Furins Fixes for 6.3 beta

[Furin]

Version 3:

- scripted back in regular costs for agents and generals bodyguard (player only), as a result the start is back to DLV level financially
- gave all faction (some were left out before) exactly the same extra starting troops (they aren't flavor or canon but work best in autoresolve for the moment cost/effectivenesswise)
- to balance that slowed down the AIs expansion a little bit, also made AI less suicidal (early game faction destruction avoided, hopefully)
- changed general 3 turn recruitment to 2 turns for infantry and 4 turns for cavalry, ships more, AI can still handle financially and doesn't need money script
- very slightly increased merc refresh rates
- moved another fort and made all fort commanders little kids (so they will stay in there for the longest possible time, when the general dies the captain will usually leave the fort)
- gave Inverness to scotland from the start
- made changes to dark age so factions who don't border the player still know what to do (kill each other)
- early growth of settlements best achieved with tax rate normal or low if governor present, high taxes lower growth more, Ai can handle as it gets extra boni
- doubled the severety of inflation in terms of costs, trying to achieve the same +/- treasury pingpong effect the negative events caused before by that, without killing the AI

Plans for next version:

- working on slowly upping refreshment rates of units and mercs
- with the current settlement_mechanics it's almost impossible to reach citadel level for castle, before the next dark age kills you again, I am trying two approaches here, with added healthbonus for buildings vs. half effective squalor for castles
- working on a basic reemerging faction script
- putting in descriptions for the stuff i found is working
- beginning to design a different americas invasion method with possibilities for all factions and different armies through different ages (long term goal)


Installation:

- The best way would be to make a copy of the folder where you installed 6.3 beta, rename it to your liking, fix the references and the shortcut like described in this thread (http://www.twcenter.net/forums/showthread.php?t=334563) and then copy the contents of my zip into it. So you can have the original 6.3, and my changes at the same time.

- Alternatively make a backup of the original files first and then copy the contents directly in your 6.3 version dlv_ext folder (not recommended)


Gameplay tips:

- Decline the money script for the AI, it does not need it. Play with everything else if you feel lucky. I recommend additional region and troops cost but not the restrictions to recruitment and building.

- Even though the time between dark ages is relatively peaceful an AI invasion is possible, especially with low garrisoned regions

- Diplomcy allows right of passage, if the faction doesn't see you as trusted ally and your reputation is not near immacualte (which is possible to reach before the first dark age, hard to regain it afterwards) it'll cost money (around 10k should do it), client states are modded out to prevent one faction getting all the money of an other faction at once (AI is saving up a lot of dough towards the endgame)

- Raising an army takes time, you need to plan ahead many years

- Try to extend your empire to easily defendable positions before the first dark age, with my changes every neighbor will come at you in full force, resulting in a multi front war that will be very hard to win. Also put some extra armies in the forts so you have replacement troops during the dark age, remember that you cannot raise an army as fast as in the original. Expect to loose regions during the dark age depending on the faction you are playing.

- the mercenary event is good, but remember that the AI will also get these mercenaries you get, mercenaries on the campaign map are cheaper but rare

[Furin]

V3

[Agis Tournas]

Good luck with your mini-mod, Furin
Subscribed

[Furin]

Thank's man, you are really building my spirit here, I am working on this crap since 6am in the morning

[Taiji]

Looking very interesting. I can see there are some very important changes in your sights.

Although I wonder about this:

-> minimum recruitment time from 1 to 3 in edu, except mercenaries, recruit priority already done

What do you mean? Maybe you are going to set units to taking 3 turns to recruit? To my mind that idea doesn't work, it just makes the game a lot easier for the player. Am I misunderstanding?

And when you say 'recruit priority already done', are you talking about my recent work as it appears in 6.3? Or new work that you have completed?

Very well done on the whole though, I see lots of good ideas

[Furin]

I meant that the recruiting priorities that were pretty random or non existent in 6.2 are there in 6.3 so I didn't have to worry about that, and I was really happy about that. A big kudos to you if it's taken from your EDU, looks like it holds it water.

Yes I am trying a method of increased recruitment times and small recruitment queues to force the AI to save money. As the refresh rate for the mercs is very low, compared to the original file, the only time when troops are available in abundance is during crusades and jihads, which makes it importand to join them. I am not sure if it makes it too easy for the player, we will see in testing, ok?

I hope the changes I did will make the campaign more active in general, but if that fails, and I have tried to achieve that a lot times already, I came up with a new idea in the meantime. Making a faction label for every faction that uses their default behavior (orthodox, muslim whatever) but on top of that sets 3 factions for each faction that they will not want to ally with and have a great priority for invading their lands (with points desire). Maybe with a rand factor so not every campaign will play the same.

I WANT WAR DAMNIT!!!!!!

[Taiji]

Oh man I feel for you, because I WANT WAR TOO!!! But recruitment over multiple turns means the AI takes a big hit vs the player. It exacerbates the problem of the AI never retraining. The player will be relatively unaffected. It was one of the things that made RRRC way too easy for me, great though it is

And anyway it strikes me that you would also be reducing the AI's capacity for war, not increasing it

Shouldn't they get more money if what you want for them is to have more money?... Perhaps I have missed the point, sorry if so..

edit:

As for my recruit priorities, thanks for the praise but if my EDB stuff is not used then the all the testing of that recruitment stuff is made useless. Basically I wash my hands of it's effect if my system is not to be used in full.

