Why am I upsetting the other civs?

[vorlon_mi]

I'm a little unsure because of the sentence structure, but swords upgrade in PTW.

But even in Vanilla, which I played for some time, I would still be unlikely to build spears vs. swords. I agree that upgradable swords is nice, but the real challenge in Vanilla is to only build as many swords as you need. Too many and it could be a waste. Frankly, the upgrade path from sword to MDI to Gurilla is pretty weak anyway.

Whether spear, pike or musketman - even rifles IMO - they are just standing around watching the battle. Meanwhile, how much gold have you spent to upgrade a spear into a Mech Infantry?

I'm not saying a handful of defenders aren't nice to have, especially at the Infantry stages. But really, if you are fighting an offensive war (even defensively), then your 'defenders' rarely see meaningful action anyway (as Ataxerxes mentioned).

Granted; I had forgotten whether swords upgraded in PTW or C3C. Thanks for the correction.

I guess I just don't build enough military. My need for defenders springs from two experiences, and based on your comments, and those of others, I should use other means to correct these problems.

1) Early war counterattacks by the AI. I've been expanding thru the REX phase, and I don't have that many productive cities just yet. Still growing, not many shields per turn. I build an invasion force, and take a couple cities from my neighbor. The AI then sends a couple of horsemen galloping into my territory (we're neighbors, right?) and threatens my towns. My previous solution: build some spears so that I don't have to reconquer my own city. Better solution: build more swords, or horsemen, and leave them behind as the home guard. They can either go out and kill the AI counterattackers before they get to my towns, or stay home and use the town's defensive bonus to survive the attack.

2) Pesky naval invasions by the AI. In the middle or industrial ages, some AI will get frisky and drop off a shipload of knights or cavs. Again, I don't want to have to waste turns reconquering my own cities, so I can't leave them undefended. My previous solution: have a single defender, upgraded along the way, to prevent an easy victory. Better solution: leave an offensive unit in the coastal towns, rather than a defender. Keep a squad (say, 3-5) of mounted units in the homeland, ready to use the roads/rails to pounce on the invaders.

Added benefit, as several of you have pointed out -- building all offensive units will bias the AI's assessment of my relative strength, and may delay or deter some of their invasions.

 

[Raliuven]

MysteryX -

I agree most circumstances favors an immediate surge of settlers to claim land, but that initial surge will flop quick without workers to connect and improve the tiles between them.

I don't know how much map size impacts these decisions. I usually play on standard-huge maps - usually standard. I play up to Emperor, usuaully monarchy. I've rarely considered surrending worker production to pop out a few more settlers for the land grab. I suppose if your perferred map is Civ heavy and land light, you may have a point.

This is a cost benefit analysis. Every worker you fail to make to improve a city results in a city that has wasted potential - and may even be totaly unproductive. Meanwhile your only productive city is pumping out more settlers to make more unproductive cities.

Or, you can produce a settler and a worker to follow it. The settler founds the city and the worker makes the city productive. Soon you have two productive cities spittting out settler/worker pairs. Then four productive cities, etc. Bonus, when you first workers have improved all the needed (neeed, not ALL) tiles, they can then bounce to the next city and excellerate the process.

Realistically the first good site I find becomes the settler factory. Second best becomes the worker or worker/warrior factory (depending on food quality). I worry about defense last. A few warriors to scout and some for MP duty/barb coverage. But I am also a builder, not a war-gammer. My production queue is not settler/worker/settler/worker or anything 'set'. I just try to maintain a balance. And if the land looks good but needs a lot of work, early investment in a worker or two that you will have for the next 500 turns - and have the improvement they make for the next 500 turns - may make more sense.

Situation dictates the response and there is no cookie cutter way to approach every game, which is why I like Civ3. Also, VC will dictate play. Since I am a builder, I am looking at the next 200+ turns. If you are going for early conquest, not much of this makes sense.


Vorlon_mi

1) my personal style is to make few military units in the beginning and fight few, if any, wars. I might start a war only in extreme circumstance. I used to try the 'sack a nearby capital' gambit all the time. Doesn't work so well at higher levels. Basic answer, (IMO) don't fight during REX.

