RF Civ Flavors 2018-03-12

[PYITE]

PYITE submitted a new resource:

RF Civ Flavors - Alters individual build preferences for AI civs

Contains individual files for each civ that set preferences that effect build orders and some diplo actions.

This is the RF version of Civ Flavor Diety https://forums.civfanatics.com/resources/civ-flavor-deity.26561/

Read more about this resource...

 

[Rajendran_P]

Pyite, the mod works great. It does make the AI very tough. First time the AI siege and conquered my city. Furthermore, the AI attacks with a lot of units. It’s nice to feel the AI is competitive.

 

[PYITE]

Pyite, the mod works great. It does make the AI very tough. First time the AI siege and conquered my city. Furthermore, the AI attacks with a lot of units. It’s nice to feel the AI is competitive.

Thanks, very nice to hear that people enjoy using the mod.

There are so many civs that I haven’t had time to fine tune each one. Feedback is encouraged to help with this process.

I tried to make a wider range of military desire between the civs so that each game is different based on who your neighbors are. This will also present situations where an aggressive AI might have a more pacifist neighbor and allow it to steam roll it.

I’m trying to get the AI to use Magnus and build early government plaza but the AI isn’t responding very well. Hopefully in future patches they will make the base desire for these increase like they did with commercial hubs in the RF patch.

 

[Leyrann]

Will you ensure compatability with AI+ once that mod gets it's big update? I would love being able to use both mods in accordance - and possibly have to scale down to something like King (while currently playing at Immortal or Deity).

 

[PYITE]

Will you ensure compatability with AI+ once that mod gets it's big update? I would love being able to use both mods in accordance - and possibly have to scale down to something like King (while currently playing at Immortal or Deity).

This mod is based (plagiarized) off the individualization file in the AI+ mod. Siesta is working on things that are way above my head and this mod is mostly just a stopgap until the hotly anticipated update comes for AI+.

I don’t know how the new AI+ will use pseudoyield values if at all. @Siesta Guru is more than welcome to use the individual leader files for tuning if they are at all useful to him.

If Siesta would like some assistance from the community tuning individual civ preferences I would be happy to help as I’m sure many others would as well.

 

[Siesta Guru]

Pseudoyields will end up mostly ignored, but can still play a role in systems that I can't override. Not entirely sure yet which those will be, but it might for example still use them on belief selection and the tiles worked by cities.

Why'd you opt for different files for all leaders in this mod btw? In case it's to prevent crashes due to missing dlc, it may be worth taking my addleaders.sql file in the latest version. That makes sure all leaders exist before trying to add stuff to them. (If you do that, make sure to have the mod load after

And if anyone is curious about current compatibility between our two mods, I don't think there's fundamental issues, but it will be adding the two changes on top of each other. So if both Civ Flavors and AI+ make Mongolia build more units, the end result will be that it makes a lot more units. If you don't mind that then I don't see any real issues. I haven't tested though, there's a small chance that variable name reuse messes some things up.

And yes @PYITE once I've got the core of the new system ready, I think I could use some help with tweaks and civ-based changes. Getting that right will take a lot of time since it has to be done from scratch, but it should end up being much easier to tweak than it currently is.

 

[ZeroSuitSenpai]

Gonna give this a try on Emperor as Egypt

Will report back...

Edit: The AI is wrecking me with this mod on lol. Could be placebo effect I guess; but I find I need to have many more troops around to be safe. Certainly seems more challenging!

Gilgamesh backstabbed me

 

[PYITE]

Why'd you opt for different files for all leaders in this mod btw? In case it's to prevent crashes due to missing dlc, it may be worth taking my addleaders.sql file in the latest version. That makes sure all leaders exist before trying to add stuff to them. (If you do that, make sure to have the mod load after

Because I’m a newbie who didn’t know what a xml file was before I got this game and it just seemed easier to be able to have all the leaders files open with a tab and make adjustments as I ran autoplay.

I haven’t worked with SQL or load orders at all. I’m guessing I can just copy what you did (my go to move) in AI+ with your addleaders.sql to incorporate it into my flavors mod so people without DLC can use it?

I was also wondering if you could explain to me how the new standing army value and number pseudoyields work?

Thank you so much for the efforts you put towards AI+
This mod just started as an effort towards adding the DLC leaders to AI+ and snowballed into this. After one of the dev patches, AI+ started causing ranged units to try to heal while still in range of enemy ranged units. So I decided to make this a full mod figuring that you weren’t going to be interested in doing much more until they released the dll. I wouldn’t expect a professional painter to remain interested if he were only allowed to use his fingers to paint. I however am at finger painting level when it comes to modding so making this was fun for me. I definitely wasn’t trying to step on any toes (I hope you saw your name in the credits) and I look forward to enjoying the fruits of your labor on the AI+ rewrite.

