Getting Rid of ICS - No free centre square

[Aeson]

Originally posted by notyoueither
Yes Aeson, but you're twisted.

BTW. How many Settlers out of Huts in the first 2000 years?

Only 1 Settler. On turn 6 if I remember correctly. I'm pretty sure that 1 Settler is the limit from huts using the 'fixed' 1.17f .bic file. The Settler speeds up expansion by about 15 turns.

That max cities of 512 is just about the only real limitation on ICS (other than map size of course)... been razing cities like mad trying to keep room for my Settlers!

[notyoueither]

Ahhh.

New Improved ICS. Now with Ethnic Cleansing!
Get Your's Today.

I wonder what they'll do to patch that.

[Grrr]

The so manyith time we've discussed this. BAD IDEA. Civ is ICS. Civ without ICS is stupid.

[Dominae]

I'll just jump in here for a quick comment. I think the simplest solution in this case is the best one: put an upper limit on the number of cities your empire can have based on government type (and, of course, map size).

This solution would leave the game essentially unchanged for those of us who never employ ICS (assuming the max number of cities figures are well-tested), and completely eliminate ICS for everyone else.

The max number of cities would be low enough to prevent ICS but high enough to give all players something to work with (I like making Worker factories, for example). Democracy and Republic would allow the most freedom in this regard, meaning that builders would be advantaged for buidling cities (which seems right).

Note that the cap is on the number of cities that you can build, not that you can have; a Settler can never build a city beyond the max, but you can gain cities in other ways (conquest and culture).

Any solution that changes the rules of the game simply won't happen, because the rules are already really good (and well-tested): the rate of growth of cities is perfectly balanced to keep the game interesting in the early game. Slowing down the pace of the game is a job for QA, and I think they've already done their job.

So, my vote is for number of city caps.


Dominae

[Grrr]

Dominae is right, and the way he's thought it out, it won't be like CtP in which on will lose lots of happiness in the late game. However there needs to be a government which doesn't have this cap, otherwise, in the late game, it becomes impossible to get that resource.



Another solution is to have the cap lifted by population. In that, if you have an average city size of 6, you can build cities, but if your average city size is 1 or 2 you can't. This wouldn't apply to the early game though.

[Dominae]

Grrr, I was thinking that under Democracy there would be no cap to the number of cities, but even that could become sort of a problem. The game goes along fine until Democracy, then BANG you build Settlers like crazy and backfill your entire empire. The advantages of doing this are clear, making the strategy a no-brainer. Essentially, ICS would be delayed until Democracy.

I think the solution rests on the exact numbers for the caps. Take a game for an average player who only builds cities 4 tiles apart (on average), and count the number of cities after expansion is over. Increase this number by, say, 25%. Then this cap prevents ICS completely, but gives opportunities to those players who want to "suck the land dry" (I miss The Hive).

Tweaking the numbers to be just right could result in the same game, just without ICS.


Dominae

[Dominae]

I guess if you really wanted to ICS sometime (after all, some players do like ICS), you could tie the "no city cap" to a Small Wonder or something, which would only be available sometime during the Industrial Age (preferably later).

Anyway, this is just wishful thinking, because major changes like this simply do not belong in patches.


Dominae

[Grrr]

Or just put it in the editor. That would make it easy enough to change.

[Austin]

Originally posted by Aeson
Just a little screenshot for anyone who thinks ICS isn't a bit overpowered. Deity, 315 cities in 500AD.

This is EXACTLY what I am talking about. In just about every single Civ game I've played the entire world is carpeted with cities barring one or two isolated islands. This simply did not happen in the real world till the industrial age because there are a lot of barriers to urbanization that can only be solved with more advanced technology, or really really favorable conditions that are not all that common.

Under Civ a civilization barely out of the stone age can carpet the entire world with cities given enough time (and it actually doesn't take very long). This simply could not have happened in the real world, and it's boring to boot.

What we should be seeing in the ancient and medieval eras is not being able to plunk a city anywhere the hell you want (except a large swath of mountains), but for there to be limits to what land is useable. As you advance your technology and get more advanced social models (governments) land that was previously unuseable becomes open for colonization.

So instead of the game being "landgrab, then exponential growth, then fighting because the world is full" there should be smaller more progressive landgrabs as technology advances. What we have now is not only unrealistic, but BORING.

