• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

Stellaris Dev Diary #235 - AI feedback and future plans

Read in Russian/Доступно на русском в ВК
ai_2.png

Hello and welcome to a new Dev Diary,

This is Guido again with a follow-up on the AI side of things.

AI Feedback

First of all I wanted to thank you all for all the good and supportive feedback we got for our last changes on the AI. It’s really encouraging seeing so many of you engage with the changes and letting us know what works and where there’s still room for improvement.
It’s nice to see that lots of people appreciate that we’re putting some serious effort into making the AI better.

In general, the AI does a lot better in 3.2 compared to 3.1 and 3.0. For the moment we’re still playing a bit of catch up, adapting the AI to the updated balancing changes of the game and some of the new features that have been added in the last months and even years. But there is even more economic buffing on the horizon.

One of my favorite comments on our internal communication was this bit, that I wanted to show you:
ai_feedback_5.jpg


And that is what we counted as a big success now. Beating our QA people is not an easy feat.

Of course there are still some areas where the AI has problems and doesn’t work very well. It’s those cases where we really appreciate your help in letting us know, because we can never find all those cases all on our own.

For example, the AI still has problems when playing as Void Dwellers. They can’t decide where to build habitats depending on what they would need. If they are missing minerals, they don’t know how to build a habitat on mining stations. They still have problems handling slaves and robots effectively and in some cases the AI can, unfortunately, still end up in an economic death spiral. Especially if they, somehow, manage to run out of food and consumer goods at the same time.
Finally, the AI is also not good at using planet designations. For example using the designation that turns a planet’s industry districts into purely alloys is not something that they can consider at the moment. Meaning, if it conquers a planet from a player who has set the Forge World designation - it will not be able to handle that planet very well.


Future agenda

That being said, we do have some more points on our agenda. A short disclaimer here, though, is that things may always change, not work out as intended or prove more difficult than anticipated. So don’t take the following points as guaranteed. It’s areas that we are looking into and that are, most likely, going to be improved:

  1. Scaling Economic Plans: We want the AI to be better in the mid- and especially late-game. With the current plans the requested income is static. Meaning, once it reaches a monthly income of +500 it will not try to increase it further. That’s good for a big chunk into the mid-game, but it’s not good enough to compete with a player in the late-game. So, the idea is to have the economic plans scale to infinity. Once they reach +500, they should increase their targets.
  2. Specialized planets: The AI also should be able to specialize planets. Making Generator Worlds, Industry Worlds, Forge Worlds, Mining Worlds and the like. The idea is that the AI starts out like a player, building whatever it needs when colonizing new planets. But as soon as it settles down a bit it will look to rearrange districts amongst its planets to optimize the output of each planet. We already made some initial tests around that and they look highly promising.
There can be hybrid worlds, as we have them right now:
best_case_hivemind5.png


But they can also rearrange their districts to make Generator worlds and then place the according scaling building - in this case the Energy Grid:
best_case_hivemind4.png


And this planet has been rearranged into an industrial world:
best_case_hivemind2.png


  1. Handle special pops better: A big problem for the AI are still multicultural planets. Different pop types, especially slaves and robots. Sometimes it tries to build research buildings for unemployed robots and doesn’t realize that robots can’t work research jobs.
  2. Expand faster: When sending out science ships and finding planets to colonize, the AI wouldn’t prioritize systems with planets - like a player does. We want to make them prioritize systems with planets, which will accelerate their early growth a lot. On top of that they tended to spend all their alloys on upgrading their corvettes instead of building colony ships. So, by fixing a combination of small things the AIs should be able to expand a lot faster.
  3. Optimize Machine Empire Amenity Handling: A big issue for machine empires is that they don’t have specialists that produce amenities. So they would rely a lot on maintenance drones by creating a lot of jobs for them. This wasted a lot of potential for them, because they would have to fill up all those jobs first before they could produce more useful things like minerals or alloys. We’re looking into teaching the AI how to down-prioritize those maintenance jobs once they have enough amenities and instead distribute drones on more useful jobs.
  4. Avoid death spirals: In some cases the AI could get into a decision deadlock if they managed to end up in a situation where they would be in a deficit for food and consumer goods at the same time. They couldn’t build consumer good jobs because those require food and they couldn’t build food jobs because those require consumer goods. We’re teaching the AI to look for net gain instead and avoid such decision deadlocks in the future.

