Regions, Population and Improvements - model ver. 0.3

[VetLegion]

Yes, that is problem. I would like to avoid putting religion, work or ethnicity on every tile.

Having only one type of tiles (very small ones) begs for a larger structure to hold data

[amjayee]

Lately there has been a lot of debate in the region and poll threads about world population and region issues. I gathered here some ideas that would be good in my opinion. I tried to make some good compromises between realism and gameplay. Judge for yourselves:

Human players' people would be the only people with organized government. So, all the civs in the game would be human-controlled.

Most tiles would have a population. These would represent the hunter-gatherer tribes, nomadic tribes and early farmer tribes. They would usually be small groups of people, each tile containing several of them, and they would have a tribal organization where every village has its own leader. So, they would show to the player only as a number of people in the tile, and would act only as minor resistance when the player tries to conquer the tile. The people of the tile don't have a "national affiliation" (there is no nation they belong to) so after a short period of time after the conquering of the people, the people will be counted to the population of the player's nationality.

This is not entirely realistic, but it would be simplest. But; I think we could create a system, in which the original inhabitants of the land, previously scattered, could form a simple "semi-civ" to oppose the human player. In GGS The Book, the organization of people was divided to four types: bands, tribes, chiefdoms, and states. In the beginning of the game, all people would be organized in tribes. Human player's people would form a state. (Or perhaps we could start with a chiefdom, and evolve it to state when the player has gained enough land.) Facing a threat from external force, the tribes in an area could form a chiefdom; they would have an area of land, and they could organize military force against the human, but nothing fancy. Historical examples of this are the Native American resistance, and the Celtic Gallian resistance against Romans.

I will now explain a little the organizational terms. Tribes would be very simple and inorganized. They would be controlled by a strong man, a local chief or warlord. There could be three kinds of tribes: hunter-gatherer, nomadic, and farmer. The type would have only some economical importance, and would mainly decide how much food the people can extract from the land, and thus how many people the land can support. Farmer tribes would of course have some simple farm improvements on the land. We don't need to have the nomads moving visibly on the map; that, too, would be mainly an economic term. The nomads would have animals with them, and would live mainly on lands suitable for herding. Of course to model mongol invasions and such, we could have a system where the tribal people can form chiefdoms also without player affecting; several tribes would join forces, and the chiefdom could organize some military. The effects could be similar to Great Migrations; mongols forced other people in front of them to move westwards etc.

So, the next thing would be mobile populations. These would be like armies; they would have settler units and simple military units. Perhaps we need a system to store the technological level of the people in some area? I mean, many inorganized tribes did have the knowledge of bronze and iron working. So, the technological level would decide how strong military units the mobile tribe has. Of course they would be no match for organized army, but could cause harm if let unguarded. So, basically the mobile tribe would be like an army, moving on the map like the units do. The player could also turn his people into this kind of unit. This is quite historical; almost always the people living in some area today have not been the original inhabitants of the area. Greece was inhabited by Asian people invading the country from North some millennias BC. Also the current people of Finland have come from elsewhere and conquered the land from the original inhabitants thousands of years ago. So, the player would have some land for him, and he could live on it for some time, meanwhile exploring the surrounding area. If better land is available, he can "pack" his people and invade the area. I don't think this would need any specific technology. Usually the people could move to the new area within one or two turns. And yes, this should be profitable, and also quite easy for an organized tribe of the human player.

The basic idea in dividing the world into regions in the beginning of the game is, that this way we could store the statistics of the people living in that area. But it's true that this is a little overcomplicating issues. Perhaps we should have, that only those areas with government would be regions. But do we need some system to store the statistics of the population living outside the regions? At least for nationality we don't necessarily need it, as I said. Also for religion, the tribes would propably have each their own animistic or polytheistic religion. Also classes don't exist, so they are not needed. The only thing that comes to my mind is the technological level. What kind of system we could use for that? Or is it needed? I think for realism, it would be good. Perhaps we could just have an area of land similar to region, but we would just store some population statistics for that area? Ethnicity perhaps (though handling that might be tricky - how the ethnologies spread etc.?), tech level and some other things perhaps... should be quite simple.

About player being a nomad civ, I think it would be good. Also, I think it would be profitable in certain conditions. Nomadic civ with government would be no match for the primitive tribes in the tiles, and also it would have such a good military it could hold back against the agriculture-based civs, until it can find a better location to settle down.

