Dirty tricks

[Dominae]

The way I use the term, a "dirty trick" is an AI exploit that isn't necessarily game-breaking. Put another way, it's those little strategies that we figure out work so well against the computer opponents, but that would rarely (if ever) work in multiplayer.

In this thread I want to outline the dirty tricks that I've frequently use. I know that all lot of this strategy has been discussed before in other places, but nowhere (that I know of) is there a central collection of these tricks. My hope is that others will post their own dirty tricks so that we all become experts at beating the AI.

1. The turn before declaring war with another civ, get some of their best techs in exchange for "per turn" deals. For instance, give 250gtp in exchange for Communism, and declare war the next turn. You won't have lost anything, and will have gained another tech. This may not always work if your reputation is already bad.

This trick works because the AI doesn't know when you're poised to attack, but only your relative military strength.

2. Sign military alliances to get all your neighbours to go to war, ideally with both sides of roughly equal strength. Do not participate in the fighting, but declare peace with your opponents as soon as possible. This effectively allows you to focus on building up your empire, while the computer opponents are focusing on a difficult war.

This trick works because the AI has no real concept of "alliance" over and above "being at war with the same civ".

3. Identify all the strategic and/or luxury resources in a neighbors territory by trading for their Territory/World map. Sign a Right of Passage agreement, and fortify any unit on any resource inside their territory not connected via a road network. Since the other civ's Workers can no longer build roads on that tile, you've denied them a resource. This trick works especially well if you've managed to previously pillage their roads during wartime, potentially denying the civ Iron or Horses (for instance) indefinitely.

This trick works because the AI doesn't understand that it should have access to resources inside its territory.

4. Restrict the movement of various friendly units with units of your own. For instance, if a civ sends a Settler to found a city in a location your were planning to claim, send a few Warriors to block its path. With three units you can usually get the AI to "shuffle" back and forth indefinitely, ensuring that the Settler never reaches its destination. Excess units can be used to build an entire fortified wall for blocking purposes, which is more resource-intensive but eliminates the micro-management of moving units back and forth.

This trick works because the AI has no concept of progress with regard to movement.

5. During your REX phase, leave small "pockets" of unclaimed land surrounded by a few other cities. The AI will eventually plan to found their own cities in those locations, and you simply let them. Surrounded by the overwhelming cultural influence of the rest of your empire, those cities will eventually flip to your side. The pockets can be chosen to be especially infertile locations which wouldn't be good for the early game. This strategy allows you to cover more ground during the REX phase, possibly denying the AI some nice city sites.

This trick works becaues the AI cannot determine in advance whether a city is prone to cultural flip.

6. Similar to #1, sign Mutual Protection Pacts to get the world into heavy war, and watch from the sidelines. Although you do have to commit some forces to the field for your Pact to kick in (by having the enemy attack one of your units or cities), you don't need to have any intention of winning any major battles. Around Communism this is especially useful, as most AI civs switch to Communism in wartime, crippling their economy.

This trick works because the AI undervalues the negative effects of going to war.

7. During wartime, use non-combat units to lure enemy forces out into the field. Foreign Workers are especially good for this, but Settler, Explorer and weak military units are subsitutes in a pinch. The AI almost always attacks the easiest target it can reach with its best offensive units. This opens up the possibility of counter-attacks by your own forces. Special care must be taken with enemy units with more than 1 movement point (especialy Cavalry), as they can potentially grab your Worker and retreat in the same turn.

This trick works because the AI cannot strategically "look ahead" beyond its current turn.

8. Use a two-pronged attack plan, but scatter your prongs in time (I'll clarify!) Split your attack force into two roughly equal groups. Attack one major target with one group. A few turns later, attack at another locations that is quite distant from the first offensive (boats may be required to get the second force into position). The AI commits most of its military strength in the initial defense ("most" meaning all units other than city defenses). The second offensive should meet little resistance as most of it is busy with the first.

This trick works because the AI simply overcommits its offensive and defensive units in defense (in other words, it sends everything it has to repel an attacker).


Whew, thanks for reading! Any additions?


Dominae

[Sheik]

This is similar to number 8:

In a recent game I had about 8 longbowman to my opponents 8 or 10 and was able to easily defeat them all by stopping a tile away from attack. Their longbowman would move forward and then I would kill them. Sometimes I would lose one but would always have one behind it to take out their original attacker.

[Jaybe]

When sharing a border with your potential enemy, declare war at THE END of your turn!