2)Most of my inner, core cities remain undefended the majority of the game unless I am in Monarchy, when I might keep outdated units as MPs. Prior to rails, I station some keep offenseive units in core areas that can cover if a AI drops a unit or two by sea. AI naval invasions tend to be weak, but not always. Back up offensive units on the warfront can either wait to eliminate the counterattack or, if the attack goes really well, join in the slaughter.

 

[Ataxerxes]

Raliuven's points are good. I want to expand on defensive units and not defending your core cities. If you go for Monarchy and have military police in internal cities and Rights of Passage (small increase in AI attitude) then you might make more defensive units.

I think he does (and I definitely do) go for Republic. When in Republic you don't need the MP and DON'T GIVE RIGHT OF PASSAGE to anyone who can reach you. The AI loves to use ROP to hit an undefended city, even at Chieftain. You can have 8x the power and they'll still do it even though they can't hold the city for one turn. Also, rememer in Republic that you can get a substantial hit from excess military due to MP.

Although I'm at heart a builder I always seem to go from one war to another to get resources.

 

[vorlon_mi]

Raliuven's points are good. I want to expand on defensive units and not defending your core cities. If you go for Monarchy and have military police in internal cities and Rights of Passage (small increase in AI attitude) then you might make more defensive units.

I think he does (and I definitely do) go for Republic. When in Republic you don't need the MP and DON'T GIVE RIGHT OF PASSAGE to anyone who can reach you. The AI loves to use ROP to hit an undefended city, even at Chieftain. You can have 8x the power and they'll still do it even though they can't hold the city for one turn. Also, rememer in Republic that you can get a substantial hit from excess military due to MP.

Although I'm at heart a builder I always seem to go from one war to another to get resources.

In most of my games, I've gone for republic and fought short(er) wars for resources. I would achieve my desired objective, e.g., take 2 or 3 cities, get the resource, and end hostilities in 8-10 turns. I would usually have trouble fulfilling my end of a military alliance, because I was done fighting before the agreement was done. And, yes, once I've driven the other tribes away from my homeland, my inner cities are not defended. Just the coastlines. Lately I have been experimenting with Monarchy, and more sustained wars, to improve my warfighting tactics. If you don't practice something, you won't get any better.

I may be starting my first war too early, and with too few units. This leaves me vulnerable to a bad streak with the RNG when I'm on the offensive, and with insufficient troops at home to handle a counter attack. As you all have said, once I can dictate *where* the attack will be -- in the AI's land, grabbing its cities -- there is little need for idle defensive units.

 

[Raliuven]

In most of my games, I've gone for republic and fought short(er) wars for resources. I would achieve my desired objective, e.g., take 2 or 3 cities, get the resource, and end hostilities in 8-10 turns. I would usually have trouble fulfilling my end of a military alliance, because I was done fighting before the agreement was done. And, yes, once I've driven the other tribes away from my homeland, my inner cities are not defended. Just the coastlines. Lately I have been experimenting with Monarchy, and more sustained wars, to improve my warfighting tactics. If you don't practice something, you won't get any better.

I may be starting my first war too early, and with too few units. This leaves me vulnerable to a bad streak with the RNG when I'm on the offensive, and with insufficient troops at home to handle a counter attack. As you all have said, once I can dictate *where* the attack will be -- in the AI's land, grabbing its cities -- there is little need for idle defensive units.

Honestly, that's about where I am. I actually prefer Monarchy, but I am a recent convert to Republic due to my streak of trying Diplomatic wins. Diplo/spaceship favor Republic, while Dom/Con favor Monarchy - at least for me. Like you, I don't balance my attacks well and I always come up one or two units short on a major offensive. I like Monarchy because happiness is so much easier to control. But as I become more accustomed to Republic, I actually rather prefer it's versitility. I've found that any time the game offers you a change not to think about something, it will be handled for you - very ineffeciently handled.

One thing that does sway my opinion is the lay of the land. If there are few luxuries around, I dread Republic.

I also NEVER sign ROP. Ever. I can't think of the last time I did that. I do use MPP sometimes, but usually for manipulative purposes when the UN votes is approaching. I know there is a time and place for everything, I am just way to insecure for that kind of trust.

