AI behavior : religions, research, alignment

[Calavente]

ah , ok...

% influence?? of the city pop or on the next pop that will be born ?

 

[Vittra]

You could always summon Hyborem yourself, and then build the mercurian wonder and switch to them.

 

[Lord Yanaek]

Well, there is an easier way, just open the editor and add Infernal Pact to one of the evil civs. However, even if this could allow me to play Basium, it won't help against the religious block.

I double checked, there is no way you can tell how many influence every religion has in your cities, no mouse-over on religious icons

Adding trade as a prerequisite to open-borders didn't help much. Well, it helped the Octopus Overlords but not the Order nor the Veil. They both come too late. Order was once again founded by Beery, Veil by Perpentach but he was already Leaves. I managed to research it myself and gift it to Os Gabella who was the only civ without a religion this time so at least there is ONE Veil civ (it's one more than in the previous game), still 0 Order.

I have another idea, but it's harder to mod (well at least for me )
Change the spell used by level 1 disciples to a "can only be used in cultural borders" spell and make religious techs non-tradeable. This way you can't trick an AI by gifting him your religious tech (and AIs will no longer bother you with theirs), and you can only use your tier-1 missionaries in your own territory. Maybe there could be 2 different missionaries (even hidden with the same name), one available as soon as you have the religion so you can put your religion in your own cities, the other only available with a higher tech could put religion in foreign cities. This could give more time to the AV and Order and help reduce the "big early-religion block" symptom.

Or was it intended to be this way??

 

[xienwolf]

Yeah, couldn't remember which thread this discussion was in. The mouseover only shows you what effect that religion is having on your city, and if the religion isn't founded, what tech needs researched to discover it.


I like the idea of missionary only being able to convert in your own city, but that could again be a "Favor: Human" situation, since I do not think that the AI is smart enough ot gift a missionary to other AI to cause the religion to spread.

 

[Lord Yanaek]

Euuurghh, didn't thought about this
Is there a way to make some units non-giftable?

 

[Grey Fox]

I think the religion specific techs shouldn't be tradeable, cause as it is, if you trade one of those techs away, the other civ gets a disciple of that religion.

 

[Lord Yanaek]

That was already discussed. However you could still gift them disciples if you have Open Borders, unless there is a way to make some units non-giftable.

 

[Grey Fox]

That was already discussed. However you could still gift them disciples if you have Open Borders, unless there is a way to make some units non-giftable.

Yeah but that can be self-controlled. I believe the problem is mostly with how the AI spreads the religion.

 

[Lord Yanaek]

OK, i need help from someone who know how to edit civ 4 xml files.
There is no spell to spread religion, instead there is a "religionspreads" node in civ4unitinfos.xml filled for disciple units. I didn't saw anything that suggests it's possible to change it to allow it's use only inside cultural borders but maybe there is something hidden somewhere else that controls how this ability is used but i didn't find it

 

[Chip56]

Maybe limit the autospread of religions to own cities or disable it at all?

 

[Calavente]

that would be the death of people that do not want to spend beakers researching religion's techs....
I like it to be able to recieve new religions from trade routes.. and the auto-spread of religion helps bringing money .. for the player but also for the AI.

as 3 religions are acceptable by auto-spread per city, the issue of religious predominance should be of less importance than in vanilla.
but we are looking for a reverse question...
is it because religion tech are outside of any other tech path ... as a difference with vanilla or BTS ?? you go for religious tech, only to get religions, not in the path for another goal.

maybe religions tech should be re-inserted inside the tech tree ? so AI civ won't have to research a tech on a dead-end path ??

 

[MagisterCultuum]

I still think that it is more important that we weigh techs by how useful/appropriate they are to a civ at the current time. For instance, a religious tech should become far less valuable after the religion has been founded, although not worthless because it still grants a disciple. Alignments should definately play a big role in how the AI views researching techs (at least the religious ones and way of the wise, way of the wicked, malevolent designs, and righteousness), to the point that the AI would never research techs of very different alignments or even accept them as gifts (although could perhaps be bribed into taking them). With such alignment weights, the inability to trade the techs of religions that would change alignments should be removed.


I also think that the religious techs should be shortcuts ("or" requirements) to some of the latter game techs, so they don't all become dead ends.

(What I would really like is for religions to be able to modify the amount of research needed to research other techs, making thematically appropriate techs faster and inappropriate one slower, but I don't know how that could possible be implemented.)

 

[xienwolf]

There is a mechanic in game for great people to assign free points toward specific tech categories. If you can find that decision process and RP boosting mechanic, you car build it into the religion.

Just have the Holy City Wonder also grant one-third of each thematically appropriate tech's Research valve for an easy approach.

RoK = Great Engineer
FoL = Great Philosopher?
OO = Great Bard
AV = Greet Sage
Order = No ides. Great General can't rush tech.

 

[zxcvbnm]

I think FoL is the bard type, their shrine is after all built with one unlike the OO's

 

[xienwolf]

Yes, but OO is cultural theme, so the tech which a bard can research will be the tech that match them better (festivals and whatnot).