This applies when you are superior in strength. Deploy in such a way as to leave an open passageway towards, but not all the way to, a weakly defended city fairly close to the front. Especially if the enemy has rails and any substantial "offensive" troops, he will march into your land with a "Stack of Doom" (SOD, 30 to 90+ units in one stack). This will trigger any MPPs that you have (better they be allied with you than with him). Surround this stack with strong defensive units and entrench. Now use your pre-deployed land and sea invasion forces.

His SOD is isolated where you can leave them to rot or slowly whittle them down while you take over his country.

--
Dominae, your "2. Sign military alliances to get all your neighbours to go to war ... declare peace with your opponents as soon as possible. ..."
will leave you with a reputation hit unless you wait to declare peace for 20 turns after the alliances are signed. Doesn't invalidate it, but is something to be aware of.

[Minmaster]

"This trick works becaues the AI cannot determine in advance whether a city is prone to cultural flip."

im not so sure about this with the latest patch. the AI seems to take a lot of his units outside a city that has its borders surrounded by another. my feeling was that the AI is anticipating a reversion. kinda like how human players leave their units outside a newly conquered city sometimes to prevent the loss of all units inside the city.

[zeh]

I agree with all you points of "This trick works because the AI ...".
A sugestion can be sent to firaxis exposing this flaws in AI behavor.
I fell very bad when I see the AI doing this very stupid actions, and I think is not that hard t implement this kind of behavors.

[alexman]

Nice post, Dominae, but I don't really feel that points 1 and 2 are exploits. They are one-time (OK, maybe two-time if enough time passes) deals. After you ruin your reputation, nobody will easily make per-turn deals (or alliances, for #2) with you, not even civs that you have not (yet) betrayed. This not only diminishes the repeat value of this dirty trick, but it also takes out your ability to export your luxuries.

I would never risk my reputation just to get Communism, for example, especially if I'm about to go to war and beat Communism out of them for free! (I hope you didn't mean that you wanted to actually switch to Communism before going to war! )

[jshelr]

This used to work pre-patch, but I consider it a bush-league exploit (meaning that I have not had to use it lately).

Suppose you have four cities


A ------------- B



C ------------ D


If an very large AI stack is headed for B. You can fortify units in B and expose A with no defense. The stack will shift toward A. On your own road/rail system, you are much faster. So, before they get near A, you can fortify units in A and expose B. They will then shift back. Things get really silly when you lure the stack within the box. Without much trouble, you can keep them going in circles by exposing the cities in turn around the box. That can last forever, or until you build enough units to successfully deal with them. I hope this no longer works, but the AI's choice of "weakest available target" suggests it does.

[Dominae]

Jaybe, you're right that you'll get a reputation hit if you make peace so soon. You could potentially wait 20 turns, which isn't very long at the beginning of the game (where the trick is most applicable). However, the reputation hit isn't so bad as you'll be planning to lay low for a while anyway, and there will be plenty of time to make friends after the AI civs are finished their stupid war (started by you!).

alexman, I consider anything an exploit that works against the AI every game (if even just once), but would rarely fool a human opponent. According to that definition, numbers 1 and 2 are "exploits". You could argue that 2 works on people, but not often; a human player isn't likely to sign a Military alliance and switch into a wartime economy just because they're now "at war". And yes, the Communism example is bad, so just replace Communism with a tech that you wouldn't mind getting a reputation hit to obtain (Military Tradition, perhaps!?).


Dominae

[Theseus]

Originally posted by Dominae
1. The turn before declaring war with another civ, get some of their best techs in exchange for "per turn" deals. For instance, give 250gtp in exchange for Communism, and declare war the next turn. You won't have lost anything, and will have gained another tech. This may not always work if your reputation is already bad.

This trick works because the AI doesn't know when you're poised to attack, but only your relative military strength.

Not an exploit, as I've had AI civs do it to me!

2. Sign military alliances to get all your neighbours to go to war, ideally with both sides of roughly equal strength. Do not participate in the fighting, but declare peace with your opponents as soon as possible. This effectively allows you to focus on building up your empire, while the computer opponents are focusing on a difficult war.

This trick works because the AI has no real concept of "alliance" over and above "being at war with the same civ".

I dunno... I often have distant civs declare war on me, and never do anything about it. I'm pretty sure that at least some of these "wars" have NOT been due to bankruptcy, so why do the AI civs do it? If it's just to mess me up, including through additional AI civs entering the fray via alliances, then this would not be an exploit.

