The early religions

[Iceciro]

Doesn't work. I know personally I don't give OB to Ljo/Khazad because even if they don't have their religions, the moment they get them they will spam you with disciples to attempt to make you convert - they do the same thing to other AIs. They're limited in Python to only a few - but that's only at any one time and they'll just spam you slowly.

I know, personally, I don't try to give other AIs my key religions - especially RoK because the temples are very useful even if you have another religion as state. The AI is still looking at religions as a diplomatic tool as opposed to the game-strategy changers they are.

I wonder, is it possible the AI sees the modifiers RoK and FoL give and swaps to it for the health/money as opposed to say, the culture of the OOs?

In standard Civ IV having a religion you didn't found is only good for a slight culture/1 happy face but spreading your religion is fantastic due to religious spying and gold from the Shrine, not to mention the Spiral Minaret which can almost fund your empire alone. Just isn't the case with FFH - for example, on a water map as the Lanun I want to be the only people with OO because Krakens are insane in the membrane.

 

[Lone Wolf]

Has anyone experimented with making those weights more aggressive? ie greater magnitude than +25 or -25? (Excepting -100 in specific cases of course)

I did. It does have some effect, but Disciple spamming still is annoying.

But I think that the Open Border thresholds for most leaders are too low. (<OpenBordersRefuseAttitudeThreshold> tag in LeaderHeadInfos.xml) I haven't played yet with OB thresholds increased, too busy with other stuff.

Remember, thresholds mean that the leader agrees to OB one attitude level above their threshold.

 

[solanacea]

Personally, I think passive spread of religions should be virtually eliminated until that religion's wonder is built -.

This wouldn't work, spreading religion with disciples is beyond AI's ability. If there is no passive spread, it will be entirely up to the human player to spread the religions as he pleases.

I agree that open borders and trade routes should have more influence on religion spreading. Also, I like the threshold idea for first religion switch. Perhaps, this should be a custom game option. You select "Abhorent Religions" and each civ has a religion it will never switch to. Sheaim can be hard-wired to go for AV every time; for most civs obviously some randomness is a good thing but it is tolerable if one civ acts predictably every time.

 

[onedreamer]

other (simple fix and real root of the problem IMHO): remove the free disciple from discovering the religious tech if the civ wasn't the first to discover it. Also, I play with Aggressive AI option turned on and I don't have a lot of problems with early religions.

This wouldn't work, spreading religion with disciples is beyond AI's ability.

This must be the most certainly wrong statement ever made on these forums.

 

[Iceciro]

Yep, AIs can and will and are perfectly capable of spreading religions with disciples, and will do so at the earliest opportunity once their own civ is covered in their state religion, they will spread it to anyone with OB with them.

 

[MacMoney]

Yep, AIs can and will and are perfectly capable of spreading religions with disciples, and will do so at the earliest opportunity once their own civ is covered in their state religion, they will spread it to anyone with OB with them.

Yeah, they tend to do this with other AI, but I've never seen them spread it to my empire. I don't know why. I've only played about ten or eleven games so far though.

 

[MetalMilitia-]

Ok, time for another test run. This is the changes Ihve made:

-Passive spread is turned off for all religions. Its not perfect, but it gets rid of the randomness.

-All evil civs have Religionweightmodifiers against all good religions. Likewise all good civs have against all evil religions. All modifiers have been doubled. This should get rid of the conquered cities problem.

-The "OpenBordersRefuseAttitudeThreshold" has been upped to cautious for all civs. This means that the AI wont agree to open borders when asked unless they have a attitude of greater than cautious.

This way, the negative attitude towards someone of diffrent alignement will act as a block against open borders, I hope.

 

[Iceciro]

I'm interested to see what the religions look like when you're some turns in, Metal.

 

[Gelvan]

cool, could you post your files, so that I don't have to do this myself?

 

[MetalMilitia-]

Will do. I have not set any weights against FoTL for most civs, however the elves wont be able to spread it into evils lands unless they lose a city.

My first game didnt get very far because i spawned next to clan of embers and shieam...I think ill have nightmares about pyre zombies and orcses.

 

[MetalMilitia-]

This rar contains a folder named assets, extract it into your FFH2 folder and press overwrite.

2 files will be overwritten

CIV4LeaderHeadInfos.xml
CIV4ReligionInfo.xml

Try it and let us know what you think. Note that any changes will probably be undone by applying the latest patch. Works with patch v.

 

[Gelvan]

great! thanks a lot!

I was playing Grigori right now, but somehow I warped my Adventourer production, so it's time for a new game.

 

[Gelvan]

Ok, I played a game with your files and the following settings
Spread Rate = zero
Shaeim, Svartalfar for the Evil side
Hannah, Kazhad for the Neutrals.
Ljolsfalfar (me), Bannor, Varn Gossam for the Good part.
Tiny world map Erebus Continents with fewer tundra, high cohesion (pangaea), low sea level.
No Vassals, No Technology Brokering, Permanent Alliance ok.