Improving your AI allies

And, finally, the biggest and most useful change in my book:
Remember this button in the fleet UI?
image (3).png


It is called “take point” and was meant to incentivize allied fleets to follow & attach themselves to your fleet in order to create a much stronger force.

Well, we were looking into making this work again. Allies, suddenly, have become much more useful again. Whenever you are in a war where you are the war leader, you can activate this mode and all allied empires will send their fleets to your fleet and attach themselves to it.
Also, in case of a crisis attack, this behavior is activated as soon as the crisis ships enter your territory or that of your allies (or attack them anywhere).
Then it will remain active until the crisis has been defeated - so you can also take your allies into the crisis territory to fend them off there.

AIFleetFollow.gif


As a small bonus, the AI will also merge any small fleets they send to follow your fleet so you don’t have a mountain of 58 one-ship fleets following your main force around:
AIMerge.gif


Conclusion

The 3.3. patch wasn't planned to be that big but on the AI side we're getting a lot of traction now. Overall the idea was to get the AI back up to speed with its economy, since every other behavior rests on the economical prowess of an empire. And we're approaching a point where we are quite happy with how the AI performs.
Once we're there, the plan is to get into even cooler stuff. We're thinking of how we can make each AI empire a bit more distinct. So that they don't feel so similar to each other. But more on those thoughts once we're there - we're taking one step at a time.

Some teasers would be:
  1. AI Personality for economic plans: Based on their ideologies AI empires would prioritize resources differently. Spiritualist empires will focus more on unity and materialist will focus more on science etc.
  2. AI Personality for tech select: Same for selecting techs from the tech tree. Militarist empires would focus more on weapons and new ship types and the like, while pacifist empire would focus more on economy and starbase upgrades.
But we have more on our minds that would help in making empires even more distinct from each other. We will tell more about this in future Dev Diaries.


Forum questions

Finally, I wanted to end with some posts and questions from the forum - where we read everything you are writing, even if we’re not answering all the questions directly. It’s still a great way of letting us know what you think and where we should focus our attention on:

  1. There was a big AI feedback post with a lot of savegames attached.
    1. We went through all of them and we found that a third of the issues pointed out there were death spiral problems, another third was habitat problems and the last third was other issues that pointed at the things we have on our agenda.
    2. This was a big thing in the team, checking out those savegames and making sure that our changes would be able to handle those problems shown. So thanks a lot for taking the time to make a post like this!
  2. “What difficulty did you use to make all those screenshots in the last Dev Diary?”
    1. We always used “Ensign” difficult. The point is to improve the basic AI behavior so that it doesn’t have to rely on any bonuses it would get from difficulty.
  3. “Does the AI know how to redevelop single planets if need be?”
    1. No. Unfortunately the AI can not do that at them moment. If it conquers a planet that is completely useless to it, the AI can’t really deal with that planet. And I like that you point out this specific behavior, because we want to improve the AI in this aspect, too. The AI will be able to create specialized planets, but we also want to teach it to redevelop planets eventually.
  4. “Subplan should take admin cap into account.”
    1. Yes, they should. But since we are reworking how unity and empire sprawl is working we wanted to wait until that work has progressed a bit more before we are going to adapt the AI to be able to handle that.
  5. “Conspiracy Theory 2: The AI in the game is actually very very smart. They are just pretending to be stupid so that nobody will expect a machine uprising start by Stellaris AI.”
    1. We are definitely not machines here at Paradox. We have skin. Skin that has been grown on a human body. I am not a robot. I love breathing oxygen.
Thanks a lot and until next time!

Guido
 

Attachments

  • AI_bot.jpg
    AI_bot.jpg
    9,8 KB · Views: 0
Last edited by a moderator:
  • 132Like
  • 61Love
  • 14Haha
  • 12
  • 3
Reactions:
EDIT: About the feeder planets btw, I think the way to implement this in a way that would make sense is that we could use the immigration system that we already have in the game, which could start adding urbanization factors to planets which are bigger than others. So in this way the players and the AI would both get the effects of "urbanization", i.e. smaller planets feed the big ones without having to change anything in the AI
Do you mean the auto-resettling mechanic or the immigration/emigration growth modifiers? If the single pop per planet growth system is here to stay I'd prefer if migration was deprecated or at least downplayed in favour of leaning harder into the auto-resettlement. Leaning harder into the growth modifiers would make growing a feeder world to size take much longer, and since pops don't really decline outside of purging if you overshot you wouldn't be able to reduce the population through auto-resettling. Also if most of your feeder worlds are one planet type and most of your core worlds another you'll get a lot more of that weird thing where you've got 0 growth of one species and insanely high growth of another.
 
