Stellaris Dev Diary #142 - Sectors

[DreadLindwyrm]

I simply fail to see why the sector assignment has to be automated... Why make it so difficult for no reason at all? You are pretty much creating your own issue here.

Someone might want to use the sector tool to assign a set of systems to be specialized. (like a science sector).
Someone might want to create a sector to then release it as a vassal. in this case one might want to avoid splitting his empire in 2 due to how hyperlanes decide the connections even though he might still have everything connected by gateway.

Someone might just want to ease the burden on his micro management.

You originally names this tool sectors which is fine.
But remember what real space sectors are supposed to be; This set of systems is to far away from this set of systems to be governed together and/or share resources efficiently due to distance and technological limitations. In this regard vassals represent the real idea of sectors much better.

What I'm trying to say is you named a real handy tool 'sectors' and it was handy and nice but now you try to better implement it to be actual sectors (I assume) and are ruining the perfectly good tool which you just so happened to name that way.

Perhaps you could do some renaming of tools/features and/or just stop and think for a second about what exactly it is you are trying to achieve.

I don't want to sound like a know-it-all as this may seem, but I'm trying to help here. I think something went wrong and I want to shed a light on what could be the issue here.

Thanks for reading,
Best of wishes.

Theres a difference between a sector (part of your country, given some autonomy) and a vassal (not part of your country, has a lot more autonomy, has independent military and economy). It's kind of like the difference between a US state, and a puppet regime overseas.

Shifting what is currently sectors to being vassals would cause problems for those empires that cant have vassals. It also makes any CB that gives vassals significantly stronger, as effectively it makes them part of your empire immediately.

 

[SacremPyrobolum]

Just get rid of geography and let us make our own sectors. Give us a planet cap of worlds that can be in our direct control and then a planet cap of how many colonies can be controlled by a single sector so you don't have people just making one big sector.

 

[Bvision]

Theres a difference between a sector (part of your country, given some autonomy) and a vassal (not part of your country, has a lot more autonomy, has independent military and economy). It's kind of like the difference between a US state, and a puppet regime overseas.

Shifting what is currently sectors to being vassals would cause problems for those empires that cant have vassals. It also makes any CB that gives vassals significantly stronger, as effectively it makes them part of your empire immediately.

I don't suggest we turn what we have now as sectors in vassals I simply wanted to highlight that from my point of view the dev team seems confused without they themselves knowing that they are.
We have admin cap which illustrates the issue of over extension, we also have sectors which by definition are supposed to combat this very issue, we also have vassals which are pretty much very autonomous sectors. Perhaps they could take a step back and think. the previous system where the sectors had their own stockpiles and autonomy over what they did and paid taxes and had representation in factions or something else I vaguely remember was far more 'sector-like' then the current version where you get all resources and simply give them build points and tell them to specialize in x. This time however you have no power of how the sector is shaped, which planets belong in it and which don't. What if you have a system with 2 planets, one with rich minerals and one with great science boosts. you can't exactly tell each of them to do a different thing... yet, planet designation may solve that issue.

So my question is: What are you trying to achieve? Make them more like actual sectors would be or make them more like the handy tool it could be? automate development of the planet and give it a specialization?

I suggest you remove the big problem where players have no control of the shape and start work on a new expansion which reworks diplomacy, sectors/admin cap, subject empires, alliances etc.

Having the ability to control the shape and size of sectors doesn't seem to grant any exploitative benefits as far as I know so why not? It's also not necessary for any major mechanics besides the governor leader which I'm sure can be solved by perhaps forcing the player to have x amount of sectors based on total administrative points owned or something.

 

[DreadLindwyrm]

I don't suggest we turn what we have now as sectors in vassals I simply wanted to highlight that from my point of view the dev team seems confused without they themselves knowing that they are.
We have admin cap which illustrates the issue of over extension, we also have sectors which by definition are supposed to combat this very issue, we also have vassals which are pretty much very autonomous sectors. Perhaps they could take a step back and think. the previous system where the sectors had their own stockpiles and autonomy over what they did and paid taxes and had representation in factions or something else I vaguely remember was far more 'sector-like' then the current version where you get all resources and simply give them build points and tell them to specialize in x. This time however you have no power of how the sector is shaped, which planets belong in it and which don't. What if you have a system with 2 planets, one with rich minerals and one with great science boosts. you can't exactly tell each of them to do a different thing... yet, planet designation may solve that issue.

