No announcement yet.

Empire efficiency

  • Filter
  • Time
  • Show
Clear All
new posts

  • Empire efficiency

    (long post)
    Hi all. I've been lurking here for a while, and would first like to thank all the members for the many useful strategies and ideas. Now for my $.02.

    There's an article over at CFC in the War Room about how the demographics work (when I figure out how to post a link, I will). I was intrigued by this, and have been exploring ways to employ this as a guide to how efficiently I'm developing my empire. I was driven to this during a regent game I have underway (large, random map, 7 AI civs, Celts by random choice). I ended up stuck on Gilligan's island (80% water, pelago, normal climate, I think). I was unable to find any land except for two small islands directly north of mine (which I was never able to get to until Astronomy-I lost 3 galley/ settlers before giving up). I was also unable to reach any other civs (I sank 5 galleys trying to break through a sea lane to the SE of me, but it only opened into ocean).

    I finally made contact around 1200 AD, when an American galley came in sight of my shore. I was in the ancient ages, while the other 3 major civs were well into the middle ages (Japan, Spain, and as I would soon find out, the Iriquois were even more backwards than me). I was in solid 4th by most measures, except for a few demographics: rating-2, population-2, GNP-3 and productivity-3).

    Also, this island was essentialy the only major landmass on the map directly on the equator (except for the Iriquois).
    The land stats were:
    4% forest (2/2 plains/grass)
    28% plains
    2% grass
    35% jungle
    21% mountain
    11% hill
    1% lakes

    Bonus/luxuries/ strategic:
    2 whales, 2 fish, many BG (once uncovered)
    7 ivory, 7 silk, 4 gold
    2 iron, 2 coal, 1 oil, 2 uranium (no rubber )

    One hut. Which gave me barbs. All in all, a developmentally challenging start (though not as bad as Aeson's iceberg game. Once developed, though it had many advantages.

    All of this intro has a point. I was able to become the contending world power (2nd to America) during the early industrial ages for one reason: high productivity. Despite having the 4th total area, I dominated all of the productivity measures (GNP, shield output, and productivity). This led to start developing what I think are a few useful metrics for monitoring the development of an empire using the demographics as an aid.

    To get these measures, you need to gather a few numbers first:

    -land area (la) from the demographics/100 is the number of tiles under your control)
    -number of worked tiles (wt=# of pop. icons + # cities)
    -productivity number from the demographics (p).
    -total food shortage (fs)

    Using these figures:
    -land usage: lu= wt/ta
    -average tile production: atp= (p+2*cities-fs)/wt
    -%tile efficiency: %te= atp*100/n (n=4 prior to railroad, 6 after)

    There are a few interesting (IMO) implications from these numbers. The first regards land usage. During the early game, close city spacing will give higher LU numbers, as your cultural borders will be smaller than with a loose or OCP city spacing. While REXing, the LU will tend to stay low and flat, since the land area is steadily increasing, while wt is increasing at about the same rate because of settler/ worker builds (~.2 for an empire of 1 pop cities with no cultural expansion). Also, steady cultural expansion will tend to even out the effects of pop. growth. Again, city spacing is a consideration here as there will be fewer "cultural holes" in your empire with a close city spacing than with a looser city spacing, so the cultural expansion will mostly have an effect at the borders of a tightly packed empire, while the first expansion of an OCP empire will have a significant effect everywhere. LU for a tight spacing empire will also tend to peak faster. A low LU can also be an indicator that you're a brutal dictator, building you're empire by crushing the people with endless forced labor .

    The ATP number is also a useful indicator, especially during the early game. It shows how well you're using your workers. Since the top number in the formula tells the total food, uncorrupted commerce, and unwasted shields, it implies that workers probably shouldn't spend too much time developing those places in your territory that are completely corrupt. This has to be balanced with the primary consideration of getting a military road network hooked up, along with strategic and luxury resources. It will also be reflective of how resource rich your territory is. Again, the effects of city spacing style show up here. A broadly spread empire will have lower ATP because of the relatively higher effects of corruption/ waste. This needs to be balanced against putting cities in strategic locations like chokepoints, as well as those cities placed in crappy areas with the hope of finding strategic resources there later.