I'd like to see some strategic choices more involving than "nail down every last hill and dale with a settler". As it stands now there is usually little choice between developing an existing city and founding a new one. If you choose to build up a few highly sophisticated cities, and the AI chooses instead to spew settler in all directions you have to follow suit or you simply get overwhelmed.

Perhaps ICS isn't the right term for what I'm describing, but I think that this "Infinite Urban Sprawl" is a problem.

Austin

[dunk]

Wait, is everyone saying that the Earth wasn't pretty much fully populated, by which I mean settlements all over the planet by 1 AD or so? I think it was. The corruption system in Civ III was implemented so that 90% of Aeson's cities are useless.

I think everyone here agrees that the AI doesn't build enough city improvements. That's the reason the player is "forced" into ICS. If the AI was programmed to focus on improvement rather than expansion, ICS wouldn't be necessary and we could all play nice, leisurely games.

BTW, you only have to do ICS if you're trying to get a high score, which IMO is nonsense.

Oh, and BTW, you can win the game with one city, even on Deity. Read through the board.

[Austin]

Originally posted by dunk999
Wait, is everyone saying that the Earth wasn't pretty much fully populated, by which I mean settlements all over the planet by 1 AD or so? I think it was. The corruption system in Civ III was implemented so that 90% of Aeson's cities are useless.

It was populated, but not by a dense cluster of urbanization, which is the problem I have with it. Even the Roman Empire at it's height, 90% of the population lived in farms and villages. That's how I always view all those squares in Civ that are within your cultural boundary but are not part of a city.

I think everyone here agrees that the AI doesn't build enough city improvements. That's the reason the player is "forced" into ICS. If the AI was programmed to focus on improvement rather than expansion, ICS wouldn't be necessary and we could all play nice, leisurely games.

Still, the ability would remain for both AI and human to simply carpet the world with cities from day one.

BTW, you only have to do ICS if you're trying to get a high score, which IMO is nonsense.

Oh, and BTW, you can win the game with one city, even on Deity. Read through the board.

And the challenge there is deliberately letting the AI carpet the world, thus increasing the difficulty for yourself by increasing the mass you are fighting against. This isn't a new strategy setting, it's simply weighting the game in the AI's favor.

Austin

[Aeson]

The corruption system in Civ III was implemented so that 90% of Aeson's cities are useless.

Actually, those cities can produce a worker in 10 turns or a Settler in 30. Far from useless. And as the Iroquois I can switch to despotism every 40 turns, pop rush 300 Mounted Warriors (disconnecting Iron and/or Saltpeter where necessary), then switch back. Corrupt cities serve many purposes.

Consider an 'unlimited expansion' map (every city site with at least 2 tiles of 2 food or better, no actual map boundaries). Let every city be corrupt. Each of these cities can produce a Settler in 30 turns. In the 540 turns that means the number of cities can double 17 times. 2^18 = 131072, and another batch of 131072 cities would produce a Settler before the end. Of course there is the time needed for each Settler to get to their city site, which limits the turn around time in most cases. I think it still shows how valuable corrupt cities can be over the course of a game.

Combined with rushing Settlers (mostly by gold), my productive cities, and demanding cities from the AI, I can usually double my number of cities every 20 turns. Of course thats only to the point of domination, free territory, or max cities limit.

Wait, is everyone saying that the Earth wasn't pretty much fully populated, by which I mean settlements all over the planet by 1 AD or so? I think it was.

There are still vast tracts of land unsettled by humans on this earth. The smallest city in Civ III terms is 10,000 people, which many of the settled areas wouldn't even match. There weren't many settlements of over 10,000 people in the BC's.

BTW, you only have to do ICS if you're trying to get a high score, which IMO is nonsense.

Some people just enjoy maxing out efficiency in games. I've always tried to 'perfect' any game I've played at least once, just to see where the limits are. Still a lot of improvement left in figuring out Civ III

[dunk]

It's true ICS is a very powerful strategy. And, yes, the cities are far from useless for a pop-rusher (which I never do). But, I don't play that way. I only put cities in nice city spots. I've stopped playing on huge maps because there are simply too many cities and too much territory. Even large maps are too big for me. I've only had problems with the AI building cities on my borders two or three times. But, that's life. You can win Civ III all you want, but it's much more fun to play it without overexploiting the weaknesses, IMO.

[Deornwulf]

I agree with Dominae and think that the number of cities you can build should be limited by the type of government you employ. Doing so could add the Colonial Age to the game as a specific time period instead of during the entire game.

