Do you build your city where the AI recommends?

akqjt

Chieftain
Joined
May 18, 2006
Messages
2
First-time poster here, Civ4 newb in general (just got my first win on Noble).

When you're thinking about where to move a settler to build a new city, do you guys use that blue circle recommendation or ignore it? Sometimes I will be looking at the yields/resources and think that one square is a great place for a city, but I don't get a blue circle. Sometimes I see a blue circle, go towards it, and think that an adjacent square would actually be superior to the circled square. Should I use my own judgment, or trust that the AI knows what it's talking about?
 
I'd say go with your own judgement. Sometimes the AI recommends some really weak locations.

What I always look for when building a city, especially my first city is an abundance of resources, and access to water nearby. I like to build my initial city where there are at least 2, preferably 3 good resources nearby. As for the water, it's good to have my initail city near the ocean, or a river that flows into the ocean in order to have lots of trade routes with other civs, and it's much easier to connect your own cities which gives your first city (usually your capital) access to most other cities resources without having to build roads all the way across the map
 
Yeah, I find it kinda weird when the AI recommends that I put a city one block off of the coast. I can't really think of any reason that would be better to building it on the coast, unless there was some great resource or squares outside the "fat cross" that I would not be able to access otherwise.
 
My personal view is that blue circles indicate the optimum location in terms of resources (which may include unrevealed resources). Resources are not the sole factor in city location: riverside is handy, on the coast or two tiles away, number of workable squares in the fat cross and so on. Very first city (4000bc) I tend to follow the blue circle advice coz I don't know the surrounding world, later on I use it as a minor indicator.
 
I rely on the blue circles less and less as time goes on.

The AI, as near as I can tell, only recommends city locations one at a time. That is, it doesn't take into account potential future city locations (it does factor in existing city locations).

I've just started "dotmapping", which is where you take a snapshot of the map (usually with the grid and resource indicators on), then leave the game to plan where the cities should go using a graphics program (even Paint will do) to draw in the "fat crosses". This is almost certainly something the AI does NOT do.

You may also have additional reasons the AI cannot fathom for a city site. You may be trying to seal off some territory, for example. Or a certain resource may not be as valuable to your tactics and strategy as it is to the AI.

And yes, the AI is infamous for building cities one tile from the coast, leaving it with a bunch of tiles only able to produce 1 food and 2 commerce, if that.

The more I play this game, the more I find I have to rely on my own judgement.
 
I have a related question: Do AI factor in resources that have not yet been discovered?

I ask this because many times the AI will recommend a really bad location, but then later that spot turns out quite good once coal/uranium/aluminium have been discovered...
 
No. There was a thread about this sometime ago, and a subtle bug was uncovered in an earlier patch by dedicating testing by some forum user. The user created a bunch of maps with symmetry in WorldBuilder to test AI settling patterns and found some anomalies. Soren took note of it and fixed the "leak" of information into the AI's algorithm for picking cities, so it's supposed to be fixed as of 1.52, I think. The AI cannot see future resources that are unrevealed.
 
I always build my first city on the startposition to make sure that other civs don't get ahead of me. After that I always get confused about where to build cities: go for the best places to block other civs or go for blue circles.

Being a newbie I am not sure what the best choice is, so most of the time I try to find a blue circle that is close to another civ, building the first line of defence.
 
I think the AI reccommends at least resonable locations for your city and is worthwhile to consider; however, If you see a hill with a forest on it (3 production hammers) or a hill without forest (2 production hammers) then found a city on that square, providing that it is not on a special resource because then your home city square will have 5 total resouces instead of 4!, which is 2Food, 2Production, 1Coin -- instead of 2Food, 1Production, 1Coin; this is just something that ive picked up with experience playing the game.
 
The blue circle locations are usually decent - but you can get some really crappy ones as well. I use it as a general guide, most often, but but it says one square, and I think another would be better (Such as one over, so it's on the coast) guess who I go with? Me, of course, over the computer. It's a matter of opinion, so there's no real right or wrong in this, but generally if you use the circle as a guideline, not an absolute "Must found in blue circle" rule, then you'll be ok.
 
TROOPERBLUE - it's not the forest on the hill that gives you the extra hammer of production, it's the fact that it's a "plains hill" - a hill on a plains tile - that gives you the extra. It's always worth considering plains hills into the equation as you get the hammer and the 50% defense from the hill.
 
The other posters have made some good points, but one thing I'll add in that I've noticed about the computer's "blue circles": in addition to prioritizing resources more than many human players, the computer seems to like to find sites that have balanced hammers, commerce, and food. However, not only does this often goes against city specialization, it tends to undervalue the ability of slavery to convert food into production.

I always pick my city sites far ahead of having a settler in the area to even know where the blue circles would be. However, I don't completely ignore them - if I see a blue circle in an unexpected spot I'll definitely give it consideration and see if there's something there I overlooked.
 
Sometimes if I can't decide whether to settle at Site A or at Site B one tile away (due to fog I carelessly left unrevealed, or overlap), I move my settler close and see if the computer recommends a location.
 
I read some where that the AI also take into account defensive bonuses? when factoring were to put a blue circle.
 
Aye.. I'm sure the factor in the defence bonuses. AI just love to build cities on hills. Its a shame they don't factor in the defence bonuses of surrounding tiles. Almost all their cities seem to have a forested hill next to them, from which I can launch my attacks.

They also don't factor in the "settle on the crappest tile" tactic. If theres a single desert square surrounded by grassland, a human is likely to settle on the desert square so they don't have it in their fat x. The AI seems to work in the complete opposite way, because they love having tons of desert ice and tundra in their fat x.
 
The AI gives decent tips: but my two main problems with it are:
1. Putting a city one square off the coast. This causes the coast squares to be terrible one food plots for the whole game. Someone might have mentioned it, but it drives me crazy.
2. The blue circles never correspond to an overall plan of city placement. The computer is picking the best place for that single city, regardless of what other cities it screws up.
 
Acceptable_Loss said:
I have a related question: Do AI factor in resources that have not yet been discovered?

I ask this because many times the AI will recommend a really bad location, but then later that spot turns out quite good once coal/uranium/aluminium have been discovered...


Yeah, it doesn't normally recommend a plot with a resource on it. Sometimes it does, but normally it does not.
 
I usually use my own judgement on city placement. If that ends up being the blue circle then I build there. I really don't use them at all.
 
Top Bottom