Europa Universalis IV - Development Diary 12th of April 2022

[Gnivom]

What about natives and the new world? When can we expect fixes?

1.34. If you mean them attacking Colonial Nations all the time.

 

[necro84]

What about natives and the new world? When can we expect fixes?

 

[Mindel]

About the sleeping AI allies issue: I was extremely annoyed about this happening to me, but knowing that it's happening because the AI is acting in its own interest actually makes me feel much better about it.

 

[augustuscaeser2]

I have a few issues with combat, which I will try to articulate here. But first, I really like the change to regiment deployment that keeps non-cannons out of the back row -- this was a much needed buff to the AI and reduced an annoying kind of army micro management. This change alone made me excited about 1.33 at first.
But, morale damage barely has an impact now. It used to be that battles ended when one side ran out of morale and was forced to run away. Morale damage is so low, that now armies that should have been stack-wiped escape almost unharmed, or last so long they get reinforced. This makes most battles far less decisive. (There is an exception here: having tech 6 over an enemy is insanely OP now. But other than tech 6, it is really hard to deal damage now).
To address this, I did a run where I stacked discipline and damage modifiers (just Orleans -> France -> rev. France, naval hegemon). Even with 140% discipline, 20% fire damage dealt, 40% artillery damage from back row, and 20% artillery combat ablity, and a full 2 more morale than the next best army, I can't get stack wipes. Why? Because I kill the entire front line and they do not run away. Since the back row takes less morale damage now, their artillery keep firing and damage my troops -- and there is nothing they can do in return. In fact, I am not sure the cannon even takes morale damage in that situation, since I am not sure if 0 man regiments take morale damage, and they don't have the daily morale hit anymore.
I do sometimes still get stack wipes -- by killing the entire enemy army in the first 12 days (usually only when they don't have artillery, or few artillery).
Faster battles are more fun, because they require you to have better reinforcement timing -- meaning that a tactically-skilled player can still win against an enemy that heavily outnumbers them (like a huge coalition), and have fun while doing it. Personally, but I think I speak for many with this, there is little in this game more satisfying than winning a war based on skill and troop quality that based on numbers looked impossible.
Similarly, I think the daily morale damage to reserves should be increased, but further reduced by 80% professionalism. This will penalize over stacking again, but less so if you have taken the time/mil points to build up professionalism -- and built up your manpower to not need to slacken.
I uploaded the save to Skanderbeg here https://skanderbeg.pm/browse.php?id=0a56b7&view=battles. Notice how low the warscore from battles is, and how even lopsided battles rarely end in full stack wipes.

EDIT:

Adding screenshots of a battle with rebels that had the same issue. Multiple fully dead infantry stayed in the battle for at least three days. Units that are reduced to 0 strength or 0 morale should be removed from the battle as soon as they do, regardless of whether or not the first 12 days have past.

 

[RMS Oceanic]

Thanks for the update!

My experience with the current combat is that morale damage slows to a crawl for reasons I haven't yet grasped, and this means battles last considerably longer and sometimes the casualties can be atrocious as a result, and this is before artillery. Clever maneuvers doesn't account for as much as before when there is more time to reinforce, and the rate of paddleball armies where you have to chase them back and forth has increased. It all feels more sluggish as a whole, and I'm unsure whether that's good or not. Maybe if the mechanics felt more transparent I'd be better able to manage. Also my understanding is the 80% bonus for Professionalism, half morale damage to reserves, is now literally useless, so if these combat tweaks are to remain that probably needs replaced.

On the bright side I'm not sure if this is a result of changes in 1.32 or 1.33, but assaulting forts feels more effective as a strategy, actually less costly in manpower than before, especially with proper micro. Handy in the early days when you're otherwise getting 10 42% ticks in a row..

 

[Gnivom]

About the sleeping AI allies issue: I was extremely annoyed about this happening to me, but knowing that it's happening because the AI is acting in its own interest actually makes me feel much better about it.

I just want to clarify that this doesn't mean every single case of "AI doesn't help player" is intentional.
I'm aware of at least two significant AI bugs that cause this as well, both of which will be fixed in 1.34.

One of these issues happened only to AIs fighting with a human war leader. But in general, AI doesn't treat the player differently on Normal and Hard. There are a few cases, but they are supposed to be very minor.