These are my ideas. Any refinements to them? At least I think, that we need the tile population and the simple chiefdoms for realism. Also they should not be too difficult to make.

[VetLegion]

quote: So, they would show to the player only as a number of people in the tile

So you think we should not have tile population have any properties? That is a possibility.

[amjayee]

Vet: I think we don't need the pop properties in tiles very much, and it would be hard to keep track of them. That would simplify things. When joined to an existing population, the people would just gain the sate properties as the new population they belong to. But of course some kind of tech level would be good. Does anyone think of anything else we would need for tile populations?

[VetLegion]

That seems ok, but

It would work for conquering "disorganised" tiles, but how about enemy tribe (civ) tiles? Then, they become just your people, no unrest.

So, if we allow tile by tile conquering, it becomes necesary to have more data on a tile.

Also, questions of splitting the region, merging regions and expanding a region are still unanswered.

[The Joker]

Hmm. This place has been awfully quiet the past weeks time. Sad.

But anyway. I think Amjayee makes more sence than all the rest of us (as usual ) in what he is saying. Disorganized hexes would only need to store the amount of people on them.

However, this should not mean that by conquoring them they would become of your nationality just like that. Even though we don't have to keep track of nationality in their case, they would still "get" a unique nationality when conquored by a civ. This would mean that you could not just conquor vast areas of land in very short time and hope to keep it. And when conquoring other civ's hexes these people too would keep their nationality and cause unrest and problems for vast amounts of time.

I am aware that the region definition thing is still an unanswered question. But I think that it is one that we can overcome, if we really work on it.

Although I have not have much time to check my sourceforge emails lately, I have seen the discussion of whether squares or hexes should be used. And in that discussion I vote for hexes. I think that they give way more sence than squares, removes the unfair diagonal unit movement and generally just looks better for terrain. Their only drawbacks are the dificulty of calculating their area (a 50 km hex has an area of some 600 sq km, or so I recall) and the potential dificulty of programming them. But the first is really not a big problem to me, and if you can get over the second I think we should stick to hexes.

One last thing: since I am stuck on these iMacs I have not had a chance to see the new UI in action. So could someone please give me a brief describtion of it? What does it do? Is it cool? Tell me, tell me!

Now if you will excuse me I am going to Alpe D'Huez (don't know about the spelling, though - those damn French) on a skiing trip the day after tomorrow. YEEEHAAA!

------------------
"If you are to hurt someone you better do it so good that you don't have to fear revenge."
- Machiavelli

GGS Website

[amjayee]

Vet: that is true. But of course we could make it so that if you conquer a tile belonging to an organized tribe (civ) they would keep their national status from the old region... Though if it's a multi-national region, how can we know which of those nationalities is occupying that particular hex? Hmmm... do we need to know? I don't know if need to know or not.

Then also Joker makes sense in his message. Damn it, I know it would be good if we could store at least national info for each tile... IT's just that it would be so hard to control, and it would take so much memory... We need to decide what level if excactness and realism we want.

I have a possible solution to this issue, but it would change the whole idea quite radically. I remember Leland was referring to something similar a while ago. What if we separate nationalities, religions and such completely from the region system? We would have _nationalities_, (ethnical nationalities, not the civ nationality )that have an area assigned to them. Quite like the region's area. For each nationality we could store some info. The areas of different nationalities could overlap. Then, we would have a regional population which collects info about the people living within its area and stores it within itself. Class info and such would be stored only at the regional level (for simplicity). This would of course complicate things, and might bring some new problems, but this is one possibility.

This is a very problematic thing. Nationality and many other population properties are quite essential in the game. We need some realism in them. But we must not make the system very complicated, or it will be difficult to handle and thus would not serve its purpose. Also using too much memory and computing power should be avoided.

Then one possibility would be to make it so, that only one nationality can "own" the tile. We would store the "owner" of the tile either as an area of tile belonging to the same nationality, or directly in each tile. The latter system would take more memory. The "owner" of tile would be the dominant nationality in it, propably until that nationality ´has vanished from that region by the acts of the conqueror (in the case the tile is a conquered tile). There could live people also of other nationalities, but they would not be counted. Just the amount of people would be stored for each tile, and the amounts of different nationalities in the regional population. We would just not know excactly where some of those nationalities are situated... this is another possibility.

I hope we can get some sense in this. I will think of this some more.