 

[CommandoBob]

ROPs and MPPs I avoid. I do get others to sign in on Military Alliances, though. I offer gpt mostly, sometimes a luxury or resource and hardly ever (now) a tech, though it depends on how well I think the AI will fight our common foe. If I really want them in the war, gpt is the best way to go. If they keep fighting after 20 turns I'll keep paying.

On Huge Pangea maps, if I can get 12 others to gang up on the 'Bad Guy' one or two of them make it their personal quest to take the Bad Guy down. Which is really nice if the Bad Guy is far away.

 

[GamezRule]

RoP is useful from time to time, but generally, it is better to make exceptions and sign one then vice-versa.

 

[bigFRANK]

I had a nice ROP with Germany in an early game, land border was completely blockaded by my troops (about 10 squares all fortified) ROP let me send a couple of dozen cavalry through to see off the Greeks, the Germans kept expelling the Greeks (knights mostly) who tried to come and pay me a visit.

 

[MysteryX]

I had a nice ROP with Germany in an early game, land border was completely blockaded by my troops (about 10 squares all fortified) ROP let me send a couple of dozen cavalry through to see off the Greeks, the Germans kept expelling the Greeks (knights mostly) who tried to come and pay me a visit.

Right of Passage is alright in a circumstance in which you can guarantee that you are the only one in a position to abuse it.

 

[Raliuven]

I had a nice ROP with Germany in an early game, land border was completely blockaded by my troops (about 10 squares all fortified) ROP let me send a couple of dozen cavalry through to see off the Greeks, the Germans kept expelling the Greeks (knights mostly) who tried to come and pay me a visit.

The only problem with this is that the Germans are now in a position to claim all of that nice Greek land - or you now have a divided nation. I don't sign ROP to 'get at' an enemy because that usually means my holding it is at the whim of the AI. Likewise, signing an ROP so that a nation can cross your boards to attack another Civ leaves you vulnerable at the whim of the AI. Its lose-lose IMO.

In my experience, the AI rarely ever asks another AI to leave. It is extremely annoying because even if they are your gracious allies, they will ask you to leave the moment you set one foot over their terriroty. As far as I am concerned, if a Civ is allowing a nation to pass through it to attack me, they are also my enemy - it just might not be declared yet.

 

[MysteryX]

The only problem with this is that the Germans are now in a position to claim all of that nice Greek land - or you now have a divided nation. I don't sign ROP to 'get at' an enemy because that usually means my holding it is at the whim of the AI. Likewise, signing an ROP so that a nation can cross your boards to attack another Civ leaves you vulnerable at the whim of the AI. Its lose-lose IMO.

My understanding of his situation was that (a) Germany was not at war with Greece, so not attacking Greece (though I suppose there is a risk that Greece gets tired of being asked to leave and declares war on Germany too), and (b) even though Germany has the right to come on his land, they didn't have the ability because his troops were physically blocking any path.

 

[bigFRANK]

Both right, the Germans and the Greeks didn’t fight in this scenario and the Germans had no way to use the ROP to move across my territory, but I did end up with a split civilization (well at least until the modern era dawned when I squished the Germans). Although the split civ wasn’t necessarily a great idea it gave me 10 or 11 cities that I used primarily as science farms to get the technological edge which later let me beat the Germans.

 

[Raliuven]

Both right, the Germans and the Greeks didn’t fight in this scenario and the Germans had no way to use the ROP to move across my territory, but I did end up with a split civilization (well at least until the modern era dawned when I squished the Germans). Although the split civ wasn’t necessarily a great idea it gave me 10 or 11 cities that I used primarily as science farms to get the technological edge which later let me beat the Germans.

That was the point I was making. Either you keep the cities and have a divided nation or you raze the cities and Germany, at war or not, gets the pick of the land. Divided nations can be irritating, complex monsters. I prefer to keep wars regional and ignore distant rivals. I'll kill any units that wander into my area. Meanwhile, I focus my strength to take down a nearby rival that I can incorporate into my empire - perhaps even add productive cities.