 

[JaxElite]

I can't say i am not disappointed. While there isn't anything obviously wrong with what @Pavía said i somehow expected a lenghty post with different parts about what each team did in the last 2 months where there was complete radio silence. Instead it is an ultra-short diary consisting of "we did fixes", which is great and of "pitch us ideas and feedback about the combat system although half the forum in the last month consisted of little else". Of course there are times in development where there is not much new content to show but there is always an interesting dev diary to write. Show us the work process of a content designer from research to implementation for example. Get a programmer to write a small but in depth article about an annoying bug and how it was found out and fixed. Get a senior dev in talking about how new hires are worked in. Pull the most frequented suggestions from the suggestions forum and write down some thought experiments about implementation. There is always something to talk about.

But this is just stepping on stage after the audience was waiting for 2 months and shouting "keep waiting".

P.S. please flag the QA team in the bug report subforum as developers, so that we can actually use the "show developers response" button

 

[JackNotLantern]

1.34. If you mean them attacking Colonial Nations all the time.

They also automatically annex provinces of colonial nation when they form federations. No war, they just get provinces.

 

[TolHydra]

Any chance we are given the "Repay Loan" button again? It's so much clunkier to mentally calculate our debt since the change. (especially in the case of checking interest rates!)

 

[SeekTruthFromFx]

I can't say i am not disappointed. While there isn't anything obviously wrong with what @Pavía said i somehow expected a lenghty post with different parts about what each team did in the last 2 months where there was complete radio silence. Instead it is an ultra-short diary consisting of "we did fixes", which is great and of "pitch us ideas and feedback about the combat system although half the forum in the last month consisted of little else". Of course there are times in development where there is not much new content to show but there is always an interesting dev diary to write. Show us the work process of a content designer from research to implementation for example. Get a programmer to write a small but in depth article about an annoying bug and how it was found out and fixed. Get a senior dev in talking about how new hires are worked in. Pull the most frequented suggestions from the suggestions forum and write down some thought experiments about implementation. There is always something to talk about.

But this is just stepping on stage after the audience was waiting for 2 months and shouting "keep waiting".

tl;dr: Abusive forumites is why we can't have nice things.

A few years ago (roughly 2015-18), all the PDS teams used to post a dev diary every week, except Swedish holidays. If they didn't have any new bugfixes, features or data to unveil, then they would write about the development process, just as you suggest. I enjoyed those DDs. But there were always many comments from forum users complaining that "this is isn't a proper dev diary", and the threads often got derailed into abuse. Nonsense like "my day was ruined because PDX lied about a dev diary" and so forth. So the devs understandably thought 'why should we continue giving nice things and getting nastiness in return?'

Fortunately the older dev diaries are still available (e.g. those on the DLC development process and pricing). It's also worth checking the diaries from other PDS games in the same era, as a lot of the processes are the same (though today's post shows that Tinto in 2022 has its distinctive approaches as well).

P.S. please flag the QA team in the bug report subforum as developers, so that we can actually use the "show developers response" button

Good point. This is an automated time-saving feature, so why not use it?

 

[Mindel]

Yes, more under-the-hood improvements, please. EU4 is already great. It just needs to be able to execute its greatness with less stumbling. I think the dev team is making an excellent decision in where to focus its manpower and resources right now.

 

[Humlepojken]

I feel that wars are much less fun and more sluggish now. Quality in armies feels much less important and instead quantity is what its all about. Over all the changes in 1.33 were good but i played a few games the first week but havn't touched the game since. A bit sad about it after 4k in EU4 but battles that takes 1-2 months just removes to much of the fun. If it stays this way I will probably just revert to 1.32.

Also a bit heavy on the forts still. Either need some changes to siege speed or make it more expensive to build and upkeep forts.

 

[WishPig]

1.34. If you mean them attacking Colonial Nations all the time.

There's also a strange interaction where non-native countries can take tribal land (instantly settling it).

 

[Tempscire]

I want to focus on why battles are so long and stack-wipes are much less common in 1.33.3 and how this can be changed. I think the discussion can be more fruitful, if the reasons and some possible options regarding the problem are known and clear.

There are four main points increasing battle length and preventing stack-wipes.

(i) The first point is mostly unrelated to the combat changes in 1.33. There is a very old bug which makes it impossible to stack-wipe shift-consolidated armies. Regiments with 0 soldiers never get deployed but still contribute to the morale of an army. Therefore, the morale of an army with zero strength regiments can never be reduced to zero and the it only loses a battle, if there are no units to deploy. As units can currently never retreat before day 12 (not through having 0 morale nor 0 regiment strength), this means the army can not be stack-wiped due to reaching 0 morale. Note that units not retreating before day 12 is again a very old mechanic and has nothing to do with the 1.33 changes.
The reason why this is far more of an issue for the player in 1.33 is that the AI learned to consolidate before each battle.