About tile shapes; hex and square are quite equal. Hex is a little more difficult to program, and it would look a little more confusing because it is a more complex shape than a square; squares have less realistic movement, but that is not an issue since we are planning to not have tile-by-tile movement. Overall, squares are now my favored tile shape; they are neat-looking, as easy as possible to program, and the most convenient. So I think they are quite appropriate for us. But if everyone else insists, hexes can be used too. But I would prefer squares.

And have nice there in France!


[This message has been edited by amjayee (edited February 01, 2001).]

[amjayee]

Yesterday in our meeting we got some new ideas for the region system. I will summarize a little those ideas and possibly refine on them. I promised to send this earlier; sorry that I couldn't do it until now. But here goes.

Region system would be re-defined. We would now have regions, that are solely governmental elements, and populations, that contain people data. Each government region could have a number of populations, each having their own area; each tile can belong to only one population, so they cannot overlap. Region borders (political borders) would be freely definable, but population borders the player could not change. When a population is divided by a political border, the population would be split into two of course, so each population can belong to only one region. But of course the split populations would remain quite similar for a long time, so within certain time the populations could re-unite if the get a chance - i.e. if they belong to the same region again. Even after being split, the national affiliations of the populations would remain, so that might cause problems for the conquering player.

This system would solve all the problems dealing with players changing region borders; since populations remain intact, no actual benefit can be received from changing randomly the political borders. Also this way we can create all the population features we have wished for. With this system, the world would be inhabited from the beginning, and it would be covered by populations; so we would always know all the stats for all people in the world. If the populations are not governed, they would show simply as "Celtic tribes" or something like that. Regions on the other hand are created only by the players, and they contain some populations in them, and govern them. I was earlier concerned with handling production, but that would be no problem actually, if we split the population along the political borders. We would only need a system for handling the population spreading and other things like that, but they should not be too tough.

With this system we can make all kinds of cool features. Including:

-hunter-gatherer, nomad and farmer tribes outside player's regions and as extra challenge in conquering

-possible uprisings of primitive tribes; each nationality would have a region. In a dangerous situation (threatened by the player) or for other reasons, like lack of food, they could unite under one strong ruler for a short while. This kind of group could organize military and better defense. This would be similar to the barbarian unrest in civ2, but with some more realism.

-spreading of diseases would be more realistic, and using germs as weapon would be easier.

-etc... I will think this more when creating the model.

I'm quite assured that this system would be good. I have made a just simple summary of it here, but I will refine on it for the model. If we now can agree on this, I could have the model ready soon! Please comment on this and make new suggestions and ideas, so we can finally decide on this system; it is after all one of the most basic things, and will affect most other models.

[Leland]

I think this solution is quite elegant and appropriate. VetLegion asked me about the city node proposal I made three weeks ago, but as amjayee pointed out back then, it was a little bit too complicated. However, my previous idea and this population model aren't mutually exculsive: I believe that one of the benefits of the new model is that it can be refined later on, for instance, by giving the city network a more prominent role. But for now, just having regions and populations is sufficient, since those are the elements that the player interacts with.

I'm glad that his particular problem seems to be solved, but it would be too optimistic to think that everything is settled. Among open items are

- how to handle migration
- what rules apply to the formation and/or evolution of populations and the land area they occupy
- how to balance the number of populations so that they neither become unmanageably numerous or assimilate into uniform masses
- the rules applying to population properties
- the role of cities and urban populations in the new model
- 1001 other issues I haven't thought of.

I'm a little bit concerned about the initial state of the earth: if we would be realistic, each tile would have its own population and that is unacceptable. So we'll have to group primitive populations (those that are not part of any civs) into large "population pools" from which the players conquer new territories/subjects. The military action by nomadic tribes can be modelled as a combination of spontaneously arising military units (strong leader) and migration within the area where the tribe originates: the unit serves as a magnet which attracts people to migrate. Since migration is assumed to be relatively free within populations there is no risk of having a military unit without a population, unless it decides to attack another population such as your friendly neighbourhood civ.

From the gameplay point of view the regions would be far more visible than the populations. However, it should be possible to have a population map which could be used to determine appropriate region borders. Maybe this could be done automatically in order to spare the player from tediously drawing lines across continents. But of course this should be optional, and the player could draw the borders himself and split the populations as he desires. I'm not quite sure how visible foreign populations should be though... maybe it depends on the tech level, maybe there could be different stages: no information, rough statistics of overall population, detailed statistics of overall population and rough statistics on regional populations, detailed statistics of national and regional populations and rough properties of individual populations, population borders well known, and complete information. Or something similar.

