2 questions

[guermantes]

hello,

1) Is there an obvious reason as to why in this screenshot the AI does not propose to me to found the new city in the tile where I am standing? My access to resources seem to me to be better in the tile I stand in.



2) In the following screenshot Alexander declared war on me by coming in an unmanned galley from a different continent. He sat there outside my city for 280 years just looking at me (I tried twice to offer 10 turns of peace but he just frowned and said I had been behaving badly) until he finally came with another ship with 5 horse archers which I whacked speadily with my pikemen. He sued for peace 2 turns later.

Was the AI strategically observing me for 280 years first to see what I was doing or was this just bad AI ? (My pikemen were mainly brought in from afar when I saw him coming so he would not have seen them from his boat during the 280 years.)

[Ijuin]

I can't speak on the galley, but I think on the city founding problem, the computer is looking at the productivity of the city as it would be BEFORE you chop down the jungle, thus it considers the two adjacent tiles to be more productive sites. The computer takes potential future tile improvements into account, but does not do so for forest/jungle chopping--it assumes that the jungle will stay.

[faustus]

From what I've seen in other posts on the subject, and what I've observed myself, the AI when recommending city sites takes into account information you dont have. Quite often I've trusted it and founded a city where it told me to, and found horses, or copper or later coal or oil pop within the city borders that would have been outside if I'd founded where I wanted to (or vice versa, so frustrating to watch copper pop just outside our capital's fat cross. I dont know if anyone has proved this yet, but I've observed anecdotal evidence of it, and seen other people talk about it.

[zabrak]

In general, whenever you observe seemingly irrational behavior from the AI, and you ask yourself, "Is the AI being really clever or really dumb?", the answer is the latter.

I'd found three cities, incidentally: on the first pic, one NW of your settler as a commerce city, and one W of your archer as a production city, and on the second pic, two E of your scout.

[joncnunn]

On that screenshot, I think it would depend upon how close you are to Iron Working. If Iron Working is a long ways off, a coastal city would be much better than both your current tile and the two the AI is sugesting. (That is, if founding a city is worth it at all now)

Now Post Iron-Working, both AI spots have the advantage of not having three permantely subpar tiles (as water tiles worked by non-coastal cities are)

In addition, both of the AI sugested spots presumably have less jungle -> less unhealthiness.

[cynyck]

one NW of your settler as a commerce city, and one W of your archer as a production city

Yes, zabrak, I was thinking the same thing - two cities. But I would put the commerce city on top of the northernmost elephant. I see two other elephants and I usually only camp one if I have more than one. I hate to trade strategic resources unless its to a "vassal" civ that I left with one city and am grooming for tech trades. I always farm all extra elephants. But, I am only playing at Prince and have no idea if this is feasible on higher levels.

By founding on top of the elephant, I get the elephant (even though I lose the production bonus camping gives) and can farm the others. I also get three more sea tiles for commerce.

My experience is extra strategic resources result inevitably in an agg civ demanding it in tribute. I refuse, because I am not going to give them the horses to build horse archers to then attack me with to secure the horses for themselves, and that results in "annoyed" which eventually results in an attack, which always comes at that point when I have neglected my army for ten turns.

I seem to do that a lot - ignore my army for ten turns. Alex likes that. So does Monty. And Ghengis. I just hate the way they show their gratitude, though.

[zabrak]

While it's true that our friend would get three more sea tiles, and leave nothing unworked, those three are single-commerce only, and will never do anything more than feed themselves. However, he (she? I'm sorry) would give up the fruit tile for that city under your proposal - the sugar and wheat provide enough extra food to work the three (four, if one is taken from Texoco) hills in the production city I proposed. The extra food for the coastal commerce city we're discussing is more valuable - specialists do much better work for commerce purposes than production, at least until factories and power plants. And as you mentioned, there's also the camped ivory bonuses that would be lost as well.

I think you'll find that placing a city on top of a resource will hook up that resource to your trade network, provided you have the appropriate tech (Monarchy for wine, Agriculture for wheat/rice/corn, etc.), which means that your point about not having extra resources available to avoid demands is moot. If you're really worried about having to refuse AIs, just don't build a road to more than one.

The other thing I've been thinking about relates to guermantes' other post about food trading - I try to plan in advance to whom I'll be trading, and only build as many roads as necessary, for the reason you mention. The exception is meat resources -- cows/deer/pigs/sheep -- which I always hook up. This is because a) the unhealthy cap is much less painful than the unhappy cap, b) supermarkets don't come until pretty late in the game, c) for trade purposes, the AI treats all luxury/food resources as though they're of equal worth. Getting silver for pigs is a sweet deal - 2 happy points for virtually every city, plus that AI likes you more, and the AI only gets one health point out of it.

