View Full Version : Power of Creative Civ AI's; and Wang Kon...?


DMOC
Feb 04, 2008, 08:19 PM
Test games:

1. Comparing Kublai Khan and Wang Kon; Monarch, Continents, Normal speed (http://forums.civfanatics.com/showpost.php?p=6472536&postcount=34)

2. Seeing the power of Suryavarman II and the strangely slow expansion of Wang Kon; Monarch, Continents, Epic Speed (http://forums.civfanatics.com/showpost.php?p=6510650&postcount=44)



Okay, this has been a little bit strange to me.

I don't know if any of you have experienced this, but doesn't it seem that creative civs are ALMOST ALWAYS top contendors in games that they play in?

I was going to take the time to find as many games as I could from my own saves and online the forums (there are plenty that I have seen to reinforce my thought -- trust me), but I just cannot do that tonight. I might do that another time (show links to all games with creative civs being a top contendor), though.

Take, Louis XIV for example. That guy has always been a pain for 1. Expanding rapidly, and 2. Cultural victories. I played two games on monarch and emperor recently, with Louis XIV on another continent (this way I didn't have much influence on him) and he was almost close to a cultural victory when I won space races in both games. He always seems to have a huge landmass, for some reason (and he's not imperialistic). I remember my second emperor game that I played recently with Louis on another continent with Hammurabi and Victoria. Hammurabi and Victoria each ended up with about 6 or 7 cities while Louis XIV got SIXTEEN! And in a replay of the game, I found out that only one war was fought on the whole continent before I entered the fray and did some bribing. Louis razed 1 English city and...nothing else happened of major results in the war. I checked the 3 capitals of the 3 AIs, and they were perfectly balanced! How did Louis get so many more cities than Victoria (who is Imperialistic) and Hammurabi (who is organized so he can afford more cities)?

I can think of the top of my head several games with Louis XIV -- or other creative civilizations! -- being a strong contedor: aelf's second immortal challenge (Louis won the game and even vassalized Wang Kon; I will get to the weakness of Wang Kon later); Obsolete's immortal challenge where Willem van Oranje (Cre/Fin) conquered a whole continent; another Obsolete game where Hatshepsut (Cre/Spi) completely outexpanded Roosevelt on their continent; Catherine (Cre/Fin in warlords i think) in Sisutil's mehmed game... Catherine (Cre/Imp in BtS) in madscientist's first Shaka role play game where she completely wiped out De Gualle from the game...Zara Yaqob (Cre/Org) in pretty much EVERY game I play (oh yeah, and Zara in one of flouzemaker's games where he vassalized MONTEZUMA!)... I am sure there are many other games that I could use as an example.

The last creative civ I want to touch on is Kublai Khan (Agg/Cre). In my experience, he is the only agressive AI who has a dominant army AND a tech lead as well. Even Ragnar doesn't seem to match Kublai. I don't have saves for a game in which Kublai Khan won by once again, taking over a whole continent on one of my emperor games (I play continents often which means I do not influence what AI's do on the other continent until optics). I do remember, though, how Kublai Khan in Snaaty's current Don Deity game handled his continent quite well, vassalizing Shaka (I believe) and perhaps vassalizing another (I don't remember). I am hoping to play several more random maps while preselecting some creative civs to verify my belief (I have, believe me, played many games with creative civs to justify my belief).

So why is it that creative civs are often top contendors? Pericles, too, is a top contendor for cultural victories like Louis XIV and he is creative as well. Is it because the AI's are slow to pop borders so they have less land initially? I don't really know how creative civs do well in this game. Does anyone else have this nagging feeling of how creative civs seem to do better (in most cases -- yes I know that tundra starts will nerf any civ but an average start...).

The second part of my thinking that occurred for the past week was the weakness of Wang Kon (Fin/Pro) as an AI.

Now, at first, I was thinking, WHAT?!? This guy should be one of the best AI players! Financial is self-explanatory...and PROTECTIVE! He should be able to fend off AI axemen attacks with uber longbows due to his teching!

But that is not the case as I have been noticing often. From time to time, whenever I see Wang Kon in games, he NEVER seems to do well and is usually defeated or on the bottom of the scoreboard. I have had that experience as well in my games (once again--these were games where Wang was on a different continent and not isolated so I had nothing to do with early rushes or anything like that).

Some games which Wang Kon has not done well in:

1. Aelf's immortal challenge again, as mentioned above, where Louis XIV (Creative civ!!!) vassalized Wang Kon.

2. Flouzemaker's Suleiman game where Wang Kon was eliminated early. Now, you may argue that Flouzemaker rushed him, but from reading the game, De Gualle of France did almost all the damage to Wang Kon so if De Gualle and Wang Kon were by their own on the same landmass they were playing on, De Gualle probably would have vassalized him.

3. 3 of obsolete's wonder spam games, where Wang Kon was vassalized by catherine in one and (in his current without the mids one) was eliminated completely (looks like Montezuma work, and if you are arguing that Wang Kon is eliminated because he has agressive civs nearby, take a look back in my post about Zara Yaqob -- yet another creative civ -- who vassalized Montezuma when they were on their own island). In a third one, Wang was eliminated but the human player (obsolete) was directly involved so we can ignore that.

4. There was a story in the stories/tales forum relating to Korea, but the maker terminated it because when he made contact with Wang Kon, what do you know? Wang already vassalize early.

5. One of my recent games with Wang on another continent who was defeated completely by Mehmed II.

I have browsed through countless games and I am sure that I saw several more games with Wang Kon on the bottom of the scoreboard, so I'm wondering, what's with him? He's not like Tokugawa (I can understand why Tokugawa never seems to be a top contendor, but Wang Kon...?).

My last thing to say is, does anyone agree with this? Does anyone notice this tendency? If anyone can provide saves, or links to other threads with dominating creative civs or a weak wang kon, or maybe a reverse, with wang kon ruling the world, I'd like to know. And yes, I KNOW that various factors such as starting land, etc affect how civs do, but this tendency.....

And I would also like to know why creative civs seem to do so well...I would rather have a financial trait.

Howard!
Feb 04, 2008, 08:37 PM
The French struggle...

I won a game through a space race in 1750 with the french on an Earth 34... The French are awesome to use in that map because Europe is so compact. Just declare war and take... I took the English by culture.

