News:

And we're back!

Main Menu

Victoria 3

Started by Syt, May 21, 2021, 01:46:04 PM

Previous topic - Next topic

Tamas

One thing I'd like to see but I am sure its going to be DLC material at best, is linking foreign affairs with internal politics. IGs demanding wars as political movements would be pretty neat.

Syt

Still bothered by some of the AI passivity. Since I was working from home today I thought I'd run a few observer games with the AI economy mod and the Lotus AI aggression mod: https://steamcommunity.com/sharedfiles/filedetails/?id=2886293750

Quote****** Make sure you load LotusAI AFTER ARoAI in mod load order in Paradox launcher! ******

### General AI Behavior Changes ###

# The AI should in general declare wars much more often than it did before and be much better at creating the games main formable tags.

# The AI should be much more reasonable/realistic in what it demands in war goals. At least more player-like. Particularly, there should be less border-gore. The AI will actually use the Transfer Vassal cb, which was simply not implemented in vanilla ai.

# The AI should attempt to pass popular laws more frequently.

# The AI should be much more aggressive with it's generals, recruiting more generals and splitting armies into smaller stacks so it can both defend & attack.

# The AI will naval invade much more frequently and aggressively than before, and should not get stuck in wars because it has to do a naval invasion and can't figure out how.

# The AI will seek to annex vassals that it can and will seek to subjugate weak countries that it can.

# The AI should be less inclined to sit in a stalemated war forever without white peacing.

# The AI should be quicker to accept peace when it has lost a war.

### Game Balance Changes ###

!>Major Balance Change<!
# Because White Peace is fundamentally broken and unfixable in mods, wars will now end when one side reaches 0 war support. Some related values have been tweaked to account for this. It is now predictable when a war will end: whoever reaches 0 War Support first will capitulate.

# Some values related to the automated battle system have been adjusted.

# Dominions no longer break when their overlord does not defend them. (This included civil wars) This is the primary cause of independent Dutch East Indies or East India Company, which imo is flavorfully awful and should be extremely rare in this mod.

# Added a hidden decision that only the AI can take that will prevent a new civil war from firing while already at war. These nested civil wars are the primary cause of indepedent revolter tags throughout the course of the game, and are the primary cause of late-game bordergore with tiny weird indepdendent countries popping up everywhere. The game simply does not handle these nested wars well; I think as a player however they are more managable, and it is super cheesy to be able to use this button as a player, so it is restricted to just the AI.

### Game Start Changes ###

# Removed all starting treaty ports except Macau. (They are just no longer treaty ports, the tiles are still owned by their respective overlords)

# Parma, Lucca, and Modena have been removed from Austria's Custom Union because Austria will very often retain a deathgrip on them even if Italy forms. This helps Italy integrate all of the minors more consistently.

# The borders between the Mexico, the US and Canada have been adjusted to allow for the AI to consistently recreate clean borders. (and also allows for player USA to get everything historic Mexico in 1 or 2 wars without getting truce locked) There is slightly less uncolonized land now, but typically the US has to spend more of it's time filling that in now with less of Mexico/Canada helping.

# Mexico & the USA now start at war along with Texas to help with the border situation & issue of Mexico backing down & trucelocking.

# Hokkaido now starts fully owned by Japan.

# Mindanao in the Phillipines now starts fully owned by Spanish Phillipines, because the AI very rarely chooses to ever colonize the last bit of the Phillipines and that small tag adds nothing meaningful to the game.

# Western Australia now starts fully owned by Western Australia, because the AI very rarely (never?) chooses to colonize this bit of Australia and that small tag adds nothing meaningful to the game.

# Added Russian claims to the 2 central Asian states that are fully enclosed in the Caucuses.

# The level 40 Barracks in Lower Egypt has been reduced to level 10 to help Ottomans not get curbstomped quite so badly and consistently.

# Mexico has been given two level 5 naval bases because as AI it sometimes liked to declare war on Caribbean countries without realizing it doesn't have the boats to do so. NOTE: The root issue here has since been addressed and the AI shouldn't declare overseas wars if it doesn't' have enough boats, but I left the naval bases for Mexico for now.

# Liberia is no longer a protectorate of the USA; This is the only way I could figure out how to stop USA from colonizing in West Africa & getting involved in African colonial wars. (they are forced to have an interest in that region because of the protectorate)

### Journal/Event/Decision Changes ###

# The Heavenly Kingdom war now starts instantly after choosing a decision in the corresponding event, and Heavenly Kingdom gets an Annex wargoal on Qing.

# The Alaska Purchase & Canadian Border Treaty both now have no relations requirements other than not being at war. This is mainly to help the AI, thoug it technically makes things easier for the player it is trivial.

# The American Civil war (when it starts by event) now give both sides an annex cb. This is because it will never spawn with clean borders.