Great and Small wonders could be added to allow an empire to grow beyond the governmental limit.

Distance from capital should also be a limiting factor. One should not be able to start marching a settler unit across the board in the beginning of the game and build a city half a world away from the capital and still be considered a part of your empire. Perhaps this distance limit could even be extended to units. (Many space based civ games already have this feature). In terms of realism, it's very silly to think that a unit can instantly report on the terrain no matter how far from the nearest city that unit may be.

Colonies become much more important in the early game if empire size is curtailed. As a civilization advances and is able to expand farther, the importance of the colony diminishes.

For the empire cap to work without slowing the game down too much, the city radius would also have to be more flexible (like in CTP 2), expanding to fill the available space, maybe out to the same distance as the culture border.........but each square should belong to a specific city. The idea of switching workers around to free up those squares shared by cities to maximize their usage is just plain silly.

[Dominae]

I think we should all be clear as to what ICS actually is.

ICS is not expanding until your borders fill all the land mass in your continent (or multiple continents). If this were ICS then the AI would be guilty of using it in every single game; the AI has been programmed to get as much land as it can within its borders.

Loosely, ICS is building cities in a tightly packed fashion in order to exploit the fact that, in Civ3, more cities is better than fewer cities. This will always be true, unless the game dynamics are drastically changed.

Aeson's pop rushing example is perfect. Playing the Iroquois, I can maybe pop rush maybe 15-20 MWs in one turn (in a normal game). With the same land mass and using ICS, Aeson could probably crank out around 50 (at the same point in the game). I don't think this is a problem with pop rushing.

By the way Aeson, I'm not getting on your back here, just using you as an example because you're the only one that admits to using ICS.

Gaining more territory, even in the BCs, is fun. Rapid expansion will remain part of the game, whether it is historically accurate or not. The fun aspect of ICS, on the other hand, is dubious (hence the point of this thread).

The fact is, ICS is not really a problem as long as MP isn't around. If you don't like ICS, don't do it. If MP ever gets here, prepare to see a lot of urban sprawls if ICS still exists.


Dominae

[Aeson]

Only 1 Settler. On turn 6 if I remember correctly. I'm pretty sure that 1 Settler is the limit from huts using the 'fixed' 1.17f .bic file.

I just have to correct myself here. Playing on a Deity game earlier today I got a 2nd settler about 15 turns after my first. I checked the .bic and Settlers weren't selected as available to barbarian chiefdoms. So it's not impossible to get 2+ (even on Deity), just extremely unlikely.

By the way Aeson, I'm not getting on your back here

No worries

[ALPHA WOLF 64]

I always wonder why people think that government alone can control a population. I doubt a democracy 2000 years ago would have survived any better than any other form of government. IMO its the technology that has allowed civs to grow and still be fairly controlled. Railroad, telegraph, telephone, radio, could a large empire be run without these?

(not a flame, just pointing out a a perceived logic flaw)

[realpolitic]

New settlers cost 2 population, so you're paying for the center square anyway. You're better off building granaires, barracks (if you're militaristic), troops etc., until you have a size 7 city. If you don't like the AIs ICS, build less cities, and take the rest. The Aztecs seem to be the best for that.

[Deornwulf]

Alpha Wolf - It's just the easiest fix to the problem. I had not even thought about it being a technology of and unto itself. Most space based games make empire size a matter of technology so why not in Civ3.

You do bring up another pet peeve of my own....ancient governments bear little resemblance to their modern namesakes and yet Civ3 has it all boiled down to five generic flavors.

[nbarclay]

Keep in mind that the "cities" in Civ 3 aren't JUST cities. They represent not only the city itself but also the people working the land for tens of miles around. Basically, the entire area takes its name and identity from whatever happens to be the largest town or city in the area. (Or at least that's the view I think makes the most sense.)

Following that logic, "towns" (size <=6) in Civ 3 would be heavily rural, while "cities" and "metropolises" would be increasingly urbanized. Depending on the culture, a size one "town" might even be more a collection of affiliated villages rather than what any of us would normally think of as a town. Conversely, in the modern era, many cities that would have been huge in the ancient world are merely outlying areas of one metropolis or another, not deserving of a separate identity (or the micromanagement that would go with it).

I also think it makes perfect sense that if you want lots of land, you have to either get there first or fight for it. The only thing I see that looks seriously flawed from a "realism" perspective is that there aren't dozens of minor civilizations or tribes that are already there and have to be conquered, assimilated, or wiped out if you want their land. (No, I don't regard Civ 3's barbarians as a particularly credible substitute.)