This starts to fall apart when you talk about island boards. Even so, I like to keep my lands close so I can move units with short battlelines.

Obviously it worked in this scenerio. I'd be cautious in the future and on higher difficulties.

I actually played a regent game as england once (random settings) where my scouting warrior came across berlin with only one warrior in the early game. I must have made a straight line right to their captial. We were on a donut world and berlin was on the other side of the large lake. I took berlin and it was a 'second empire' for a long time. Very corrupt, but it was a small map so it actually worked okay. I didn't manage to connect the two by land for a very, very long time. It was an interesting game, but difficult. Berlin (with neighboring captured german cities) was always in danger from 2 other Civs and I was ferrying units across the lake to protect my 2nd, rather corrupt empire. In hidesight, I probably would have won faster by abonding the german cities I captured, shipping the slaves back, and concentraing on other closer and productive conquests/improvements.

 

[shl7070]

From my experience regarding AI dogpiles, they happen if:

1) I'm extremly weak.
2) I'm extremly strong.
3) I REXed with swordsmen instead of settlers, they remember this for eternity.
4) I negotiated peace treaties with the "my forces approach your cities" way, the one who entered such negotiations will try to find allies next time and those will willingly agree as they know that those 50 tanks didn't disappear after the treaty and they don't want to see them near their capital.

If I'm playing arch they also know that my stacks aren't present everywhere and will use this info.

 

[agonistes]

In answer to the original question...

I play on modded maps larger than 200x200 with 31 civs. I've actually lowered the hostility level, and even so war gets out of control. No civs maintain relationships for long periods of time, its dog-eat-dog.

I've noticed no rational for why civs go to war. I only play on regent, so am always the biggest dog on the block, and even so any civ will start a war with me.

If you are at war with one civ, that civ can get any other civ to start a war with you if the first civ can make an offer the second civ can't refuse, be it gold, techs, or something else. I have yet to find a definite way of making a civ happy enough to make sure they never go to war with you.

In the extreme case, I may have 100+ cities, and the other civ may have just one, and they may still go to war with me even if I am giving them gold and techs every turn.

As for indications that they are going to start a war, the most obvious is troops headed in your direction, particularly if you gave them right of passage and they start moving towards your capital.

Just recently I discovered that if I have some fortresses on my border, and empty my troops out of them to upgrade without replacing them on the same turn, if another civ has troops nearby they will race in to take the fortresses, and declare war when I ask them to leave.

Situationally, I suspect this can be a tactic used to replace the 'funnel of death' tactic. If occupied fotresses tend to be left alone, emptied fortresses do the opposite.

As to the above advice about offensive units being better than defensive ones, I disagree to some extent. In the long run, building alot of spearmen is very efficient. They upgrade to mobile infantry eventually. All of the offensive units that are not artillery hit an upgrade wall until you can build tanks. In a tight war, I'd rather have two offensive on a stack of defensive, instead of a lesser number of more expensive offensive units. And if you are taking cities rapidly, it is cheaper to buy defensive units to hold the cities.

As to which offensive units I favor, horseman-knights-cav. The ability to attack then retreat in the same turn trumps the slight defensive ability of single move units imo. You can pick off injured enemy units this way without risking your own troops. Make the shield investment for some faster offensive troops, but build up heaps of artillery and defensive troops. A ratio of 1 fast, 1-2 artillery, and 2-3 defensive is good. Attrition wins wars. And don't underestimate the value of a fotress that puts your artillery in a good tactical position.

 

[agonistes]

I too avoid RoP and MPPs.

I usually only will sign a RoP if I need to move through a neighboring civ quickly and I have no airports and using a navy is too slow.

MPPs tend to lead to wars I am not interested in fighting. I have never experienced a positive benefit from being in a MPP that could not be had using a lesser investment of potential reputation loss.

