First, warmongering is obviously quite powerful due to the ease of creating and maintaining an army - and I voted it is easy because at the easier difficulties it is the most piece of cake (you don't have to tech the whole tree or sit ending turn forever). I guess I'm discounting the broken AP system, but I don't even consider that a real win...Warmongering is also obviously the easiest to set up a win based on initial settings (choosing Inca, Rome, choosing your opponents, Pangea, etc...)
In response to Skallagrimson (and others) I have to point out that unfortunately the reason the AI isn't "smarter" in many cases is because of purposeful, huge imbalances in favor of the human. In fact, I would wager most posters here would take for granted the following points I have here (in fact, as I've said before, these are some things that I'd like to mod - maybe if I keep saying it I'll stop being able to makes excuses for not learning how to and doing it yet):
1) The human has zero diplomatic limitations, both contrary to AI competition and realism in the game. I never played Civ before CivIII, and of course the games had many fewer features back then. However, I remember hearing of one thing, that most players seemed to think was a great change in future versions; however I wish we still had this mechanic: in democracies the player's "Senate" could control what your civ had to do. With all the advances of Civ IV I'd love to have an expanded mechanic on this - serious (like +5 :angry apiece) penalties for religious war (and not the AP system, mind you, I'd be happy with completely overhauling/eliminating most of it) attacking another Friendly civ, etc... The human can currently ignore all of this (it's just a game) while all of the AI diplomacy is built around it - so that's essentially part of the AI not playing to win.
2) The AI likewise still does not seem to recognize what the real "threats" to it are. A heavy reliance on the power graph is part of this, and things like being forced to be friendly for "sharing religion." Also, as many less experienced players aren't aware (in fact, I didn't know the exact details before coming to CFC but suspected as much) there's also the warmonger respect factor, which relatively empowers warmonger AI's.
However, I think there's an incredibly easy fix out there for this (in terms of concept) and this could actually fit right into the system. The simple idea is: All AI's get suspicious (diplo penalty) with any other civ who has had success in x number of aggressive wars (capitulating vassals or eliminating someone, is one place to start)- and a simple message like "-X We can't trust your violent ways"). Of course this matters little on something like a dual map; however in a more normal game by the time the human (or somebody like Monty if he's succeeding) has killed other civs, ALL remaining civs get a penalty and grow suspicious. Currently this can all be avoided ("You declared on our friend is sad and even broken in the case of vassals"). Conquest obviously becomes a lot harder, just by letting the AI overcome "Hey, Earthling the Hopeless has killed the last 3 civs in a line towards me, but we share religions so let's be best buds."
3)This last point isn't so much about the AI/human relation but is something that I still find, how should I say, unrealistic in civ - units are way, way cheap in upkeep. In fact, in marathon they are even cheaper to build, giving the human a huge advantage in warmongering there. All throughout real history the maintenance of a standing army was a huge cost (even into modern times this can be said for many countries). But in civ, you can fund a ridiculous number of troops basically by the output of a single city in your empire. This isn't a simple change but a system which made all military, for all players, more costly (and likewise unit strengths and such may be adjusted) could have some use. I'd put a much greater emphasis on drafting/levying - in fact, one thing I missed in CivIII that I never had a problem with was that you could draft a much larger number of troops (modern times).
My idea, then, is maybe increasing flat upkeep rates, but then giving certain units like archers (city defense line) or a building for cities that allowed units to "garrison" for no upkeep would help in ancient times - the defensive army is cheaper than the swords. By the modern era I'd make rifles/infantry more mass-draftable - in contrast the real strong units (and more costly) would be tech like tanks, planes. Also, I think this could also be a fix to removing inflation entirely - you may be paying 5 gpt per tank but at least it's an honest choice - if you're a total pacifist you could create a high-tech utopia free of inflation. Overall, at any rate, getting a massive army (for both the human and AI) wouldn't be so cheap - but at the same time if someone does have a strong offensive army it would be much harder to counter.
In response to Skallagrimson (and others) I have to point out that unfortunately the reason the AI isn't "smarter" in many cases is because of purposeful, huge imbalances in favor of the human. In fact, I would wager most posters here would take for granted the following points I have here (in fact, as I've said before, these are some things that I'd like to mod - maybe if I keep saying it I'll stop being able to makes excuses for not learning how to and doing it yet):
1) The human has zero diplomatic limitations, both contrary to AI competition and realism in the game. I never played Civ before CivIII, and of course the games had many fewer features back then. However, I remember hearing of one thing, that most players seemed to think was a great change in future versions; however I wish we still had this mechanic: in democracies the player's "Senate" could control what your civ had to do. With all the advances of Civ IV I'd love to have an expanded mechanic on this - serious (like +5 :angry apiece) penalties for religious war (and not the AP system, mind you, I'd be happy with completely overhauling/eliminating most of it) attacking another Friendly civ, etc... The human can currently ignore all of this (it's just a game) while all of the AI diplomacy is built around it - so that's essentially part of the AI not playing to win.
2) The AI likewise still does not seem to recognize what the real "threats" to it are. A heavy reliance on the power graph is part of this, and things like being forced to be friendly for "sharing religion." Also, as many less experienced players aren't aware (in fact, I didn't know the exact details before coming to CFC but suspected as much) there's also the warmonger respect factor, which relatively empowers warmonger AI's.
However, I think there's an incredibly easy fix out there for this (in terms of concept) and this could actually fit right into the system. The simple idea is: All AI's get suspicious (diplo penalty) with any other civ who has had success in x number of aggressive wars (capitulating vassals or eliminating someone, is one place to start)- and a simple message like "-X We can't trust your violent ways"). Of course this matters little on something like a dual map; however in a more normal game by the time the human (or somebody like Monty if he's succeeding) has killed other civs, ALL remaining civs get a penalty and grow suspicious. Currently this can all be avoided ("You declared on our friend is sad and even broken in the case of vassals"). Conquest obviously becomes a lot harder, just by letting the AI overcome "Hey, Earthling the Hopeless has killed the last 3 civs in a line towards me, but we share religions so let's be best buds."
3)This last point isn't so much about the AI/human relation but is something that I still find, how should I say, unrealistic in civ - units are way, way cheap in upkeep. In fact, in marathon they are even cheaper to build, giving the human a huge advantage in warmongering there. All throughout real history the maintenance of a standing army was a huge cost (even into modern times this can be said for many countries). But in civ, you can fund a ridiculous number of troops basically by the output of a single city in your empire. This isn't a simple change but a system which made all military, for all players, more costly (and likewise unit strengths and such may be adjusted) could have some use. I'd put a much greater emphasis on drafting/levying - in fact, one thing I missed in CivIII that I never had a problem with was that you could draft a much larger number of troops (modern times).
My idea, then, is maybe increasing flat upkeep rates, but then giving certain units like archers (city defense line) or a building for cities that allowed units to "garrison" for no upkeep would help in ancient times - the defensive army is cheaper than the swords. By the modern era I'd make rifles/infantry more mass-draftable - in contrast the real strong units (and more costly) would be tech like tanks, planes. Also, I think this could also be a fix to removing inflation entirely - you may be paying 5 gpt per tank but at least it's an honest choice - if you're a total pacifist you could create a high-tech utopia free of inflation. Overall, at any rate, getting a massive army (for both the human and AI) wouldn't be so cheap - but at the same time if someone does have a strong offensive army it would be much harder to counter.
. That is why military wins with a lot of civs and/or in higher levels are harder........