As for terminology, I believe that we need to make a distinction between statistical properties and concrete properties. The word "statistical" suggest a set of properties that are derived from a lower level: this regional and national populations would be statistical because their properties are determined by the concrete populations. In order to avoid confusion, I'll define the terms as follows (based on amjayees definitions):

Population: a group of people living in a specified area. Populations are disjoint: no two populations may occupy the same tile. Population classifies it's people with different properties to model the behavior of the people in the game situations.
Region: the basic unit of government; region consists of one or more populations. It and its people are governed by a single authority, which is in turn responsible for a higher level authority. Region is a supergroup; depending on the status of the region within the empire and the way it's governed, the regions can be states, provinces, colonies, protectorates etc.
Regional population: statistics describing the properties of all populations in a given region.
National population: statistics decribing the properties of all populations within a civilization.
Tile population: number of people living in a tile.

(It should also be noted that the word "population" can refer to either a group of people or the number of the people.)

Now for some comments about your initial proposal.

Tile population: Why is it necessary to know exact number of people occupying a tile? The land area of a 50km across tile is about 2000 square km, surely it's unimportant whether there are 4678 or 4679 people living in it? In another thread S. Kroeze pointed out various population densities of nomadic peoples; if I recall correctly, one person per a couple of square miles was common, though there were exceptions. However, for the purposes of this game I believe that knowing the number of people in each tile with the accuracy of, say, 200-250 people is sufficient. This way two bytes can express 13 to 16 million people. Hmm... maybe it's not enough. I'm not sure about the population densities of modern cities, but aren't they in the same ballpark?

Another solution would be to use different multiplyers for urban and rural populations, maybe by using one bit of the number for this purpose. This double standard would make migration a little bit tricky, so I'm not so sure if it should be used.

The advantage of having a fixed number of bytes instead of varying number is clear to all programmers: referencing an array is a lot easier if it's uncompressed. Naturally, we'll be using various compression schemes for network and save data, but those are programming and not design issues. I apologize if I have bothered you designers with such trivialities.

Improvements:

I think improvements should be clear distinguished into two groups: tile improvements (those that have a clear position on the map, for mines, harbors and very important monuments) and region improvements (such as public schools). There is, however, a problem with roads and canals: should they be tile improvements (two connected points on the map, or a path of adjacent tiles) or regional ones (a general "road infrastructure")?

Also, the problems formerly with populations and regions are now present as regards to regions and improvements. If a region has a water supply improvement, you chop off a piece of the region and connect it with a neighbouring region, does the neighbour gain water supply as well? Tile-based improvements don't have this problem, obviously, but it is clear that regional improvements are needed.

Another thing I'm unclear of is, how customizable are individual improvements? Will/should there be an "improvement workshop" where the player can build his churches and schools and military bases, or should the customization be achieved by grouping improvements? For instance, can the school system be made militaristic by coupling it with military bases, or religious by coupling it with temples? (Just some wild ideas...)

Micromanagement (a response to Joker):

It's true that people playing SMAC rarely use the automated units or governors. At least I don't. But in GGS I think the option of micromanagement should still be present, but not as a default. If a player wants to micromanage, a bunch of difficult statistics will be thrown to his face and it should be possible to get away from it with one click of the mouse. In the beginning of the game it is particularly problematic to keep the players from touching the micromanagement option: there is inevitably less to do on the macro scale, so bored players will explore the details and may not want to let go once the game progresses.

I think it's not a bad thing to have some micromanagement on demand: I for one would like to build my capital region more closely than other regions, but I would gladly let the AI take care of all military actions. Some other people want to do the opposite. It's not a bad thing to have this sort of options; however, to make the game playable and fair to those who do not want to have certain levels of management, the game mechanics and AI should be balanced so that there is little to be gained from fiddling with the details.

Conquering regions:

Well, the new population system solves this issue. When a chunk from another civ or a nomadic tribe is conquered, it splits off into a separate population within whichever region is closest to it. Maybe by default it could be it's own region. If you conquer back an area, the people would not automatically become one with their original population because the whole event of being conquered and then re-conquered surely changes the population properties enough to distinguish them from the unconquered people. How this happens exactly is to be defined later.

About the next version:

Perhaps the model should be cropped to a pure population model and leave the region stuff to the government model? I anticipate that just the population property stuff will be quite complex if you dig into it. Also, anyone writing models should pay special attention to the connections with other models. I believe that population model is, after map and terrain, the most fundamental thing in the game. Group interaction model and government models are built on top of this, so the questions which are addressed there should be outlined in this model.