 

[Siesta Guru]

Because I’m a newbie who didn’t know what a xml file was before I got this game and it just seemed easier to be able to have all the leaders files open with a tab and make adjustments as I ran autoplay.

I haven’t worked with SQL or load orders at all. I’m guessing I can just copy what you did (my go to move) in AI+ with your addleaders.sql to incorporate it into my flavors mod so people without DLC can use it?

Took me a while to figure out that was possible too. Yes you can just add the addleaders.sql file and add it to your .modinfo. Then place the loadorder bit in your modinfo in the same way I did.
If everything works fine you shouldn't see anything pop up in database.log regardless of whether you have the DLCs on.

I was also wondering if you could explain to me how the new standing army value and number pseudoyields work?

Haven't tested with them yet since they're fairly new, so I don't know for sure. But based on the description my best guess is that it'll increase its desire for combat units if the amount of units is lower than a sum based on PSEUDOYIELD_STANDING_ARMY_NUMBER and city populations, with the total effect being related to PSEUDOYIELD_STANDING_ARMY_VALUE. Having a lower NUMBER may mean it tries to target smaller armies.

This mod just started as an effort towards adding the DLC leaders to AI+ and snowballed into this. After one of the dev patches, AI+ started causing ranged units to try to heal while still in range of enemy ranged units. So I decided to make this a full mod figuring that you weren’t going to be interested in doing much more until they released the dll. I wouldn’t expect a professional painter to remain interested if he were only allowed to use his fingers to paint. I however am at finger painting level when it comes to modding so making this was fun for me. I definitely wasn’t trying to step on any toes (I hope you saw your name in the credits) and I look forward to enjoying the fruits of your labor on the AI+ rewrite.

Hah, no worries, it's always good to see new modders and I don't mind it if someone uses some of my material. It looks like you mostly ended up doing your own thing too anyway, using mine as an inspiration at best
Not entirely sure what you mean with the ranged units thing, AI+ doesn't really change individual unit behavior. Is that still happening for you?

 

[ZeroSuitSenpai]

Okay after getting wrecked a few times I managed to pull out a win with Cleo (cultural) on emperor. I have no idea what is placebo vs. what isn't, as I'm not sure what this mod changed, but I have noticed some very encouraging things.

Instead of building something completely nonsensical when I declared war, the AI would reinforce their troops. I was especially pleased to notice the AI responding properly to a combined religious + military assault by building troops and inquisitors. I've noticed plenty of times before this mod the AI beginning work on something totally stupid while being attacked (e.g. starting a wonder in a city about to die under siege).

It seems, in general, a fair amount easier to get a religion on emperor using this mod.

 

[PYITE]

Okay after getting wrecked a few times I managed to pull out a win with Cleo (cultural) on emperor. I have no idea what is placebo vs. what isn't, as I'm not sure what this mod changed, but I have noticed some very encouraging things.

Instead of building something completely nonsensical when I declared war, the AI would reinforce their troops. I was especially pleased to notice the AI responding properly to a combined religious + military assault by building troops and inquisitors. I've noticed plenty of times before this mod the AI beginning work on something totally stupid while being attacked (e.g. starting a wonder in a city about to die under siege).

It seems, in general, a fair amount easier to get a religion on emperor using this mod.

Nice to hear the AI beat you a couple of times, sorry to hear about your cultural victory

You definitely hit on a couple of the major things this mod focuses on; building larger standing armies, building more units/walls/encampments when under attack and not having all civs being so overly religiously focused.

You have to really pay attention to the military scores because civs that have low scores because of losses will quickly replenish when given a chance. I tried to vary the military emphasis so that each game would be different depending on neighbors.

It will usually be much easier to found a religion but will vary greatly based on who is in the game. Some civs will not care at all about religion, others will be obsessed with it and many are in between. I find it makes things more challenging in general because for the most part the human player is the one responsible for keeping the AI from winning. In vanilla it’s normally a religious spreading stalemate where there is little chance the AI will get more than 2 civs converted. If only 2 of 8 civs are really focusing on religion and one gets taken out by Shaka it’s gonna be up to you to keep the AI from winning a religious victory.

Admittedly the problem is that sometimes it might be to easier for the human player to win a religious victory but you are going to need to build military as well or you are bound to be assaulted.

 

[ZeroSuitSenpai]

I just had a game in which only three civs (including myself) pursued a religion at all. The last great prophets remained unrecruited. This seemed odd.

 

[PYITE]

I just had a game in which only three civs (including myself) pursued a religion at all. The last great prophets remained unrecruited. This seemed odd.

That’s what I was explaining in my last post, military and science civs will almost completely ignore religion. That is actually one of the major factors that make many of the civs more challenging to play against. If the AI is not going for a religious victory they really don’t benefit much at all from founding a religion besides preventing the human player from getting one. Getting a religion spread to their cities for free can actually be much better than having one of their own.