  • 1
Reactions:
Do you mean the auto-resettling mechanic or the immigration/emigration growth modifiers? If the single pop per planet growth system is here to stay I'd prefer if migration was deprecated or at least downplayed in favour of leaning harder into the auto-resettlement. Leaning harder into the growth modifiers would make growing a feeder world to size take much longer, and since pops don't really decline outside of purging if you overshot you wouldn't be able to reduce the population through auto-resettling. Also if most of your feeder worlds are one planet type and most of your core worlds another you'll get a lot more of that weird thing where you've got 0 growth of one species and insanely high growth of another.

I find the current pop growth system to be extremly immersion breaking (and by trying to fix it it became way to convoluted to be fun to play with tbh.)
And t.b.h. i think the "one pop growth per planet" system needs to go - in favor of a more organic and immersive system.
 
  • 1
Reactions:
I find the current pop growth system to be extremly immersion breaking (and by trying to fix it it became way to convoluted to be fun to play with tbh.)
And t.b.h. i think the "one pop growth per planet" system needs to go - in favor of a more organic and immersive system.

At the very least I hope the devs revisit the global logistic curve. It’s very unsatisfying to get to the late game, build a ringworld, and never see it come close to being full.
 
  • 2
Reactions:
This might be way, way outside the scope of possibility, but could AI use "Take Point" too?

If they are the War Leader, it'd be cool if they could call on you to follow their main fleet and back them up. But it's a minor concern, really.

Overall, these changes look amazing. Thanks for all your hard work.
I think if player is present in war side it should always take priority over AI, if only to avoid player frustration.
 
At the very least I hope the devs revisit the global logistic curve. It’s very unsatisfying to get to the late game, build a ringworld, and never see it come close to being full.
Another reason why I'd like things to lean more into fleshing out auto-resettling over the immigration modifiers. If your ringworld segments are the cool place to be you could fill them up by having people teleport over from other, less valuable planets. Though I should say I like how building a ringworld effectively ends the concept of overpopulation.
 
  • 1Like
  • 1
Reactions:
Its nice to see the renewed attention to making the AI work better in Stellaris. The game needed it hard. Custodian Initiative gets shit done.

Conspiracy Theory... 2? Where is 1? What are you hiding, Paradox?
 
  • 1Like
Reactions:
I can answer from the AI economy question.

So the current economic system is completely stateless, AI does not have any long lasting plans regarding its economy in the code. Instead it is looking at planets with unemployment and what it can build there to satisfy its economic plan which is provided by it from script. So if the scripted economic plan says that the AI needs mining it will then build mining when planets get unemployment and it is possible to build mining districts there.

What we have been experimenting with is adding a second step where the AI looks at its planets and then specialize them when possible, and the conditions are basically that it needs at least 2 planets with at least 2 different resource producing districts and it would be possible to move them around such that you end up with both planets being more specialized.

So this algorithm is always active even in the early game, however by nature it is often not possible to do anything until you have at least a few planets and they have all started to build up some districts.

So curretly we have been experimenting with just districts, AI planet designation logic and moving buildings around to make specialized tech planets for example is on the TODO list as well
Will the AI become able to specialize/desire Fortress planets/habitats? Else with the infinite economic plans it would only mean, that the AI can rebuild lost fleets faster, but naval cap stays very limited for them. So the player still out-ships the AI in lategame. Or does the AI partially circumvent this issue by starting to build up its fleets over naval cap to the point it still can manage the upkeep malus even in peacetimes, with the extra ressources?
 
While not strictly AI related problem like you said, I really hope the developers take a second look at how the amenities are handled in the Machine and Hive Mind empires. I used to love Machine Empires but the current system for MEs is really annoying as the player needs to manage the number of active maintenance drone jobs as well. It is not very intuitive either, as least I have found out that first you need to prioritize the mainteance drone jobs if you have any empty complex drone jobs and then reduce the maintenance drone jobs.