Bak to the topic. The same can be said for sitting bull. He always screws himself over. If you start near him, and refuse one of his random demands, no matter my power he has always decleared war on me... Wait, did I say war, I meant an invasion by his one warrior!

Normally he ends up as a one city vassal (anywhere on the globe) and yeah... He is the bastard who always buggers up my attempt to change the world the environmentalism in order to lauch my attack of creative jewlers inc. executives upon the world [damn commi-nazis!]. That's right, he always votes never to my environmentalism proposal!

DMOC
Feb 05, 2008, 05:34 AM
Yeah Sitting Bull is another weak AI player, I've noticed. Although he did do well in a Dutch immortal game or something like that I recall.

More comments?

Sjaramei
Feb 05, 2008, 06:05 AM
Gilgamesh is insane too, if he is in the game you can be sure he will get a huge empire.

Howard!
Feb 05, 2008, 06:26 AM
I think the thing is that some computers need some really specific circumstance for them to do well. I remember one game where Monty the space race... He was near a domination!

Unthinking Pain
Feb 05, 2008, 08:37 AM
Suvarayman (sp?) also seems to do well.

I think the AI doesn't do a great job chopping out Monuments and getting border pops fast enough, and that can slow their whole game down.

VirusMonster
Feb 05, 2008, 08:56 AM
na, I think the many examples you gave do not prove your point... 100s of games are being played and written up in these forums; some of them would have to be won by creative leaders, that is all.

Creative is a somewhat good trait at Deity Difficulty however. It helps you to block AI settlers from crossing you territory, because building those monuments and waiting for border expansion takes some time.

von Krysiak
Feb 05, 2008, 09:21 AM
Yeah there are many AI players I could mention:
Napoleon is a maniac, war junkie and is very likely that he will hve a huge empire - so is Catharina and Gilgamesh.
Monty is just an another example of a neighbour you dont want to be getting
But MOST dangerous (at lest fairly early in the game) are Genghis Khan, Wang Kon - them guys expand like crazy and in the effect their technology lacks behind. But if they start a war before their technology is outdated they will make short work of anyone.
In Most of my games I always see Hannibal, Zara Yaqub, Isabela as most advanced nations plus the combination of any of the above leaders due to their large territory and population...

Iranon
Feb 05, 2008, 09:40 AM
Traits matter little compared to personality... and as it happens the Creative AIs tend to follow a balanced and sound strategy:

Catherine, Gilgamesh and Suryavarman are opportunistic warmongers, being a menace without screwing up their economy by default.
Pericles and Louis are the usual suspects to go for a cultural victory, which can be quicker than the usual spaceship approach especially if there was a lot of fighting in the game.
Zara Yacob and Willem are first-rate empire builders who don't totally neglect their military as others often do; Zara is quite good at getting a big empire without being backwards and Willem just plays to win... nobody else managed to surprise me as much in the late game (beeline military techs and competently starting world wars, a switch to a cultural victory strategy, unexpected naval invasions...).

I haven't really seen Kublai Khan do anything amazing so far (for some reason, the Mongols are usually far away from me if they are in the game), and Hatshepsut seems a total push-over.

***

If there is something in CRE itself rather than the associated personalities... maybe it's because the AI is completely pants at picking good city sites and CRE prevents them from shooting themselves in the foot too badly?
People who played more games with random personalities might be able to verify this.

von Krysiak
Feb 05, 2008, 09:49 AM
Traits matter little compared to personality... and as it happens the Creative AIs tend to follow a balanced and sound strategy:

Catherine, Gilgamesh and Suryavarman are opportunistic warmongers, being a menace without screwing up their economy by default.
Pericles and Louis are the usual suspects to go for a cultural victory, which can be quicker than the usual spaceship approach especially if there was a lot of fighting in the game.
Zara Yacob and Willem are first-rate empire builders who don't totally neglect their military as others often do; Zara is quite good at getting a big empire without being backwards and Willem just plays to win... nobody else managed to surprise me as much in the late game (beeline military techs and competently starting world wars, a switch to a cultural victory strategy, unexpected naval invasions...).

I haven't really seen Kublai Khan do anything amazing so far (for some reason, the Mongols are usually far away from me if they are in the game), and Hatshepsut seems a total push-over.

***

If there is something in CRE itself rather than the associated personalities... maybe it's because the AI is completely pants at picking good city sites and CRE prevents them from shooting themselves in the foot too badly?
People who played more games with random personalities might be able to verify this.

I totally agree with you on Zara Yaqub - last game I was playing on Huge map/Epic I conquesred one player and then had Zara and 5 other more or less backward civs left there. While I wanted to build up my economy and techs Zara beelined his military techs so he remained on EXACTLY same lvl as me AND managed to vasalize every other civ on the continent so my tanks had to wage war against him and hordes of longbowmen and knights as well!

JustinianVII
Feb 05, 2008, 01:25 PM
Pericles is a culture monster. Creative/Philosophical means the free +2, faster Great People, cheaper Libraries, Theatres, Odeons (which gives +3 culture), Universities...it's a bit easier for him.

pi-r8
Feb 05, 2008, 02:14 PM
Suvarayman (sp?) also seems to do well.

I think the AI doesn't do a great job chopping out Monuments and getting border pops fast enough, and that can slow their whole game down.
I think this is the reason why the AI does well with creative leaders. Often I'll see them with multiple size 1, 0 culture cities, all doing nothing but ruining their economy. Creative just fixes one of the AI's largest problems.

Also, I don't agree about Wang Kon. Sure he's not the best AI, but he's not too bad either. I had one game where he vassalized 3 other AIs, then came after me with the largest modern army I've ever seen.

DMOC
Feb 05, 2008, 02:18 PM
I think this is the reason why the AI does well with creative leaders. Often I'll see them with multiple size 1, 0 culture cities, all doing nothing but ruining their economy. Creative just fixes one of the AI's largest problems.

Also, I don't agree about Wang Kon. Sure he's not the best AI, but he's not too bad either. I had one game where he vassalized 3 other AIs, then came after me with the largest modern army I've ever seen.


really? wow, that's a first for me.