(ii) The second point is about artillery taking less morale damage in the back row and again regiments not retreating before day 12. As this was also spotted in this post,

I have a few issues with combat, which I will try to articulate here. But first, I really like the change to regiment deployment that keeps non-cannons out of the back row -- this was a much needed buff to the AI and reduced an annoying kind of army micro management. This change alone made me excited about 1.33 at first.
But, morale damage barely has an impact now. It used to be that battles ended when one side ran out of morale and was forced to run away. Morale damage is so low, that now armies that should have been stack-wiped escape almost unharmed, or last so long they get reinforced. This makes most battles far less decisive. (There is an exception here: having tech 6 over an enemy is insanely OP now. But other than tech 6, it is really hard to deal damage now).
To address this, I did a run where I stacked discipline and damage modifiers (just Orleans -> France -> rev. France, naval hegemon). Even with 140% discipline, 20% fire damage dealt, 40% artillery damage from back row, and 20% artillery combat ablity, and a full 2 more morale than the next best army, I can't get stack wipes. Why? Because I kill the entire front line and they do not run away. Since the back row takes less morale damage now, their artillery keep firing and damage my troops -- and there is nothing they can do in return. In fact, I am not sure the cannon even takes morale damage in that situation, since I am not sure if 0 man regiments take morale damage, and they don't have the daily morale hit anymore.
I do sometimes still get stack wipes -- by killing the entire enemy army in the first 12 days (usually only when they don't have artillery, or few artillery).
Faster battles are more fun, because they require you to have better reinforcement timing -- meaning that a tactically-skilled player can still win against an enemy that heavily outnumbers them (like a huge coalition), and have fun while doing it. Personally, but I think I speak for many with this, there is little in this game more satisfying than winning a war based on skill and troop quality that based on numbers looked impossible.
Similarly, I think the daily morale damage to reserves should be increased, but further reduced by 80% professionalism. This will penalize over stacking again, but less so if you have taken the time/mil points to build up professionalism -- and built up your manpower to not need to slacken.
I uploaded the save to Skanderbeg here https://skanderbeg.pm/browse.php?id=0a56b7&view=battles. Notice how low the warscore from battles is, and how even lopsided battles rarely end in full stack wipes.

I will not explain this again. Note that this only became a problem in 1.33.2.

(iii) The third point is about the removal of daily morale damage. As on low tech levels the daily morale of 0.03 can constitute a significant portion of total morale damage (especially with bad rolls), this increases battle length. In the early game this is quite noticeable, while in the late game the effect is negligible.

(iv) The pip change lead to situations with significantly more defensive pips than offensive pips affecting morale damage. This is mostly the case for late game artillery, as it contributes defensive to the front row. This reduces base morale damage significantly and makes battles quite slow.


Knowing what is causing these long battles, we can now explore possible solutions to these points:

(i) As this is clearly a bug, the solution here should also be clear namely to fix it. I.e. regiments without troops should not affect average army morale.

(ii) Now this is trickier.

a) One option could be to return to full morale damage to the back row. This is really easy, one just has to change one define. The problem is that the intention behind this define change is gone. If this is problematic or not, is up to debate.

b) One could allow regiments to retreat before day 12. But this would only partially alleviate this issue, as artillery would still enter the front row and stay there for a decent amount of time (as it still has 60% morale left). Stack-wipes would still be a lot harder.

c) Another idea would be to have some form of morale hit when artillery enters the front and combine this with b). This would further reduce the problem.

d) A more experimental approach would be to base army morale only on infantry and cavalry morale. Therefore, an army would retreat as soon as all infantry and cavalry regiments have been reduced to zero morale. This would allow to keep back row morale damage the way one likes and still have the benefit of instantly retreating artillery. A side effect would be that pure artillery stacks have zero morale and therefore immediately get stack-wiped, if there are enough enemy troops or retreat after 12 days if there are not enough enemy troops.

(iii) One can reintroduce daily morale damage to everyone. But this would go against the original intention of the change. To penalize overstacking.

a) One option is to reintroduce daily morale damage to all regiments while adding additional daily morale damage to reserves.

b) A different approach would be to increase the base dice roll. Maybe only for morale casualties.

(iv) I think the least invasive way here is to change certain pips. Especially to reduce the defensive pips of late game artillery.

 

[Bestiliand]

The "Repay Loan" button was useful.

Also, the Defensive AI that build forts is a good new but, not understanding the Battle mechanichs, i think the battles should not last longer. In this game, the idea in wars is to gain local superiority, catch clueless armies separated from the rest, and gain an advantage. Also, it's fun to get to catch the enemy separately. If we remove stackwipe, and if the battles last longer in the end any attack on a defenseless army becomes a battle of all the armies of each side, that's less fun and a disadvantage of players who concentrate on separate enemy armies.