And finally, about version numbers: I've always been more accustomed to thinking of 0.1 as the first step instead of "one tenth finished". So, in my opinion the drafts/proposals could be number 0.4, 0.5, ... , 0.9, 0.10, 0.11, ... ad infinitum without worrying about finishing. When you reach a version which everybody can agree with (well, the majority at least), say 0.X, you just label it as 1.0, upload it to the website and consider it done regardless of X being 9, 29 or 229.

Not that it looks very probably that drafts above 0.9 will be needed... I'm pretty sure we can reach a concensus of the basic functionality before that. This thing has to be programmed sooner or later anyway.

Leland

[VetLegion]

Amjayee and Leland,

Good work, and seems ok now.
If I still think that in the morning, we have a deal

[VetLegion]

Ok, verified

[VetLegion]

quote: Tile population: Why is it necessary to know exact number of people occupying a tile? The land area of a 50km across tile is about 2000 square km, surely it's unimportant whether there are 4678 or 4679 people living in it? In another thread S. Kroeze pointed out various population densities of nomadic peoples; if I recall correctly, one person per a couple of square miles was common, though there were exceptions. However, for the purposes of this game I believe that knowing the number of people in each tile with the accuracy of, say, 200-250 people is sufficient. This way two bytes can express 13 to 16 million people. Hmm... maybe it's not enough. I'm not sure about the population densities of modern cities, but aren't they in the same ballpark?

excellent!
I agree with this alot, 250 people +/- is not a big thing, except in the army.

anyway, should we vote on this, or on Design Doc with it?

Anyway, someone put up a vote!

[amjayee]

I agree that we don't need the excact number of people. Anything to reduce the complexity even a little!

I really should finish this model. I'm afraid that cannot be done this week, because I have an important deadline to meet, but then ít will be the next thing to do.

Edited:

Ok, my school work is done, and I will start going through this model again. Hope to get some results soon, during next week I guess.
[This message has been edited by amjayee (edited February 15, 2001).]

[S. Kroeze]

quote: But of course some kind of tech level would be good. Does anyone think of anything else we would need for tile populations?

Dear Amjayee, Chrispie, Joker, Leland, VetLegion and others,

I made a list of all data which -at least in my opinion- should be kept up at hex level in some way/somewhere, directly or indirectly:

• terrain type
• to which macroregion it belongs to: here could be stored information about climate of the region and its domesticable plants and/or animals
• population size
• the dominant nationality/ethnicity
• the dominant religion
• the prevailing production mode
  I would suggest ten different strategies: 1. hunting/gathering 2. pastoralism 3. hoe cultivation 4. plough cultivation and six 'mixed economies', i.e. combinations of 1 and 2, 1 and 3, etc; this would make transitions between them possible
• whether the population is sedentary or nomadic (of course pastoralists are always nomadic while plough cultivators are always sedentary)
• whether the population consists of unorganized 'bands' or one organized 'tribe'
• an indication of the degree of bellicosity of the natives: this could be simple: peaceful/warlike or in degree
• perhaps some indication of the tech level of those natives
• improvements like irrigation, roads, mines, fortresses, cities etc; I agree with the Joker that the original list was too extenxive

Secondly, I would like to react on the proposal of Leland:

quote: Tile population: Why is it necessary to know exact number of people occupying a tile? The land area of a 50km across tile is about 2000 square km, surely it's unimportant whether there are 4678 or 4679 people living in it? In another thread S. Kroeze pointed out various population densities of nomadic peoples; if I recall correctly, one person per a couple of square miles was common, though there were exceptions. However, for the purposes of this game I believe that knowing the number of people in each tile with the accuracy of, say, 200-250 people is sufficient. This way two bytes can express 13 to 16 million people. Hmm... maybe it's not enough. I'm not sure about the population densities of modern cities, but aren't they in the same ballpark?

Of course I agree with him that it doesn't matter whether there are living 1 million people in one tile or 1,001,000. On the other hand, I think it might have consequences whether 10 people or 1,010 are living on one tile. And there is an other problem I foresee:

'Thus the population figure of 100,000 for the whole of the Middle East at 8,000BC given by Carneiro and Hilse (1966:178) as the basis of their calculations seems eminently viable, representing as it does 2,000 groups 50 strong; or much more likely 4,000 groups with a mean membership of 25, the ethnographic 'magic number' for band size (Lee and DeVore 1968a:245-9). Taking 100,000 as a base population and rates of percentage annual increase ranging from 0.07 to 0.50, Table 4.2 emerges.

