terrain if he wants, but the tiles must be "legal" new region tiles; not too far away, and so forth.

Emigration is needed, as Elmo pointed out; otherwise, we end up with completely stationary and isolated regions. As I said, we would have two kinds of immigration; from countryside to cities and between regions.

Joker:
For the beginning of the game issue I believe that the player should start with just one hex. His region, his capital. At least that should be so if the player starts at the beginning of the tech tree.

I think that in the beginning the size of regions should be very, very small. So if the player is just a bit expansionistic he will need to split his civ into more regions pretty soon.

I think regions would be conquored hex by hex. When the capital is conquored the region should go into "riot" mode or something, which will mean that the rest of the region would be easily conquored. Then the army could be ordered to use one turn to conquor all the hexes within it's action range. This would hurt the army a bit, but this amount would be severely reduced if the region is in riot mode. This would make conquest realistic, and still use Amjayee's brilliant unit system to make it fun as well.


In my opinion this thread contains many excellent ideas on Regions, Cities, Expansion and Migration. I hope Leland will use them for the new Design Document. It seems my extensive citations killed the discussion which was definitely not my intention! I am sorry about it, but only wanted to help or convince people.

Especially Migration is extremely important and barely covered at all in the Population thread. In my opinion migration by hex does also make sense, at least in prehistoric times when people didn't own a land rover yet, nor Elephants or Camels!

And: Thank you again, Vetlegion, for accepting the hunter-gatherers!

[Darkstar]

Wow... S. Kroeze, LOTS and LOTS of info!

re: steppes raiders
Unorganized? Pardon? Just make them AI controlled. You don't need much computing to handle them. They are either happy or they aren't... Happy and satisfied, don't raid. Unhappy? Raid. Overpopulated (not enough foord)? Raid. That by itself will help keep pop down.

re: What we know
If you were Aztec, you'd DIE. They were dead as soon as they came in contact with the Spanish. Germs. It just took a while to run through them. The Spanish didn't kill enough by their own hands to even measure, from the stats I've heard quoted on the history shows about it.

Anyways, I KNOW you can model anything in software. It's just a question of working out the algorithms.

It seems to me, that in the beginning, an expansionist player is going to have LOTS of regions. Loads and loads. As many as they can cram in. And keep cramming. Why? Because that's how they get more production, more research, and more credits. Period. Each and EVERY player is going to do it. ICS. You still have it. Players like to stack the deck in their favor. If I, as a player, know I can send in an army to that border hex cluster, that is beyond my current administrative control, and then order a outpost built... then order a fortified administrative center... I'm going to do it. Period. Why? Because that spreads my empire. In TIME, the local populous will come around to my people's view... although my people there will also come around a bit to theirs. But my civilization is growing. I'm gaining land and all it's resources... and elbow room for my people.

City versus region... it seems to me, that your regional node center is nothing more then your Building of Power. For a military power, that's your forts. For a religious based power, that's your temples. For a corporate based power, that's your new branch. So... game start, you get ONE Control building. As time passes, you instruct more to be built. Maybe there on the coast, to better control/administrate the port.

Since tech is a major player, the further away a tile you control is from the nearest Control building, the more it COST to admin. And the less effective your orders. A drag... when the fastest for of travel is walking, your no cost range is small. As the cost of admining too large an area over your tech climbs, you loose valuable resources (credits and pop) in handling it. I would think Control buildings would attract workers, so you generate migrational attractors in your pop model.

Now, the reason I was mentioning those are... control facilities, not cities, and not other things, are you main instrument to control a region. MOST of the time, you are better off placing in in a population center. But there will be times when you want to reduce costs or gain better utilization (reduce wastage/local corruption) by adding a new building. If each Control Building has a basic rating, the computer will be able to easily decide that the bigger nodes are the branches that the little low ranked leave CBs flow into.

When a player orders a different arrangement, you incur a COST of moving the people admins around, as well as the cost that would be for expanding the local facilities. More people, more economic opportunity is now at that node... larger attractor. This would balance out the loss of those people elsewhere.

The main problem I have with all this is... nomads. What is their control building/node? Is it a person (Leader) ? Or just a mobile "facility". Figure that out, and you have the key to handling nomad civs just like any other.