# Great Britain now has a copy of Netherlands decision to give territory to East India Company that the AI will automatically take.

# Added a decision to Netherlands that will annex all of Dutch East Indies subjects into DEI, and turn DEI into a puppet. It requires Nationalism, and the AI will take it.

# Added a decision to Great Britain that will annex all of British Raj subjects into British Raj. It is available once the Raj tag is activated fpr East India Company. (usually like 20-30 years into the game the AI gets it)

# These new decisions I feel are necessary due to the backwards mechanics behind annexing vassals. The AI does not understand that it needs to reduce relations with its overlord in order to be able annex its own subjects, and barely understands needing to reduce reduce relations with the subject it wants to annex. The mechanic itself is counter-intuitive, because good relations should be a good thing, and in games like EU4 you annex vassals with good relations, not poor ones.

# Added a decision for Great Britain and Netherlands to automatically re-vassalize an indepdendent Raj or Dutch East Indies. IMO there is no reason these tags should ever be independent if there is no mechanics or flavor for them becoming countries run by the native population, they just sit there being a nonsensical independent european colony. What's worse is the reason they become independent is usually because of nested wars causing automatic peace deals.

# Reduced Colonial Investment level requirement of British Raj decision from 5 to 3.

# Britian now has an additional decision to directly annex vassals in Malaya, unlockable by the same requirements as it's Raj decisions.

# France now has a decision to annex Algeria if it is France's puppet, again for the same reasons & because France starts with the Algiers state they are clearly intended to take Algeria & did historically (maybe not in 1836, oh well). The issue was Britain/Spain like to Transfer Vassal CB Algeria away from France and then it become a really ugly border due to Algiers state still belonging to France.

# Russia now has a decision to annex the 3 central asian minors if they are vassals of Russia. This is for the same reason as the others; the AI doesn't actually understand how to annex vassals, and just pushes the normal "annex vassal" button on the very rare occasion that the conditions for it have been met by chance. NOTE: I added this before fixing the Conquer State issues, and now Russia typically will take them directly instead of puppeting them, but left this there in case.

General observations: yes, AI is a lot more active and will add more wargoals to their demands. France and England will constantly fight over Africa and to a lesser extent South and South East Asia. Russia gets fairly involved in Asia and the Middle East. Additionally, other countries may join the fray. Sometimes Spain might grab holdings in Asia, sometimes the Netherlands. Sometimes Germany grabs some of their neighbors, sometimes they get invested in Africa.

What I dislike is that some "historicity" is achieved by basically giving the AI cheats. US can't manage to take their historic gains from Mexico? We'll trigger a war that gives them the right goals! France never annexes Algeria, and Russia never annexes their Inner Asian puppets? We'll give the AI a decision to insta-annex them! Raj and Dutch East Indies break free a lot? Give AI a decision to revert that! In all three games Canada, Australia, Scandinavia and Australia formed. In none of them was there much happening in the Balkans besides the occasional revolution in one of the small states. France and GB fought each other in Africa basically as often as their truce timer allowed. Usually one would attack a smaller country while the other one joined the target in combat. It's on one hand fun to watch, but on the other might be a bit too much.

Here's three screenshots of three different games played till 1865 (first one) and 1866 (next two):







You can see some variance. Sometimes the Brits go after Persia, sometimes the Russians. In one game NL conquered Belgium and chunks of China. In another, Spain conquered SE China. In one game, Brazil lot its initial wars, and Bolivia-Peru never united while Russia got into Africa early. But you also see Scandinavia, Canada, Australia, Raj in ever single game.

I let the third one run till 1920:



Americas look fairly historical (thanks to railroading/cheats for the AI). USA is a presidential oligarchy, still has slavery. Germany went communist, Russia is a presidential dictatorship. Russia took Japan and Korea, Persia, Sikh Empire ...  :ph34r: Italy seems to have a hard time forming up, even though the countries were in good relationship, allied etc. Not sure I like the colonial powers outright annexing so many small states - more of a personal preference, I guess. With the AI mod, the economies are producing electricity, radios, telephones, planes ... still, Russia was still using line infantry, whereas France, Germany, A-H and GB were using squads, planes (or cars) and siege arty. US was a step behind the Europeans in that regard.

Bonus question: GB is #1 Great Power, but who's #2? It's Austria-Hungary (before Germany and France, with US #5 and Russia #6 - Ottomans became an unrecognized power). In fact, A-H  has the highest GDP and SOL (28 something), ahead of the UK. A-H fought in many wars, but usually just joining plays in progress and not starting ones themselves. So their war goals, if the existed, were generally quite modest, and often reparations. Their only possession outside of A-H is the Sinai.

Btw, China became a Great Power and has a few minor colonies in Africa ...