Nathan

[chiefpaco]

Why not make it so that no cities can overlap tiles? Or a minimum spacing enforced?

[nbarclay]

In regard to game balance, Civ 3 has four basic things that a city can produce:

1) Shields
2) Gold
3) Population
4) Culture

There is also a fifth relevant factor that is largely a derivative of the third, Score.

The corruption mechanism in Civ 3 keeps ICS from creating serious imbalances in wealth and shields, at least anywhere near a nation's core. Closer packing helps in the early game, but hurts later on as cities don’t have room to grow and cannot disband without losing whatever investment has been made in city improvements. Those two factors more or less even out, depending on playing style. There is some danger from the fact that in areas of high corruption, five ICS cities can produce five gold and five shields where one large totally corrupt city would just produce one of each, but it would take a lot of extra ICS cities to add up to the equivalent of one extra core city (especially a heavily improved core city).

Population is another matter entirely. Numerous small cities can grow population far more quickly than one big one can. That's not too horrible in and of itself, but as Aeson pointed out, the game leaves loopholes through which that population can be transformed into things of significant value. Pop rushing can add a dramatic percentage increase to the production of totally corrupt cities. Using totally corrupt cities to produce workers and settlers allows the good cities to focus on other matters. And after the discovery of Nationalism, the draft can turn population into conscript riflemen, infantry, or even mechanized infantry.

One way to fix these problems would be a simple check that if a city is over, say, 75% corrupt, it cannot pop rush, draft, or build workers or settlers. Basically, the same corruption that somehow mysteriously takes away practically all the city's gold and regular production would also render those other means of extracting value from the city useless. Voila, the loophole Aeson exploits disappears. (For balance reasons, it might be necessary also either to ban using gold to rush build in such cities or to provide a “remote pop rush” mechanism so despots and communists can assign the population and happiness penalty for pop rushing in an almost completely corrupt city to a less corrupt city where the penalty means something. In the latter case, the idea would be that since the locals are so corrupt, the work crew has to be brought in from outside.)

Culture is also a problem because the number of cultural improvements you can build is directly determined by the number of cities you have. That makes tight city packing and/or conquest absolutely vital to an early culture victory. Something seems just plain wrong to me when conquering half the world (and then rush buying cultural improvements in the conquered cities) is the easiest way to achieve a "cultural" victory.

I can think of two possible (and probably mutually exclusive) solutions to that problem.

(1) Have corruption affect the amount of culture produced, probably using a floating point number instead of an integer, so highly corrupt cities produce almost no useful culture. That would shift the cultural focus more back to the core where it belongs.

(2) Allow the building of multiple copies of each cultural building in a city, probably with the additional copies affecting only culture. Thus, building a second or third temple or library in a core city would produce just as much cultural value as building one temple or library in a corrupt city. The numbers for each type of cultural improvement would have to balance, e.g. you cannot build a third temple in a city until you have two of every other cultural improvement you know how to build.

I don't think any of these fixes would be all that hard to put in a patch (if Firaxis does any more after the upcoming one), and the only play styles they would seriously impact are ones that are way out of balance to begin with in order to take deliberate advantage of loopholes in the rules. That’s a lot more realistic than trying to rework the whole food system at this late date.

One last thing: I think the whole "score" mechanism in Civ 3 is way out of kilter. It would be much better suited to a game called "Domination" than to one called "Civilization." The benefits of ICS for score are merely one symptom of vastly larger problems.

Nathan

[nbarclay]

Originally posted by chiefpaco
Why not make it so that no cities can overlap tiles? Or a minimum spacing enforced?

So what happens when the theoretically ideal place for a city is two tiles deep in a mountain range, or buried under a few dozen yards of ocean? Sometimes cities need to be placed abnormally close together to take good advantage of the available land due to positioning of mountains or oceans, or maybe just for the benefit of building an early city along a river.

But decisions for such placement ought to be regarded as a compromise to be accepted reluctantly, not as a desired outcome. The real problem is game mechanics that turn things upside down so that dense city spacing that ought to be regarded as an unwelcome compromise becomes desirable.

Nathan

[Dominae]

Originally posted by nbarclay
One way to fix these problems would be a simple check that if a city is over, say, 75% corrupt, it cannot pop rush, draft, or build workers or settlers.