3. Identify all the strategic and/or luxury resources in a neighbors territory by trading for their Territory/World map. Sign a Right of Passage agreement, and fortify any unit on any resource inside their territory not connected via a road network. Since the other civ's Workers can no longer build roads on that tile, you've denied them a resource. This trick works especially well if you've managed to previously pillage their roads during wartime, potentially denying the civ Iron or Horses (for instance) indefinitely.

This trick works because the AI doesn't understand that it should have access to resources inside its territory.

Exploit for sure.

4. Restrict the movement of various friendly units with units of your own. For instance, if a civ sends a Settler to found a city in a location your were planning to claim, send a few Warriors to block its path. With three units you can usually get the AI to "shuffle" back and forth indefinitely, ensuring that the Settler never reaches its destination. Excess units can be used to build an entire fortified wall for blocking purposes, which is more resource-intensive but eliminates the micro-management of moving units back and forth.

This trick works because the AI has no concept of progress with regard to movement.

Allowable exploit, as the lack of ZOC and the inability to BAR frigging settlers from crossing your land is absurd.

5. During your REX phase, leave small "pockets" of unclaimed land surrounded by a few other cities. The AI will eventually plan to found their own cities in those locations, and you simply let them. Surrounded by the overwhelming cultural influence of the rest of your empire, those cities will eventually flip to your side. The pockets can be chosen to be especially infertile locations which wouldn't be good for the early game. This strategy allows you to cover more ground during the REX phase, possibly denying the AI some nice city sites.

This trick works becaues the AI cannot determine in advance whether a city is prone to cultural flip.

Not an exploit, but a choice. Who says your territory has to grow like The Blob? (Also, you left off doing the same thing but with the intent of militarily taking those towns )

6. Similar to #1, sign Mutual Protection Pacts to get the world into heavy war, and watch from the sidelines. Although you do have to commit some forces to the field for your Pact to kick in (by having the enemy attack one of your units or cities), you don't need to have any intention of winning any major battles. Around Communism this is especially useful, as most AI civs switch to Communism in wartime, crippling their economy.

This trick works because the AI undervalues the negative effects of going to war.

The AI civs do this too.

7. During wartime, use non-combat units to lure enemy forces out into the field. Foreign Workers are especially good for this, but Settler, Explorer and weak military units are subsitutes in a pinch. The AI almost always attacks the easiest target it can reach with its best offensive units. This opens up the possibility of counter-attacks by your own forces. Special care must be taken with enemy units with more than 1 movement point (especialy Cavalry), as they can potentially grab your Worker and retreat in the same turn.

This trick works because the AI cannot strategically "look ahead" beyond its current turn.

Exploit.

8. Use a two-pronged attack plan, but scatter your prongs in time (I'll clarify!) Split your attack force into two roughly equal groups. Attack one major target with one group. A few turns later, attack at another locations that is quite distant from the first offensive (boats may be required to get the second force into position). The AI commits most of its military strength in the initial defense ("most" meaning all units other than city defenses). The second offensive should meet little resistance as most of it is busy with the first.

This trick works because the AI simply overcommits its offensive and defensive units in defense (in other words, it sends everything it has to repel an attacker).

Good strategy.
______________________________

Good idea for a thread Dominae.

You left out one, though... buying ancient era Workers!

:ducks from Arrian:

[Arrian]

Bah!

Ok, since Theseus brought it up (and I knew he would before even opening this thread), I will submit my latest little "trick." I am not entirely convinced it is an exploit, but Theseus thinks so. I offer it up for discussion:

Buying AI workers early in the game. They typically cost 25-28 gold a piece. Depending on when you manage to buy them, this can be immensely powerful. First, if boosts your civ, as you have an extra unit of population early (I add them to cities, I do not use them to actually work). Second, it really hampers the development of the civ you got the worker(s) from. It's cheap, and the AI will *always* sell you a worker, regardless of their attitude or your reputation.

However, you *do* have to pay for them, and getting them is VERY hit-and-miss, because you can only do it when the AI has a worker sitting in its capitol. Additionally, if you use them instead of adding them to cities, their former owners get mad at you (documented by firaxis that having slaves from another civ will anger them). This can get you into an unwanted war.

In the last game I played out, I purchased roughtly 6 workers from the AI in the ancient age. 1 was very early on, and appears to have helped cripple China (they did start in a jungle too). It helped me. The others helped replace the pop points used in my ReX phase. That + excellent starting spot (ask Theseus) = kickass results.