Calculations of the population of the Near East at 1,000-year intervals for various rates of increase, assuming a base population fo 100,000:

% increase / 8,000BC / 7,000BC / 6,000BC / 5,000BC / 4,000BC
0.50 : 100,000 / 14,700,000 / impossibly high
0.30 : 100,000 / 2,000,000 / 40,000,000 / impossibly high
0.20 : 100,000 / 737,000 / 5,440,000 / 40,100,000 / impossibly high
0.15 : 100,000 / 448,000 / 2,000,000 / 8,970,000 / 40,200,000
0.12 : 100,000 / 332,000 / 1,100,000 / 3,650,000 / 12,100,000
0.10 : 100,000 / 272,000 / 738,000 / 2,010,000 / 5,450,000
0.07 : 100,000 / 201,000 / 405,000 / 816,000 / 1,640,000
(source: Carneiro and Hilse 1966:178)

Significantly, if our generous starting level of 100,000 is cut in half, the picture does not qualitatively change -rates of growth above 0.1 per cent per annum on a sustained basis are impossible, otherwise colossally large populations emerge before the advent of cites. Even with the advent of cities over half a millennium later than the terminal date of the tables, we find the following population estimates by Adams (1981:90) of the then most densely populated parts of the Near East: the Nippur and Uruk regions of Mesopotamia. In round numbers they amount to a mere 60,000 in Early and Middle Uruk times (c.3,500-3,300 BC), and attain 110,000 for the combined areas only in the First Dynastic period, well into the third millennium.

The inference to be drawn is that the lowest rate in each case, that of 0.07 per cent, is the most realistic, with 0.1 in the nature of a ceiling. Carneiro and Hilse conclude:

"First and foremost, the increase in population that occurred during the Neolithic Period was not 'exceedingly rapid'. It was, in fact, only on the order of one tenth of one percent per year. For a village of 100 this rate of increase is equivalent to a net gain of only one person over a ten year period. Yearly population increments, of course, increased as the Neolithic progressed; yet these increments never became very large. Only by virtue of the fact that they were accumulated over several thousands of years did they culminate in a population of considerable size."
(Carneiro and Hilse 1966:179)

Assuming an exponential growth model (growth at a constant rate, resulting in a population expanding geometrically in size through time), the time it takes for a population growing at 0.01 per cent per annum to double is 6,931[/b] years; at 0.03 per cent per annum 2,310 years, and at the likely 'ceiling' level of 0.1 per cent per annum the doubling time is 693 years, only sustainable for short periods and/or small areas. For comparison the doubling time of an annual growth rate of 3 per cent, currently experienced by some Third World countries, is only 23 years, and at 1 per cent, which seems like a big reduction in rate, only 69 years.

Carneiro and Hilse's rate of 0.07 per cent that I think is most reasonable, is, of course, one spanning thousand-year intervals. From what we have seen of doubling times relative to annual growth rates, for the early Holocene something like 0.01 per cent per annum growth rate is by far the most plausible, shifting gear to the region of 0.03 per cent with the onset of the Neolithic, even conceivably reaching 0.1 per cent for short bursts in certain localities.'
(source: C.K.Maisels:'The Emergence of Civilization',1990)

Now let's assume we have a tile with 2,000 inhabitants. With a relatively high growth rate of 0.1%, next year there would be living 2,002, neglecting migration. But due to rounding off this number would finally be stored in the game data as 2,000. So as a result, all population growth stops and becomes actually impossible!

My solution to this problem is to give the population size of a tile as a number of four figures (a,bcd), multiplied by a factor of 10: 10¹, 10², 10³ etc. I hope this could work without occupying too much memery.

Best wishes!

S.Kroeze
[This message has been edited by S. Kroeze (edited March 19, 2001).]

[amjayee]

I don't know about the population growth thing... it is a problem and has to be solved. Though your suggestion sounds like it would take too much memory. Of course for really small populations we could store the amounts x1; when the amount grows to exceed the boundaries, it could automatically change to x10, then x100, x1000 etc. We'll see could this work... If we use this system, I think we would have to use 2 short ints per tile, and that would not help us save memory...

About the tile properties, it gave me an idea; perhaps we should have geographic regions, that could store the areas that shares the same terrain properties? Perhaps also other system to store the areas with the same climate? This could help us to save the memory needed for the map.