Stellaris Dev Diary #142 - Sectors

[Lucriiduo]

Oh thank god

 

[HighKings_Ruin]

"Going forward however, I'd like to increase our interactions with you, our community. While we want to have a more open communication, we want to avoid over promising or disappointing you if ideas change radically."

*Clap Clap Clap*

This is what I've been saying. Communication is critical.

 

[GinAndJuche]

If all the planets in sectors are by default set to automated but can be switched to manual, I think it would be nice if only the planets the player deems important enough to be on manual to show in the outiner, or a new tab just for them. At some point a player might not even want some of their earliest planets in the 'core sector' on there anymore because they aren't important planets anymore.

 

[exi123]

I dont think that configurable templates of planets wont work because of how different the amount districts can be. Districts have their own housing, at this point it doesnt matter if a planet has 8 or 10 of the needed ones. Its the same with ameneties: It doesnt matter if the planet has a few pops more or not, who cares about a little bit stability more or less if its in a good stand at all.

The most interesting thing would be to actually plan how a planet "should" look when its "finished". At the actaul state of the game you have mostly no clue how you end up with that fully build planet. And thats why many players look so often for their planets: They want it to build the way they like it.
Planet-planning pre 2.2 was:

- Conquer or colonize planet;
- Get those 5 or 10 pops on it to upgrade the capital building;
- Queue all buildings on the tiles where you want them;
- dump it in a sector to let the ai upgrade the building when possible
- forget the planet for the rest of the game

And now? Even your core planets are maybe not fully grown or build in the late midgame, so you have to babysit them until the last thing is build. And mostly even more when you have to resettle pops because of the growth further the capacity of the planet...

 

[AlbertKerbin]

First off- this system sounds excellent, and I can’t wait to play with the finished version!
I’ll offer my thoughts and ideas, for the two cents they’re worth.

Events for sectors-
Seriously, guys. Events are your strong right arm, and you need to be using them. Sectors don’t have much in the way of events specifically for them right now (I have yet to encounter one, but, to be fair, I haven't managed much playtime in 2.2 )

I want my sectors to be able to rebel against my empire, decide to have wars with each other, and occasionally be as much of a problem as the outside empires. Double this, if I happen to be playing a Feudal Society, which, like Aspec, I believe should be Sector-related, not vasal-empire-related.



Sector range-
Everyone has probably suggested this already, but sector range should probably be affected by Ascension perks and Civics, much like Core Sector Size was back in 1.0-1.9.

Also, those of us with apocalypse should have access to a civic that grants a major buff to core sector size while we have a working planet cracker. (Bonus points if we get a “fear of this battle-station-” achievement… )

Also, possibly a decision for Habitats in a Sector Capitol system that converts them into an Orbital Administration Structure, which, while reducing the output of the habitat itself, increases the size and efficiency of a sector.


Trade-
Obviously, the trade value will accumulate at Sector Capitols before heading to your Capitol planet. There should be some benefit to trade value that only makes it as far as a Sector Capitol, since it is affecting the local economy.

Sector Initatives-
I’m not sure what form this would take, but I would really like to see an interaction option where you can essentially tell a Sector “All right, I want all your planets to have at least one precinct house on them,” and it will prioritize building Precinct Houses on all the planets which don’t have one.


Mini-sectors
This seems like a strange thing, but it should be possible to designate a single system as a sector capitol that can’t and doesn’t have any other planets around it.
Particularly useful for things like Holy Planets, since having a sector automatically colonize one…

 

[Ter.Yeah]

I have to admit I haven't read through everything written in the 8 pages up until now, but what @grekulf wrote initially seems very promising for me.

I'd like to challenge a couple of things though; If sectors were primarily meant to alleviate micro, - I personally would just like to have 1 "sector". (Sector is a bad name in this instance, but I hope you know what I mean). With the allowance for planetary specifications, I don't see the need for multiple sectors beyond "realism" geographically.