[cynyck]

While it's true that our friend would get three more sea tiles, and leave nothing unworked, those three are single-commerce only, and will never do anything more than feed themselves. However, he (she? I'm sorry) would give up the fruit tile for that city under your proposal - the sugar and wheat provide enough extra food to work the three (four, if one is taken from Texoco) hills in the production city I proposed. The extra food for the coastal commerce city we're discussing is more valuable - specialists do much better work for commerce purposes than production, at least until factories and power plants. And as you mentioned, there's also the camped ivory bonuses that would be lost as well.

I understand your point. Two things. I was wrong, it will only be two additional sea tiles because we are already on the northern coast. And I forgot that the second level sea tiles are only one commerce. I almost always play a financial leader and also almost always go for the Colossus, giving me three commerce in those tiles.

I think you'll find that placing a city on top of a resource will hook up that resource to your trade network, provided you have the appropriate tech (Monarchy for wine, Agriculture for wheat/rice/corn, etc.), which means that your point about not having extra resources available to avoid demands is moot. If you're really worried about having to refuse AIs, just don't build a road to more than one.

I am missing something. Perhaps not, perhaps I was simply not clear enough. This is what I was trying to say - found on top of the elephant and get the resource but not the production (elephants and ivory but not the hammer and commerce is it?). Then I would farm the rest. But I do see that you are correct, I could also camp the elephants but not road the tile. However, farming gives me the extra pop so I guess its a case-by-case decision.

c) for trade purposes, the AI treats all luxury/food resources as though they're of equal worth. Getting silver for pigs is a sweet deal - 2 happy points for virtually every city, plus that AI likes you more, and the AI only gets one health point out of it.

Yes, I have noticed that. I love the reverse also - getting a health resource plus some coin in exchange for giving one of my extra gold. Which is why, as someone else pointed out in another thread, it is always best to not accept an AI trade offer without negotiating first.

[zabrak]

The Collosus is a neat wonder for exactly that reason. However, it's elegantly nerfed, since on the water maps on which it's most valuable you'll be racing for the tech that obsoletes it: Astronomy.

I can see the comparative advantage of farming ivory when it's on plains, since there are circumstances when a city's got no food resources or farmable grasslands or floodplains, meaning every tile has to pull its own weight or it doesn't get worked. That's far from the case here; on top of which, our friend will have to wait until Civil Service, and then chain 4-6 farms out to that elephant.

My trading priorities, from the most likely category to give up to the least, are:

Meats
Seafoods
Incense
Grains
Grocer's
Dyes
Market's
Jewelery
Strategic - Buildings
Strategic - Units

There's some tweaking, of course - the AI's too dumb to realize that Fur/Ivory/Whale are useless one turn before they're obsoleted, or if I've got the Eiffel Tower, Iron is a harmless trade in the late-game.

One last note - there's a narrow window before an invasion in which it makes sense to trade some big-ticket happiness resources to an AI. You let their cities' population grow, then when you declare war, those extra citizens are either useless or force a suboptimal reallocation of tiles worked. But when you take the city and restore those resources, you've got more citizens with which to work.

[cynyck]

One last note - there's a narrow window before an invasion in which it makes sense to trade some big-ticket happiness resources to an AI. You let their cities' population grow, then when you declare war, those extra citizens are either useless or force a suboptimal reallocation of tiles worked. But when you take the city and restore those resources, you've got more citizens with which to work.

Boy, you keep bringing up such interesting points. Interesting for me, at least. I do like the idea of trading with an AI civ just before hammering it so that it really hurts that civ when we go to war and the trades are cut off. I also try to do this with a civ I think is about to attack me. And you are absolutely correct about the narrow window. It has to be long enough to allow them to grow, for sure.

This is another consideration for trading. In Guermantes' "Diplomatic Trading" thread I was trying to make the point that there is so much to consider when trading. But I am afraid I was not very clear.

One other point about elephants. I cannot think of another resource that gives you a production bonus if you develop it, along with a happiness bonus (for the ivory), and a strategic bonus in that you can build war elephants. Although, iron and copper have an added kick also because they will speed up certain wonders. Marble and stone speed up wonders but do not have any military value.