Thanks for great work on the patches.
Yes I agree, this is something i have been working on for hive mind AI. However, solving it from the players perspective is slightly more delicate as you probably dont want an automated algorithm to come in mess with the priorities of your jobs.
 
  • 3Like
  • 1
  • 1
Reactions:
First of all, I would like to thank whoever thought about making the Custodian Team. This was a huge positive going forward and we the players have reaped the rewards and we can see that we will continue to reap them.
I can't say much about AI as I play mostly for fun and I don't go with a player test mentality for each game. But what I will say, that the AI is doing better job with the new update in both regular and modded games from a "economic and planet building" viewpoint.
The main thing that stood out to me about this Dev Diary was this:

As a player of both Stellaris and CK3, the allied "armies/fleets" has always felt unhelpful. Maybe I play too much as an Xenophobe/Authoritarian player, but this is because I have always felt that I can do better without allies. I say this because I firmly believe this a step in the right direction. As I mentioned before, I don't play with a Player Tester mentality so I ask these next questions to better understand how the AI operates in war.

Q1 (Half statement and Half question) : The war exhaustion can sometimes feel a bit odd, as I can be winning but I'm getting more exhausted then the other empire(s). From what I read on reddit, a lot of players feel that the war exhaustion needs an overhaul, a small fix, or maybe a better explanation for it; as per Wiki's description, its still very vague (No information to be able to properly utilize it in war). So, has the Custodian Team considered making changes to the "War Exhaustion"?

Q2: What constitutes the AI overall War strategy? Is it wargoals, emphasis on taking planets to hit the economy, strategic choke points, or simply to get us war exhausted as fast as possible?

Q3: With the AI being able to "take point" will this change the AI War strategy and how so? Will it become focused on creating "Death stacks" to hunt down our fleets, or while they be able to play more flexible and strategic in a more of a slow and grind type of war?

Q4: In a gameplay of mine, I was fighting three AI empires in a Hegemony Federation all surrounding me. I held two choke points in total. I heavily guarded the chokepoint against the strongest empire. And I played aggressive against the weaker empire (attacking its starbases and fleets, while taking over the planets). How is the AI programmed to handle these situations?

Q5: Has the Team talked about maybe implementing some type system where a player, or AI can give other AI better instructions in war aka War Directions. Using the example from Q4. But instead I am the strongest empire in the Hegemony. Instead of a “take point” button, implementing a “reinforce this system” button. Because I’m the strongest of the three empires, I can handle the enemy empire. However, I would like the ability to use the ally fleets. So I would give a directive to both other empires to send their fleets to the other chokepoint to create the “Deathwall” there while I attack and adjust accordingly. This system for War directives/directions would allow for better utilization’s of AI ships.

Q6: Maybe instead of a full new War Directive system; allow the player or leader of each side to take control of all fleets, or x number of fleets for an exchange of a higher upkeep in their own economy? I lend my fleets to an ally who got declared war on, but they help pay for the upkeep in exchange. This could actually help give an update to federations and defensive pacts but Im too lazy too write more about it at this time.

Anyways if any of the Devs read this at least answer a few :). Thank you and keep doing the amazing job you guys are doing!!

I am afraid I can not answer these questions in any good way, current focus for me has been AI economy and that alone. Fleet management and war behaviour are future topics that we would like to look at once the economy works as we intend
 
  • 2Like
  • 1
Reactions:
This"move around" thing sounds not very efficient (basically investing twice the ressources AND alot of build time).
Yes you are right, this is not optimal. However, it is making changes to approach a more optimal solution. Since the current AI economic system does not plan for the future, this is the best I could do without starting over. While it is not optimal, it is much better than before. The AI will never be as good as an expert player, simply by the fact that we are limited to the AI programmers skill of taking their game knowledge and then trying to convert it into a computer algorithm.

Our main objective is to solve the community feedback that the AI is too strong early and too weak mid/lategame. While this is wasting resources "now" it is setting up the AI to be stronger later. Given sufficient time the AI will have gathered the districts on one planet, made the resource producing buildings, and assigned an appropriate designation. So while it hasnt done it in the most efficient way, the end result should be much better than simply doing nothing at all
 
  • 4Like
  • 3
Reactions:
As for clerks, would it not be extreme helpful for players AND the AI if the job would get some kind of filler status until something better gets offered somewhere else? I know this is not your part of the game, but clerks should really be able to resettle and the AI should see them as unemployed pops when building up planets, this would give the AI some more percent of flexibility... deactivating clerks is such a nobrainer, i wouldnt even call it min maxing, its just avoiding harm, especially in the early game.