Although this may solve some problems with ICS in Civ3, somehow I don't think players will very much enjoy the idea that cities that are already almost "useless" can't even help themselves in the simple and obvious way (creating Workers to improve the land and pop rush some key imrovements). I pop rush all my Temples, but this is especially important for corrupt cities because they are usually on the edge of my empire where I want my borders to expand. With the outcries of the severity of corruption in Civ3, I can't imagine many players learning that corrupt cities are going to be even more "worthless".

Given that, I'm not sure that preventing very corrupt cities from doing the things you mentioned would even disable ICS. The "loophole" isn't the pop rushing, it's the efficiency of production centers (no matter how corrupt).

Originally posted by nbarclay
Something seems just plain wrong to me when conquering half the world (and then rush buying cultural improvements in the conquered cities) is the easiest way to achieve a "cultural" victory.

I don't see this as a problem. As a avid fan of the Japanese civ, I often conquer half the world, then rush a lot of cultural improvements to put me over the top, rank-wise. Again, the problem is the efficiency of production centers (and consequently the benefit of having more than fewer).

Originally posted by nbarclay
One last thing: I think the whole "score" mechanism in Civ 3 is way out of kilter. It would be much better suited to a game called "Domination" than to one called "Civilization." The benefits of ICS for score are merely one symptom of vastly larger problems.

Domination is the most difficult victory type (I suppose Conquest is equally difficult). Early conquest is the most difficult part of the game, IMO. Giving the highest rewards to these achievements seems natural. It would have been interesting to have a game where peaceful victories were just as difficult as aggressive ones, but Civ3 isn't that game. "Interesting" here does not mean "fun"; I think all civ games should have a bit of conquest in them, otherwise you're playing SimCity.


Dominae

[Austin]

Originally posted by Dominae
I think we should all be clear as to what ICS actually is.

ICS is not expanding until your borders fill all the land mass in your continent (or multiple continents). If this were ICS then the AI would be guilty of using it in every single game; the AI has been programmed to get as much land as it can within its borders.

True, perhaps "Infinite Urban Sprawl" is a better way of describing this.

Loosely, ICS is building cities in a tightly packed fashion in order to exploit the fact that, in Civ3, more cities is better than fewer cities. This will always be true, unless the game dynamics are drastically changed.

And this is why there is a second problem. You can't fill the world edge to edge with urbanization untill your technology is fairly advanced. There should be significant gaps in the world map later on in the game. Instead the whole world is a suburb by the Middle Ages.

This is boring. The first part of the game is pure landgrab, then the rest is pure mass mass mass.

Aeson's pop rushing example is perfect. Playing the Iroquois, I can maybe pop rush maybe 15-20 MWs in one turn (in a normal game). With the same land mass and using ICS, Aeson could probably crank out around 50 (at the same point in the game). I don't think this is a problem with pop rushing.

By the way Aeson, I'm not getting on your back here, just using you as an example because you're the only one that admits to using ICS.

Gaining more territory, even in the BCs, is fun. Rapid expansion will remain part of the game, whether it is historically accurate or not. The fun aspect of ICS, on the other hand, is dubious (hence the point of this thread).

Personally I find it about as fun as ICS is. Eliminates any challenge from the game, since you just spew spew spew, and then rush rush rush.

The fact is, ICS is not really a problem as long as MP isn't around. If you don't like ICS, don't do it. If MP ever gets here, prepare to see a lot of urban sprawls if ICS still exists.


Dominae

Yup.

Austin

[Austin]

Originally posted by nbarclay
Keep in mind that the "cities" in Civ 3 aren't JUST cities. They represent not only the city itself but also the people working the land for tens of miles around. Basically, the entire area takes its name and identity from whatever happens to be the largest town or city in the area. (Or at least that's the view I think makes the most sense.)

Thing is, for most of history those people working the land don't produce anything above their own immediate consumption needs. You need urbanization and concentration in order to get productivity above today's needs, and it is this "above and beyond" production that is what is available to make temples, army units, conduct research.

And until fairly recently you could not simply plunk down a settler just about anywhere and expect a city to naturally grow. You look at where cities first got established, and you needed very favorable circumstances.

As technology progressed, this became less and less of an issue, which progressively opened up more turf for urbanization.

