Quantitative Analysis of Civ Traits

[mharmless]

Even if you tried to qualify the extra sheilds from industrious, you are looking at 13 vs 12 sheilds per turn. Not very signifigant for wonder building.

A great leader is INSTANT. 300+ sheilds, right now, when it matters. This time factor must be included, the time value of that wonder is not insignifigant. 300 turns earlier is not counting it twice... that 300 turns would not have benifited from the wonder/palace otherwise. It is too large a block of time to ignore.

[nato]

rah - Yeah I'm pretty up front that it is in no way close to perfect. Just talking and trying to get a vague idea. I'm not going to try to perfect it ... but I will try to get what conclusions and info I can out of it! Learning from other people is why I am at this forum, like you allude to.

6-10 changes! Wow even when I'm religious I don't do that. Either I'm in full war mode (communism), or if its a quick grab, I stay in democracy and try to end it quick ... maybe I am being too pedestrian though.

One note about that, the faster government change is a tough advantage to try to maximize. You still lose 1 turn of production ... so in a way by trying to maximize that advantage, you are maximizing 1 lost turns for yourself.


mharmless - I see what you are saying about the wonders, but including it in the value for militarstic would still be double counting it's value ... the shields to build it should represent that. Just like I don't credit Industrious with whatever the extra shields actually build, but only the shields themselves. Oh well.

Cool points on the elite units being better than veterans. I'd have never thought of that. Thanks for the insight.

I really would like to know how many extra leaders Militaristic tends to get you. 4-8 wow ... I get about 3 with France ... I had figured 2 extra, maybe its more like 3-5.


player1 - Good point about selling techs. I did consider including it but had trouble deciding if I should. I was unsure if non Expansionistic civs couldn't still develop the tech normally and sell it.

You're right I probably should have included it ... wonder how much money you get for an early tech on average, and how many guys you can sell it to on average.

As to free cities ... I've played a lot of Civ3 and have seen only one Settler/city from a hut, which I got with Romans. I didn't figure it occured often enough on average to include. I could of course be wrong though.


Dale - It seems to me non-Expansionistic civs can explore fast too ... they just have to pay for extra Warriors to do it. They have to pay a lot of shields early to match the Scouts. Otherwise it's nothing other civs can't do ... they just have to pay more to do it, at the very start of the game.

Just how I see it, could be wrong.

Oh and to answer THAT , like I said above I think everything has a value in gold/shields you would be willing to trade it for, and that is it's value. Again just how I see it, could be wrong.

You guys are probably right, there must be more to Expansionistic ... it can't be THAT inferior to the other traits. OTOH it seems to be the least popular and regarded as weakest, so maybe it is not a good deal. Just trying to find out exactly what.


vmxa1 - Ok good points. Actually those were a lot how I felt when I was France only!

Libertarian - Thanks. About the workers though, I think they can be measured.

A non Industrious civ can do the same thing as an Industrious civ simply by building two workers for each of the Industrious civ's one. Therefore, the value of the faster workers is one free worker per worker built. How you use the workers, and all those intagible benefits, are irrelevant ... you can use non Industrious workers the same ways as you described ... you simply need twice as many. The bonus to Industriousness is in not having to build twice as many, which is just saving the costs of the workers. No intagibles about that particular thing.



Overall, I think most things really can be quantified ... I am just not smart enough to think of everything and figure out how to account for it all. I just find the quantification attractive because it is so revealing ... if one civ trait is in fact better, I would like to know. If, for instance, Industrious isn't as fantastic as it seems, that is good info.

In any event, this all has helped me to see where the advantages of the traits are coming from and just how much they may actually be worth.

[Libertarian]

It really is a good analysis, Nato. As you disclaimed, however, their are so many intangible factors. Consider the French, for example. Commercial and industrious.

Thanks to lumberjacking, you can greatly increase your science expenditures because a small army of workers can get your improvements completed or nearly completed in a single turn. They can swarm around all over the place, moving from city to city, building libraries, universities, cathedrals — practically anything and everything except wonders. I sometimes have science set at 80% and don't skip a beat. Especially as the game progresses, I routinely achieve the four turn minimum for tech advances.

In the very late game, this can be exremely valuable as you're trying to build up brand new cities taken either by conquest or culture. Plus, the workers can quickly build a dynamite infrastructure of roads and railroads that make defensive war a breeze. Zzzzappp! You move your unit from one end to the other instantly to defend key tiles. All the while, these roads are contributing to internal resources (shields, food, etc.).

