Certain Civs Always Fail?

Seconis

Chieftain
Joined
Aug 14, 2009
Messages
7
I've played around 25 games of FFH2 and about 10 more using Orbis. They're both wonderful mods, but it seems that certain civilizations always do absolutely poorly. The Clan of Embers, the Mechanos, the Doviello, the Illians, the Khazad, the Luchuirp, the Sidar, and the Scions are always either killed off within the first 100 turns or they never rise about 10th place. I usually play a large or huge map with 12-16 civilizations on Monarch difficulty.

Is this just the way it goes for these civilizations? I never have a hand in the fall of their civilizations as I tend to take a non-interventionist role for the first 200 turns or so. Is there something that I'm missing? Is there a patch that improves the way the computer plays these teams?

Along the same lines is that I always do much much better than the computer does and I'm an average game player at best. I play the Calabim, Scions, Banor, and Ljosalfar and I'm always, unless I unleash an extremely powerful beast from a ruin, in either 1st or 2nd place usually with a 20% point difference. Are these simply the most powerful civilizations or am I, once again, missing something?
 
The AI does better with some then others. I've seen the Khazad, Scions, Clan, and Doviello do rather well but the Luchuirp just suck eggs when the AI runs them, not really sure why, I need to learn how to mod the AI.
 
The next patch should improve the AI somewhat. The AI is also dependent on relative start locations. If you get an aggressive and strong early game Civ (e.g. Clan, Doviello) next to a weak early game Civ, you may find that the strong Civ wipes out the weaker Civ and captures enough territory to continue to be a threat throughout the game.
 
Try one of the AI mods, either Wildmana, Skyre's mod, or the one in my sig. They all improve the general and civ-specific AI tremendously.

I'm not sure, but I think the leaders enhanced mod helps for Orbis.

As for the civs, there is usually a specific reason why they fail in the base game:
1)The Clan does not use its best feature, the Warrens and usually commits economic suicide by casting its worldspell.

2) The Doviello have a similar problem with their worldspell and they get hurt by being very similar to the Clan without the Warrens building.

3) The Illians have the problem that they usually start building Samhein really early and thus do not expand/improve their land properly.

4) The Khazad have a nasty block on building settlers unless they have enough gold in the treasury, which the AI does not understand so they expand very rarely.

5) The Luchuirp have the problem that they don't retreat their workers, don't understand how to use Barnaxus and generally plaster their land with useless workshops.
 
Some other civs that struggle in the standard AI's hands, generally:

Sheaim: Human players tend to adopt very specific tactics when playing them, spamming Pyre Zombies and death mana nodes, but the AI plays them like any other civ and ends up suffering from their lack of any useful economic bonuses. It'll also pour resources into building Planar Gates (extremely expensive buildings when they first become available) when the armageddon counter is low and it has few (often zero) buildings in the city that would spawn creatures.

Amurites: Aside from the general crappiness of the AI's magic use, three words: No Tech Trading. They won't even trade techs if they're friendly with another civ!

Infernals: There are other issues, mainly with the AI understanding of the Fallow trait, but the main problem is that the AI simply doesn't expand aggressively enough. It's not uncommon to see Hyborem sitting in Dis the entire game, doing nothing. Unsurprisingly, they get left behind rather quickly when they do this.
 
I'm not sure, but I think the leaders enhanced mod helps for Orbis.
It does, even if very slightly, mostly using your tweaks adapted to Orbis...

Anyway, it's great to have this list of things the AI can't do correctly. It will help me tweaking it. Especially since some are very easy to change... Like For the Horde worldspell, easy to disable until the civ is economically stronger.

What's the thing with Luchu and workshops? Isn't there a thing with workshop at all? I never build them but the AI seems really fond of them...

Also, you may want to point that the techs beelines are awful too. IIRC, Sheaim and Amurites will beeline Sorcery when they get KotE... It's a hard beeline and it makes little sense for the Sheaim; or else they should rather beeline Corruption of the Spirit.
 
In one of my latest games Braduk had 3 Gold(!), 2 Pigs and one other food resource in it's fat cross... it took the AI years to build Pastures, let alone mines. In the end I used the world builder to give various AIs lots of improvements because they didn't manage to farm their corn etc.
It really has serious problems with the economy. But that's a known fact..
 