So my question is: What are you trying to achieve? Make them more like actual sectors would be or make them more like the handy tool it could be? automate development of the planet and give it a specialization?

I suggest you remove the big problem where players have no control of the shape and start work on a new expansion which reworks diplomacy, sectors/admin cap, subject empires, alliances etc.
have the ability to control the shape and size of sectors doesn't seem to grant any exploitative benefits as far as I know so why not? It's also not necessary for any major mechanics besides the governor leader which I'm sure can be solved by perhaps forcing the player to have x amount of sectors based on total administrative points owned or something.

There was an exploit with the old manual sectors involving just dumping everything you colonised into one huge sector that took up basically the whole galaxy.

Forcing the player to have a minimum number of sectors based on empire sprawl would have issues depending on how many empty systems and multi-district planets they have, and would be hard to balance so that it works for both wide spread empires with few pops per planet, and tall dense ones with a lot of pops (and thus districts) per planet without causing all those dense planets to effectively have to be their own sectors.

Also sprawl isnt exactly over extension, since its linear and can be outrun by having planets produce more than they cost in term s of efficiency.

 

[ItThatAltersSanity]

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

Regarding the first point, I think the main objective is "make colonies sort themselves out" though a stretch goal may be to give sectors some escort ships to suppress piracy themselves. The establishment of a new colony is less frequent compared to expanding by building an outpost. As such, I believe it's 'acceptable micro' for players to designate every colony with one economic direction. I do not think an entire sector needs a designation for its economic direction... unless that's tied to a governor character. If flavoured as "this governor and their personal staff lean towards e.g. mining minerals and forging alloys", then it makes sense that a whole sector strives for one type of economic output.

In other words, the governor could determine how the AI automates colonies in a sector. That would then let the AI balance the colonies within it to be self-sustaining (or not) - e.g. a research-oriented governor would still build mining districts on appropriate worlds but then build consumer goods instead of alloys with those minerals.

For the second point, I believe the first step should be sector-specific events of a socio-political nature. It need not be full on rebellion, but something that's limited to the pops/colonies in a sector or maybe hostiles spawning - e.g. pirates or "a gene-monster experiment gone wrong". That latter example need not be at the fleet-level, it could be armies spawning on colonies to keep the garrison armies busy (and re-construction after collateral damage).

A future step for the second goal would be to make them feel different depending on the empire's ethics. Right now sectors care not if an empire is Militant-Authoritarian or Xenophile-Spiritualist. I don't know how ethics-switching would mesh with this but having sectors mean different things to different ethics (in addition to being administrative blocks) would very much make them feel unique to a given empire.

If I may switch off my feasibility-limiter and make an outlandish suggestion: what of CK2 style De Facto vs De Jure sector borders? Each empire would have their own idea of how their sectors should be shaped and deviations from that would affect diplomacy (maybe even a CB). Add in "De Jure drift" and make sources of strategic resources add modifiers to a sector to create something that feels alive instead of feeling like pages on a spreadsheet. Other things could add sector-modifiers too, e.g. 'wonders' of cultural value.

 

[TiberiousKirk]

Sector geography
The current plan is to have systems be automatically added to a sector within range. If a system could belong to two different sectors, it should be possible to nudge them to decide which sector they belong to. This important for players being able to set a sector geography that looks good to them in their game.

Moving sector capital will also redraw the sector, and could potentially remove or add new systems to it. You cannot add systems to a sector if they are outside its range. Systems must also maintain cohesion to a sector, so it's not possible to cut off parts of a sector....

....This is also a good chance for you affect this great game, so I hope an open discussion will lead to some constructive exchanges.