Just some odd thoughts from the new village idiot...

[VetLegion]

If you were Aztec, you'd DIE. They were dead as soon as they came in contact with the Spanish. Germs. It just took a while to run through them. The Spanish didn't kill enough by their own hands to even measure, from the stats I've heard quoted on the history shows about it.

no, I would not die there is a good chance that I would have landed in Spain earlier then 1500.
But in the situation where they come first, I would not let them land, or if they landed I would quaranteen the area. Many cities in the period (harbours especially) had quaranteen areas, all over europe. Aztec cities would have them too, if I was there.

the argument that we cant reproduce behaviour of the historical leaders (and many events that are result of it) stands. That behaviour was always lacking knowledge of future (obviously), and often it was irrational, selfish, and non competitive.

Anyways, I KNOW you can model anything in software. It's just a question of working out the algorithms.

I know. We are making a game where you can force different outcomes on history, and history is the outcome of your actions. Player can not be defeated by the Spanish if he is Aztec, like it is a history replay. Instead they both lead their countries in directions they choose... and they will both have knowledge of tools at their disposal during the whole game-time.
One of the results of introducing Germs in game will be Germ Warfare between players: breeding, spreading or even trading germs and germ defenses.

It seems to me, that in the beginning, an expansionist player is going to have LOTS of regions. Loads and loads. As many as they can cram in. And keep cramming. Why? Because that's how they get more production, more research, and more credits. Period. Each and EVERY player is going to do it. ICS.

This is the underlaying philosophy:
Players who expand will benefit from it. Players who expand more will benefit more. But process of expanding will require greater skill and cunning and be more difficult then in up to date civ games. Civ depends on two mechanisms (unhapiness and corruption) to supress ICS. You beat the first with Hanging Gardens and the second you dont care about , since capitol city (where your science is) has 0 corruption.

In GGS, both unhappiness and corruption will be harder to suppress. Also, initial founding of a city will have a cost, as will its maintenance. And 2 cities of size 1 will never outproduce a city of size two. Actually, you answered yourself:

As the cost of admining too large an area over your tech climbs, you loose valuable resources (credits and pop) in handling it. I would think Control buildings would attract workers, so you generate migrational attractors in your pop model.

this is among the lines I am thinking (and many others I think)

... control facilities, not cities, and not other things, are you main instrument to control a region.

and roads, but I like the term, so roads can belong to set of control facilities. I think amjayee (?) previousely used the term Region Upgrades or simmilar.

MOST of the time, you are better off placing in in a population center. But there will be times when you want to reduce costs or gain better utilization (reduce wastage/local corruption) by adding a new building. If each Control Building has a basic rating, the computer will be able to easily decide that the bigger nodes are the branches that the little low ranked leave CBs flow into.

Sounds very reasonable.

When a player orders a different arrangement, you incur a COST of moving the people admins around, as well as the cost that would be for expanding the local facilities. More people, more economic opportunity is now at that node... larger attractor. This would balance out the loss of those people elsewhere.

yes, the cost will be a major inhibitor to some actions: region founding, changing of borders, switching centers...

Regions (city that is administrative center and it surroundings) are going to be expensive to found and maintain, and will have the optimum area depending on technology (communication technology), infrastructure (roads) and population density.

That means that if you have and island that supports 8 distinct regions you better have the exact number. More than optimum and you face too much bureaucracy (costs you money) and any less you face ineficiency (money not collected, corruption, etc)
The "optimum number" will not be shown to you on a menu and will not exist in game.
Instead you are challenged to manipulate your adm. centers, their borders and roads to achieve greatest efficiency per region. If one region barely collects enough taxes to support its own administration & infrastructure, perhaps it is time to merge it with a bigger one? Or to build more roads so that countryside may reach the center and you may tax the distant peasantry?

What does the player do?
The number of administrative centers is not limited, but is costly as each of them needs basic upkeep. Only taxes after this upkeep is substracted are elivered to state treasury.
Size of a region is also not limited, but it depends heavily on infrastructure (roads). You may assign a large area but only have roads that reach one third of it. In that case, your tax collecting efficiency will be mere 33%. Waste!
Frequency of change of borders or centers is also your decision, but it will have a cost which will inhibit too often changes.