So the AI doesn't really expand their borders is what I was thinking about when I wrote this. I also didn't seem to notice much of a difference in the AI personalities so maybe the Creative civs are by chance good traders with a strong military.

I really should do a demo game, perhaps I will post one up here preselecting some AI's.

AfterShafter
Feb 05, 2008, 02:32 PM
All the Ai's are on Noble, and on Noble, traits aren't nearly as game breaking as they are on higher difficulties. Civs with great traits have notably less of an advantage on Noble since everyone is pretty damned good. Thus, a guy who is a very, very good leader/Civ like Wang Kon of the Koreans does reflectively poorly compared to other Civs who might not be as good.

Also, as was said, personality matters more. Wang just doesn't go for the jugular... Mansa Musa has the personality he should have - total ass kisser, keeping him diplomatically well off, absolute tech monkey.

DMOC
Feb 05, 2008, 10:10 PM
Well a lot of games that I noticed were not on noble difficulty level, they were actually on emperor, immortal, and so on.

On noble I understand that AI traits aren't game breaking and therefore that's why AI's often do not become extremely powerful on lower difficulties but on immortal difficulty in my games and the games I notice on the forums, Wang keeps getting his ass kicked.

morchuflex
Feb 06, 2008, 05:51 AM
Hello.

I think creative is underrated. To experience how good this trait is, the best way is to play a lot of creative games, then revert to a non-creative leader. Talk about suffering! :cry:

After playing Louis a lot, and doing decently on emperor, I decided to play a non-creative leader. It was a real pain, because all the great city sites I had in my vicinity demanded border expansion to shine (that is, the key resources were always TWO tiles away). Three of my cities had huge potential, but I first had to build monuments, and it took forever, because the adjacent tiles contained no food resources (hence no whipping) and little production.
And after the monuments were completed, I still had to wait 10 more turns for the expansion to occur.

Of course, newbies will say something like "why didn't you build Stonehenge" or "why didn't you found a religion"... But this is Emperor level, which pretty much excludes early religions except for financial leaders or those starting with mysticism. And I don't want to rely on the dubious habit of always building the same early wonder. Furthermore, I play archipelagoes, and I need my capital's forests for the absolutely essential GLH.

Overall, I think creative is a great trait, especially on water maps, when you depend a lot on seafood and need to hook these fishes ASAP, not in twenty-five turns.

Still_Asleep
Feb 06, 2008, 10:18 AM
Hello.

I think creative is underrated. To experience how good this trait is, the best way is to play a lot of creative games, then revert to a non-creative leader. Talk about suffering! :cry:

After playing Louis a lot, and doing decently on emperor, I decided to play a non-creative leader. It was a real pain, because all the great city sites I had in my vicinity demanded border expansion to shine (that is, the key resources were always TWO tiles away). Three of my cities had huge potential, but I first had to build monuments, and it took forever, because the adjacent tiles contained no food resources (hence no whipping) and little production.
And after the monuments were completed, I still had to wait 10 more turns for the expansion to occur.

Of course, newbies will say something like "why didn't you build Stonehenge" or "why didn't you found a religion"... But this is Emperor level, which pretty much excludes early religions except for financial leaders or those starting with mysticism. And I don't want to rely on the dubious habit of always building the same early wonder. Furthermore, I play archipelagoes, and I need my capital's forests for the absolutely essential GLH.

Overall, I think creative is a great trait, especially on water maps, when you depend a lot on seafood and need to hook these fishes ASAP, not in twenty-five turns.

Chopping works, even outside your cultural boarders. In the rare case you dont have a food ressource for whiping OR a hill for a quick mine OR 1-2 trees for quick chopping, you should consider settling in the slightly suboptimal spot to get a decent start - but all those circumstances rarely kick in at once.

Creative is a fine trait, but I come along without it pretty well, playing 18 civ pangea on emporer/immortal.

Supr49er
Feb 06, 2008, 10:29 AM
Chopping works, even outside your cultural boarders. In the rare case you dont have a food ressource for whiping OR a hill for a quick mine OR 1-2 trees for quick chopping, you should consider settling in the slightly suboptimal spot to get a decent start - but all those circumstances rarely kick in at once.

Creative is a fine trait, but I come along without it pretty well, playing 18 civ pangea on emporer/immortal.

If the forest is between you and a rival, mouse over the chop icon to see who will get the benefit of the chop. I have chopped only to see a rival city suddenly complete their build. :mad:

jray
Feb 06, 2008, 10:36 AM
If the forest is between you and a rival, mouse over the chop icon to see who will get the benefit of the chop. I have chopped only to see a rival city suddenly complete their build. :mad:

Yeah, that's the stupidest game mechanic ever! It would be one thing if it was an automated Worker, which we all know are morons. But this is just a case of sheer laziness and insubordination. "Uh, we have to climb a hill to take this wood back to Antium. Let's just take it to Edirne instead."

ChristofferC
Feb 06, 2008, 02:07 PM
Well a lot of games that I noticed were not on noble difficulty level, they were actually on emperor, immortal, and so on.

On noble I understand that AI traits aren't game breaking and therefore that's why AI's often do not become extremely powerful on lower difficulties but on immortal difficulty in my games and the games I notice on the forums, Wang keeps getting his ass kicked.The thing is that the AI:s always play on Noble difficulty, no matter what difficulty the player chooses. If all the AI:s got the same bonuses as the player on all difficulty levels, there woulden't be much difference between settler and diety.

Diamondeye
Feb 06, 2008, 02:19 PM
It's a complicated issue, because it does not only rely on the traits. Most Creative leaders hae a really good AI personality. And Wan Kon has a really bad one.

Oh, and concerning Sitting Bull, he is nearly always a loser because he is the way he is, but if he is left alone for some time (not isolated, just not war), he can get to be a massive enemy.

jray
Feb 06, 2008, 02:21 PM
The thing is that the AI:s always play on Noble difficulty, no matter what difficulty the player chooses. If all the AI:s got the same bonuses as the player on all difficulty levels, there woulden't be much difference between settler and diety.

I have no clue how multiplayer with AI's works, but at least in single player, there is no such thing as an "AI Difficulty." All of the AI modifiers are completely dependent on the human's difficulty level. For example, if the human is playing Monarch, then all AI's get free Archery and an extra Archer and a 20% work rate bonus. If the human is playing Emperor, then all AI's get free Hunting+Archery, free extra Archer and Scout, and a 50% work rate bonus. All AI bonuses/penalties are built in to the human difficulty level.