So whaddya think? Exploit or not?

-Arrian

[Theseus]

I'm just giving you grief... I do it too.

Speaking of it as a strategy, don;t forget that adding the slave for pop is best done in your capitol.

Something just occurred to me; I bought a worker from China too... at what speed does an originally industrious slave work? Has anybody actually ever checked this?

[Dominae]

Theseus, concerning the Military Alliance and Mutual Protection Pact "tricks" ('exploit' is such a dirty word!), my point is that they're almost always effective in some form or the other in games against the AI. Human opponents will not allow themselves be "used" so easily (and if so, probably have some hidden agenda). Also, the fact that distant civs do not seem to attack may very well just be because they cannot reach their targets. For instance, in AU-106, I was at war with the French for many many years, but their troops were never sighted within 4 tiles of my shores.

Arrian, purchasing early Workers is definitely a "dirty trick", so much so that I can't believe I forgot to mention it. 30 Gold each?!? Sure, I'll take those guys off your hands! Although it's tedious, I now check every turn in the trade screen whether an AI civ has Workers for sale. If you make sure that your neighbors are at war, you can buy Workers from friendly civs more frequently, because they fortify Workers in cities when their land is threatened: if their capital is under siege, you can almost always "save" a few Workers.

I don't know if this is such an exploit that it needs to be fixed, but I offer two suggestions. Obviously, you could raise the cost of foreign Workers (150-200 Gold seems about right). Also, you could make trading Workers similar to trading resources, in that it is only possible if the two civs' capitals are connected via roads.


Dominae

[alexman]

Originally posted by Theseus
... at what speed does an originally industrious slave work?

At the speed of any other of your captured workers.

If you are industrious OTOH, then your captured workers work as fast as one non-industrious non-slave worker.

[BillChin]

Here are a couple more:

1) Say two AI empires have a mutual-protection-pact. For this example make it America and England. The human player wants to go to war with one of them but realizes that both will make for a nasty war. At whatever cost, make a MPP with one, move troops into the other. When asked to leave refuse. When the human player's troops are attacked the new MPP kicks in overriding the old MPP.

2) Here's a nasty one for warmongers: say an empire is nearly defeated but has a few island cities far away from the human player units, or a few hill top cities that will be difficult to take. Make peace demanding these cities. If a player does not care about reputation, he can immediately redeclare war and take the last city. The hit on reputation is the same as going to war after making a gold-per-turn deal.

3) The AI is drawn to attack the least defended human city. A human player can exploit this by leaving a city empty. This draws the AI forces to it, but if the city is far away or in the interior, the attack force will not arrive in any reasonable time.
- Bill

[jshelr]

Buying workers: deal between consenting adults and not a dirty trick.

Buying workers for nearly 30 gold each when relatively new techs go for 80 offered, 72 bid seems to be a bit high, if anything. I don't think this is much of an exploit and, while I do it when I see the opportunity, I would like to see a bit of cost-benefit analysis that both includes the city growth limits and still proves it's highly profitable to the buyer.

I agree that the AI does many of the dirty tricks described. They sometimes do sign alliances and then never show up -- even powerful civs. They often sign alliances and then make peace very early. They also dangle bait during combat, hoping you will take it. Favorite bait is the warrior or spearman at the edge of your visual range as bait for the knight, with a stack of longbows waiting out of sight to hit the knight.

[Arrian]

It would seem there is some difference of opinion on worker buying.

I think we can agree that this tactic is most powerful when done very early in the game. For the sake of discussion, let's define "very early" as pre-2000BC.

After that, though it can still be powerful, I do not believe it can possibly be described as "game breaking" or "unbalancing."

I see potential merit in the "no trading workers until there is a viable trade connection" idea, Dominae. On the other hand, I am loathe to surrender an effective tactic until I'm sure it's an exploit. And I am still not sure about it.

I think each of us has a sense of where the line between "effective strategy" and "exploit" lies. All effective strategy comes from our human ability to out-think the AI. There are many tactics which the AI cannot deal with, but not all are commonly considered "exploits." For one thing, the AI is not very good at protecting key cities and/or resources from a determined human invasion. It will defend the resources lightly or not at all, and it will defend the cities as it does all others (dependent on size, I think). A human would rush everything he/she has available to the defense of a strategically important city. Yet no one advocates that we stop targetting the AI's weaknesses when at war.