Following that logic, "towns" (size <=6) in Civ 3 would be heavily rural, while "cities" and "metropolises" would be increasingly urbanized. Depending on the culture, a size one "town" might even be more a collection of affiliated villages rather than what any of us would normally think of as a town. Conversely, in the modern era, many cities that would have been huge in the ancient world are merely outlying areas of one metropolis or another, not deserving of a separate identity (or the micromanagement that would go with it).

This would be subsumed under those tiles that are part of your cultural borders but not being worked in a city radius, since those little village clusters are'nt going to contribute any above and beyond production to your civilization.

I also think it makes perfect sense that if you want lots of land, you have to either get there first or fight for it. The only thing I see that looks seriously flawed from a "realism" perspective is that there aren't dozens of minor civilizations or tribes that are already there and have to be conquered, assimilated, or wiped out if you want their land. (No, I don't regard Civ 3's barbarians as a particularly credible substitute.)

Nathan

Well the minor tribes is a whole other issue. I think a good substitute would be allowing barbarian huts to produce any military units that are available to a majority of the civs in the game. So if 4 of the 8 civs in the game have ironworking, the huts start producing swordsmen. If 4 civs have musketeers....and don't bother with resource restrictions (since then you would have to have barbarian workers making colonies, trading, and at that point they might as well be civs themselves).

Austin

[Austin]

You know, I wonder if one possible way to work some of this in is to reduce the food value of grassland when it's unimproved, and then increase the value you get when it's irrigated (so that an irrigated grass square still gives the same amount as before). This would put more of an emphasis on developing existing areas rather than spew spew spew. In that context workers are worth a lot more overall to your civ verses settlers than before, and rewards a builder strategy more than a settle in all directions one.

Austin

[planetfall]

The problem is not the center square. The real problem is the lack of support for megacities. Why are there cities all over the map?
1. If we don't plant a city the AI will
2. A city will only produce from within it's 21 tiles
3. The more production you have the higher your chances of winning.

What would happen if there were support for up to 5 megacities per civ? A megacity is one that expands the workable tiles beyond the 21 city tiles. Ideal would be a city that could expand 2 more times.

Alternative:
Scenario today
Key: T=tile, B=bonus tile, R=resource

B T T T B T R

Today there is no way to get both bonus tiles and the resource in one city. Thus you must either plant 2 cities or give up bonus/resource.

Possible Scenario
With megacity support, you could plant one city centered on the middle 't' and grow to include all three special tiles.


Now if there was some mechanism to stop AI ICS and we had megacities, the cities would look more like cities and not like a continent of development.

If the concern is more visual, allow us to improve a tile by improving with a park/sports park and give us shields for the park. That would eliminate the cities of 20 tiles all mined and irrigated.

[chiefpaco]

Originally posted by nbarclay


So what happens when the theoretically ideal place for a city is two tiles deep in a mountain range, or buried under a few dozen yards of ocean? Sometimes cities need to be placed abnormally close together to take good advantage of the available land due to positioning of mountains or oceans, or maybe just for the benefit of building an early city along a river.

But decisions for such placement ought to be regarded as a compromise to be accepted reluctantly, not as a desired outcome. The real problem is game mechanics that turn things upside down so that dense city spacing that ought to be regarded as an unwelcome compromise becomes desirable.

Nathan

I understand your point after reading your later post. It won't solve the overall problem of some factors being subject to ICS penalties (corruption) & others not (culture).

Overlapping tiles idea was pretty bad. However, right now, the mimimum spacing of cities is 2. Why not make it 3? Wouldn't it help?

[Mark_Everson]

Originally posted by Austin
Thing is, for most of history those people working the land don't produce anything above their own immediate consumption needs. You need urbanization and concentration in order to get productivity above today's needs, and it is this "above and beyond" production that is what is available to make temples, army units, conduct research.

Hi Austin:

I don't think this is right at all. First of all, its a game, and so mimicing reality is a bonus only when it adds to the fun. But lets suppose realism is the point. I don't think people Anywhere, rural or urban, produced that much above subsistence until the "modern" world starting in 1500 or so. Urban dwellers might be more productive on average in ancient times, but the bonus is nowhere near large enough to make up for that fact that the population of most 'civs' is 95+% rural in general.

But back to the main point, moderately dense population covering the globe is more realistic than not, aside from mountain ranges, desert, tundra, etc. And I think it could make for a Better game. The big defect in civ in this regard is, has been said here before, the lack of smaller countries that can fill the space without being a major player. Of course some of these could become civs in the real world, but that seems beyond the Civ scope.