Workers also can be a great assistance in wartime, literally laying down the roads and railroads as your army advances. They can swarm a newly conquered city and build it up to full capacity in just a few turns.

Now, combine all this with the extra commerce and lower corruption (albeit not a lot lower) and that would account for the reputation the French have as a building Civ. But remember that you can build not only buildings, but armies as well. Yes, you get leaders and promotions less frequently, but you do get them. And you can crank military units out fast and furious with a sufficient hoard of workers and enough cash. Especially with democracy.

In a recent game, I was ambushed by the Aztecs. They just attacked out of the blue! Using my workers as a sort of collective tool and my extra income, I was able to crank out a force sufficient not only to repel the Aztecs, but to follow them back into their territory and pillage the hell out of everything they had. It was many many years (because of their aggression) before war weariness set in for me, and I had to sue for peace. But by then, Montezuma was willing to give me nearly everything for peace, since I had destroyed practically all his roads, mines, and irrigation.

Interestingly, when I finally withdrew (my people were starting to get really pissed — "All we are saying is give peace a chance." ), I surveyed the former great Aztec empire. Lordy, it was weird looking. Littered with size 6-12 cities and scarcely a road to be found. They had to start all over. Montezuma remained furious the rest of the game. But I digress...

[mharmless]

I usualy get four to eight leaders as a millitaristic civ. Usualy play as the Romans, and about the only time I'm not fighting sombody is that lull between the death of the first civ I contact and locateing other targets.

[vmxa1]

This is fun. I should say I will put universities and labs in every city I can. What I am saying is that Labs do not become available until modern ages and I can only get them in a few cities before game is won, hence not much value (I do not need more culture by then as I will have more than all others combined). A tank is more useful by then as I am getting tech as fast as I need anyway. Universities are in the middle and are up in core cities, but it is getting late by that time.
Military, I have so many elite troops with out it, why should I concern myself? Leaders? They are rare and I did not do any better with military. Since I am at war 80-90% of the game and have battles a plenty, I get lots of elites (start with vets (barracks)).
Religious and the 1 turn anarchy gambit, who needs it? I mean so many civs favor Monarchy, it is a useful for a very long time. I switch once to Monarch and once to either Rep or Dem, depending on which one my civ does not like. I did say the religious is a good trait and if you want to play a civ that uses it, that will be fine. I just do not see any big value in Military or Expansion. A scout is not much of bonus in most maps. You will likely get to the same huts before the AI with a warrior as the scout (maybe on a huge map it pays off). Finding the other civs will occur a little later, no big deal. Exposing the map faster is not a worth while thing either as you will not get settlers there any quicker.

[player1]

Plus, if you are expansionistic, don't trade communications.
Be MONOPOLIST.

You get a tech, and then you sell it (and only you) to all others, etc...

[Dale]

Expansionist

From reading the forums, I think a LOT of people miss expansionist's BIGGEST advantage.

Scouts = Faster exploration.
Faster exploration = Faster Contacts.
Faster Contacts = Faster trades (resources/techs).
Faster trades = Faster science (either from commerce or the techs).
Faster trades = Early attacks stalled (Other civ is happier with ya).

In the early game, this can be very important. In some situations, the added diplomatic/trade benefits to trading has saved my butt from early death. At least long enough to get my temples/spears out there.

How do you quantify THAT advantage in gold/shields?

[player1]

You didn't take in account free cities and early contact profits in calculations for Expansionistic civs (especcialy on larger & emptier maps)

In my own expirience I know that they are much more usefull then you think.

Especially since you can keep up in expansion with AI.

As anybody remember games in civ2 when one player starts with 2 settlers (other with one) & wins much easier.

Expansionistic is similar to that.

[mharmless]

I counted the full wonder because in the early game (for me at least) the only way I get wonders is if my leader allowed me to get it immediately. Even if it could be had via building, the time it takes to make it should be factored in, and then the time value of that wonder can be added to the pile too. If colosus takes your archtypical city 100 turns to build, than the leader has caused 100 extra turns of effect on that wonder. Same for all wonders.

With the forbiden palace, I didn't caculate it's value for the entire game, but just for the 300 turns it would take a backwater berg to construct it. With the leader, it is done 300 turns earlier and turns that entire area into a powerhouse immediately. Those 300 turns are then all productive in all those cities, hence the insanely large sheild value.

