1. We have added a Gift Upgrades feature that allows you to gift an account upgrade to another member, just in time for the holiday season. You can see the gift option when going to the Account Upgrades screen, or on any user profile screen.
    Dismiss Notice

City Specialization: how do you decide early?

Discussion in 'Civ4 - Strategy & Tips' started by rmunn, Mar 1, 2010.

  1. rmunn

    rmunn Chieftain

    Joined:
    Aug 7, 2009
    Messages:
    59
    I'm currently trying to move up from Noble to Monarch (and simultaneously trying to cure myself of a bad habit of auto-reloading when things go wrong, but that's another story), and I've come to realize that one of my weaknesses is city specialization. Once a city is fully-grown and I've mined the hills and cottaged the land, I have no trouble knowing what to do with it, but I have a hard time looking at the bare, unimproved land and picturing the fully-grown city in my mind's eye. Should my workers start putting mines on hills and workshops on land to make this city into a production powerhouse later? Or should I put windmills on hills and farms on land and make it a GP farm? Or should I be cottaging up every flat bit of ground I can find, building only just enough farms to support the rest of the tiles?

    I've read http://www.civfanatics.com/civ4/strategy/cityspecialization.php and found it very useful (especially the +F and +P methods of evaluating the resources in a BFC). But even knowing the technique, I can't seem to apply it. How much surplus food should I look for in a GP farm? How much production should a city's BFC have for me to turn it into a hammer powerhouse?

    Early city placement is especially tough for me: I can cruise along playing pretty quickly, making decisions about worker use and production queues without getting too bogged down, but my entire game grinds to a halt every time I build a settler. I spend at least ten or fifteen minutes looking at dotmaps, tying to find an optimal city placement, before I send the settler out, only to repeat the process over again with my next settler, second-guessing my earlier decisions! (E.g., if I'd settled that previous guy a couple tiles left, I could have put the gems and the gold mine in the same city without overlap, but now I have to do major overlap to make this commerce city work).

    How do you decide, when looking at a bare map, where to settle? What rules-of-thumb do you follow in deciding to pick this spot rather than that other spot 1 tile to the NW? What sorts of +food or +production values do you look for when you draw dotmaps? And how do you pick between cottaging every flat area in sight, or farming those areas instead? (And if you do cottage every flat area in sight, how do you decide that this should be a science city with Libraries/Universities/Oxford rather than a money city with Markets/Banks/Wall Street?)

    (Edit: Deciding where to build an Academy with my early GS is another hard one for me. That should obviously be the same city in which I plan to build Oxford... but I never know, that early in the game, which one I'm going to want to make my science city 200 turns later).

    Advice from experienced players much appreciated. :)
     
  2. Jet

    Jet No, no, please. Please.

    Joined:
    Mar 16, 2006
    Messages:
    2,364
    As a rule of thumb, most food resources or clusters of food resources get a city.

    Re specialists - look for multiple food resources.
    Re production cities - late production cities, it doesn't matter a lot. Early production cities, look for food, hills, and production resources.

    There are different ways you can play, but find the article named "Cottages!" It has more rules of thumb for cities for one generally solid style of play.
     
  3. Yared

    Yared That Guy

    Joined:
    May 21, 2009
    Messages:
    4,516
    Location:
    Hufvudstaden


    A Great Scientist farm I had as Shaka. That's generally how it should be in terms of food - 2 clams, 1 banana and some chain irrigated farms. Of course, you should try to find a spot with at least decent production capabilities, because you need to build things in that city too.

    After that university and then Oxford that city would be producing something like 340 beakers. I chose to settle all my great scientists; you can choose to bulb them if you need it.
     
  4. Mutineer

    Mutineer Deity

    Joined:
    Feb 20, 2006
    Messages:
    3,715
    City specialization pretty mach depends on need and change in course of the game.

    Early GS example, well if your capital is reasonably good commercial cite 99% of the time first academy go there. Otherwise, consider other ways to use GS. In many cases, like specialists economy for example, academy is not best use. Extremely early GS is better of settled.
    A bit late light bulbing philosophy tend to be a good option. Later bulbing education/liberalism astronomy chemistry, sci method physics could be a good idea, depends on situation and your plan. More late Golden age become better option.
     
  5. rmunn

    rmunn Chieftain

    Joined:
    Aug 7, 2009
    Messages:
    59
    I can't seem to find this... Google search ignores punctuation, so I'm getting all kinds of results that mention cottages, but not the article you recommended. Would you have a direct link you could share?
     
  6. VoiceOfUnreason

    VoiceOfUnreason Deity

    Joined:
    Dec 5, 2005
    Messages:
    3,663
    Using the Advanced Search option allows you to (a) search on titles only, and (b) restrict your search to the Strategies and Tips section and (c) restrict your search to articles started by DaveMcW.... You will quickly learn that Cottages!! is his particular battle cry.

    Here's the thread.
     
  7. mabraham

    mabraham Deity

    Joined:
    Aug 9, 2007
    Messages:
    4,027
    Location:
    Stockholm, Sweden
    You make a GP farm out of a cluster of food resources with a mine or two. Farms supplement that, it won't create it. The Shaka example in this thread is nice, but I would have maybe put an academy here, planned academy+Oxford on some polycottage river town, and settled all the great scientists there. That gives a higher return for longer, at cost of a lower short-term payoff.