 

[shl7070]

It's hard to detect signs of war, the only sure sign is large enemy stack going inside my own territory and includes attackers (at least half of the units).
Settler pairs are an indication that the NOT intend war. If they can settle anywhere on the map they won't declrare usually.
If they go inside my territory but have legitimate objective and on the same time offer MPP means they want to war agaimst a third party.
When using the ask to leave option they declare if they intended war from the beginning. Checked that in 5 cases, saved before asking to leve, then ask, then reload and not ask (and not move units), when they reached their objective they declared. Of course if I reinforced the objective they changed the direction of the stack and sometimes didn't declare when asked to leave.
Less trustworthy indicators are checking thei cities, if seeing 70% tax which indicates unit overuild then war is looming (usually) but the victim could be anyone.
Seeing AI sign MPPs or even more so signing embargoes against me then war is looming for sure.
MPP is not always with me in sight but an embargo is directed at specific civ and is clear indicator of intensions.

Alliances are signed after war had started, signing alliances indicates weakness of the warring AI usually, they will try to win alone if they think they can but will sign alliances if they think they can't. Several times saw backstabbing alliance in which an AI civ signed alliance against me and then asked for peace so their ally will take all the burden of war.

 

[Raliuven]

One side note is that seeing a settler/spear pair is not a definate sign the AI will not start a war. That pair may have started out to plant some forsaken tundra town on the other side of your border. The AI will still attack even though the settler/spear pair are doomed. And they will keep sending them even if you are at war.

Seeing JUST a spear and settler is nothing to worry about - other than to keep them away from YOUR land.

 

[CommandoBob]

Alliances are signed after war had started, signing alliances indicates weakness of the warring AI usually, they will try to win alone if they think they can but will sign alliances if they think they can't. Several times saw backstabbing alliance in which an AI civ signed alliance against me and then asked for peace so their ally will take all the burden of war.

You should be signing alliance against your target civ, even if they started the war. And not just one or two but everyone you can. That limits the options of your primary opponent and could break a trade route of Iron or Saltpeter, which is a plus for you.

I have found it more effective to offer gpt deals for Military Alliances. Sometimes swapping a tech will get you an ally, but that doesn't keep them in the war. Gold is always good, followed by strategic resources (Iron, Horses and Saltpeter usually) and lastly luxuries. Gold is best, since you have some control of how much you send. Sometimes it is cheap, too, like 10 gpt. Iron and Horses will help them make units which they will send out to attack the common foe. And sometimes they just need some liquid courage (Wines) before they go into battle. Whatever they need, if it is reasonable, give it to them.

If they make peace, that ends whatever you are sending them. That rarely happens with a gpt deal. They want what you have to offer.

If the foe is close to you, then you are probably going to be forced to do much of the warring. Not always, but usually. But even if you set back and go defensive on that front, one or two AIs will take the Alliance personally and do all the work for you, depending on the map and ease of access and such. Which is really nice when your targeted AI is halfway around the world.

Generally, avoid whatever MPP/ROP/MA deals the AIs offer to you. But always try to get the AIs into MAs when you see a long war or a war with a bigger nation.

 

[shl7070]

You should be signing alliance against your target civ, even if they started the war. And not just one or two but everyone you can. That limits the options of your primary opponent and could break a trade route of Iron or Saltpeter, which is a plus for you.

I have found it more effective to offer gpt deals for Military Alliances. Sometimes swapping a tech will get you an ally, but that doesn't keep them in the war. Gold is always good, followed by strategic resources (Iron, Horses and Saltpeter usually) and lastly luxuries. Gold is best, since you have some control of how much you send. Sometimes it is cheap, too, like 10 gpt. Iron and Horses will help them make units which they will send out to attack the common foe. And sometimes they just need some liquid courage (Wines) before they go into battle. Whatever they need, if it is reasonable, give it to them.

If they make peace, that ends whatever you are sending them. That rarely happens with a gpt deal. They want what you have to offer.

If the foe is close to you, then you are probably going to be forced to do much of the warring. Not always, but usually. But even if you set back and go defensive on that front, one or two AIs will take the Alliance personally and do all the work for you, depending on the map and ease of access and such. Which is really nice when your targeted AI is halfway around the world.

Generally, avoid whatever MPP/ROP/MA deals the AIs offer to you. But always try to get the AIs into MAs when you see a long war or a war with a bigger nation.

Thats OK, I understand

I meant AI-AI alliances against me or a third AI