If you were matching the Great Person to the religion, then great merchant would fit RoK, since they have a money theme to their temples. But you aren't, the great people are simply the reference to find a function for filtering the tech tree. Production related Techs are filtered under the great merchant, hence that is what fits RoK.

Really, I do not think that the great philosopher will match well with any single one of them that well, but I can't think of which great person is able to advance your research down the Nature loving side of things (Tracking, sentry and whatnot).

Ideally, we can find the section of coding for these Great People, and it simply has a list of which tech they are tied to, so we can do a similar function, but with a customized list, and fit the religions to their own setup. But if we cannot make sense of and customize the function, we can at least tap into it and utilize this slip-shod match up.

 

[Nikis-Knight]

I have another idea, but it's harder to mod (well at least for me )
Change the spell used by level 1 disciples to a "can only be used in cultural borders" spell and make religious techs non-tradeable. This way you can't trick an AI by gifting him your religious tech

Really though, teching to trade takes about as long as teching to most religions (maybe one tech sooner...) so I think the issue is rather that no ais really persue the later religions as some do the early ones.

 

[Lord Yanaek]

Maybe limit the autospread of religions to own cities or disable it at all?

I don't think there's any trouble with the auto-spread. The trouble is with massive missionaries rushes (and the AI can do this, in my last try on Sureshot's map, some 20-25 turns before open borders became available to most civs due to trade being finally traded between AI civs, Arendel had a massive stack of something like 15-20 disciples in her city (it's a one-city map), so as soone as she had open borders she was able to send one to spread her religion. Another trouble is with AI human players offering their religious tech to AIs in order to trick them to their religion even if the AI don't want open-borders. An AI will never refuse a tech gift, so you could easily turn every evil civ neutral by offering them runes.

maybe religions tech should be re-inserted inside the tech tree ? so AI civ won't have to research a tech on a dead-end path ??

I don't think that's the main issue either. Their is no trouble with leaves and runes, making open-borders available later gives enough time for OO to be founded even on such a map, but Order and even more the Veil comes way later (i noticed Veil comes even later then Order) and often too late on this map. It would be less of an issue on larger maps where AIs can build settlers (so elves wouldn't stack missionaries), still i think it shows something is not perfectly right with religions (not sure what)

I still think that it is more important that we weigh techs by how useful/appropriate they are to a civ at the current time. For instance, a religious tech should become far less valuable after the religion has been founded

so I think the issue is rather that no ais really persue the later religions as some do the early ones.

Well, changing the AI might be the best solution, but it's clearly waayyyy above my possibilities I was trying to find a work around that wouldn't distract the Team from their work on Shadow

 

[Calavente]

maybe make once again missionnaries limited in number as in vanilla ??
like 4 per civ at the same time... it would limit overspreading..
(just an issue with engranging xp for futur priests and using early disciples units to fight...)

 

[Lord Yanaek]

Yes, in vanilla missionaries were just that ... they didn't even have a combat value IIRC (i wonder why i'm talking about vanilla using past times ) In FfH, they can become priests later, they are healers for your troops on campaign ... they have so much more uses limiting them to 3 or 4 could be an issue.

I would rather see something that prevents them from spreading religion to foreign cities too early in the game, or reduce the chance to spread it depending on civs religions weighting (i always found it was sorta silly you could as easily spread the order as the veil in a Sheiam city without religion, while they should listen veil preachers more than order ones)

This would probably require a switch of the spread religion action to a new spell however, so more troubles for the team and probably new bugs as well

 

[xanaqui42]

I still think that it is more important that we weigh techs by how useful/appropriate they are to a civ at the current time. For instance, a religious tech should become far less valuable after the religion has been founded, although not worthless because it still grants a disciple.

This (changing the value of a technology depending on if it can still found a religion) is done (since Vanilla Civ), save in the couple of cases where the FfH II python code presently forces a religion choice. Whether or not the change is large enough is another issue. There are, in my opinion, 2 major problems in tech choices in FfH II:
1) The base weights for technologies are sub-optimal.
2) The situational modifiers for technologies are bad, and in many cases, missing.

Alignments should definately play a big role in how the AI views researching techs (at least the religious ones and way of the wise, way of the wicked, malevolent designs, and righteousness), to the point that the AI would never research techs of very different alignments or even accept them as gifts (although could perhaps be bribed into taking them). With such alignment weights, the inability to trade the techs of religions that would change alignments should be removed.

Interesting proposal. I wonder, however, if religion isn't more important (from a technology decision perspective) than alignment.

I also think that the religious techs should be shortcuts ("or" requirements) to some of the latter game techs, so they don't all become dead ends.

(What I would really like is for religions to be able to modify the amount of research needed to research other techs, making thematically appropriate techs faster and inappropriate one slower, but I don't know how that could possible be implemented.)

It can be implemented, yes. My initial thought would be to add religion-specific flavors to each technology, and multiply the beaker cost by such numbers as a percentage (largely because this would require a small amount of C++ code). However, I think that the problem is that if the modifier is significant, there will be players who flip religions just to get the bonus. If they aren't significant, there's little point in implementing them.