I'd also think it would be nice if you could specify at least two things on a planet:
- The focus: (Ie: Forge world, Tech world, Agri world etc). (This was mentioned already).
- Spending (Should the planet be allowed to leech off the rest of the empire, should it be balanced and self-sufficient, or should it be a surplus producer). With a setting like this, the overall resource production for the empire could be further tailored when needed, otherwise left alone after setting the initial setting).

For me, sectors are a necessary evil, and I think personally that the best changes that can come to sectors is to make them as predictable and effortless as possible.

 

[Yagami913]

Just copying a planet doesn't work though.
You might not have enough of a particular district to fill it out, you might want production districts to be kept rather than city districts (for a given "pattern"), or you might not have a match in what features are available. Or indeed a planet might have a modifier.

And of course you have to have built the planet you want to copy up in the first place whereas "build me a science planet" (or a mining planet, or an agri-world, or whatever...) can be triggered from a new galaxy without having to manually setup a world in the first place.

I respectfully disagree, maybe you and i play completly differently but from my perspective: What am i doing in the first 50 years if not building up my first couple of planet??? If im just let the sector ai do everything its terribly suboptimal and terribly boring but then again its just me.
Btw because the build up takes time in my mind u dont need to "finish" a planet before give to the to copy because this copy always gone be behind (ofc there are some edge case)
On regard to the issue with the different planets, its not that complicated. What is important is the build order, what is the raw resourse and the manufatured resourse. Its completly possible to copy an other planet. Ofc if u want to copy an energy planet to a 20 mineral disctirct planet that is your problem. Im sure this way the result give much more statisfatory then the sector ai ever can done. But if you still not see my point hey i not against to use a "simple" sector ai if u want.

 

[NilsFabian]

I am repeating myself but a good compromise between auto sectoring and create your own sector might be applying an admin cap on sectors, preventing one big sector for the rest of the galaxy as i did in earlier builds.

 

[Riktol]

The goal

• Sectors should help to alleviate the player’s need to micromanage everything
• Sectors should feel like a more unique part of the player’s empire

The way sectors are generated makes them too broad to automate, even within the same system you can have planets or habitats which have different functions. There is no point having an agriculture sector if one of the planets has 2 slots for agriculture districts but 20 for minerals. Either sectors should not determine planet/habitat automation behavior or planet/habitats get added individually to sectors. I'm guessing you don't want to do the latter but the former begs the question: why do we have sectors anyway? At the moment it regulates the number of leaders you need and limits how far a single leader's bonuses can be applied.

If the goal is to reduce micromanagement then setting goals on a planet/habitat scale is the way to go.
I think the following should broadly cover things:

1) Pick an action taken for unemployed pops:
1a) automatically build districts or buildings to create new jobs when there are unemployed pops
1b) automatically attempt to resettle unemployed pops onto another planet (either a specific one or one chosen based on habitability and job availability).
1c) no action

2) Set a target unemployment rate: a setting for point 1, maybe you want 5 unemployed pops before it takes action, maybe you want the planet to have a negative unemployment rate and constantly have 2 jobs available for pops to take as soon as they 'pop'.

3) Building queue for buildings and districts: either a strict "build this in this order" list or a "wishlist" system where the planetary AI picks what it thinks the best option is at the time. This should allow placement of buildings in the queue even if there isn't enough population (they get skipped until there is enough) and automatic queuing to remove tile blockers. (Are they still called tile blockers?)


With these options, planets and habitats should mostly be "set and forget", you work out what you want the planet to do, you set a few settings, queue the districts and buildings and 40 years later you have what you wanted. Maybe you forgot that crime increases with population or didn't include enough amenities (I do both aaaallll the time), you can just go and tweak it when you notice the problem.

You mentioned that you think there are too many sectors in late game, I agree. You could get rid of them entirely. Instead of setting and managing budgets for a dozen of sectors, you could allocate a portion of your galactic stockpile for building, if multiple planets are waiting for the same material the one waiting for longest gets it first.