That means that regions may change over time. Your original island maybe was very flat and was divided to 8 regions in begining because of infrastructure criteria, people cant travel very far to market their goods or pay taxes.

In late game, with fast railroads and highways, the inf. criteria may get weaker as people and government can travel faster and further, making bigger radiouses possible. So the importance may shift to population density criteria: where you dont want too much people on one administrative center, or you want to have more then one big city.

So you may end up with 3 regions from original 8, or even only one.
You will need to take into consideration that administrative centers are attracting population (trade and economy are population properties) more then ordinary city does. The region capitol will therefore be by far the biggest and most developed city in it.

You should want to have more then one region to support city growth. Assigning administration centers in GGS is almost equivalent to building cities in Civ.

I think that it can be quite a challenge to administrate your civ, without too much complexity.

[S. Kroeze]

In my view this was a rather important thread.
That's the reason I bring it to your attention again.

[Jon Shafer]

Being that all this is related to the map, our first order of business right now (*ahem*), I think I'll respond. Well, from what I could tell from the meeting yesterday, not much in the way of deciding things was accomplished yet... either that or all progress made was erased, so I'll start over (that's waaay too many words to read!). I'll cut down on the number of words so as not to continue the previous pattern of this thread.

My Plan:

Regions. Yes, we all know what these are. User-definable borders are a must. The easiest way to handle regions is to have "City Populations" and "Rural Population". Rural population is the population of the capital city of the region + all the other people who live in that region but not in cities.

Cities. This issue must be dealt with carefully. Obviously, there should be a capital city for each region. I feel the rural population should be lumped in with the capital city's population for manageability (as I've stated previously). I feel that the leader may invest in the "urbanization" of a region, and when that value reaches 100%, then a new city is formed out of the rural population of the region. Maybe this isn't a absolute accurate simulation of how things happened, but I believe it's the best way to manage things

Why Cities? Cities were havens for culture, commerce, and business. Cities will generate more income, and will also produce more goods faster (easier to get it out to people from a city, etc.), than the rural population does. The actual merits of the city can be argued about later, but that's my twist on things.

This ties into my map model... I should have it done for the meeting next week. Find as many people as you can to come!

[S. Kroeze]

Originally posted by Trip
Being that all this is related to the map, our first order of business right now (*ahem*), I think I'll respond. Well, from what I could tell from the meeting yesterday, not much in the way of deciding things was accomplished yet... either that or all progress made was erased, so I'll start over (that's waaay too many words to read!). I'll cut down on the number of words so as not to continue the previous pattern of this thread.

My Plan:

Regions. Yes, we all know what these are. User-definable borders are a must. The easiest way to handle regions is to have "City Populations" and "Rural Population". Rural population is the population of the capital city of the region + all the other people who live in that region but not in cities.

Cities. This issue must be dealt with carefully. Obviously, there should be a capital city for each region. I feel the rural population should be lumped in with the capital city's population for manageability (as I've stated previously). I feel that the leader may invest in the "urbanization" of a region, and when that value reaches 100%, then a new city is formed out of the rural population of the region. Maybe this isn't a absolute accurate simulation of how things happened, but I believe it's the best way to manage things

Why Cities? Cities were havens for culture, commerce, and business. Cities will generate more income, and will also produce more goods faster (easier to get it out to people from a city, etc.), than the rural population does. The actual merits of the city can be argued about later, but that's my twist on things.

This ties into my map model... I should have it done for the meeting next week. Find as many people as you can to come!

I hope you are not offended but I disagree almost completely!
I am sorry to say.

Actually you are re-introducing the CIV-city concept most of us have learnt to hate thoroughly.

Yesterday we had a discussion whether we should have 'fixed provinces' or not. For the time being, I would prefer fixed provinces -because this would make all things much easier.
I can live with the possibility of having influence on borders of provinces, BUT it was never our intention to make provinces completely user-definable. It is so irrealistic! No one would be able to create a well-centralized political structure when different parts are seperated by mountain chains.