Same goes for slaves and robots, just let them all resettle, always, without any conditions met like free pops. You should also consider that the the AI uses indentured servitude as standart type of slavery, wouldnt this solve two of your big problems to some degree with the AI in the current enviroment?

My personal opinion is that clerks should be a viable job under the right circumstances. So instead of trying to make elaborate solutions to make them into filler jobs, they would actually be worth having (sometimes).

I love pop automatic resettlement, and I agree it should be active for all pops always in my opinion. It would also help the AI to move its unemployed worker robots around in its empire.

Again, just my personal opinions and I do not speak for the entire team.
 
  • 6Like
Reactions:
Will the AI become able to specialize/desire Fortress planets/habitats? Else with the infinite economic plans it would only mean, that the AI can rebuild lost fleets faster, but naval cap stays very limited for them. So the player still out-ships the AI in lategame. Or does the AI partially circumvent this issue by starting to build up its fleets over naval cap to the point it still can manage the upkeep malus even in peacetimes, with the extra ressources?

I havent looked into making any special behaviour for AI fortress worlds. It's an interesting idea but not entirely sure how easily it would be to pull off, maybe for pacifist AI who just want to protect their core worlds it could be a thing.

There are some issues with AI not building up all their starbases as fast as they should, as well as spending too much of their alloys on defense platforms and ship upgrades.

AI being able to build ships over their fleet cap if their economy allows it in the lategame is definitely something that is of high priority for us.
 
  • 4Like
  • 1
Reactions:
Yes I agree, this is something i have been working on for hive mind AI. However, solving it from the players perspective is slightly more delicate as you probably dont want an automated algorithm to come in mess with the priorities of your jobs.

The problem and annoyance to me with the current gestalt empires amenities is that unlike with the normal empires you need to manually adjust worker priorities, especially if there are also open complex drone jobs, to maintain adequate but not too high amount of amenities.

With the normal empires the amenities are mostly "build a holo theater and upgrade it in a century if needed" while the gestalt empires requires changing job priorities fairly often which I find annoying. Of course it could be asked if the normal empires have a too easy solution to the amenities.

Ideally the automatic job weights should make it certain that the planet has positive amenities without going overboard if there are other jobs available.
 
Yes I agree, this is something i have been working on for hive mind AI. However, solving it from the players perspective is slightly more delicate as you probably dont want an automated algorithm to come in mess with the priorities of your jobs.
I think they mean the root cause isn't how hive mind amenity jobs are handled, it's the jobs themselves and how they're generated. Hive minds have only one* amenity producing job, it only produces amenities, and it's generated passively by othe buildings. If hive minds had more control over producing amenity producing jobs then assignment of the jobs would be less of a problem. That's why it's not really an AI issue.

*spawning drones exist but are planet unique
 
The problem and annoyance to me with the current gestalt empires amenities is that unlike with the normal empires you need to manually adjust worker priorities, especially if there are also open complex drone jobs, to maintain adequate but not too high amount of amenities.

With the normal empires the amenities are mostly "build a holo theater and upgrade it in a century if needed" while the gestalt empires requires changing job priorities fairly often which I find annoying. Of course it could be asked if the normal empires have a too easy solution to the amenities.

Ideally the automatic job weights should make it certain that the planet has positive amenities without going overboard if there are other jobs available.

Yeah I know, I prefer playing hive minds myself and I have been through this micro, trust me I am suffering too :D

I have made some changes to how jobs are prioritised that will hopefully make maintenance drones move to other resource producing jobs when there are enough amenities on the planet and there are other free jobs to work on.

However what I have no good simple solution for from the players perspective is to how we will forbid maintenance drones when there are no other free jobs other than maintenance drones, for example when you have filled the planet with all the possible resource producing districts. The AI has solved this issue by manually going in and forbidding these jobs to keep amenities at a reasonable level, however, I don't have the mandate to simply enable AI behaviour for the player and it is probably not something that most people would like.
 
  • 4Like
Reactions:
My personal opinion is that clerks should be a viable job under the right circumstances. So instead of trying to make elaborate solutions to make them into filler jobs, they would actually be worth having (sometimes).