So I'm curious, how does this work in multiplayer?

Silence101
Feb 06, 2008, 02:42 PM
Creative is a powerful early game trait - extremely powerful. It helps AI's establish a strong position in the early game. A strong early game position compounds as the game continues... assuming that the available land decent, Creative AI's will have more land, and thus more production, commerce, and resources, giving them the ability to expand, conquere, tech, and build faster. This can create a sizable advantage by the mid-game and later.

jray
Feb 06, 2008, 03:30 PM
Creative is a powerful early game trait - extremely powerful. It helps AI's establish a strong position in the early game. A strong early game position compounds as the game continues... assuming that the available land decent, Creative AI's will have more land, and thus more production, commerce, and resources, giving them the ability to expand, conquere, tech, and build faster. This can create a sizable advantage by the mid-game and later.

Agreed. The high-culture cities also make the AI less likely to become the target of a human. I think Gilgamesh is perhaps the ultimate AI opponent, being Cre/Pro with high-culture cities defended by CG1-D1 archers. Gotta think twice before axe-rushing that. While he doesn't tech as fast as Financial leaders or settle as fast as Imperialistic ones, he does tend to form a stable core in the early game, resist and save money from those Ziggurats. Well, I guess he does also have a couple weakness: he spams Vultures that are only really valuable on offense, in lieu of more protective Archers, and falls prey to Wonder addiction sometimes.

Sian
Feb 06, 2008, 04:13 PM
well ... i'm in the middle of a Huge Marathon game as Etiophia on Noble (easymode though since i manged to axe my two neighbours before they got archers) ... and the only AI that can somewhat keep up with tech is Wang Kon even though his starting area is bad (only one worse is Mongolia who started on a 5 tilte pennesula with a mountain blocking the rest of the island) ... okay it might help him a little that his placed in the middle of everyone and founded Budism which have become the dominant religion ... only ones not budists are me, Greek (confusists together with me) and Gandhi (hinduist owners)

of my knowlegde Wang Kon will do fine as long as he doesn't have a warmonger or massivly explanding as nextdoor neightbor

d.a.oconnell
Feb 06, 2008, 06:29 PM
I find Zara to be one of the best AI's in the game. He generally always does very well unless I early war him. Creative and Organized... he's a tech monster and generally grabs a lot of land. Anytime I meet him I try to wipe him out asap.

Sian
Feb 06, 2008, 06:33 PM
... and the longer you can stall Astronomy the better are you looking in terms of Cultural victory ... 25% might not be much for a newly born city ... but if you'll get some cultural going early on they are really helpful

TheLastOne36
Feb 06, 2008, 06:46 PM
I think is the expansion. All civs that get land quickly end up strong.

Which means creative and impieralistic civs. But for some reason Im[ieralistic civs never seem to want more and more cities(except for joao)

Joao is civ 4's top city rexxer! and he's always at the top of the table to with creative civs.

Maybe i'll make a game with all creative civs, and joao.

DMOC
Feb 06, 2008, 08:01 PM
I was thinking of doing some tests this weekend or during president's week. I could fire up some random CONTINENT maps, standard sized with the typical seven AI civilizations. I will probably preselect Wang Kon for almost every game I play along with creative civs. I will also check worldbuilder after everyone has settled their capitals to see where the civs are. If Wang Kon or any creative civ that I have selected is on my continent, I will either destroy their capital and recreate the capital in the other continent with the same units as well so gameplay is affected minimally, or move my capital to the other continent. I want to see what happens with creative civs and wang kon on other continents.

So maybe, game 1 could be like this:

Wang Kon, Kublai Khan preselected to start on their own continent, along with 5 other non creative civs. Optimal placement would be wang, kublai, and a middle of the pack non creative AI (Peter works very well for this). Of course, they will all not be on my starting continent. Speed will probably be Normal to speed things up, and difficulty will range from Noble to Immortal (don't know what difficulty level is optimal.

Then maybe I could do Wang Kon and ... Louis XIV/Pericles/Hatshepsut/Gilgamesh/Willem/Zara etc.

I could also pair up any creative civ along with one warmonger (Shaka for instance) on their own continent to compare with non creative civs to see the result.

This is all, assuming, time permitting. And yes, I will use worldbuilder for every game I test to check for any obviously overpowered starts (I may very well remove any gold mines from capitals) to provide an even playing field.

Diamondeye
Feb 07, 2008, 09:38 AM
I have no clue how multiplayer with AI's works, but at least in single player, there is no such thing as an "AI Difficulty." All of the AI modifiers are completely dependent on the human's difficulty level. For example, if the human is playing Monarch, then all AI's get free Archery and an extra Archer and a 20% work rate bonus. If the human is playing Emperor, then all AI's get free Hunting+Archery, free extra Archer and Scout, and a 50% work rate bonus. All AI bonuses/penalties are built in to the human difficulty level.

So I'm curious, how does this work in multiplayer?

My guess is that the highest difficulty dictates what the AIs start with, and anyone below noble still get their bonuses.

I find Zara to be one of the best AI's in the game. He generally always does very well unless I early war him. Creative and Organized... he's a tech monster and generally grabs a lot of land. Anytime I meet him I try to wipe him out asap.

Yeah, ZY is dangerous, although I find him a bit inept at diplomacy - last game I played he founded Taoism and converted, effectiely disconnecting him from techtrades with other nations. He even had had the Christian AP (although I took it :devil:).

I think is the expansion. All civs that get land quickly end up strong.

Which means creative and impieralistic civs. But for some reason Im[ieralistic civs never seem to want more and more cities(except for joao)

Joao is civ 4's top city rexxer! and he's always at the top of the table to with creative civs.

Maybe i'll make a game with all creative civs, and joao.

I have noticed this aswell. Try playing as Joao II yourself and really REX. Did it once on a LAN (buddy was Hatty), and I went city-crazy, hitting 30%:research: with courthouses. Played well, nobody tried to bully me and I actually defeated India and Vikings later on.