With regards to automating space station construction, an "auto explore" style option would work for resources already within the empire, to automate construction of starports outside your boarders I think the current shift click to queue is pretty much ok, the problem is pre-payment which causes overpayment of influence (because of the extra cost based on distance to your borders) and stops you from queuing if you are short of influence. If you removed pre-payment and simply had the constructor pause if it didn't have enough resources and resumes when you do, then while not exactly automated, the expansion of your empire could be pretty smooth.

I don't really have anything to add about making sectors more unique.

[edits: minor clarifications]

 

[DreadLindwyrm]

I respectfully disagree, maybe you and i play completly differently but from my perspective: What am i doing in the first 50 years if not building up my first couple of planet??? If im just let the sector ai do everything its terribly suboptimal and terribly boring but then again its just me.
Btw because the build up takes time in my mind u dont need to "finish" a planet before give to the to copy because this copy always gone be behind (ofc there are some edge case)
On regard to the issue with the different planets, its not that complicated. What is important is the build order, what is the raw resourse and the manufatured resourse. Its completly possible to copy an other planet. Ofc if u want to copy an energy planet to a 20 mineral disctirct planet that is your problem. Im sure this way the result give much more statisfatory then the sector ai ever can done. But if you still not see my point hey i not against to use a "simple" sector ai if u want.

That first 50 years?
Exploring, having a war with the next neighbour over.
Handling space expansion.

In the first 50 years there's rarely enough development to be worth copying, certainly not to the point where loading and applying a template will be "better" than just building those one or two buildings.

After that, maybe - but you're still relying on the AI building to that template competently and in the right order, and taking into account which buildings it has in the template are better off being built first.


And of course the point here is to get a better AI for both supporting the player, *and* for the computer controlled empires. Getting a better sector AI will alleviate the whole "terribly suboptimal" problem.

And then theres the issue that you'll end up going back and forth copying ever more advanced versions of a given world, ending up with badly fitting generic versions which might as well be "science world", "foundry world", and so on. The downside being that you have to continually update the world templates in each game rather than having good existing ones that persist game to game; even if you do build a better template, the AI is still the problem with getting it to intelligently build to that plan.

 

[Zarpaulus]

I can see an "auto-build" command for constructors to build mining and research stations, sure.

But there might be some diplomatic consequences if they can auto-build starbases.

 

[DreadLindwyrm]

I can see an "auto-build" command for constructors to build mining and research stations, sure.

But there might be some diplomatic consequences if they can auto-build starbases.

Hopefully it cant be worse than auto colonisation annoying the Fallen Empires...

 

[Riftwalker]

I think really sectors should behave kinda like estates in EU4, they have influence, loyalty and bonuses. mix in with stuff like autonomy, and you could get a system where it makes sense to give power away so you can get more resources out of the sector. this would also kinda nerf wide play, but in a way where they can work to refine the land the acquire and still come out ahead.

 

[BlackUmbrellas]

I can see an "auto-build" command for constructors to build mining and research stations, sure.

But there might be some diplomatic consequences if they can auto-build starbases.

Specifically not going to happen- check the dev responses.

 

[Skycroft]

For me the single biggest improvement to micromanagement would be a 'staggered' build queue.

Basically I would like to be able to queue up a bunch of stuff, and then maybe there's a little checkbox next to the build queue that would make so that the planet would only start building the next item in the queue if there is unemployment on the planet.

(Or possibly when there are no free job slots. They are both slightly inefficient, but either is fine.)

That way, I could set up a planet ahead of time and only have to worry about it if there are any problems. You can just queue up stuff ahead of time right now, of course, but unworked buildings and districts use up admin cap and upkeep for no benefit if you do.