[zabrak]

It has to be long enough to allow them to grow

And short enough so that they don't become too tough a nut to crack.

I cannot think of another resource that gives you a production bonus if you develop it, along with a happiness bonus (for the ivory), and a strategic bonus

Ivory is the only resource that hits all four categories. There are quite a few that get three - horses/oil/aluminum are only missing the happy bonus, and gold/silver/gems/whale are only missing the strategic quality. Personally, I've always wanted whale-chariots as an early frigate, but that's probably because I enjoyed the worm-riders from Dune so much.

Marble and stone speed up wonders but do not have any military value.

Tell that to my Pyramid > Police State and Oracle > Civil Service powered Macemen wave.

[Urban Ranger]

I think you should move the settler up one tile for a coastal city. You get the same specials, and you get to work the sea tiles as well. I usually consider city sites one away from the coast the worst.

[Silver14]

Up one more would be right on the Gems. You get some benefit from building on top of resources, but nowhere near as much as mineing the tile and working it instead.

[Urban Ranger]

Okay, I didn't see that. Up and one to the left, then. You lose the wheat, but that's no big deal given access to the ocean tiles.

[Silver14]

Na, plop the city on the Banana. Who doesn't want delectable fruit in their town square after all?

[Yosho]

Originally posted by zabrak
One last note - there's a narrow window before an invasion in which it makes sense to trade some big-ticket happiness resources to an AI. You let their cities' population grow, then when you declare war, those extra citizens are either useless or force a suboptimal reallocation of tiles worked. But when you take the city and restore those resources, you've got more citizens with which to work.

Hmmm....but how do you know the AI won't just take the extra food from the health, and instead of growing their cities faster, use that extra food to move some people over to hill mines to get more production?

[zabrak]

If a city is working tiles in such a way that it's netting 0 food (i.e., stagnant), as occurs when it's at the happiness cap, then necessarily one or more of its citizens is a specialist or is working tiles that produce less than two food. If I raise the cap by donating a luxury resource, then in order to grow in population it must shift some citizens from those low-food occupations to higher-food tiles. In almost all circumstances,* this will result in less production and/or commerce. On top of which, if it does grow and I then take away that resource, those new citizens are now worse than worthless, because they consume food but generate nothing - that means to avoid starvation, the city must reassign even more citizens to higher-food but lower-production/commerce tiles.

*The two circumstances I can think of in which this can backfire are:

1) There is an unworked tile which is higher-food AND higher-production/commerce, which the AI is avoiding because it would give it a food surplus. For example, plains cows give 3 food, 3 hammers, while a grassland forest gives 2 food, 1 hammer - I can picture the AI sticking with the latter to avoid growth, but my donation allowing it to switch to the former.

2) You let the gift run on too long, and the AI grows enough extra citizens to start working more lower-food but higher production/commerce tiles, netting enough to build or buy an extra defender.

[Yosho]

To be clear, I was talking about donating health resources; didn't realize you were only talking about happiness resources.

Let me put it this way. An AI city has a +2 per turn food surplus, with -1 health. If you give it a health resource, will it stay at a +3 food surplus and grow faster, or will it move a person working a farm to a square with a cottage and stay at +2 food intead?

[zabrak]

Oh, I see. Well, now I just feel like a jerk.

That does raise an interesting question - I know the AI will try to stay under the happiness cap by arresting its growth. But will it go over the healthiness cap, if it has the food and happy points to support the growth? I know I would (although I'd work to lift it) since there's no downside, but the AI is, how shall we say, less than clever.

Incidentally, I'd support boosting the unhealthiness penalty to -2 food. As it is, unhealthiness is pretty toothless, and there are plenty of readily available ways to deal with it.

[clinton]

What I would do is make unhealthness -1 food for first pop over, -2 for second, and so on. So minor unhealthness is not a big issue but major unhealthyness starts to become a problem that rivals unhappiness.

Even with this system unhealthyness is still no worse than unhappyness as you choose not to grow.

This would increase the power of environnentalism slightly and provide some incentive not to clear cut forests.

[Urban Ranger]

Originally posted by zabrak
If a city is working tiles in such a way that it's netting 0 food (i.e., stagnant), as occurs when it's at the happiness cap...

Not sure what you are saying. A city can be htting happiness or health cap and still produce a massive amount of surplus food.

Unless what you mean is, when that happens, reduce food surplus to 0 by moving the people around, e.g. making some pop into specialists.