Edit: About the not loseing the unit when it takes four hits. Four elite spearmen are superior to five veteran spearmen. If attacked by five veteran horsemen, ignoreing defensive mods, then the five veteran spearman each have about a 50% chance of dying. The Elites though, they will only lose about 20% of the time. The four elites will hold the ground better than five vets (ignoreing in battle promotions to elite from vet). Particularly against horseman, as if we trade hits equaly once it loses 3 it will be at one hitpoint and retreat, wheras my spearman is at 2 hits still. If it was a vet, then they would both be at 1 hp and the horse would attack untill death (50% chance I lose the vet at this point, 0% I lose the elite). Less units lost, less resources expended. In this case, I save 20-40 sheilds that would have been spent replaceing the extra deaths among the vets.

[rah]

Please don't kill yourself making it perfect, since I doubt it ever could be. But tweaking a little provides some great food for thought.

It has already served the purpose by making us think on how to quantify it. By attempting to shoot holes in what you've done, it's made us discuss exactly some of the issues and makes us question if we're choosing the right civ, or taking proper advantage of what we choose.

For the record, if you're religious I'd expect at least 6-10 government changes if you want to maximize your advantage.
But the value can't be compared to non-religious since you wouldn't think of doing it that often. Oh... it's so hard to quantify.



RAH

[nato]

Sevorak - Thanks, I'll try to add in the upkeep costs saved tonight or tomorrow. Good catch.


Grim Legacy - Thanks, though the "infallible Civ-intuition" may be in a bit of trouble!


randomturn - Well I don't generally think of gold for building, I usually use it for science ... I guess commerce would be the more technically correct word than gold, but 1 beaker = 1 gold.

There are definitely unquantifiables ... but I think most things essentially have a gold/shield value. Like in real life, many things are not for sale ... but realistically speaking, you can use dollars as a rough measure for their value.

What I mean is, there must be some amount of gold/shields which I would rather have than the unquantifiable values of a trait ... that amount is their value in gold/shields. Its a decent unit of account.

You're right of course though, this whole thing is an attempt to put a number on unquantifiables, and it's nowhere near perfect. But by trying to translate the two bonuses each trait gives, its not unreasonable.


vmxa1 - You're right, play style counts big time. However I made an assumption that most people would put all 3 science buildings in all their GOOD cities (not the hopeless outliers). At least I would. If you don't, then you are right, Scientific's gold/shield value would be lower. Happily, that particular adjustment would be quite easy to make. (On a sidenote, give Universities a shot - 4 culture!)


Arrian - You're right, playing to the trait will increase it's value. More fighting increases Militaristic, more Temples Religious, and so on. I was trying to find which was worth more if all else was equal, or ceteris parabis as they say. This would help give an idea which trait gave you more return in a generic situation. If it gives more in that case, then when you play to the trait, it will give even more.

However some traits may well give more return if you play to them than others. What I mean is, if you go all out war, Militaristic may increase in value more than Commercial would if you went all out gold. Hard to say. I'll give this some thought.


mharmless - As Sevorak first noted, I do have to take lower unit upkeep costs into account.

However, I am not so sure I follow the point "I don't lose that unit when it takes four hits." I think having 20% extra units would simulate this at least reasonably well.

As to the wonders, very good point, and good math! However I think adding in the value of the wonder would be what they call double counting, which is an easy trap. The price of the wonder is the shields it takes to make it, not it's effects. For instance, if I counted the value of the wonder for Militaristic, then I would have to count the value of whatever the extra shields were used to build for Industrious, the extra gold was used to buy for Commercial, and so on. I think that is indeed double counting and not right to do.

Good points though, I'll have to think about it, but I do believe the wonder thing is double counting.


rah - Good points ... I think I have some answers.

For the workers, the double speed has a value ... I tried to calculate it and credit Industrious for it.

For the Religious anarchy, that is correct, if you play to the trait you will get more out of it. Again I was trying to come up with a generic situation. My one early change + one late change setup was to simulate going to either Monarchy or Republic early, and then Democracy (or less likely Communism) later.

I did say "More or less revolutions will change this value, as will size of your empire." ... I think two is reasonable for a normal game ... maybe it should be three though.

Leaders are valuable ... they are worth one wonder apiece. I credited them for the cost of the wonder. I think the 20% extra units measures the value of extra promotions, albeit very roughly.