Also, I find it’s nice being able to found a religion on higher difficulty without having to dedicate all early production towards getting it. Opens up some strategies involving having a religion that just aren’t feasible on vanilla diety.

It’s definitely different than vanilla but having all the AI spamming holy sites in the early game is a huge reason I started changing the AI preference.

Tuning these pseudoyield values is like trying to plug leakes in the dam with your fingers. You improve one area and another suffers as a result.

 

[V. Soma]

Hi, I made autoplay test on Europe map and added the yield typo-fix mod too - played 120 turns on standard

I saw that AI improves tiles - I like that
I also saw AI building wonders - I like that

Maybe a bit more military would be good
BUT also
more money is needed for the AI - maybe please check that for some AIs

like:
I saw Georgia starting a war with many pikemen, it looked like Tamar will win over Scythia,
but then in 1-2 turns all Georgia pikes disappeared - upkeep problems and disbanding...?

 

[PYITE]

Hi, I made autoplay test on Europe map and added the yield typo-fix mod too - played 120 turns on standard

I saw that AI improves tiles - I like that
I also saw AI building wonders - I like that

Maybe a bit more military would be good
BUT also
more money is needed for the AI - maybe please check that for some AIs

like:
I saw Georgia starting a war with many pikemen, it looked like Tamar will win over Scythia,
but then in 1-2 turns all Georgia pikes disappeared - upkeep problems and disbanding...?

Disbanding is the worst

Trying to balance out military vs gold income for all maps and civs is pretty tough. If I bump up the military the gold issues get worse. I find the religious civs like Tamar have the most problems with gold upkeep because the early holy site spam is expensive and they also tend to value faith over gold more than others. The religious civs are the toughest to tune.

Tamar is definitely one of the civs I have with lower military emphasis. By far the best strategy for a competitive civ opponent is the campus and commercial hub in most cities but I certainly didn’t want to have all the AI have the same basic strategy. For Tamar I went with highly religious, high value on CS influences, peaceful (favors only protectorate war) and lower military as I hoped she would build enough military and walls to fend off attacks while really pushing for a RV.

I’ll definitely take a look at adding some more gold desire but sometimes it’s just random that civs are lower on gold usually because they have fewer hexes with gold yield (working more pastures than plantations) or are limited to luxury items like jade that don’t give gold.

Abundant resources will definitely help the AI with gold. If you get a chance run an autoplay with new world age and abundant resources, there is a very noticeable difference in how well the AI performs vs standard settings.

 

[V. Soma]

Disbanding is the worst

Trying to balance out military vs gold income for all maps and civs is pretty tough. If I bump up the military the gold issues get worse. I find the religious civs like Tamar have the most problems with gold upkeep because the early holy site spam is expensive and they also tend to value faith over gold more than others. The religious civs are the toughest to tune.

Tamar is definitely one of the civs I have with lower military emphasis. By far the best strategy for a competitive civ opponent is the campus and commercial hub in most cities but I certainly didn’t want to have all the AI have the same basic strategy. For Tamar I went with highly religious, high value on CS influences, peaceful (favors only protectorate war) and lower military as I hoped she would build enough military and walls to fend off attacks while really pushing for a RV.

I’ll definitely take a look at adding some more gold desire but sometimes it’s just random that civs are lower on gold usually because they have fewer hexes with gold yield (working more pastures than plantations) or are limited to luxury items like jade that don’t give gold.

Abundant resources will definitely help the AI with gold. If you get a chance run an autoplay with new world age and abundant resources, there is a very noticeable difference in how well the AI performs vs standard settings.

OK, this problem with Tamar may be because of the TSL Europe map

In another autoplay test of this Europe Tamar again had low money, but this time she is peaceful...
AND:
in turn 200 she has NO campus even though she would have good spots for that near mountains...
But again low pop also can hinder that...

 

[V. Soma]

Abundant resources will definitely help the AI with gold. If you get a chance run an autoplay with new world age and abundant resources, there is a very noticeable difference in how well the AI performs vs standard settings.

I play a smoother immortal game on small continent map: abundant resources and new age world.
AI is great and will win the game... Korea has spaceport in 1745... with sc/t 3x than me
Australia is also very far ahead in tech...

I am Egypt and mediocre in everything... yeah, I have culture, but AI USA is better in that...

 

[japanesesamurai]

After about 200 turns into a (epic speed) game while using this mod for the first time, I'm already seeing huge improvements to the AI's tactics. It will put ranged units in city centres and use them to attack, put ranged (damaged) units behind city centres and front lines. It will use siege units to siege cities behind their front line melee units. As it should. Really cool stuff. I'm about 10 turns in a land (close border) war and am a bit worried about losing haha. Great stuff. This is deity difficulty without other AI mods.
They also stopped 'spamming' religious units. I'm curious if they actually are able to go for victory properly within the next half part of the game.

Anyway, so far: 5 thumbs up. Recommended!