I beelined Mysticism and Hunting and got a Great Prophet for Way of the Forest. I was a bit doomed by my starting location, no forest, only jungle, and I had no Bronze Working, so.. not the best start. The second city was better, but at that time the Shaeim had already 4-5 cities (ICS ICS lol) but no PZ by now. I declared war and won a city, then another one, before they were able to produce PZ. Then they came... To make a long story short: Gilden Silveric with Fire Resistance made the PZ not very frightning at all. (!) With two Fawns that had Fire Resistance as well, I conquered all cities of the Shaeim.

Hannah was a bit cautios, (I was playing Aerendel, therefore I was good), but I gifted her my nature mana, so that she eventually would open borders and recieve a fellowship disciple, Varn Gossam also declared war at the Shaeim and we both were good, therefore he converted as well.

Meanwhile the Khazad developed RoK and they made good use of it, building tons of Soldiers of Kilmorph with high promotions. They showed a typical AD&D anti-elves attitude. The Svartalfar made no open border as well and even with Songs of Autumn and Sailing there was no spread of FoL to their lands.

My goal was to get either Varn oder Hannah in a Permanent Alliance, but before Varn was convinced and the nescessary techs developed, he researched Honor AND The Order, switching to both religions over the time.

The Svartalfar researched Way of the Wicked - but Bannor had no religion at all. They were fighting against Svartalfar.

After declaring war on the Svarts, I retired.

Conclusion: To turn passive spread completely off, makes it difficult for the AI to adopt a religion, if he is not Varn, Khazad or Ljoslfar. OO was not present, Hannah even asked for the Way of the Forest tech (could be, that this was due to the huge forest presence on that map).
If Varn is in the game, Bannor has difficulties to get The Order, even if Varn has already two other religions.
It takes much time to build religions in the own empire. Faster with priesthood of course, but still.. you have to focus on it (more micromanagement).
Maybe a slower passive spread rate would be better than zero.

Still, it seems as if (aside of Varn) the AI won't beeline for religions. Bannor, Svartalfar and Lanun could have researched a religion but they did not.

I think I have to start with a bigger map and without Varn, to give them more time to develop and playing someone else than Ljolsalfar to look if the AI does more spreading than me.

 

[Lone Wolf]

Maybe the religions other then RoK and FoL should have a moderate-to-low passive spread rate.

And yeah, the Bannor often fail to found the Order.

 

[MetalMilitia-]

Well the bannor sucks at build their economy.

About not beelining religions, its not allways the best option for the AI, however with Lanun there is no real reason why they shouldnt pick up OO, maybe a bit later than RoK and Fotl are founded but it should happen. Its more a matter of tech pathing and economy. Most civs will probably found a religion eventually. Sadly I dont know where to fix that, but you can change the spreadrate in Assets\XML\GameInfo\CIV4ReligionInfo.xml. Open it in a text editor and change the value in the <iSpreadFactor> tag.

 

[Iceciro]

After they hit their first beeline target they go about researching whatever they need by AI weight, I believe. Thus you couldn't weight one of the religion techs high or every single AI would chase after it.

AIs are locked from researching religions that are vastly out of their alignment - for example good civs cannot research OO and only an evil civ can research AV - all the religions are blocked off based on alignment to the AIs - they literally cannot research them if they'd cause a radical change. This may be part of the problem more than people think, though - if the Sheiam receive RoK and swap to neutral they are unable to ever research Corruption of Spirit. In much the same way, if any of the good civs are converted to Neutral through another early religion, they become patently unable to research Orders from Heaven.

The problem is, how do you tell an AI that a religion is important - because it is - but yet not neglect economy, military, arcane, tech trees? And how do you tell which AI which religion it needs to adopt based on the circumstances, for those who do decently with multiples. It would be a boring game if every race always adopted a certain religion - but they shouldn't roll over and swap to the first thing in their borders, either. Part of the difficulty lies in re-training the AI to understand religion is no longer a diplomatic tool - it is lifeblood.

Looking at it now the tech beelines are in Python, though - you could perhaps set up a short chain of tech-beeline orders, though that would make the AIs slightly predictable - just like you know the Khazad will always rush to RoK.

Religion Beelines:
Khazad - RoK
Ljosofar - FoL

Non-religion
Cabalim - Feudalism
Balseraphs - Festivals
Hippus - HB Riding
Lanun - Sailing
Luchirip - Construction
Svalts - Hunting
Bannor - Bronze Working
Doviello - Bronze Working
Sheiam/Amurites - None, that I could see. Once either of them pick up Knowledge of the Ether, they will however, beeline to Sorcery.

Civs with AV will beeline the Infernal Pact.

Every race has a beeline set up once they have Philosophy, where they will try to nab either Deception and Corruption of Spirit (evil) or Emperyan and Orders From Heaven (good). This explains what happened to Varn in your testgame - Varn grabbed Philosophy somehow and the tech-beeline code kicked in. Same with the Svalts grabbing AV. This only occurs if (as I understand the code) no other player has the techs. Thus, once someone has AV, no other AIs will try to grab it, even if they do nab Philosophy.