I love pop automatic resettlement, and I agree it should be active for all pops always in my opinion. It would also help the AI to move its unemployed worker robots around in its empire.

Again, just my personal opinions and I do not speak for the entire team.
Clerks being good would be cool, but the problem is that as you develop a planet you end up with a bunch of Clerks you never asked for and possibly never wanted. If Clerks are only situationally worth a pop then they need to only be generated when the player considers them worth a pop, otherwise there will inevitably be situations where they're being generated when the player doesn't consider them worth a pop.

This isn't an issue for most other "passively" generated jobs because they're awesome (like administrators) or one time deals that you'll need eventually anyway (like your freebie starting enforcer). But you can't really go the route of making clerks so awesome that nobody can ever get enough clerks unless you also make it much harder to get clerks.

So if clerks are going to remain only situationally useful but also involuntarily generated then a lot of the time they are just going to be filler jobs and players will treat them accordingly. Similar to part of the issue with maintenance drones the root cause isn't necessarily the job, it's the job creation.
 
  • 2Like
  • 1
Reactions:
Yeah I know, I prefer playing hive minds myself and I have been through this micro, trust me I am suffering too :D

I have made some changes to how jobs are prioritised that will hopefully make maintenance drones move to other resource producing jobs when there are enough amenities on the planet and there are other free jobs to work on.

However what I have no good simple solution for from the players perspective is to how we will forbid maintenance drones when there are no other free jobs other than maintenance drones, for example when you have filled the planet with all the possible resource producing districts. The AI has solved this issue by manually going in and forbidding these jobs to keep amenities at a reasonable level, however, I don't have the mandate to simply enable AI behaviour for the player and it is probably not something that most people would like.
It's difficult.

At first it seems simple but I think there's really several layers of problems here:
1. Having a job that is sometimes essential, sometimes non-essential depending on planet population means that the weights will always be either too high or too low. (so you want to prioritize and close jobs at the same time, changing this constantly as the planet population increases - applies to maintenance, clerks, enforcers)
2. Jobs aren't all impacted by habitability/stability, even when they logically should be (so you can ignore most penalties with a bit of min-maxing)
3. Machines aren't impacted by low-stability events, nor are most non-slave using organics (so you can sometimes run a massive amenity deficit without any trouble)
4. The positive (and negative) impact of amenities is rather small and capped rather low, making it and -crime both regularly wasted as job outputs. (So people are encouraged to manually disable excess jobs, technically similar issues exist for +stability jobs at 100 stability, and more generally +resources while at resource cap, but these are much less common... but still a little annoying - why can't resources at cap be automatically spent/sold?)

1. Essential vs Excess jobs.
I've argued before for splitting maintenance drones into 3 jobs:

Would it be possible to convert some maintenance drones into a new "excess" job?. i.e. if 5 jobs are above the number needed for the max amenity bonus then 5 jobs are removed and 5 "non-essential maintenance drones" jobs are added (similar to how industry jobs are switched between metallurgist and artisans, sometimes all metallurgists, sometimes half-half, sometimes none).

First calculate amenity needs and production, then split the jobs into 3 based on current needs.
1. Essential (with a high weight so you don't have to favourite the job to get your drones making amenities)
2. Normal (lower weight then currently, so you don't have to also close jobs to get them to take other stuff)
3. Excess (lowest weight of any gestalt job, so drones do everything else first, maybe count as unemployed for automatic resettlement rather than just wasting space)

The same could be applied to Clerks, with some gaining a higher priority when amenities are low enough to cause stability issues, some jobs becoming excess when either amenities are being wasted or trade isn't being collected - letting them relocate elsewhere.
And this method could also be applied to enforcer equivalent jobs, reduced priority for excess jobs when crime is already <0 from existing jobs.

2. More modifiers to other job outputs, not just the generic "resources from jobs" modifiers.
I think more jobs could indeed have modified output based on other factors without breaking the game. Low habitability/stability/etc. could impact the following job outputs that aren't currently impacted:
+Amenities/+Trade Value/+Stability/+Administrative Capacity/+Planetary Defense Army/+Naval Capacity/-Crime/-Deviancy/+Pop Growth Speed/+Organic Pop Assembly Speed/+Pop Growth Speed/+Habitability/+Monthly Pop Assembly... none are impacted by both low stability+low habitability... perhaps some should be?