Another example of this is the AI inability use bombard units properly. Yet we still use bombard units (granted, they were weakened considerably several patches ago).

This is not to say that early worker buying is not an exploit, but I do want to point out the fact that if we decide any human tactic which routinely fouls up the AI is an "exploit" we reduce ourselves to playing the game like the AI. Where do we draw the line between using our strengths against the AI's weaknesses?

-Arrian

[alexman]

Think of it this way. When the AI sells its worker at the beginning of the game, it has to build another one. That means:
1) A city wastes turns of production building a worker instead of something else.
2) There are no tiles being improved until a new worker is produced.
3) Assuming no unhappiness problems (not an issue for the AI that plays on Regent and usually has a good garrison), that city remains at one less population until it reaches its growth potential. Assuming that the tile that would have been worked is a 2 food, 1 shield, 1 commerce tile, that's one less shield and one less commerce from that city until it reaches size 6, which usually happens after the REX phase is over (late ancient age).
4) REX is delayed because that city is not large enough to make a settler.

Now what player in his right mind would sell a worker for only 30 gold in the beginning of the game? Would it be an exploit if you could steal workers without losing your own? For sure. Well this is damn close. The AI values workers too low. That's why it's an exploit.

[jshelr]

I guess I'm thinking of producing workers for sale as a legitimate business for an AI civ to be in. An AI civ with, say, ten cities and a despot income of, maybe, 100 gold per turn could probably devote two cities to worker production and sell a worker, what, every two turns or so for an income of 14 per turn. Not bad. About a third of Wall Street's return without the up front $1000 gold investment. If they also use other cities to build an optimum domestic worker force, the question is whether this is a good way to get gold to buy tech. Maybe it's not that bad a method.

[Theseus]

True, but we're talking about when the AI civ has maybe 1-3 towns.

Look, anyway, I agree with Arrian... does the AI do stupid things that we don't? Uhhh, can you say "irrigation?"

Eddie Izzard has a funny lines about lies. He says there should be a scale for lies like there is for crim (e.g., Murder One, Murder Two). Lying One would be saying you ****ged someone when you didn't and Lying Nine would be saying the Holocaust never happened.

Worker buying is pretty far down on my scale, compared to say, ROP Rape.

[Catt]

The "early worker buy" tactic feels like an exploit to me for the same reason as alexman gave - an early worker is worth far more than 30 gold IMHO, and I suspect the vast majority of experienced human players would almost never sell an early worker (of course, an early worker limited to the AI worker task list and prioritization may be worth a lot less than 30 gold ). OTOH, I definitely agree with Arrian that eliminating all aspects of human play that the AI is incapable of employing just reduces us (limits us) to the AI tactics

The real decision factor for me in whether or not to use the "early worker buy" tactic is wholly distinct from "exploit or not" however: I am far too lazy to ever consider opening diplo relations with every civ on every turn. Perhaps the benefits of the "early worker buy" tactic are justified by the time / frustration of making the moves? "I'll give you 27 gold and 52 mouse clicks for that worker?"

Catt

[jshelr]

I'm sold. I'd never sell an "early" worker. And I hardly ever am smart enough to search for one to buy early either, although I occasionally stumble on the chance.

[Dominae]

Originally posted by Arrian
I see potential merit in the "no trading workers until there is a viable trade connection" idea, Dominae. On the other hand, I am loathe to surrender an effective tactic until I'm sure it's an exploit. And I am still not sure about it.

Me neither (about Worker trading being an exploit). But I think a good compromise would be to raise the cost to
at least the cost of techs at that stage of the game.

Originally posted by Arrian
I think each of us has a sense of where the line between "effective strategy" and "exploit" lies. All effective strategy comes from our human ability to out-think the AI. There are many tactics which the AI cannot deal with, but not all are commonly considered "exploits." For one thing, the AI is not very good at protecting key cities and/or resources from a determined human invasion. It will defend the resources lightly or not at all, and it will defend the cities as it does all others (dependent on size, I think). A human would rush everything he/she has available to the defense of a strategically important city. Yet no one advocates that we stop targetting the AI's weaknesses when at war.

I don't advocate denying myself any of the strategies mentioned in this thread. The reason why I used the term 'dirty trick' and not 'exploit' is that people believe exploits ruin the game and should be changed in future patches. I don't think any of the dirty tricks mentioned here should be fixed, (primarily because they would require a major overhaul of the AI system) apart from maybe early Worker buying.