[zabrak]

Yes, the latter is what I mean. I'm fairly sure the AI does this with the happiness cap; now I'm wondering how it behaves if it hits the healthiness cap but with plenty of extra happiness and food.

[Sarek]

Originally posted by guermantes
hello,

2) In the following screenshot Alexander declared war on me by coming in an unmanned galley from a different continent. He sat there outside my city for 280 years just looking at me (I tried twice to offer 10 turns of peace but he just frowned and said I had been behaving badly) until he finally came with another ship with 5 horse archers which I whacked speadily with my pikemen. He sued for peace 2 turns later.

Was the AI strategically observing me for 280 years first to see what I was doing or was this just bad AI ? (My pikemen were mainly brought in from afar when I saw him coming so he would not have seen them from his boat during the 280 years.)

Just a thought.
Had a similar situation happen to me in a game against Germany (and others). The German's declared, sent a Caravel to hang out for many, many years.
1) It negates working some of the sea tiles, I believe, by your city.
2) It prevents you from getting out into the sea yourself, unless you create multiple ships of equal value (and if the enemy is promoted, even this may not work...he may pick you off as you come out).
3) It keeps you at war with that Civ for almost as long as that Civ chooses to remain at war, which likely causes you some other problems within the game with other Civs?
4) And, in the case of my game, it disrupted my game by causing me to concentrate on it. In my case, I wasn't able to really dislodge it, so I eventually gave up and just let it sit there. BUT, many years later, the German's launched a MASSIVE sea invasion against me, and because I couldn't dislodge the Caravel way back when, they just walked onshore and tore me to shreds.

* Probably some poor play on my part, but thoughts to ponder!

Get the thing out of there!!!

[trev]

The sea tile the galley is on and all immediately adjacent sea tiles cannot be worked by your city, and therefore sitting it there is a tactic designed to starve your city. Of course the AI is not clever enough to know if it is working as a tactic.

[Ijuin]

Is that right? I was not aware that other nations' ships denied you access to tiles adjacent to them, only the tiles that they themselves occupied . . .

[Guardian]

This is called a zone of control, meaning that the ships are able to control not only the tile they're in, but also the adjacent tiles. You can try this out with one of your own ships if you like. As you look at the map, you will see little boats marking the sea tiles that are being worked. Now, as you move your ship close to an enemy city, you will see the little boats disappear as your ship moves up next to them. As long as your ship is there, those tiles are considered to be controlled by you, hence the enemy can not work them. This will also cut off sea trade if you manage to control all sea tiles adjacent to the enemy city. In other words, naval blockades can be quite effective and can often be enforced by a single ship.

[Generaldoktor]

I thought I heard some griping that Civ4 doesn't have zones of control? I know Civ3 did.

[Guardian]

Well, technically, I guess Civ4 doesn't have zones of control in the way Civ3 did. In Civ3, certain units were considered to have a zone of control (ZOC), defined as an ability to project their power into adjacent tiles. What this actually meant was that whenever an enemy unit moved into an adjacent tile, the ZOC unit would have a first strike chance and the enemy would not be able to strike back unless actually attacking the ZOC unit (ie. trying to move in on top of it). Civ4 does not have zones of control implemented in such a manner, but ships are considered to control all adjacent tiles unless an enemy ship is present. In other words, if you lose the ability to work a tile because an enemy ship is present in an adjacent tile, you can make it workable again by placing a ship of your own there. This will give you back control over the tile that your ship is in, but not the remaining tiles adjacent to the enemy ship. These waters are now considered to be disputed and can not be worked until they are either all occupied by friendly units or the enemy is gone.

[Hermann the Lombard]

The concept that a ship would have a ZOC while land units would not is a tad warped.

[Generaldoktor]

Well, I'm a little confused whether you're saying Guardian is wrong, or you're saying he's right and the designers got it wrong. Ships can sail around and project, over the course of a year in a blockade situation, their presence in a lot of territory. On the other hand, land units can spread out and cover a fair amount of ground too, depending on the perceived notion of how big the land unit is. (In WWII, I recall, 12 linear miles was considered about the best a Western-style "division" could cover, without being ridiculous, though they were often asked to cover more; I forget the figure for corps, but obviously it was much larger.)

It sounds like we should have ZOC, but Civ4 was not supposed to be a "combat" game too much, (which is funny now that the announced XP is reportedly celebrative of combat,) and that may be something they simply tried to "simplify."

I haven't noticed the naval ZOC phenom myself, but I rarely let myself get blockaded.