Civ 4: The Number Crunching Starts

For those of you interested in this sort of stuff, a CivFanatics poster, Gato Loco, has done some analysis of how the maintenance cost of cities works in this post over on the CivFanatics forum:

The nitty gritty appears to be that that maintenance starts to really hurt after 9 cities, peaks at 28 cities, but then falls again and plateaus out until 44 cities when it peaks again and then starts to fall again.

Also the distance from the palace at which maintnenace due to distance kicjs in is 4 spaces, diagonal spaces being counted as 1.5 spaces. This part appears to be a hangover from Civ 3’s distance corruption system.

*Edit to fix cinfusing typo

I just play the game by feel and don’t want numbercrunching to get the job done. Works great so far.

Agreed. I know that numbers really turn some folks on but all it does for me is slow me down trying to figure all that crap out.

“It was my understanding there would be no math.” :?

Thanks for the Link Mike, most useful Post I`ve seen on the Game anywhere.

Yes as I said I posted for people who are interestered in this sort of stuff. I know it turns some folks off,

Another useful thread from that site that fully opens up the hood on the combat system:

Important tidbits:
(Ratio is the ratio of the adjusted attackers combat strength vs. the adjusted defender’s combat strength)

Ratio value Odds of winning
>1.8 99%+
1.58-1.79 95%-98%
1.39-1.57 87%-90%
1.25-1.38 75%-80%
1.01-1.25 62%-75%
1.0 50%
0.80-0.99 25%-38%
0.73-0.79 20%-25%
0.64-0.72 10%-13%
0.56-0.63 2%-5%
<0.56 <1%

And, the value of first strike, from the related thread:

First strikes should be used for two different types of units: those that defend and the overwhelming attackers. As defenders are most likely stronger because of the bonusses, they are helped as they can fend of multiple attacks, still remaining strong. It’s very annoying to have dozens of units attack a city, where one single defender just shrugs of the attacks.

For the overwhelming attackers, it can keep you going. This especially applies to pillagers: they will remain healthy long enough to pillage that resource far inside the foreign borders, where units with the same strength but no first strikes will get slowly damaged over time to a point where an older unit can win.

Someone made the counterpoint that although in the above situations first strike allows your unit to avoid damage altogether & thus last longer, the true value of the ability comes when the two units are closely matched… then the attribute shifts the % to win from roughly 50% up to 63%.

But Spock, don’t you understand the feel of the game? All your damned Vulcan logic just destroys the aesthetic of the game. Just feel Spock!

You’re so…analytical; sometimes you just have to let art flow over you.

I have always wanted games that went the extra mile in concealing statistics from the player.

A role-playing game that merely tells you your characters is a “hardy and experienced warrior,” instead of “STR:18/32” or “his razor wit incites rage in the stupid,” instead of “INT:17” or whatever.

I tend to think it deflating to know these things. Medieval was more chaotic and unpredicatable before the relative unit strengths and modifiers propogated – afterwards, “stock” armies ruled and battles often played out like the turgid, pat, scripted battles you see in Hollywood movies.

And imagine if Morrowind was an MMORPG. Just imagining hundreds of people bouncing around everywhere maximising their strength.

Actively preventing mix-maxing is good.

I’m interested to know things like the 4-square radius before maintenance kicks in. If it just progressed at a steady rate based on distance, I wouldn’t care. But if there’s a sudden threshold, I want to know what it is.

Yeah – I think what I’m really asking for is a system that is complicated enough to not have precipitous modifiers. In an RPG, for example, the difference between 17 and 18 in a characteristic is important enough to warrant telling people about.

Trying to conceal it would just cause people to try and decipher hints. So having a much broader scale allows for more gradations, and more clever ways to convey the essential information without creating a legion of kremlinologists.

E.g., say we have a natural language description system, each strength point (or maintenance cost modifier) increases the probability of a particular word being used in the description of this state. With an “out of 20” scale for strength, or a “4 squares away” braking point for maintenance, the descriptions would merely be an annoying cipher.

But if the gradients were smoother, they become an insight. In a civ game, for instance, better advisors might provide more accurate descriptions, like “This city is far from the capital, has no port, and no road connections, and the culture is a bit foreign, so the maintenance costs of 10g is astronomical,” while a shitty administrator tells you “This city is too far from home and costs 10g a turn”

This way of doing things, I think, would be a great alternative to looking stats up on the internet.

/armchair game design

Exactly. Rather than having a 0 modifier up to 4 squares, and then a jump to one, they should just have a modifier of .25 per square, add it up and truncate the result. For one city it would be the same, but for 2+ cities within the 4 square radius there would be a maintenance cost, rather than it being free.

King of Dragon Pass did this - while there were likely mechanics governing the success and failure of the various things you would try to do, the game did a very nice job of keeping them from the player. Personally, I found this a bit frustrating, as I couldn’t really plan things the way I’d like to, but perhaps the real reason is that I’m just lousy at being a barbarian chief. Nifty game, though.

[size=1]And of course, there were the duck people.[/size]

I am the opposite. I want the numbers to be transparent but not a requirement to play. For instance its handy to know that for the Hunter in WOW agility is king. I don’t need to know the exact number crunch but if I wanted to know I could look it up. Or things like reputation numbers.

As opposed to EQ where things were so ambiguous it took hundreds of log parsing sessions to figure out "hey there is little to no difference for mana above 300 INT) or whatever.

Wow… I just read that article. I didn’t realise people did this kind of number crunching for non-MMOs.

Actually the CIv series has always been like that and the same probably applies to other popular starategy games. Just look back at the strategy articles for CIv 3 and CIv 3 on the same site.

Personally I don’t have the time or inclination to do the the research and maths nyself, but I’m interested in knowing what the numbers mean for playing the game.

Yeah, I think that’s key. I love the way in Civ 4 I can read the AI’s relationships to me very clearly, but I don’t have to sit there with a pen and paper to calculate when the best time to make this or that decision. I hated with earlier AIs that it didn’t matter what you did, they’d just randomly start to loathe you at some point. Now I can manipulate them, but in a way that is understandable and feels credible.

Actually, when it comes to diplomacy, the ability to see the +/- modifiers makes it feel a lot less arbitrary. That’s one instance where revealing the underlying math improves the experience.

  • Alan


I am the opposite. I want the numbers to be transparent but not a requirement to play. For instance its handy to know that for the Hunter in WOW agility is king. I don’t need to know the exact number crunch but if I wanted to know I could look it up. Or things like reputation numbers.

As opposed to EQ where things were so ambiguous it took hundreds of log parsing sessions to figure out "hey there is little to no difference for mana above 300 INT) or whatever.[/quote]

Something that might serve both our objectives would be a game that lets you know exactly what to do to get better, and tells you when you are done. I don’t mean “Do x to achieve y,” but just games that bother to tell you a bit about how the system works before you get going.

For example, I never know in any random RPG whether dexterity is useful for fighters. You always have to look it up. And then you have shit like “constitution,” which just makes me think of how likely the character is to puke if they see a nasty road accident.

IMHO exposing the underlying math has a tendency to result in games that are more scrutable to the casual player, as an exposed system will (tend to) be made simpler, cleaner, and more easily understood. Hidden systems in computer games have a tendency to become baroque, nearly impossible to understand without seeing the formula, and actually encourage min-maxing.

Hiding things IMHO doesn’t acheive the desired effect, and instead only gives an edge to the players willing to figure things out empirically (e.g. Dominions).