Back to what you/we desire, WAR!!! I think the AI could do with a system like the DLV settlement costs system, but the opposite, that makes sure they have more money if they are larger. Like they get 10k for taking a settlement and get a boosted income based on how many settlements they own. Something like this could improve their abilities to kick ass.

If you are thinking that they are not building the right buildings.... then they need big cash injections, not a constant trickle. It seems this simple: The trickle will help troop productions and the big lump sum will get buildings done.

But though we want war.... I wonder if we differ on this item: Endless stacks of doom are very boring. TA:TW suffers badly from this, it's a great game if you don't have to fight the same 4 battles every turn.... but that's what happens.

It might be nice perhaps to script an increase in the normal money trickle if the AI is at war with the player too, just a thought...

Anyway, carry on, you've obviously got some great ideas already. I'm keen to see what you do

[Furin]

The old RTW system was that as soon as the ai borders you they will attack, and the more units you killed the better the ai was economically, and could afford buildings and units. Also in RTW you could only recruit one unit/turn.

In MTW2 on the other hand the diplomacy makes the AI build huge stacks but not neccesarily attacks with them as soon as they could. These stacks cost upkeep and with fast recruitment and huge queues the AI comes to a point where it is broke and cannot afford the neccesary economical buildings any more, at which point it also stops attacking you, because it thinks it's weaker than you.

By reducing the time in which the AI can amass a huge army, and also forcing it to use it's money for building, and also by making the tiers long to build and expensive, the point at which the AI becomes broke economically is delayed and sometimes delayed long enough that a war starts and the old mechanic, of you helping by killing, kicks in. As the money script in it's current implementation only makes the AI more poor in the long run I try this method and look if it works. I had some good results with 6.2 but I am not the type of player that blitzes the AI. I agree that a blitzing player, with just conquering as fast as he can, will outpower his first three neighbors, but then he will come to a point where the next faction has amassed a huge army, his reputation is down and the counterattack comes (hopefully, in theory).

Anyhow, I try implementing limitations to the settlement mechanics that force the player to keep garrisons and slowly expand or face rebellion.

If I just could find out how to slow down the rate at which religious conversion happens, I have been looking for that quite a while in the workshop but I can't figure it out. It should take a much longer time to convert a region and also convert it back but I can't find the screw to turn.

[Taiji]

... I really don't like this idea of taking out 3 neighbours before meeting a challenge, maybe I misunderstood?

As for my recruit priorities, thanks for the praise but if my EDB stuff is not used then the all the testing of that recruitment stuff is made useless. Basically I wash my hands of it's effect if my system is not to be used in full. Afterall it would be untested, and where my work in full is already shown to be quite effective.

Back to what you/we desire, WAR!!! I think the AI could do with a system like the DLV settlement costs system, but the opposite, that makes sure they have more money if they are larger. Like they get 10k for taking a settlement and get a boosted income based on how many settlements they own. Something like this could improve their abilities to kick ass.

If you are thinking that they are not building the right buildings.... then they need big cash injections, not a constant trickle. It seems this simple: The trickle will help troop productions and the big lump sum will get buildings done.

But though we want war.... I wonder if we differ on this item: Endless stacks of doom are very boring. TA:TW suffers badly from this, it's a great game if you don't have to fight the same 4 battles every turn.... but that's what happens.

It might be nice perhaps to script an increase in the normal money trickle if the AI is at war with the player too, just a thought...

Anyway, carry on, you've obviously got some great ideas already. I'm keen to see what you do

[Furin]

Thank's

It's funny that you mention big infusions. What I have done is working with the kings purse which, according to the threads in the mod workshop, is corrcetly taken into calculation by the AI, not like the money created out of thin air with add_money.

Every faction starts with a 10k kings purse, then I take away 7500 from the player with the campaign script. So the player sees 2500, the AI has 10k, thats as good as a medium developed city.

Now at the americas invasion event, that happens roughly every hundred years, I raise that purse by 10 for the AI, which invigorates factions that went broke by that time. Also it makes the Aztecs develop, just when the invaders arrived, anot not build stacks of doom on their seperated continent before (thats why the upkeep of aztec units is so high).

Right after the AI event comes the Black death hits, and the player must now fight against an AI that has funds and wants to fight.

It worked really well in my 6.2 modifications, where I made the initial unit costs high but the upkeep low. For now I haven't changed the costs or upkeep of the units for 6.3, just the recruitment time and the mercs. I hope that it will work anyway, but if it does not I guess I have no other choice than to do it all again and I can already see the complains that it does make absolutely no sense to pay the same for a peasant than for a knight, and the people are right.

By the way, do you know if there is a limit to faction_labels in descr_campaign_ai_db.xml?