 

[RoverStorm]

I think you nailed everything I want. There's just one thing I'd like: an option for districts to be automated, but not the buildings themselves. See, I can't colonize a new planet and instantly build all the districts because that would increase my empire sprawl by a LOT for no return until the pops grow to actually use those districts. If sectors could auto-build districts, but not buildings, I would love that.

Different check boxes for "automate districts" and "automate buildings" would therefore be nice. Of course this is a minor issue compared to all those other beautiful changes you've listed.

Oh, and is this going to be the next expansion, or a mid-expansion update? Because sectors I feel are the one thing that truly needs immediate fixing. I'm totally cool with waiting until the sector fixes are ready, I'm just curious if we'll get them sooner than new content, or if it'll be delayed until the next major expansion?

 

[SolarDwagon]

Just had a thought on how to tie sectors in with existing mechanics
1) All systems in a sector have a factor of reduction of Administrative Capacity (potentially in exchange for not having direct control at all)
2) The factor of reduction depends on the sector strategy-if you set an overarching sector strategy like "produce X resource", you get a massive reduction in Admin Capacity usage, but planets following this might not be even close to optimal, whereas if you set a planet-by-planet detailed plan the effect would be close to nothing.
3) Taking manual control of build queues would either remove them from the sector or prevent the Admin discount for X years.
4) Sector Capitals could either be manually built with new buildings that make their sector better discounted or more effective, or have the same but under automatic control depending on sector settings-changing the capital would destroy these buildings automatically.
5) Any effects would have diminishing returns based on the size of sectors and the distance from their sector (and empire?) capital.

 

[Yagami913]

That first 50 years?
Exploring, having a war with the next neighbour over.
Handling space expansion.

In the first 50 years there's rarely enough development to be worth copying, certainly not to the point where loading and applying a template will be "better" than just building those one or two buildings.

After that, maybe - but you're still relying on the AI building to that template competently and in the right order, and taking into account which buildings it has in the template are better off being built first.

I replaied to your statement about "planet template" not available at the start but sector ai do and now you write exactly my point. In the first circa 50 year there is no need for sector ai or planet template either so there is no difference. I'm sorry if not correctly understand you please correct me. About the build order i think its not to hard to save what was the order you build up your planet, and the ai execute that when the pops growth. For me this is the perfect sector ai.

And of course the point here is to get a better AI for both supporting the player, *and* for the computer controlled empires. Getting a better sector AI will alleviate the whole "terribly suboptimal" problem.

I be very happy if the ai can be much "smarter" about the general macro of the game. But lets be honest its never be perfect morever if its too good that a problem too, because the ai need to be balanced to the average player. Just let think about this: The sector ai always do the perfect strat. Do you ever wanna build up a planet or just play autopilot always? Why do you bother play the macro game if u can not get any leverage on the ai.

And then theres the issue that you'll end up going back and forth copying ever more advanced versions of a given world, ending up with badly fitting generic versions which might as well be "science world", "foundry world", and so on. The downside being that you have to continually update the world templates in each game rather than having good existing ones that persist game to game; even if you do build a better template, the AI is still the problem with getting it to intelligently build to that plan.

Hmmm i'm sure now i did not explained well my thoughts. I dont need to update anything, when planet A has 50 pop and mineral/science world when i saw a similar planet i told the ai copy planet A to B and after that on the rest of the game when new pop available on planet B the ai build exactly the same buildigs in the exatly same order then planet A (ofc there is the district problem but that is not that big).

 

[eagletrekkie]

Having it work that way would mean we'd never get sector-based politics/loyalty/rebellions. No thanks!

I don’t understand how much this logically follows, especially if you put a cap on number of systems in a sector to avoid megasectors.

 

[Everstill]

I don't want to automate the process of building Outposts. I see expanding as being a player choice.

I understand this point, but there is one case where there is no choice:

When a Crisis destroy your outposts.

There is no choice here. A crisis fleet come and destroy your outposts, so now you have holes in your empire to fill again. This is where automation should come. So, at least remake the way the players destroy Crisis outpost.