Then you suggest every province should have a capital. When you would say 'administrative centre' I could agree, but I strongly object to the word 'capital city'.
Imagine a very sparsely populated province in the Sahara desert. Even a village with some 1,000 inhabitants wouldn't survive here, let alone a CITY!

Then you suggest that the population of capital and countryside should be combined. This really defies all logic!
The main reason we separated rural and urban population is because of differences in demographical growth. While rural populations tend to grow, urban populations tend to shrink. Without constant migration from countryside to the cities, all cities would cease to exist -before the creation of waterworks. Hygienic conditions in towns and cities were dismal.

I also have doubts -though less strong- about your 'urbanization' idea. It was always the idea the people would settle mainly according to their own wishes. Of course when a hex/tile would be rather densely populated a town should arise. But the influence of a government is rather limited. There are many examples of city foundations that were a failure.
Alexander the Great founded some thirty Alexandrias. Only a few were a succes!

Then I also strongly disagree about the greater productivity of cities. I could even argue the opposite -though I will not do so.

Most cities were bottomless pits: the populace consumed food, the elite spent their money on luxury items: conspicuous consumption.

Please do not be angry!

Sincerely,

S.Kroeze

[Jon Shafer]

Originally posted by S. Kroeze
I hope you are not offended but I disagree almost completely!
I am sorry to say.

As long as you only attack my ideas and not me, then I won't ever be offended (unless you say something like THAT IDEA WAS STUPID IT MAKES NO SENSE!!!! Well, unless it is stupid and doesn't make any sense ).

Actually you are re-introducing the CIV-city concept most of us have learnt to hate thoroughly.

Not exactly. I feel that cities should hold some importance. Not Civ-like importance though.

Yesterday we had a discussion whether we should have 'fixed provinces' or not. For the time being, I would prefer fixed provinces -because this would make all things much easier.
I can live with the possibility of having influence on borders of provinces, BUT it was never our intention to make provinces completely user-definable. It is so irrealistic! No one would be able to create a well-centralized political structure when different parts are seperated by mountain chains.

So just put restrictions on what things borders can cross. Maybe early on your borders must go along rivers and mountain chains. Later with the advent of bridges, the river restriction no longer exists. And once dynamite comes along, then the mountain restriction can go. I just feel that one of the things a leader has the most power over in his country is borders and the organization of provinces.

Then you suggest every province should have a capital. When you would say 'administrative centre' I could agree, but I strongly object to the word 'capital city'.
Imagine a very sparsely populated province in the Sahara desert. Even a village with some 1,000 inhabitants wouldn't survive here, let alone a CITY!

Okay, I didn't really mean a "capital city" per se, but I was advocating a central location for the governing of that province. If possible however, I feel that a city should pop up where the governing location stems from eventually.

Then you suggest that the population of capital and countryside should be combined. This really defies all logic!
The main reason we separated rural and urban population is because of differences in demographical growth. While rural populations tend to grow, urban populations tend to shrink. Without constant migration from countryside to the cities, all cities would cease to exist -before the creation of waterworks. Hygienic conditions in towns and cities were dismal.

Again, when I wanted the rural population lumped in with the capital city, it was basically you were controlling the province ITSELF, and less the actual things within the city and the rural areas both.

I also have doubts -though less strong- about your 'urbanization' idea. It was always the idea the people would settle mainly according to their own wishes. Of course when a hex/tile would be rather densely populated a town should arise. But the influence of a government is rather limited. There are many examples of city foundations that were a failure.
Alexander the Great founded some thirty Alexandrias. Only a few were a succes!

Maybe not the most accurate and historically accurate way to do things, but how else do you suggest cities spring up? Having a function to determine the exact population and urbanization of each tile completely within the computer would drain the CPU power like nothing else. I'm open to ideas here, I just don't have any better ways to impliment the creation of cities as of yet.

Then I also strongly disagree about the greater productivity of cities. I could even argue the opposite -though I will not do so.

Well, as I stated in the meeting yesterday, there has to be some benefit of having cities. Otherwise why would they exist throughout time? If a better way to use cities and give advantages for their existance more than defensible locations is created, then I'll be just fine with it.