Overall - does the mod help the AI do more stuff and be more aggressive? Yes. (All games above were on default AI aggression.) But IMHO it comes with unacceptable crutches to force certain outcomes (e.g. making sure Raj becomes single entity and doesn't leave GB, making sure US gets help achieving historical outcomes, ...).

One thing I noticed was that there were noticeably fewer revolts in major countries and a lot less turmoil, for the most part. Unsure which mod does it. Yes, in vanilla, the constant turmoil across the world from mid game or so is a bit much. But e.g. France and GB staying steady as a rock in all games also seems too much. Seems overcorrecting. I didn't see it with just the economic AI mod, so there might be something with the Aggression mod that does it.
I am, somehow, less interested in the weight and convolutions of Einstein's brain than in the near certainty that people of equal talent have lived and died in cotton fields and sweatshops.
—Stephen Jay Gould

Proud owner of 42 Zoupa Points.

Tamas

The turmoil thing might be a 1.0.6 I am seeing less of it even without the economic AI mod.

Syt

One thing that's annoying (and I'm not sure if it's a 1.0.6 thing or existed before) is that with war parties dropping out (getting annexed?) you often end up with a war where there's no unpressed war goals (all goals are checkmarked), but there's still belligerents left who will keep fighting till one side completely surrenders - even though neither side has anything to gain anymore.
I am, somehow, less interested in the weight and convolutions of Einstein's brain than in the near certainty that people of equal talent have lived and died in cotton fields and sweatshops.
—Stephen Jay Gould

Proud owner of 42 Zoupa Points.

Tamas

Quote from: Syt on November 16, 2022, 03:11:53 PMOne thing that's annoying (and I'm not sure if it's a 1.0.6 thing or existed before) is that with war parties dropping out (getting annexed?) you often end up with a war where there's no unpressed war goals (all goals are checkmarked), but there's still belligerents left who will keep fighting till one side completely surrenders - even though neither side has anything to gain anymore.

That was in before, modding peace desire seems to help although it doesn't entirely eliminates it.

Tamas

#695
Ok so how is the mod order supposed to be?

If I want it so that I use my own mod and the AI mod, but I want to make sure that if any parameter is changed by both, than the AI mod's takes precedent, I should have my mod on the top and the AI mod after it, right?

EDIT: confirmed as correct.

Syt

#696
You want the mod with "final say" at the bottom, if you have multiple mods that touch the same files, yes. Can still be a crapshoot sometimes. :P And some mods just won't work together because no matter what they will overwrite crucial settings for the other - though for very popular mods there might be compatibility patches.

What's annoying is mods not being updated after PDX patches. CK3 had a number of interesting mods that aren't compatible any more, though in some lucky cases someone else has picked them up and kept them updated. And I've seen some modders who basically said, "Yeah, I will skip this patch, would be too much work because I'm doing a major rework and will make it compatible whenever I'm finished." Which was posted 2 years ago. :P
I am, somehow, less interested in the weight and convolutions of Einstein's brain than in the near certainty that people of equal talent have lived and died in cotton fields and sweatshops.
—Stephen Jay Gould

Proud owner of 42 Zoupa Points.

Syt

New dev diary. Mostly about adding more oil to the map. And increasing rubber output by adding an irrigation production method for late game.

Riveting. :mellow:

(I'm glad they address bottlenecks, but this is not exactly marquee item stuff, IMO :D )
I am, somehow, less interested in the weight and convolutions of Einstein's brain than in the near certainty that people of equal talent have lived and died in cotton fields and sweatshops.
—Stephen Jay Gould

Proud owner of 42 Zoupa Points.

Syt

See this mod has an interesting idea, adding market volatility modifiers, to affect markets:

https://steamcommunity.com/sharedfiles/filedetails/?id=2889688573&searchtext=

QuoteThis mod is designed to add price fluctuations and more uncertainty to markets.
Often times, commodity production and prices are affected by unforeseen events related to weather, production, storage, delivery or lack of market information. In my opinion, markets in Victoria 3 are too stable and this mod aims to correct that.

Every year, and every five years, volatility modifiers are applied to buildings. These modifiers are grouped into four types:
- Yearly statewide volatility for most building groups (mines as an example)
- Yearly countrywide volatility
- Five year statewide volatility
- Five year countrywide volatility

This means that all mines in the same state will share all four volatility modifiers, while mines in another state, but in the same country will only share two. The modifiers affect throughput, or the ability to produce. Most of the time, short term modifiers are much more significant than five year ones, but it's not always the case.

The effects of these modifiers are fairly small, but they do stack. This means that once in a while, effects can be devastating or give a huge advantage. Effects are balanced so that globally they even out. For the individual country however, they will not be. This makes every game more unpredictable and unique.

The effects also applies to migration attraction, as well as certain special buildings like university, construction sector and ports.