[Taiji]

Yeah but I think they're all wrong in the modding forums, and Repman is right

I think that kings purse is not the way to go, for troops yes, but for buildings the amount is not enough. If you want to get buildings built with no questions, no reliance on the Artifical Idiot, then you have to give the AI enough money in a single turn for it to recruit every soldier available, and still have enough left to buy the constructions.

The problem the AI has in DLV then is being able to take a beating and bounce back, it has more troops than it can afford. So when it takes a beating it can never bounce back to before, even if it did have an income. The boosts raise armies bigger than the economy could, and the economy is destroyed by these armies.

So the AI is unable to bore me to death with unending stacks of doom, which I like. But the AI is also easy to weaken into idleness, which I don't like.

Possibly the AI just needs more frequent boosts.... I was scared to add this to BB, for the fans of course. BB is quite challenging for anyone already... I thought this might push it out of reach for too many people. They might start crying on my thread again lol

But give it a shot anyway, do what you want. Mainly I'm just trying to explain that these 'modding experts' are usually working on a mod which does not come close to DLV in terms of financial complexity, and gameplay. Personally I check their work out before I'd trust their judgement on gameplay related issues like this one.

And no I know of no limit to faction_labels... what is that about then?

Maybe a label per faction? Would be cool perhaps. Maybe religion shouldn't matter as much as it does.... interesting to think about, I wish I had some time.

[Furin]

"And no I know of no limit to faction_labels... what is that about then?

Maybe a label per faction? Would be cool perhaps"

Yeah that was my escape route, if the ai is still unwilling to attack when it is capable then I'd make a seperate label for every faction, with three archenemies for every faction (their neighbors), and once those are dealt with it would jump back to the default label (like catholic or islam).

This huge money infusion with add_money, i tried that already, it does not really work. With every 40k florins you conjure up, the first helps, but then the ai recruits units and mercs for all whats left and goes more into the debt, then the next time you would have to conjure 80k, to make up for it and so on and so forth. You might aswell give every Ai faction 1.000.000 kings purse from the start and forget about economy (and I wonder how long it would take for the Ai to go bancrupt with 1mill).

My approach does lead to the AI ammassing huge monies though, at the expence of early military power. I am pretty sure that my changes in the EDB 100% guarantee that the AI is no longer affected by negative events of markets, roads and so on, and this alone will bring a noticeable difference!

I thought of another way to fix the Ai's economy, via script, that works similar to the mechanics that are used with the mercs. You check the treasury each factionturnstart and when the balance is negative or very small you freeze the recruiting of the Ai. That would be a pretty huge script though as you'd have to take away every possible unit from every possible faction for every possible region, that's quite some lines of text there. And when the balance is positive again you stop the script that siphons units away from the recruit pool every turn. I really don't wanna write that thing though

And another idea I have in my mind for a while, is an alliance script. If faction A is allied with faction B and faction B(or A) goes to war with faction C the ally is automatically set to war. With that you could have a Bismarck like domino effect especially as the AI currently loves to make alliances all over the place.

[Taiji]

And another idea I have in my mind for a while, is an alliance script. If faction A is allied with faction B and faction B(or A) goes to war with faction C the ally is automatically set to war. With that you could have a Bismarck like domino effect especially as the AI currently loves to make alliances all over the place.

I think that is a setting somewhere... Perhaps I am mistaking something else... wish I could remember where I saw it...

edit:

descr_campaign_ai_db.xml

	<trusted_ally_enemy_auto_war bool="true"/>

Try that maybe

[Furin]

Sweet! I didn't know that. I think I'll switch that on then , let's see if it works.

Edit: Lol it's right at the beginning of the campain ai file too, and I never saw it...
Edit2: Ah I c now, it's in the global params, so switching it on there would result in excommunication for anyone who attacks allies of the pope... Hmm, if I take away the papal faction ability to ally with an other faction I could turn it on... What do you think?
Edit3: Just got attacked and am allied with the pope but he didn't declare war, neither did my other allies, somehow that param isn't working as it should, ah well.
Edit4: Found the culprit, there's a typo in the global params (also in the 6.3 release and I also saw it in 6.2) that seemed to interfere. I fixed the typo and reset the tresholds for trusted ally, and now it's working

[Taiji]

Good job getting it working, nicely done

[Furin]

It was pretty crap though, it kinda confuses the Ai and makes it less effective than before.

Anyhow, I thought of another way to keep the AI's balance healthy. With a 4TPY script you could only allow recruiting on the winter turn, so 3 turns with no recruiting and then 1 when it's allowed ( and event_counter winter_turn 1).
During the summer there's only building.

[Furin]

Okay Version 1 is out, I hope I didn't forget a file . This are the settings I use for my game now and it costs me nothing to post them here, but please don't expect such sofisticated stuff as TaiJi's Battlefield balance. It's really a most bang for the buck project, minimum effort approach.

Hope you like it and remember to make a backup of the original files in case you don't.

[Furin]

I'll post a version 2 asap which will include changes to the stratmap (forts) and the scripted armies to prevent some (rare) bugs.

[LaMuerte]

So great watching demi-gods fixing bugs and improving our game

[Taiji]

Ah come on Furin, humilty is for the lesser men, the mere mortals