    More generally, how many production vs commerce cities you want is context-dependent. If you're planning wars, you need production. If you're planning to turtle and race for some tech, then minimal production is needed. All cities need and like the same kinds of things, so start by hooking up all the resources, make sure there's enough food to grow to your happiness cap fairly readily, and make specialization decisions when you actually need to make them. That way you have more of an idea how many cities you'll have before you've made your nice ring of production cities around your capital and suddenly the AI have stolen all of your planned commerce city sites (of course, now you just beat the daylights out of them... but you get the idea...)

    Several, in both cases. Any particular number would be context-dependent. Plan on a GP farm, plan on several production cities. Every city needs enough food to be useful and enough production to build the necessary infrastructure, so your initial terrain improvements won't vary hugely from city to city.

    OK, well don't make a city-by-city decision. Make a whole-of-empire decision. When you're out scouting and you see a promising cluster of food, use Alt-S to write a sign there. Later when you can see the BFC, and preferably two rings around that, move your signs around to suggest actual sites. If you do that, then you will see how your planned city sites might mesh, and can adapt single-city decisions into empire-decisions.

    Don't stress much about excessive overlap. It's not until they get to size 20 that it matters, and even then not much. You get a much higher return early from 2 cities at size 10 working 20 tiles in the midgame, than one city working 10 of them in the midgame and all 20 of them in the late game. The early return is the important thing. You can take this principle too far, of course. Each city will need lots of late-game infrastructure, and there's economies of scale for large cities... so make these decisions based on your endgame plans... if you're going to conquer the world by chariot then you don't care about overlaps (and why are you building settlers anyway?)... if you're going to win a space race, then plan for big cities with little overlap. Alternatively, interleave one early-game production city between two late-game commerce powerhouses, and give over the production city's main tiles to the others as the game progresses. Now you just leave it building Wealth or something for the last 300 years.

    Look for food. Anything else can be arranged later. Do make sure some of your sites can be arranged for early production.

    Specialization of :gold: vs :science: is only important if you plan to spend noticeable amounts of time running those specialists, or have an external source of :gold: (holy city, corporation). The slider is empire-wide. If you have two commerce cities producing 30 commerce running no specialists, you get the same return from one of them with library+market and the other with nothing as you would from one with library and the other with market. That's regardless of the position of the slider. If you do specialize, however, then it becomes much more reasonable to run some specialists in the early midgame. Cottages are eventually more productive, however.

    Oxford is easy. First consider your capital. +50% to your commerce immediately suggests many cottages and thus lots of :science: Next, consider any site with lots of flat land, food source, and a river, to get as many farms as needed for growth, and the rest cottages. Do not consider any site that will be running scientist specialists, unless you'll be finished the game by about 1200AD. (Even then, you won't get a return from Oxford that will matter!) Compare your expected output at about size 16. Rename that city "Oxford".
     
  8. mariogreymist

    mariogreymist Deity

    Joined:
    Aug 11, 2009
    Messages:
    2,201
    This shouldn't happen, if you dotmap properly. Once you've scouted the area within your settling range, you create a single dotmap to which you will adhere. When I first started, I liked to number the cities for which would be settled in which order. If you do that, rather than decide every time you build a settler, then you will have an easier time. I tend to label my cities on the dotmap too. "commerce" or "milprod" (military production) or "GPfarm" using the alt-s feature. This way, when I send workers to the city, I can queue them to build the appropriate improvements and then forget about them. Start by just marking the 2-3 very best sites. Don't compromise on their location unless you have the choice between 1 great and 2 mediocre or 3 very good cities. 1 great and one mediocre is usually better than 2 pretty good cities. (hope that makes sense) Then, once the premium sites are marked, figure out how to work the rest of the resources in your area. Mark those sites as the second wave of settlement. Then figure out how to work the rest of the tiles in your area...even if some will wait until biofarms, figure out where they go and mark them (clearly for much later settlement in many cases).

    hope that helps more than confuses.
     
  9. Tomice

    Tomice Passionate Smart-Ass

    Joined:
    Oct 5, 2009
    Messages:
    2,320
    Location:
    Austria, EU, no kangaroos ;)
    And don't forget to draw lines to mark the fatcross of each city, if this helps! I do it that way, and it helps a lot. looks buttugly, but this is after all, it's not about looks! :D
     
  10. Woodreaux

    Woodreaux Prince

    Joined:
    Sep 18, 2007
    Messages:
    357
    Location:
    So Cal
    Or if you have BUG installed, you can use the dot map feature (Alt + X), and draws pretty BFC's for you.
     
  11. Rameau's Nephew

    Rameau's Nephew Bureaucrat

    Joined:
    Aug 14, 2002
    Messages:
    262
    Location:
    Bethesda, MD
    You can even color code them. For me, white = commerce/blended cities; red = production cities, and green = GP cities.
     
  12. mariogreymist

    mariogreymist Deity

    Joined:
    Aug 11, 2009
    Messages:
    2,201
    I only color code the filler cities, so I don't accidentally settle one before a prime site. But then again, I am particularly lazy.
     

Share This Page