This mod should be compatible with all other mods.



Now, this might seem a better idea of paper then in action. Personally, I would rather have more events that affect these variables, and a system that creates these things on a large scale, and not just for one state which may or moay not have a noticeable effect. But I find the idea of having a bit more fluidity/uncertainty in markets appealing (but I'm generally a fan of having more random elements - within reason - in play that force the player to react). Let's not forget that markets are also highly affected by psychological effects that can increase or decrease the effects of an otherwise harmless event (e.g. a sudden panic in a market, or a speculator bubble). If we could get to a point where major stock market crashes can be nicely simulated ... :mmm:
I am, somehow, less interested in the weight and convolutions of Einstein's brain than in the near certainty that people of equal talent have lived and died in cotton fields and sweatshops.
—Stephen Jay Gould

Proud owner of 42 Zoupa Points.

Syt

Oh, and saw a fun post on the forums today, where a guy complained that the general leading a battle on a front is chosen by die-roll, with bonuses to commanders who have more troops under them, and that either player should be able to choose, or it should give a bonus to commanders with good skills.

Very much disagree. Commanders were often as much political as military appointees; the game models this by strengthening the IGs that generals you appoint/promote/retire belong to, and I know I've appointed generals based on IG affiliation if I needed to keep another IG down, or wanted to strengthen the one the general belongs to. If this means I appoint incompetent commander, just because they have the right membership card, then that should come to bite me in the ass during war. Alternatively, I can fire those generals (or strictly appoint based on competence). But letting me appoint morons because I can shove them aside during war times is me having my cake and eating it too :P
I am, somehow, less interested in the weight and convolutions of Einstein's brain than in the near certainty that people of equal talent have lived and died in cotton fields and sweatshops.
—Stephen Jay Gould

Proud owner of 42 Zoupa Points.

Tamas

I am struggling to balance my budget as Greece with the AI mod. I hope its because of market conditions. Like, when I joined the British market my income skycrocketed for a couple of weeks or so than came right back down. So either this is the result of nifty simulated economics (which you can largely ignore with a giant like Austria but can kill you with a tiny one like Greece) or its a broken mess. I won't look TOO deeply into it because the game doesn't make that easy, but at least I can dream its the former. :)

I am also thinking that the multi-ethnic empires whose decline and fall was a major reason for why history didn't turn out as "boring" as it often does in Vicky 3 (and even more so in 2), somehow need to be made more prone to nationalistic movements. I am tempted to seriously reduce the threshold at which secessionist revolutions start ticking (its 50 radicalisation in a region at present) but it's probably going to just create a lot of even more unhistorical clusterfuck.

Tamas

So I am running an observer game after reducing the min. req. turmoil for a secession rebellion from 50% to 30%. In hindsight I should have turned the AI mod off because it hits performance hard.

But, there was a first I have seen: mid-50s Prussia faced a big Polish uprising (which to be fair I have seen previously, it even succeeded then). But what happened this time was that while the uprising's diplo plan was counting down, some Austrian regions also joined the same Polish uprising state and now they were at war with both Prussia and Austria. I thought that was cool and didnt realise it was a thing. Rebellion got crushed eventually.

Tamas

The notification settings are hilarious. There is of course the mass migration target pop-up, but when a great power launches a diplomatic play against me, it just humbly appears on the top right corner the same way a diplo play I am not involved with in would.

Syt

That's why I switched most relevant messages to "toast" in the notifications file. That way I at least get a pop up. :P
I am, somehow, less interested in the weight and convolutions of Einstein's brain than in the near certainty that people of equal talent have lived and died in cotton fields and sweatshops.
—Stephen Jay Gould

Proud owner of 42 Zoupa Points.

Tamas

#704
Playing a Spain game, it's been fun so far I have puppet-ed most of Central and a big portion of South America to significantly boost my income, I was unsuccessfully checked by the USA once and successfully by France twice.

One thing I noticed is.. AI USA has a big war going with Mexico, the wargoals could push them to near-historic borders. Except.. the US AI won't assign any of its generals to the one and only fron the war has...

I think I can see why, the US garrison units of the front are almost as many as the assigned Mexican ones, and the defines file tells the AI to only consider a front disadvantaged if it has half or less as many units as the enemy. But... surely there must be some instruction somewhere to actually achieve war goals. It's really dumb, no wonder America never expands.

EDIT: I have changed the following values which were enough to trigger the AI into sending one of its like 5 generals into the fray, but obviously its impossible to tell from this one example how much this will break war priorities in other wars:

INVADE_STATE_CAPITAL_WEIGHT = 50
   INVADE_STATE_CAPITAL_WAR_NEGOTIATOR_WEIGHT = 150
   INVADE_STATE_WARGOAL_WEIGHT = 120