Rough payoff times and exchange rates

[ellman]

This reminds me of an analysis of Civ4 Colonization I started but never finished. Its a very simple game by comparision to traditional civ so this type of analysis is a lot easier. Basically I set the goal of the game to be increasing your population. Then I converted the "costs" of every unit and decision into gold. With everything in terms of one variable you can then compare the efficiency of each choice easily.

Here is a thread I started about it. If you don't understand Civ4 Col I could give you a crash course in it.

http://forums.civfanatics.com/showthread.php?t=326011

For example

At a tax rate of 50%, normal speed, 750 colonist cost in Europe, manufacturing cigars with your average tobacco harvester producing 10 tobacco (price of tobacco set to zero to make comparison to farming clear), a free colonist working as a cigar maker in a 6 production building makes ~12 gold per turn when the price of cigars is 8(add 3 or subtract 3 for each change in price). A free colonist farming working a 4 food tile makes 7.5 gold per turn.

[luddite]

I'm morbidly curious to see how I rated there in the 19 graph. My general impression is that I managed to be in the top of the pack most of the time in GNP/tech, at the expense of expansion. In my defense, first game.

Here you go :P Looks like your growth was also fairly constant at 3.5-4% each turn. However there was a long stretch from 14-27 where you didn't post any demos, so that might affect the result. And you also built some military in the middle to go after Mr. Nice Guy.


Alternatively, this might be a better way to measure growth. I combined 1.5*crop yield plus Mfg as "FpH", and then added up all the beaker costs of the techs you researched and what turn they came in on (roughly). Doing it this way shows very nicely constant growth of 4.8% beakers per turn, 3.3% food-hammers per turn.

I thought a big reason for the Oracle -> MC was that it usually locks down the Colossus for the Oracler, as it's a darned cheap wonder hammer-wise, the big expense for Colossus is the tech. I do think on a land map, a toroid if Org, or anytime with Spi CoL as an Oracle target makes a lot of sense too. Of course, the leets Oracle Feudalism, Civil Service, Theology, or Machinery.

Yeah the Colossus can be really good on the right map. My rule of thumb is: river cottage > colossus coast > non-river cottage. Using seven's discounting scheme, a river non-PP cottage is 3.5 bpt, colossus coast is 3bpt, and non-river non-PP cottage is 2.5 bpt, so that math checks out.

[zakalwe]

Historical numbers for MFG, CY, GNP, etc. should already be in the save files, since you can view them in the in-game graphs. So it should be feasible to make a tool that could grab any save game and inspect the average growth rate of all the civs in it. Or alternatively make a mod.

[luddite]

Historical numbers for MFG, CY, GNP, etc. should already be in the save files, since you can view them in the in-game graphs. So it should be feasible to make a tool that could grab any save game and inspect the average growth rate of all the civs in it. Or alternatively make a mod.

Yeah that's a good idea. I have no idea how to do that though.

[luddite]

Hypothesis: For any civ expanding onto (mostly regular) land tiles, the growth rate of their FpH will be equal to the average FpH available to them according to Novice's map tuner (ignoring water tiles).

[Krill]

What's FpH again?

[luddite]

What's FpH again?

1.5 times food plus hammers, as measured by the demographics screen.

[Krill]

Hmmm, that would give a few interesting results, comparing between different players on some of the more balanced maps.

[SevenSpirits]

I feel like adding crop yield to other demographics is double counting. Crop yield is mostly based on your current population (2 food per) though there is also surplus food thrown in. The drawback of using crop yield to measure your civ's output is it provides no indication of tile strength. Maybe you're only working unimproved grasslands and the actual net output of your population is negative.

MFG + GNP is ALSO a measure of your population/output, though it focuses on the output. And the drawbacks of using it are 1) GNP is often modified by spurious crap like whether you have a tech selected, and culture, and 2) it is blind to food surplus.

If you add your food + mfg + gnp together (perhaps multiplied by some coefficients so they are comparable in value) you are getting the net output of all your citizens plus twice your pop size in food (plus some GNP artifacts). Since the average citizen's net output is less than 2 food, more than half the value you've calculated is noise.

Oh and multiplying gross food by 1.5 seems like it will give very misleading results.

[luddite]

I feel like adding crop yield to other demographics is double counting. Crop yield is mostly based on your current population (2 food per) though there is also surplus food thrown in. The drawback of using crop yield to measure your civ's output is it provides no indication of tile strength. Maybe you're only working unimproved grasslands and the actual net output of your population is negative.

MFG + GNP is ALSO a measure of your population/output, though it focuses on the output. And the drawbacks of using it are 1) GNP is often modified by spurious crap like whether you have a tech selected, and culture, and 2) it is blind to food surplus.

If you add your food + mfg + gnp together (perhaps multiplied by some coefficients so they are comparable in value) you are getting the net output of all your citizens plus twice your pop size in food (plus some GNP artifacts). Since the average citizen's net output is less than 2 food, more than half the value you've calculated is noise.

Oh and multiplying gross food by 1.5 seems like it will give very misleading results.

Well on a micro level you're correct- a corn farm is obviously better than working 3 unimproved grassland tiles, even if they both produce the same total crop yield. But I think it's safe to assume that no one is going to do that- or, if they do, they're going to grow very slowly. On a macro level, we already use crop yield and mfg to estimate how powerful civs are, so it makes sense to combine them.

I think adding together total food + mfg works well for these purposes- it lets you estimate how fast your civ is going to grow in the future, and therefore estimate the exchange rate between immediate value and per-turn value. It's a lot easier than figuring out how to value beakers vs. food vs. gpp, anyway.

And on a micro level it makes sense also. You can very easily convert 2fpt to 3hpt by switching from a farm to a grass hill. You could also convert 3fpt to 4hpt on a plains hill, but that's a bad trade so most people avoid it. Or you could convert roughly 13 stored food to 20 hammers through slavery, so that's another good reason to count total crop yield instead of just the surplus.

More importantly- it gives plausible results, and there's no easy way to figure out the total number of citizens that someone has on each turn.