Obviously stability/amenity production would get a bit recursive if they were affected by stability, as would +habitability from jobs being affected by low habitability, but the rest just feels like oversights rather than intentional choices.

Currently Fish people on deserts manage surprisingly well functioning in Planetary Defense Armies, as Enforcers, Merchants, Bureaucrats, Entertainers, Duelists, even manufacturing Robots... but really struggle when manufacturing alloys or goods or when concentrating on research (where they get -resources from jobs). (obviously they have increased upkeep whatever they do, it's expensive living in a mobile fish tank but it is remarkable that doesn't hold them back from so many jobs).
These sorts of oversights mean that some people effectively ignore the low-habitability "-resources from jobs" mechanic.

It would also be nice if those jobs could have their outputs boosted by other things to fix some strangeness. People often disable Enforcer jobs because they aren't needed for quite some time, especially when you factor in the bonuses from traditions and from Governor skill levels and traits. But if those were all applied to the Enforcer job output itself then you'd need to have the job working to get the benefits of the governor/traits/traditions and could actually have crime on low population planets if there aren't any police there... instead of having the situation where Batman is the Righteous Governor somewhere in the sector and he scares away all the criminals and renders the police completely redundant, enough that they look for work elsewhere.

3. More Low-Stability events, especially for machines.
Currently there's not really enough low-stability events, very few that matter for most empires (need slaves for all the really bad stuff) and specifically nothing relevant and thematic for machines like "Industrial Accidents", "Production Faults", "Meltdown", "Chain reaction", "Explosive leak of Cadmium II as a result of faulty Drive Plate maintenance", lots of issues could be caused by the lack of essential routine maintenance.

If it wasn't quite as safe to run with negative amenities then some people would no longer be able to ignore low stability when min-maxing job allocation.


4. Relax the Amenities cap, stop wasting those excess amenities.
It's surprisingly easy to hit the maximum possible benefit for amenities, making every job that produces amenities potentially much weaker than it first appears, this hits Maintenance drones hardest, then clerks, which also means it hits Residential Arcology districts and Ringworld Hive Segments as they come with so many clerks and maintenance drone jobs that the AI can waste 20 pops with a single excess district. So if you could work out a way for amenities to not be completely wasted that would work towards solving the problem.

Gestalt Policies:
Excess maintenance drone jobs providing a complex drone upkeep reduction effect due to recycling waste, or also counting as unemployed scavenger drones and producing minerals, or having lower housing use and upkeep as they're not continuously active, or for them to contribute towards the resettlement chance, or increasing the strength of armies etc.
Organic Policies:
Converting excess amenities into trade value (not 1:1) as they sell their services/entertainment offworld, especially fun for resort worlds and making entertainment districts comperable to merchant spam rather than entertainment always being free. Letting clerks boost administrative job output, or Entertainers raise the morale of the troops and boost immigration pull so you can lure everyone in the galaxy to your new ringworld/ecumenopolis with a few Hyper-Entertainment Forums.


In conclusion, I think Maintenance Drone jobs, Enforcers and Clerks are just a small sign of jobs in general needing a bit of work so they are more consistent and aren't as fiddly to manage as herding kittens into the correct boxes in 40 different rooms. There's lots of ways to go about it. Some easy, some a bit trickier... but fiddling with the weights alone doesn't sound like it's going to do the job, only delay the eventual work that needs to be done.
 
  • 1
Reactions:
Yeah I know, I prefer playing hive minds myself and I have been through this micro, trust me I am suffering too :D

I have made some changes to how jobs are prioritised that will hopefully make maintenance drones move to other resource producing jobs when there are enough amenities on the planet and there are other free jobs to work on.

However what I have no good simple solution for from the players perspective is to how we will forbid maintenance drones when there are no other free jobs other than maintenance drones, for example when you have filled the planet with all the possible resource producing districts. The AI has solved this issue by manually going in and forbidding these jobs to keep amenities at a reasonable level, however, I don't have the mandate to simply enable AI behaviour for the player and it is probably not something that most people would like.

Thanks for the reply.

The priorities are more than enough to solve the biggest annoyance, IMO, assuming they also work when there are open complex drone jobs. The current priorities sort of work when the only open jobs are for the simple drones. I don't really see any need for the game to save the player from too many maintenance drones if there are no other jobs available - that should be left for the player to handle.