    Finally, the %TE can tell you how well developed your empire is (obviously related to ATP), and how much "productive terrain" it contains. My choice of 4 and 6 as numbers comes from the "typical" output of standard terrain, and is especially important under despotism. Excluding bonus tiles (and rivers, except for flood plains), most useful or productive terrain will generate four resources if properly developed:

    (cracker has a similar analysis at CFC in the opening plays article in the war room)

    grass (mine/rd.): 2f/1s/1c
    plains (irr./rd.): 2f/1s/1c
    forest (rd): 1f/2s/1c (grass/plain/tundra)
    hill (mine/rd.): 1f/2s/1c
    lake: 2f/0s/2c
    flood plain (irr/rd.): 3f/2c ("best" basic terrain output under despotism, unless you've sat and watched your pop. wither away due to disease 3 turns in a row which happened to me in AU101)

    Less useful:
    Mountain (mine/rd.): 0f/2s/1c --3
    tundra (mine/rd.): 1f/1s/1c --3
    desert (irr/rd): 1f/1s/1c --3
    desert (mine/rd): 0f/2s/1c --3
    coast (before harbor): 1f/0s/2c --3
    jungle (rd): 1f/0s/1c --2 (grass after clearing)
    sea (before harbor): 1f/0s/1c --2

    Once out of despotism, flood plain, hills and mountains improve, with hills becoming 1/3/1 (5), flood plains 4/0/1 (5), and mountains 0/3/1 (4).

    A similar analysis post-railroad shows the majority of productive terrain types going to 6 total output. So having a high %TE tells you that you have a decent mix of the "good" terrain types (or crap with a lot of bonus tiles ), it's properly developed, and that waste and corruption are low overall.

    There are probably a lot of other implications to be gleaned from an analysis of the demographics, and the relative importance of all this greatly depends not only on your playstyle, but on your relative skill as a player. It also depends on whether or not you subscribe to the notion that productivity and resource management are the basis for any sucessful game (I do, and I imagine that it becomes even more important at higher difficulties facing the AI's production bonuses). I've found the utility of doing this not in the actual numbers themselves, but in their changes over time, and at significant points in the game (getting the pyramids, or Hoover's or building you're FP, and watching your productivity explode, getting a better government, etc.). It can also be an object lesson for those trying to improve their game. It's probably clear that I favor a close city spacing. The value of this was driven home to me by my Gilligan's island game, where I had little choice but to make maximal use of the land I had (in the end, I think there's only two unused mountains near the south end of the island). With city infrastructure in place (and I don't have police stations all over yet), corruption is ~5% of GNP. I also think it can help the beginner to evaluate the impact on decisions they make in the course of the game, or to guide choices (gee, should I go for nationalism or steam power? ).

    Anyway, sorry for the longwinded diatribe, it's just something I've been toying with, and I hope somebody out there finds it useful (if you do, ask me sometime about the info I'm trying to squeeze out of the domestic advisor)

  • #2
    -number of worked tiles (wt=# of pop. icons + # cities)

    Well this is true if all citizens are working. If any are specialist it will not be accurate.


    • #3
      Good call. I knew there'd be something I'd miss. It should be pop. icons+#cities-#specialists.

      One thing I meant to add before was that you will have a low score with this close placement, as the territory is smaller (until you start conquering). I saw this mentioned in the Ralph's camp placement thread as a possible advantage in MP as your opponents will consider you less of a threat.


      • #4
        Anybody every watch Emeril Lagasse on the Food Network?

        Pork Fat Rules!!


        ATP RULES!!

        (and I agree... tight city placement encourages maximal use of the best tiles (assuming they are worked properly))

        Not sure I understand %TE well.

        Excellent post.
        The greatest delight for man is to inflict defeat on his enemies, to drive them before him, to see those dear to them with their faces bathed in tears, to bestride their horses, to crush in his arms their daughters and wives.

        Duas uncias in puncta mortalis est.


        • #5
          Thanks Theseus! A good review from you tells me I'm on the right track. The %TE is just a rough guide to how well developed all your worked tiles are. If it's near or better than 4 under despotism, it probably means you're only working developed tiles, have low corruption, or a load of bonuses (or some combination thereof).

          And yes, porkfat does rule! My mom's baked beans wouldn't be the same without it !

          Edit: I just realized that I put in 6 for the post-railroad divisor, when it should be 5. Grass, plain and mountain will give 5, hills will give 6, fp will give6, and tundra and desert will give 4 (probably no jungle or forest left by that time unless you're keeping it around for aestetic reasons or it's on a tundra, or you desperately need shields or a defensive bonus). So I guess that indicator gets a little less reliable in the industrial age, but it's still pretty meaningful under despotism (esp. for a newbie to the game like me).


          • #6
            The chief advantage of these formulae is that they make the player think and evaluate. I micromanagement is the key to victory, then having targets is the key to micromanagement. %TE is a useful formula the rest of us can manage to. We were doing thisanyway, but without a metric. Way to go!
            No matter where you go, there you are. - Buckaroo Banzai
            "I played it [Civilization] for three months and then realised I hadn't done any work. In the end, I had to delete all the saved files and smash the CD." Iain Banks, author