I think my assumptions on the game, 25 good cities basically, are reasonable. I am definitely open to suggestions if ppl think that is not a good average. It is supposed to be a normal game that gives all the traits a level field to show what value they give under normal circumstances.

"it is most important that you take max advantage for any characteristics that you choose."

You are dead on here - that is a great conclusion to point out. Doing that basically increases the gold/shield value your trait is giving you.


Thanks for reading and taking the time to make comments everyone! I do appreciate it.

[rah]

Good work, but of course I'll add my 2 cents.

Industrial, only is real useful early. Most workers in my empire are chain gang, while my workers are adding to pop in conquered cities.

Religious, I will change goverments quite a few more than 2 times in a game if I only have one turn of anarchy. Great for taking advantage of those war lulls and slipping into democracy, then switching back to com for fighting, or rushing those temples and units after a major offensive. Better to rush buy with those foreign citizens then just wasting them by starving them.

Militeristic, Yes those extra leaders are big. and field promotions early in the game can sometimes spell the difference.

In all, the QUANTIFICATION is good, but some of the assumptions (like the game) could be tweaked depending on how you play. But it is most important that you take max advantage for any characteristics that you choose.

RAH
Good effort.

[mharmless]

I'd have to argue that millitaristic is undervalued here. Having an elite unit is not the same as having 20% more of that kind of unit. First off, I'm not supporting .2 more units, and more importantly I don't lose that unit when it takes four hits.

Next, in regards to leaders. Lets assume you get four leaders over the course of the game. If you get your second arround 1 AD and use him to rush a forbiden palace, then you will be increaseing the output of 10-15 cities from nothing to 20-100% production value. Call it 300 turns earlier than otherwise, and by your numbers they would be all size 12 cities. You list production as 1 sheild 1 gold per citizen when abstracting this, so lets do that.

15 cities X 12 production X 2 (to count gold too) - 2 (existing 1 gold 1 sheild) = 358

358 (gold/sheilds) x 300 (turns) x .6 (average yeild after corruption) = 64,440


And above all, lets not forget the time/value of precious precious wonders. If you get the pyramids as a direct result of a leader, then that leader has cause growth rate to double. Assumeing to goes from 10 turns to 5 (20 food needed to only 10, +2 more than needed to support population), and that you get the pyramids with your first leader, and that you rush build with citizens.

In this case, you are getting some 500 turns of 10-15 cities being able to rush build (40 sheilds) every 5 turns instead of 10. We go from 50x15 rush builds to 100x15 rush builds. This means the pyramids have yeilded you an extra 30,000 sheilds because you got them first as a direct result of being a millitaristic civ.

Or hell, colossus. If you get that from a leader than you are getting 12 extra gold in that city for about 300 turns (average shelf life). 3600 value.

You can see the point. Leaders are extremely valueble and a millitaristic civ will get them more frequently and earlier than a non millitaristic.

[Arrian]

One man's library is another man's barracks...

It is, in fact, all relative.

Vmxa1 - I play differently from you. Nearly ALL of my cities have both a library and university (if they lack one, they're building it) and later, research labs. The science improvements were very important in CIV I and II, but even more so in CIV III, due to culture.

Nato - good effort. I think the beauty of the game, however, is that your attempt at quantifying (down to gold/shields) the civ attributes was an impossible task. People play differently, each game develops differently, etc. This is a good thing. Someone who values the militaristic trait is probably going to make use of it, and fight, for example.

Personally, I think scientific and religious are the two best, with commercial and industrious coming in 3rd and 4th, respectively. But that's just me, and I can imagine particular situations where the other traits might just be pretty useful.

-Arrian

[Ray K]

Originally posted by vmxa1
I want to say that Scientic value looks good, but in reality is much less. This is due to the large percentage of the labs and university. These are not all that commonly built. Labs are likely to be in few cities as they come so late in the game. Universities will be some, but not most cities and in fact even libraries are not in every city. This really blunts its value, although I still like it.

I think it depends on your style.

Right now, I rush-build temples, then build libraries, harbors, aqueducts, marketplace, universities in about that order.
About midway through the conquest, courthouses come right after temples and cathedrals come after aqueducts.

Universities are a must-build for all of my cities by the time they're size 12. With all of the rush-building in the newer cities, most of my empire has universities by the Industrial era.