Most cities were bottomless pits: the populace consumed food, the elite spent their money on luxury items: conspicuous consumption.

Again, I feel there has to be some benefit to building/running cities.

Please do not be angry!

As long as you can argue sensibly about legitmate issues, then I won't get angry. Thanks for more of your input.

[Leland]

I suppose I could share my personal gut feeling on this issue: trying to get populations, provinces and cities fit together somehow is not as easy as it may seem. Well, at least not the way we've thought it should be done. There are certain important tradeoffs to be made, and trying to find a sensible compromise between them drove me to a nervous breakdown and gave me lifelong trauma:

1. Complexity vs. realism:
   The best solution would be to hold all the population information on each tile, so that in effect there would be a "city" (as in, a center of production) everywhere there is people. However, like Trip pointed out, this is insanity until we all have supercomputers at our disposal. So we need to find some way to lump popolation together. In Civ, this is done by associating all population to cities. In fixed-province games all the population is associated to provinces. And what I think Trip suggested was that we have a hybrid of the two and store them both individually. All are nice solutions, purely from population model's point of view.
   
2. Autonomous vs. Player-controlled population:
   In Civ, due to the population being associated to cities, it is essentially fully controlled by the player. This is something that we have been trying to get away from, and to give populations a reasonable autonomy, so that player's role is more akin to a herdsman who tries to manipulate people than a God who just snaps his fingers.
   
3. Autonomous vs. Player-controlled provinces:
   Now, here's where the problems arise. We want political constructs like provinces to be under player control (with certain limitations like geography, of course), so that player can adjust or negotiate borders as he chooses, and possibly divide and join his own provinces. The rationale why we need this is to allow the gameplay to scale easily from pathetic stone-age tribes to global superpowers without an exponential increase in micromanagement. Compared to Civ, where every city lasts more or less forever in one place once founded, dynamic provinces represent a fundamentally different kind of philosophy.

Okay,so what exactly is the problem with these issues?

It all boils down to how we are solving issue number 1, i.e. where do we associate the population data. There are the three alternatives mentioned above (provinces, cities, both) as well as the option to make populations totally independent from political structures (which has been our approach in the past). What are the consequences of each design choice?

Population associated with provinces:
This was what someone initially suggested ages ago when trying to figure out the first population models. However, reconciling population autonomy and player freedom turned out to be a little bit tricky. For instance, if you have two provinces with their respective population data and you want to join them into one, then there is information loss when, for instance, life span or education level gets averaged between the two old populations. This is highly undesirable because splitting provinces is a political action, and should have no concerete bearing on the properties of the actual population. This can be solved either by taking away the player's ability to mess with provinces (i.e. make all provinces fixed like S. Kroeze seems to advocate), or by coming up with some sort of primitive provinces which the player can join together into larger entities. In the latter case, there is still the question of population autonomy and dynamic change... for instance, if a lot of french people migrate over the border to Germany, will the border eventually change to reflect the increased French dominance in the area?

Population associated with cities:
This would work much like in Civ, except that cities need not be directly under player control. Cities have fixed locations, but possibly variable radius so that they can grow and shrink (and preferably vanish altogether), but not move from one place to another.

Population associated to both cities and provinces:
I am not sure how Trip's idea is supposed to work exactly, but my guess is that there is a separate city population and province population, and they work by different rules.

Population is totally independent:
Which means that there should be some sort of moderately complex simulation of population dynamics that has nothing to do with provinces or cities. And make no mistake, it would be complex. Populations tend to migrate from one place to another, split up and assimilate smaller populations, among other things. Once all this is figured out, we'd still need to come up with different mechanics for provinces and cities, which are still based on the underlying population model. This is, in my humble opinion, the most fascinating route but unfortunately at least my own brain power ran short when trying to come up with a solid population model.

* * *

Hmm. I'm not sure if I was able to illustrate the problem... it could be that I am just too dumb to figure out a working model. But this could be because I was looking at the problem from a wrong angle all the time... perhaps we need to toss away the ideal of autonomous population altogether. I am looking forward to reading Trip's suggestion for the map/population model.