Diamondeye
Feb 07, 2008, 09:39 AM
I have no clue how multiplayer with AI's works, but at least in single player, there is no such thing as an "AI Difficulty." All of the AI modifiers are completely dependent on the human's difficulty level. For example, if the human is playing Monarch, then all AI's get free Archery and an extra Archer and a 20% work rate bonus. If the human is playing Emperor, then all AI's get free Hunting+Archery, free extra Archer and Scout, and a 50% work rate bonus. All AI bonuses/penalties are built in to the human difficulty level.

So I'm curious, how does this work in multiplayer?

My guess is that the highest difficulty dictates what the AIs start with, and anyone below noble still get their bonuses.

I find Zara to be one of the best AI's in the game. He generally always does very well unless I early war him. Creative and Organized... he's a tech monster and generally grabs a lot of land. Anytime I meet him I try to wipe him out asap.

Yeah, ZY is dangerous, although I find him a bit inept at diplomacy - last game I played he founded Taoism and converted, effectiely disconnecting him from techtrades with other nations. He even had had the Christian AP (although I took it :devil:).

I think is the expansion. All civs that get land quickly end up strong.

Which means creative and impieralistic civs. But for some reason Imperialistic civs never seem to want more and more cities (except for Joao)

Joao is civ 4's top city rexxer! and he's always at the top of the table to with creative civs.

Maybe i'll make a game with all creative civs, and joao.

I have noticed this aswell. Try playing as Joao II yourself and really REX. Did it once on a LAN (buddy was Hatty), and I went city-crazy, hitting 30%:science: with courthouses. Played well, nobody tried to bully me and I actually defeated India and Vikings later on.

InFlux5
Feb 07, 2008, 12:00 PM
I think it's fine for some AIs to be weak, especially if they were historically a small or weak civ (e.g. Korea, Native Americans.)

CHEESE!
Feb 07, 2008, 12:44 PM
Well France under Louis XIV wasnt that super-powered was it?

DMOC
Feb 09, 2008, 11:22 AM
Okay, I rolled up a test game this weekend. I was Ramesses, game speed Normal, difficulty monarch, and I ended up alone on my continent with Gandhi. On the other, 5 civ continent, Wang Kon and Kublai were quite close, although Wang had a REALLY good floodplains start with iron in his cross... Kublai was kind of boxed in by Washington so Kublai took 4 of his cities. Frederick was the strongest of all the AI's, vassalizing Washington as well as Peter who was nerfed by a tundra start.

Overall, from this game, I see that Wang Kon does do well, provided the conditions. I kind of judged him too soon, I guess. We'll see in a few more test games.

Screenshots:

http://img338.imageshack.us/img338/705/koreaandmongolia0000ey6.jpg

http://img338.imageshack.us/img338/4702/koreaandmongolainsideviqo1.jpg


By the way...Kublai's score was, I think, a hundred points or two hundred points above Wang. He wasn't near to Frederick, that's for sure. And me...I had between 5000 and 6000...grabbing that bottom island really helped. (The screenshot is cropped to keep my name private.)


The AI rankings were as follows:

1. Frederick. I dont' know how, but he somehow vassalized Washington AND Peter...impressive. And he gave me a run for my money in research. Frederick vassalized washington by taking 2 of his cities...and Kublai actually took 4 American cities and didn't vassalize him...hm....

2. Wang Kon. So he does play well. I never saw his full potential since Kublai asked me to join him in a war vs Wang so I nuked him 13 times as evident from the screenshots provided.

3. Kublai. Had a mediocre start without horses nearby and was locked into wars with Washington and Wang. He probably would have lost to Wang in his wars even though they traded cities MANY times but I gave him help.

4. Peter, who got nerfed by a tundra start.

5. Washington. Absolutely horrible, wound up with 3 cities.

6. Gandhi. Down to 2 cities and was my vassal.

Diamondeye
Feb 09, 2008, 12:34 PM
Well France under Louis XIV wasnt that super-powered was it?

He was a very good general, I have heard. Although he lived a little too luxurious, so the peasants weren't very happy with him. Among his luxurious expenditures, Versailles and it's constant maintenance is a rather good example.

DMOC
Feb 09, 2008, 12:56 PM
He was a very good general, I have heard. Although he lived a little too luxurious, so the peasants weren't very happy with him. Among his luxurious expenditures, Versailles and it's constant maintenance is a rather good example.


Actually, I'm not sure on wether France was superpowered under him. Louis basically tried to curtail the power of everyone but himself--hence the absolute monarchy that he is most famous for. I don't think he was that good a general because whenever he fought wars, there would be several contries allying themselves into an anti-France coalition (remember the War of the Spanish Succession?). If you compare him to, say Bismarck, Bismarck always made sure to isolate his enemy when waging a war. Bismarck's second war verses Austria is a pretty good example of this; Bismarck got Russia to agree on neutrality in the case of a war and he also got Napoleon III of France to stay neutral by providing vague promises of territory.

What I think of France is that it would have definitely been better off without waging the highly costly wars Louis waged.

Weird thing is, though, the current Louis AI goes for cultural victories and not domination. Actually, no AI really goes for domination, but you get the picture.

TheLastOne36
Feb 09, 2008, 04:08 PM
I think it's how creative/stonehenge civs settle:

Peter is not creative, but in my game he had stonehenge, and this is how he settled:
http://i25.tinypic.com/2v309qq.png
(ignore my cheat of giving myself an explorer, and i'm using a map that puts floodplains on any tile next to a source of fresh water(lake or river, much more realistic)
Yah, he settles like a human. The Creative culture bonus/per city and the stonehenge's free monuments let the city expand it's borders before the ai settles another city. This makes it so that the AI doesn't settle crappy cities that overlap with other cities just so they fill up tiles quicker.

Joao however has crappy city sites but alot of cities, he get's his high score from the amount of cities he has, not from what his cities produce. It's just because he's the best AI rexer in the game.