 

[PurpulaPhoenixum53]

I always enjoy honesty and openness from PDX on DDs. As for feedback, the only thing that comes to my mind is Natives AI being super blobby and surviving into the late game, not because they were good, but because they start outnumber the colonists armies by significant odds, which with the combat changes means that the Colonial Nations get murdered every time. I like that you guys are going to go back and relook at the Native AI, so some suggestions.

- Native AIs should be passive towards the colonizers early on. (IRL Indigenous Chiefs clearly understood the benefits and drawbacks to Europeans, but at the end of the day the benefits were better than the drawbacks.) In game, have most natives have a "threatened" attitude towards colonial nations and colonizers, until they start to tread into the tribal land (most natives should be willing to allow one or two provinces to be colonized, but once half of their tribal land is lost then they will become aggressive. AI Hostility should also be influenced by the European's colonization policy, with more aggressive colonization policies being met by more aggressive natives, and vice versa.

- Rework Smallpox. Right now there is no massive die off of the native population and large portions of North and Central America are still Indigenous culture. Colonial Nation AI should rarely accept native cultures and should seek to aggressive culture shift any non-European cultured provinces to a European culture group. This would require a relook at Colonial Nation's Idea and Traditions, which should've been done with the Colonial Nation rework (It doesn't make sense for a royal colonial to get legitimacy from a republican ideal) Overall, smallpox should be represented in massive dev loss and serious debuffs that make native armies melt away before the Old World's armies (Those major Native mega-cities should become depopulated as heck after smallpox burns through them)

- Attrition. While its less rare for European or other armies to march half way across the world, it doesn't add up that an army 60K Frenchmen would still have 60K Frenchmen if it marched from Lyon to Multan. Have attrition scale with distance from home/occupied provinces. I would also increase the max attrition (I feel like 10% would work and be pretty severe late game for mega stacks).

Edit: Formatting and a word

 

[vaLor-]

Hi, I haven't seen this addressed much, but since 1.31 (IIRC) a change to armies caused the splitting mechanic to not work properly. Before, if you had a 16k stack and would repeatedly box-select+split army 16 times you would have 16 1k stacks. Now (and in 1.32) if I try to do the exact same sequence, I end up with a small amount of split armies, and unable to split further without clicking the button manually:

Please fix this.

Also going to plug my culture-related discussion thread: https://forum.paradoxplaza.com/foru...ian-should-be-in-the-byzantine-group.1519290/
I think there's some good stuff in there, quick fixes too.

 

[Metz]

@Pavía

Any chance that colonial nations that expand beyond their colonial regions, create a new colonial nation? It is a bit problematic to see a colonial nation (ex: Colombia) have land in the Caribbean, Mexico, and even enclaves in North America proper. I think instant expansion of a colonial nation from one region into another should be land that goes to either the overlord or becomes a new colonial nation in that region if the province minimum is achieved.

Also the Portuguese AI should strive for going to Brazil more than Spain.

 

[Belgarath]

I'm taking this as an invitation for some more radical suggestions for combat than what many have been saying so how about:

• Battles should take a day. I could accept maybe 3-4 days just to allow some reinforcing and spreading out the calculations, but this would be a compromise.
• Battles should have no RNG. I don't think the RNG improves the game experience in anyway even if it makes it more historically accurate. At least having less RNG would be an improvement.
• 1 technology difference shouldn't make a huge impact, but 10 should. 1 unit current tech troops should be able to stack wipe 20 vastly outdated ones. But 20 v 20 of only slightly different techs should be pretty close. Perhaps a debuff for being behind the military curve that rapidly stacks.

And for a truly radical suggestion:

• There shouldn't be separate unit types. Army composition should be adjusted by ideas, investment, technologies, professionalism, etc. not an individual recruitment basis. If you unlock artillery you can instantly deploy it in your armies, but at an increased maintenance cost. The amount of artillery could then be increased for further cost, the caps being changed by technologies and skills. Similar could be for horses to what percentage of the army they can make up, hordes and strong nobility could lead to higher percentage of horses in battalions. This could even be used to enable other unit types to be in armies (conscripts vs professional soldiers, elephant or camel cavalry, grenadiers)

RNG should certainly play a role, it did in history and should do in EU IV. Eu IV obviously is not a simultaion (or even close to that), but in my opinion a element of RNG is beneficial to the gameplay. I do not necessairly fancy the current weight of RNG, but that is another discussion.