Give peace a chance

[FrustratedPoet]

I think that is a very good idea alexman. You're right that this would be the best to test the build prefs because the human wouldn't be messing with the AI civs too much.

[Arrian]

Oooh, I agree. Great idea, alexman. We could have a "trade/wealth/science" block, a "happiness/trade/wealth" block, and a "production/happiness/trade block. Or something like that. 2 AI's per category = 6 AIs. Leave the Germans or Zulu alone. And see how each "block" performs.

-Arrian

[Theseus]

The human player approach to automated playtesting... very cool.

alexman, I posted on player1's mod thread to get some input on putting his mod together with the build preferences concept.

[alexman]

I volunteer to create the savegames for this.

Just let me know if you want to use player 1's mod, what features of that mod you don't want included, and what map settings and civs you would like for the game.

[Theseus]

I don't know what was happening before... I cross-posted a bunch and never even saw it. Now I get it... Americans as peacenicks is great!

So, GA triggers:
* Industrious: Pyramids, Great Wall, Hanging Gardens, UN
* Expansionist: Colossus, Lighthouse, Copernicus, Magellan, SETI

I'd probably shoot for Colossus and Hanging Gardens.

Originally posted by Arrian
I'd say emperor, standard, pangea, 70%, wet, cool, 3 billion, barbs to "sedimentary."

I agree, except for pangaea... I'd suggest continents, 60% water, and 7 total civs. This will give a little breathing room (which, btw, Soren said would help the emergence of killers), and make map trading a MUST.

Originally posted by Arrian
Opponents would then be France, India, China, Iroquois, Greece, Germany, Zululand.

Lemme think about this in the context of your "block" idea.

[Theseus]

OK, the way I look at it, AI civ characteristics break up into four main pieces: cultural (incl. of happiness, scientific, and culture preferences), monetary (wealth and trade), military, and growth. Of those, I think we are feeling our way towards the idea that killer AI civs will really come from a core focus on either cultural or monetary.

So, starting with Arrian's most recent suggestion for AI opponents, but knocking it down to only six, I would suggest getting rid of Zulu and Greece, and adding Russia.

Thus:
France - culture / money
India - culture / money
Iroquois - culture / growth
Germany - culture / military
China - money / military
Russia - money / growth

This sorta gives us three blocks:
1) both culture and money
2) culture with something else
3) money with something else

It's pretty hard to characterize this way, as there is spillover (for instance, Russia also has a science preference, which is cultural).

Anyway, I think that's a good list.

[alexman]

Question: Are we modding the rules for this game (units, etc), or are we just changing AI build preferences?

My feeling is to change only the build preferences so we don't introduce too many extra variables in the experiment.

[Theseus]

Good point, although I don;t think player1's changes are enough to make much of a difference.

I'm happy to go either way.

[Arrian]

I concur - the less changes, the better.

Theseus, using the (revised) build preferences Alexman posted in the AI Power thread, all civs would have "production" flagged. Ok, that leaves three other preferences each. I disagree with the addition of Russia and dropping Egypt. Drop Greece, keep Egypt (how many expansionist civs do you want in this thing?)

India: Happy, Wealth, Trade
France: Happy, Trade, Culture
the rich, cultured civs

Egypt: Growth, Wealth, Happy
Iroquois: Growth, Happy, Explore
The "Growth" civs

China: Trade, Culture, Defensive units
Germany: Offensive units, Science, Trade
The militaristic trader civs - one "builder" one "warmonger"

[Arrian]

It would be funny if Germany and China ended up alone together... a militarist set to build lots of offense vs. another set to build defense. Heh.

It would be nice to be on the same landmass as the Iroquois, so we will at least get the diplo bonus (cultural group). Correct me if I'm wrong, but you still get that little extra edge in AI attitude even if "culturally linked starting locations" is off.

-Arrian

[Theseus]

OK, that's fine. But I didn't drop Egypt.... it was never on the list!

[Arrian]

To.... many.... threads.... crossposts.... new ideas.... ahhhh!

My bad. I should have said: drop Greece, drop Zululand, add Egypt (from my original list).

-Arrian

[Theseus]

It's cool...

I really like what we are ending up with.

Go alexman!!!

[Arrian]

The more I think about it, the more daunting it seems, though. Particularly on Emperor, because only your 1st citizen will be "born" content. W/O units... OUCH. We will probably be running 30% luxuries or even more for a good chunk of the ancient age.

Gotta get those luxuries at all costs. Can't capture them later - gotta get 'em now! I haven't played like that in a looooooong time.

Hmm, I see granaries right off the bat in this one (something I've never done). Scouts find luxuries, workers build roads in that direction, settlers race the AI's. On Emperor... with happiness issues the whole way. This is gonna be hard. Pray to whatever you believe in that we don't end up next to Bismarck.

-Arrian

[Arrian]

Here's the other huge issue:

Golden Age. The Industrious wonders:

Pyramids
Great Wall
Hanging Gardens
Hoover(right?)
UN

The Pyramids on Emperor? Maybe, but it would most likely kill your expansion. Great Wall? The worst wonder in the game - a military wonder in a game where we can't build units? Hanging Gardens - the best shot, I think, plus it provides happiness, but the AI likes this one too. After that, boys and girls, it's a long way to Hoover or the UN.

So unless you think you can pull this off w/o a golden age until Hoover, you must plan on building the Pyramids, GW or HG. The expansionist wonders are better spaced out (I will definitely take a shot at the Colossus if possible).

The main problem with the Pyramids/GW/HG dilemna is that building any of them almost assuredly knocks out the Great Library (especially the Pyramids and HG). The GL would be enormously helpful, but is it worth a golden age? Decisions, decisions...

-Arrian