Therefore, Creative/Stonehenge civs settle better sites that pay of at the end, get more resources, and don't have overlapping cities.

lovetramy
Feb 09, 2008, 11:40 PM
Cathy is the best AI , always top score in all games I play with her ... Cre + Imp = mass land always
and she has nice personality too ( kind of :D )
Sitting Bull definitely the worse . In my recent Prince game he has more land then other civ . Ok its 1800's and he researching .... Education :D dont know what the hell he doing though . but my tanks and bomber like it :D

Lance of Llanwy
Feb 10, 2008, 12:07 AM
Actually, I'm not sure on wether France was superpowered under him. Louis basically tried to curtail the power of everyone but himself--hence the absolute monarchy that he is most famous for. I don't think he was that good a general because whenever he fought wars, there would be several contries allying themselves into an anti-France coalition (remember the War of the Spanish Succession?). If you compare him to, say Bismarck, Bismarck always made sure to isolate his enemy when waging a war. Bismarck's second war verses Austria is a pretty good example of this; Bismarck got Russia to agree on neutrality in the case of a war and he also got Napoleon III of France to stay neutral by providing vague promises of territory.

What I think of France is that it would have definitely been better off without waging the highly costly wars Louis waged.

Weird thing is, though, the current Louis AI goes for cultural victories and not domination. Actually, no AI really goes for domination, but you get the picture.
France was, for a very long time, by far the strongest European state, as evidenced by its ability to run roughshod over basically the whole continent a century after Louis XIV's time. England and the rest of the European states were chiefly interested in maintaining the balance of power, always precarious with such a mighty state as France in their midst, and this is why France often ended up fighting broad alliances(supporting France, naturally, running entirely counter to balance of power). Bismarck was able to manage it not only because was a peerless diplomat, but also because France was traditionally isolated.

As for Louis XIV himself, he was a control freak precisely because of a rebellion when he was something like 10-12 years old that forced him to flee the palace in Paris. He never got over that, and his desire to punish Paris for his insolence and simultaneously tame the French nobility led to him building Versaille, complete with archaic rules of etiquette, and such odd priviledges as witnessing the King shave. Strictly speaking, it did, indeed, drastically reduce French stability, for the small price of millions of dollars and reducing the noble class to skill-less buffoons jockeying over who would escort the king to his bed-chambers every night. He also revoked an earlier royal edict of tolerance for Protestants. He became downright puritanical(in the Cromwell sense) towards the end of his reign. Mechanically, he laid the framework for perpetuation of absolute monarchy in France, but his successor and second grand-son, Louis XV was frankly a terrible monarch, mostly because he spent his time organizing his own private brothel and anything that wasn't ruling the country....

Thrasybulus
Feb 10, 2008, 02:10 AM
Not long after BtS came out I kept track of all of my results in similar games (Large map, 11 AIs, Hemispheres, Monarch, no tech brokering). At the time I was winning on Monarch only about a third of the time so these results aren't unduly influenced by civs getting whacked for startiing out next to me.

18 civs appeared in at least 3 of these games and their average rankings at the end of the game are:

Top Tier (average 1st to 4th place): Zara Yaqob, Pacal, Willem, Joao, Mansa Musa

Mid Tier (average 5th to 8th place): Bismarck, Justinian, Pericles, Gilgamesh, Tokugawa, Hammurabi, Augustus

Bottom Tier (average 9th to 12th place): Mehmed, Gandhi, Charlemagne, Ragnar, Isabella, and dead last....Montezuma.

Of course score isn't everything. I think Willem's worst result was one in which he won a cultural victory.

It would take a huge amount of results to really make anything of this and some leaders are obviously suited to certain types of maps, but I might start doing this again on Emperor on which I rarely win.

BalbanesBeoulve
Feb 10, 2008, 05:16 AM
Frederick managed to do well militarily and even get vassals? That's pretty surprising. In most games I've played with him he always ends up with a tiny, weak empire.

DMOC
Feb 10, 2008, 09:22 AM
Not long after BtS came out I kept track of all of my results in similar games (Large map, 11 AIs, Hemispheres, Monarch, no tech brokering). At the time I was winning on Monarch only about a third of the time so these results aren't unduly influenced by civs getting whacked for startiing out next to me.

18 civs appeared in at least 3 of these games and their average rankings at the end of the game are:

Top Tier (average 1st to 4th place): Zara Yaqob, Pacal, Willem, Joao, Mansa Musa

Mid Tier (average 5th to 8th place): Bismarck, Justinian, Pericles, Gilgamesh, Tokugawa, Hammurabi, Augustus

Bottom Tier (average 9th to 12th place): Mehmed, Gandhi, Charlemagne, Ragnar, Isabella, and dead last....Montezuma.

Of course score isn't everything. I think Willem's worst result was one in which he won a cultural victory.

It would take a huge amount of results to really make anything of this and some leaders are obviously suited to certain types of maps, but I might start doing this again on Emperor on which I rarely win.


Tokugawa was actually on the middle tier? For him, that's comparable to Hatshepsut winning by conquest. :lol:


Frederick managed to do well militarily and even get vassals? That's pretty surprising. In most games I've played with him he always ends up with a tiny, weak empire.


Yeah, I heard that he's pretty weak usually like most peaceful AI's. I don't know why Washington vassalized to Frederick; he should have vassalized to Kublai Khan because Kublai captured more American cities.


Anyway, I'll make a test game sometime next weekend if I can. And I'll be more specific as well instead of just posting the endgame screenshots.

InFlux5
Feb 11, 2008, 11:24 AM
It would take a huge amount of results to really make anything of this and some leaders are obviously suited to certain types of maps

Exactly. To answer this question you would need an extremely large data set. There are so many variables that go into any game of civ, it would be extremely hard to attribute a civ's performance in any game primarily to their traits, or their starting land, or their personality, or the civs they started next to, etc.

Then there's the added difficulty of how to actually rate a particular AI in any particular game. Game score is largely irrelevant, as someone made clear with their example of a Dutch cultural victory despite a low score.

Still, as a general measure you could just go with score. You'd still need a huge number of sample games to begin to look for any correlations.

DMOC
Feb 19, 2008, 10:01 AM
Okay guys, a second test game. This was actually bit of a surprise since I did NOT preselect any AIs. I was playing as Pericles and attempting a specialist economy when, to my surprise, I found both Wang Kon and Suryavarman II in this game. So I played it out and decided to post a summary of the game.

Background: Monarch difficulty, Epic speed, Continents map, Standard size, 8 total players (7 AI), and there were 4 civs on my starting continent and 3 on the other continent.