Most AI mods created something to push the AI toward researching the right techs depending on nearby resources...

... but I do wonder why they don't without our intervention? Would it be because they have too high military flavors or something? Or is it again the beelines?
 
well, the short answer is that the BTS techtree has a different structure. less choices, faster to get techs mean that mistakes in the tech choice aren't that bad for the AI.
 
From what I've read the economic mod which forces the AI to research economic techs first makes them much more competitive, so I guess that's it.
One could have a look at the beelines of certain winners (like Balseraph) and compare it to other civs. With beelining for Festivals they probably hook up some important techs automatically.
Others which go for Iron Working (Clan, Bannor, Khazad) always stink because you don't need anything apart from mining... and without food you can hardly work mines anyways.
One could also have a look at the techs itself and if they enable buildings. I always see the Clan building craploads of units and wondered if they just don't have access to buildings due to the Metal Path.
In the same game out of 8 civs or so I met not one bothered with researching calendar. Which means no Festivals, which means no Market (don't know if Markets come with Festivals at FFH), which means no money for expansion and could be why Balseraphs are one of the strongest civs.
 
I have tried to justify strategies involving starting to tech different things than Agriculture, for instance going directly to Mysticism, but I always find myself wanting that food from the very beginning. I don't see how other techs can beat the power of Agriculture at the beginning of the game (unless you play as the Scions, of cource). Thus I think that the AI should always do this first as well. The economic mod should be incoroprated in further updates of the basic game.
 
I think one of the best idea that no one seems to recognize (from the sage Magister) is to force Hyborem to stay only in defiled lands, make him hyper aggressive, and boost him BIG TIME. My suggestion to add to this would be to have hell lands expand into good culture at AC 95 plus.
 
I have tried to justify strategies involving starting to tech different things than Agriculture, for instance going directly to Mysticism, but I always find myself wanting that food from the very beginning. I don't see how other techs can beat the power of Agriculture at the beginning of the game (unless you play as the Scions, of cource). Thus I think that the AI should always do this first as well. The economic mod should be incoroprated in further updates of the basic game.

Mysticism is very powerful as a first is some civs hands. I always play the Sidar, and I usually go mysticism first. Early Great mage, plus the early elder councils boost military and science well and early. (pop Knowledge of the Ether for enchanted blades) is great for the Sidar.

Anyways, Wildmana makes the Infernal use the fallow trait great. They make huge armies, and quickly have cities of size 50!
 
Mysticism is very powerful as a first is some civs hands. I always play the Sidar, and I usually go mysticism first. Early Great mage, plus the early elder councils boost military and science well and early.

I agree, but since you need to first research it and then buy the Elder Counsil and then use one pop to start adding up GPP, it would take you a very long time to get het first important worker. Thus I think that in the long run, an early worker is better than an early GM.

It would be fun to read a "scientific" essay about this :p
 
I agree, but since you need to first research it and then buy the Elder Counsil and then use one pop to start adding up GPP, it would take you a very long time to get het first important worker. Thus I think that in the long run, an early worker is better than an early GM.

It would be fun to read a "scientific" essay about this :p

I build a worker first or second usually anyways. Now that I think about it, agriculture is always my first tech... if there is anywhere that can be irrigated. Of course, I also always play end of winter, so usually, forests are better than plains or grassland in the early beginning.
 
I have tried to justify strategies involving starting to tech different things than Agriculture, for instance going directly to Mysticism, but I always find myself wanting that food from the very beginning. I don't see how other techs can beat the power of Agriculture at the beginning of the game (unless you play as the Scions, of cource). Thus I think that the AI should always do this first as well. The economic mod should be incoroprated in further updates of the basic game.

This thread is very old and the issues are already dealt with. In fact the AI always researches Agriculture first unlees it gets it for free (on higher difficulties). You can see the achievement of the various AI mods in that Civs that used to perform very poor by the AI have been made much better. Kuriotates is a good example since the rapid Expansion strategy, the only thing the old AI was good at, just doesn't work for them.
 
Top Bottom