The fact is, most civs aren't setup to beeline any religions. The two religions that people bring up in this thread causing trouble are, unsurprisingly, the only two setup to be raced to by the AI.
As another note, you wouldn't want to set up some religions for a beeline at all, or at least not without some heavy coding/beeline this, this, this, this work.
Honor/Deception are wayyyy up the tech tree, and honestly, beelining for Orders From Heaven is going to drive your economy into the dirt as you miss out on improving your land and have to move all the way to Code of Laws/Way of the Wise. AV is up there to a degree too, but at least Way of the Wicked coughs up slavery. OO isn't very bad with Fishing/Mysticism, but Honor/Deception are just out of the question.

 

[Turinturambar]

Something that I did in .34 worked very well, namely removing the founding of the religion away from the religious tech itself.

I had the following techs as founding techs:

Runes of Kilmorph: Mining
Octopus Overlords: Fishing
Fellowship of Leaves: Hunting
Empyrean: Philosophy
Council of Esus: Trade
Order: Code of Laws
Ashen Veil: Knowledge of the Ether

I also reduced the passive spread rate for the earlier religions.

Note that almost all of the benefits from adopting the Religion still come later when you research the tech that was previously required for founding.

The results were that all Religions were founded pretty quickly and usually by thematically fitting civs, due to them having a leg up in the race with their starting tech or their AI weights.
So all Religions spread to a few citys passively. This resulted in the AI weights actually working, because the difference in spread is small. Leading to a pretty thematic adoption of religions.
Of course there are problems that need to be solved, but the basic mechanism worked very well.

 

[Gelvan]

@Turinturambar
This sounds like a good idea to solve the problem. In Vanilla religions get founded, because the tech is needed anyway. I don't believe the Vanilla AI is able to beeline a religion? (?)

@test
I made a new test with the files above and zero spread rate.
This time a bigger map with these civ's: Clan, Shaeim, Svarts, Ljoslfar and Khazad (and Elohim, but they were wiped out at turn 10 or something like that).
I was playing Grigori to look what happens with my cities, if I do nothing to give them a religion.

The result was, that Clan founded Veil but quite late (~turn 260), and even later they researched Infernal pact. Ljoslfar found FoL (surprise!) and later they found The Order. They never spread The Order anywhere beoynd the holy city.
Khazad found RoK and spread it to Svarts, and Shaeim, converting them.

I killed the Shaeim (again and again and again) this time with Grigori Heroes and Resist Fire (seriously the Shaeim have not a chance against this promotion at all). They found no religion. (couldbe they found CoE but I could not find the Holy City in the Worldbuilder, so... no idea).

Interestig was, that all my Grigori Cities had either RoK or Leaves (even with zero spread rate), but due to closed borders with the Clan not any Ashen Veil. Order was never spread, Empyrian was found by me, and I could not build Temples.

The Elves spread something else: cities! They have had many more than I possibly could have built. Their Economy must have had an enormous boost by the Songs of Autumn..

 

[Verily]

Something that I did in .34 worked very well, namely removing the founding of the religion away from the religious tech itself.

I had the following techs as founding techs:

Runes of Kilmorph: Mining
Octopus Overlords: Fishing
Fellowship of Leaves: Hunting
Empyrean: Philosophy
Council of Esus: Trade
Order: Code of Laws
Ashen Veil: Knowledge of the Ether

I also reduced the passive spread rate for the earlier religions.

Note that almost all of the benefits from adopting the Religion still come later when you research the tech that was previously required for founding.

The results were that all Religions were founded pretty quickly and usually by thematically fitting civs, due to them having a leg up in the race with their starting tech or their AI weights.
So all Religions spread to a few citys passively. This resulted in the AI weights actually working, because the difference in spread is small. Leading to a pretty thematic adoption of religions.
Of course there are problems that need to be solved, but the basic mechanism worked very well.

One problem is that the Calabim will always get to Code of Laws before the Bannor since the Calabim are programmed to beeline Feudalism (while the Bannor beeline Bronze Working and always have a disaster of an economy). I'm not sure how this could be solved without changing the Bannor beeline, and even then Flauros would always beat them.

Also, Philosophy is a bit early for Empyrean. Maybe Way of the Wise for Empyrean and Way of the Wicked for Council of Esus?

 

[Turinturambar]

One problem is that the Calabim will always get to Code of Laws before the Bannor since the Calabim are programmed to beeline Feudalism (while the Bannor beeline Bronze Working and always have a disaster of an economy). I'm not sure how this could be solved without changing the Bannor beeline, and even then Flauros would always beat them.

Also, Philosophy is a bit early for Empyrean. Maybe Way of the Wise for Empyrean and Way of the Wicked for Council of Esus?

Yeah I modified the beelines as well (altering them so that civs go for them after having researched an economic tech, most often education).
I like Order Vampires, so that did not strike me as much of a problem.
What is the problem with Empyrean appearing rather early? Since you cannot leverage any advantages apart from the +1 happy it does not work differently compared to the other religions.