AI's on my continent: Wang Kon, Isabella, and Saladin

AI's on the other continent: Suryavarman II, Mansa Musa, Lincoln, and later Zara Yaqob in around 1700 as Lincoln's colony.


Okay, in this game, Wang Kon started on the very tip of our continent and I started in the middle area but "expanded" south when I phalanx-rushed Isabella and took her Christian holy city (surprise anyone?) so Wang had tons of room to expand. Saladin was also to the south and on the western side. The map kind of looked like this:


(X= capital, .= land)

...X (wang's)..
........
........
..........
...........
..........X (mine)
.............
..............
X (saladins)........
...................X (isabella)
................
.................

And wang had a pretty decent start as well. Coastal, with some fish, hills, and flooplains. So, with me in the middle and expanding southwards and Wang freely able to settle land, you'd think he'd do well, right?

WRONG.

Okay, after I rushed Isabella, I continued my expansion to about 5 or 6 cities. Saldin? he expanded to TEN OR TWELVE cities by around 300-400 AD! And he peaked out at 14 before I started warring him in 1000 AD.

What about wang kon? well, when I traded techs with him, I could also see the number of cities he has in the information column. Guess what? By the time Saladin had 14 cities and I had around 8 or 9 (but better quality cities than saladin which led to his downfall), wang had a grand total of...THREE!!!! Three cities by 300 AD!!!!!:crazyeye: And it wasn't like Saladin and I expanded towards him at all, Saladin and I were basically competing for land in the middle and southern portion of the continent. Wang had the entire northern third to himself, which was good land and had tons of floodplains, and he only created 2 new cities. Needless to say, by the time I had conquered Saladin, wang was hopelessly behind in tech and was easy pickings for me. Wang did eventually expand to 8 cities total (there was room for clearly 8 cities on the northern third so don't think that Wang didn't have room to expand when he had 3 cities). And to clear some questions, wang did not have a barbarian uprising event and there was only one barbarian city near him with 3 archers in it. Clearly, barbs did not slow him down, so what did?


And I haven't even talked about the other continent yet. There was one AI who was the clear SUPERPOWER on the other continent, and that was...you guessed it: Suryavarman II, another civ with the creative trait!

Now, I looked at all the capital starts. None were too powerful, although Sury did have stone, but I don't know how well he utilized it. The map of the continent is shown below in the "spoiler" screenshot. This was in the late nineteenth century when I had already conquered my own continent and was invading Sury in an attempt for domination (and Sury and I had already defeated Mansa by then):

http://i272.photobucket.com/albums/jj197/DMOC_album/Suryandwangkongame2surysland0000.jpg


Notice something in the screenshot? Aside from my cities that I got from capturing Mansa (and the three cities I captured from Sury, which you can see are the cities still in revolt) and aside from the two cities Zara Yaqob has, Sury has control of practically the ENTIRE CONTINENT!

And if you continue looking on the map, Sury's capital is to the mid-south of the continent, while...a Washington, a New York, a Philadephia...and other American cities are in Sury's possession. Is there a single American city remaining? No. Suryavarman II wiped out Lincoln from the map entirely! And he didn't even vassalize him! :crazyeye: (This is, by the way, the first time in one of my games where an AI kicked another AI out of the game--so now Sury definitely has my respect in future games.) Sury actually managed to capture two of Zara's four pathetic cities amphibiously, although since Zara was a colony spawned off the now-dead Lincoln, they were likely weakly defended. Sury managed an amphibious attack with an insane amount of cavalry using only galleys (he didn't research Astronomy at the time) and vassalized Zara. So now it was him that I had to fight in order to win a domination victory (yes I made the mistake of vassalizing Wang and Saladin).

As I did in the last game, I will do some AI rankings:

1. Suryavarman II earns first place. No surprise at all. Also, if you take a look at the screenshot I posted up, all but two of Sury's cities is building something military-oriented! This guy and Montezuma were definitely separated at birth!

2. Allright, I have some trouble here deciding who earns second. Lincoln? Nah, he got bulldozed. Mansa? Maybe he could earn this place, as he was doing allright but nonetheless, would have probably died to Sury and was fortunate that Sury hated Lincoln more. Or maybe Saladin could earn this spot. He gave me a good fight.

3. Zara Yaqob (yes he is a creative civ but since he spawned as a colony, he's irrelevant, really).

4. Lastly...Wang Kon and Isabella! I don't know which one performed worse in this game. Wang, who only expanded to three cities, or Isabella, who only had one archer defending each of her three cities in 700 BC and built THREE christian missionaries when I conquered her.

So, in conclusion, I can see more evidence of the power of Creative civ. I will definitely be preselecting Sury in some future games that I play to see how he does.

Sian
Feb 19, 2008, 01:51 PM
an interesting idea would be to start a normal game ... and then move yourself to a 1tile island circled by coast and a ring of mountains (and maybe another ring of mountains so planes can't get to you either) ... and get the game to play out by it self without joining the fray (giving yourself Satelites in the process so you can shadow the AI's), and when alls said and done see who did the best job (don't know how to do this but it must be possable in one way or another)

If it aren't possable it would be an interesting Toolmod where you could start a game with only AI's and then get it to give you a world map from every 20'th turn, and from there fastforwarding 20 turns more. Rhye's Raise and Fall should have most of the basics by its starting system

EDIT: just checked around and it should be doable with liberal use of AIAutoPlay (http://forums.civfanatics.com/showthread.php?t=174812) maybe as a proxy Civ with Satelites

DMOC
Feb 19, 2008, 05:03 PM
Hm, interesting Sian. I do that sometimes (but the only games I do that are on deity pangea map slugfests. In my last game, GANDHI out of 17 total computer players and me on my own island won a space race with about 5 vassals. Boudica was the second best AI.

Zabumon
Feb 21, 2008, 11:11 AM
I agree with pretty much everything you said in the first post, based on my experience.
Louis XIV and Zara Yacob are among the best AIs, maybe the best 2. I've seen Zara pretty often since I got BTS, and he's almost always competing with me. Always has a huge empire and good teching, and a few vassals too. In one archipelago game I played, he shared a huge continent with Cyrus and Hammurabi, who both ended up his vassals with only 2 cities at the end of the game. Had Zara been more opportunistic and taken the weak AIs on small continents he would have won domination.