What I am getting at is that none of these strategies/exploits/tricks will work against a relatively competent human player. I would never sell Workers in the early-game for under 250-300 Gold (if that). Nor would I chase after useless enemy Workers when mounting an assault. And I definitely would never ever let myself get caught in a "infinite loop" whereby I shuffle my units back and forth between two targets.

Figuring these things out brings you one step closer to beating the AI (hence the point of this thread), but gives you no advantage in multiplayer (unless bots are used). When Play the World comes out, I know I will have to unlearn many of my best "strategies".


Dominae

[Theseus]

Actually, I'll feel really bad for the noob players, getting nailed by "Fool's Mate" moves.

Can;t you see it: noob plays Egypt, maxes research to get The Wheel ("oh boy, oh boy, gonna have an early GA!!"), never has more than a few gold, gets an offer for his Worker, looks around and realized his speedy little guy has gotten ahead of needed tile improvements... "oh boy, oh boy, 30 GOLD!!"

[Jaybe]

Originally posted by Jaybe
When sharing a border with your potential enemy, declare war at THE END of your turn!

This applies when you are superior in strength. Deploy in such a way as to leave an open passageway towards, but not all the way to, a weakly defended city fairly close to the front. Especially if the enemy has rails and any substantial "offensive" troops, he will march into your land with a "Stack of Doom" (SOD, 30 to 90+ units in one stack). This will trigger any MPPs that you have (better they be allied with you than with him). Surround this stack with strong defensive units and entrench. Now use your pre-deployed land and sea invasion forces.

His SOD is isolated where you can leave them to rot or slowly whittle them down while you take over his country.

Really Stupid Question:
Has nobody yet commented on this because it was so good, so worthless, or so rarely applicable (or other)??

[Catt]

Originally posted by Jaybe

Really Stupid Question:
Has nobody yet commented on this because it was so good, so worthless, or so rarely applicable (or other)??

Other. I can think of no witty comments to your fine "dirty trick."

Catt

[punkbass2000]

I feel dumb harping on the subject of war, but I am kinda thinking that, in some ways, war in and of itself is exploitative, depending on how far down Arrians sliding scale you want to go. I mean, it's quite clear that, ultimately, the best way to win civ is an all out war machine. This has been shown time and time again, through all the incarnations of civ. The computer simply canlt manage a war too well. Though war per se may not be exploitative, I really think the early oscillating wars may be. The computer simply can't conceive of that kind of strategy. Can you imagine starting up a deity game and having a neighbour fighting oscillating wars? It wouldn't even kill you off, it would play with you for a while, essentially until you quit the game. More often than not, war is the best solution to a problem. You take what you want and that's that. You lose nothing but expendable units, typically. Despite the historical accuracy, for me the power of war is a game breaker, which is why I've been going about peaceful deity strategies as of late.

[Chemical Ollie]

Originally posted by Jaybe

Really Stupid Question:
Has nobody yet commented on this because it was so good, so worthless, or so rarely applicable (or other)??

I have done it in a few games. Artillery is useful here.

[Bella Hella]

last night, shaka sent a ship to my territory completely at random and deployed and impi and a longbowman. it was obvious that he was about to declare one of those utterly random, pointless, futile AI wars. i just didn't feel like dealing with it, so i signed an MPP with my neighbors, called shaka, told him to get the F out, and he declared war. now my nice neighbors are at war with him, and will hopefully send units over to the zulu island... and when the MPPs run out, maybe they'll stlil have some units over there and i can take over their territory which is "rightfully" mine.

[jshelr]

punkbass2000

Your comment above is appreciated. However, the game is relatively historically accurate when you do history on the 6000 year time scale. The fighting civs won. (Maybe the Iriquois shouldn't be in the mix since the Americans killed most of them and settled thier land. Ditto the Aztecs. But, OTOH, the game gives these civs a chance to reverse the real results. They didn't reseach in the Americas fast enough in real life before first contact. So, they got killed by superior tech. Sounds familar to a civ player.)

I'd like to see your attitude apply more in the post-industrial era. Unprovoked attacking past 2000 should be impossible for a democracy (somebody call George Bush, quickly) and cause big negative effects for other government types, IMO.

[vmxa1]

Historically Democracys to not attack democracys. Modern era wars do lead to collapse of the government, even those that did not lose. Germany/Japan, LBJ.