As for Louis, he also always has a big empire, good teching, but also builds a lot of wonders. He's also very opportunistic for war. Plus he's very good for a cultural victory, so he's a dangerous AI. Last time I saw him, it was in a game in which I had selected all the most agressive AIs (Shaka, Monty, Ragnar, Alex, Caesar, Tokugawa, Gengis, Justinian, Izzy, Louis, Joao, Boudica). He was my main competition and almost won a cultural victory. He was also the only AI, along Joao, not to end up a vassal. I won as Hannibal. That was a memorable game.

I haven't seen the other creative Ais enough but I did notice they play well. Catherine, I haven't seen a lot, but she was always a contender. In the few games I've seen Willem, he was tech leader. Suryavarman was always a menace, and it's true that Kubilai, while a warmonger, can also tech well. As for Hatty, I only remember one game I had with her, and she was 2nd behind me.

As for other good AIs, I'd have to say Huayna Capac, Hannibal, Augustus, Joao and Cyrus. Maybe also Pacal, Mehmet and Mansa, but they're not consistant enough. Asoka (!) and Bismarck have also managed to win space races ahead of me.

Finally, Wang Kon is certainly a contender... in terms of the worst AI. He almost always ends up a vassal (often mine though), I only remember one game where he was teching better than me for a good part of the game, but he had a small empire. Other bad AIs would have to be Tokugawa, Sitting Bull, and any of the Americans and British.

offworld
Feb 21, 2008, 10:14 PM
Zara Yacob can be considered a sort of super-creative leader. His Steles (replaces monument) give him an extra 25%:culture:

BalbanesBeoulve
Feb 22, 2008, 01:29 AM
In my experiences Catherine and Gilgamesh are always top contenders.None of the other creative civs have impressed me. Gilga always has an absolutely massive empire. he rexes aggressively, and then uses culture to expand even more. It's really annoying and I hate sharing borders with him.

DMOC
Feb 22, 2008, 06:37 PM
I agree with pretty much everything you said in the first post, based on my experience.
Louis XIV and Zara Yacob are among the best AIs, maybe the best 2. I've seen Zara pretty often since I got BTS, and he's almost always competing with me. Always has a huge empire and good teching, and a few vassals too. In one archipelago game I played, he shared a huge continent with Cyrus and Hammurabi, who both ended up his vassals with only 2 cities at the end of the game. Had Zara been more opportunistic and taken the weak AIs on small continents he would have won domination.

As for Louis, he also always has a big empire, good teching, but also builds a lot of wonders. He's also very opportunistic for war. Plus he's very good for a cultural victory, so he's a dangerous AI. Last time I saw him, it was in a game in which I had selected all the most agressive AIs (Shaka, Monty, Ragnar, Alex, Caesar, Tokugawa, Gengis, Justinian, Izzy, Louis, Joao, Boudica). He was my main competition and almost won a cultural victory. He was also the only AI, along Joao, not to end up a vassal. I won as Hannibal. That was a memorable game.

I haven't seen the other creative Ais enough but I did notice they play well. Catherine, I haven't seen a lot, but she was always a contender. In the few games I've seen Willem, he was tech leader. Suryavarman was always a menace, and it's true that Kubilai, while a warmonger, can also tech well. As for Hatty, I only remember one game I had with her, and she was 2nd behind me.

As for other good AIs, I'd have to say Huayna Capac, Hannibal, Augustus, Joao and Cyrus. Maybe also Pacal, Mehmet and Mansa, but they're not consistant enough. Asoka (!) and Bismarck have also managed to win space races ahead of me.

Finally, Wang Kon is certainly a contender... in terms of the worst AI. He almost always ends up a vassal (often mine though), I only remember one game where he was teching better than me for a good part of the game, but he had a small empire. Other bad AIs would have to be Tokugawa, Sitting Bull, and any of the Americans and British.

Haha yeah Louis's cultural wins are quite annoying if he can get away with them.

As for the other bad AI's you said, definitely Tokugawa, Siting Bull, and the Americans (get this--in both of the two test games I have posted, the Americans were one of the worst. In game 1, Kublai and Frederick completely whittled down poor Washingtion and in game 2, Suryavarman II actually wiped out Lincoln completely :crazyeye: ). Dunno why the American leaders don't seem to do well. Tokugawa is quite obvious, though, at why he is horrible.



Zara Yacob can be considered a sort of super-creative leader. His Steles (replaces monument) give him an extra 25%:culture:

Possibly, although his inherent +2 culture is probably the best reason why he expands quickly. Steles might play a role in future, cultural battles between cities.


In my experiences Catherine and Gilgamesh are always top contenders.None of the other creative civs have impressed me. Gilga always has an absolutely massive empire. he rexes aggressively, and then uses culture to expand even more. It's really annoying and I hate sharing borders with him.

Catherine, I can agree with you. I can't say the game with Gilgamesh since I have never played a good length of time with him in one game. I will definitely use him in a future test game.

SimonL
Feb 22, 2008, 10:46 PM
Wang Kon vassalized a bunch of people in my last game (large map, like 10 AIs) and was the board leader when I called it quit towards the end. I was Alex and was stranded on a desertic island with J. Caesar who I had wiped out early... But I was on a rock... so.. yeah.

paydro
Feb 22, 2008, 11:50 PM
I would be very interested in an elaborate experiment on best leaders. Obviously with all the permutations of the game it would be very difficult to do, but ideally you'd do 7-civ standard size, standard speed maps of various kinds (start with Fractal or Continents and move on), and note final scores and victors.

I would actually expect to have rather limited correlation between scores and victory chances. In my experience the best expanders (Joao, the Egyptians, etc.) are the ones with the best scores, even if they're not the best civs. Usually the computer wins either culturally or space race, so I wouldn't be surprised if smaller countries won frequently.

Anyway, it would be a fun way to find out the most effective leaders, civs, and on another level, traits. Traits would be hard, because they might be brought down by weak members that have nothing to do with their trait. Tokugawa, for instance, bears poorly on agg/pro, even though his real problem isn't his traits, but his AI.

At the very least you'd know what the strongest AIs are for non-Domination games...