I personally prefer the sector geography system that allowed the player to draw their own sectors entirely. I do not mind if the computer wishes to have a default geography in mind, but I want to do more than move planets on the edge of two sectors. In short, let the governor AI control the sectors' planets, civilian ships, local militias, and stations (with the possible exception of military starbases) and give the Player control of the sectors' settings, number of systems, and borders.

 

[skytta]

I like the ideas mentioned in this Dev Diary, but I have one remark and one question.

If you can set a Planet spezialization to instruct the sector AI to build specific buildings and districts then they have to create some spezializations for Habitats because they don't exist at the moment.

And can somebody explain to me why we have to use a sector focus if you can set individual planet/habitat specialization?

 

[xBolivarx]

I like to dedicate some work to the sectors even though they don’t build an ideal economy. In the Endgame I don’t mind loosing some efficiency because of them.

The problem I have is that they cannot run even a single planet reliably. At some point you always have to step in: either there is no room or no work. I’d like to be able to grant sectors strategic resources if there is a demand and to grant them a pool of unity so they can automatically apply policys to hinder pop growth as an example.

Referring to the sectors resource pool: please make them dumping the resources in the empire pool if sector pool is full. Void afterwards.

 

[metalosse]

And can somebody explain to me why we have to use a sector focus if you can set individual planet/habitat specialization?

If you do a agriworld, there is still a few district slots to use.

The sector specialization would set what should be done with those spare districts, I guess.

 

[LaDoncella]

Wouldnt be easier and more reasonable to just restore the sectors from 2.1 and add a soft cap to the maximum number of planets per sector?

 

[meneertje11]

Nice concept of how the sectors should be.
Few question:
A) Can I still micromanage every planet (I personally love large-scale micromanagement)?
B) Are there plans for the civic Feudal Society to make it worth taking (since it sort of fits with sectors IMO)?

 

[Inverness Moon]

It is an interesting thought, and something to look into. I'm not sure a "planet template" will be very easy to set up, though, considering how different planets can be. Maybe a small limit to how much a template can include, so it more easily can fit different planets?

I think there is a problem of priorities here.

I don't want something that is easy to set up, I want something that is effective and makes the AI do what I want.

I believe it is possible to design a planet template system that both provides pre-made templates that are easy to use, but also provides more experienced players the option to create their own.

At the high level. Sectors would be configured with a default planet template that is applied to all planets in that sector unless the player chooses to override the template on a per-planet basis. The default template for a sector, or the specific template for a planet, can be either a pre-made template like "Generator World" or a template custom-made by the player.

Pre-made template should be immutable, but players should have the option of duplicating them and modifying them to suit their needs. They would also serve as examples for players for templates they can create.

My idea for a planet template consists of a list of one or more build queues. Each build queue consists of one or more items (buildings or districts). You would be able to queue the same item more than once if you want multiple instances of it. An option can be provided to repeat the queue if all items in the queue have been built. The build queue could also have one or more preconditions that must be met before the queue can be used.

When the AI needs to make a decision on what to build next, it will examine each build queue in the list from top to bottom. If a queue does not provide a result, it will move on to the next queue in the list until one provides a result. Having a repeating queue as the last item in the list can be useful if the others are more specialized or conditional.

There can be several reasons a queue does not provide a result: all items in a non-repeating queue have been built, precondiitons for the queue have not been met, or there are not enough resources to build the next item in the queue.

Within each queue, items are checked in order. The AI will determine its position with in the queue by counting what items in the queue already exist on the planet. If the queue is set to repeat, then the AI will repeat this operation on the queue in order as many times as necessary. An example of this would be a queue consisting of a farming district, generator district, and mining district set to repeat. If your planet has 2 farming districts, 1 generator district, and 1 mining district, the AI will build a generator district next after repeating the queue once.

It's important to note there that items are not removed from the queue when they are built, instead the AI is figuring out the next item in the queue based on the current state of the planet.

I also think an option is required to decide what the AI does if the next item in the queue cannot be built: skip to the next item after that, or stop processing the queue and move on to the next queue in the list. This would likely need to behave different depending on whether the cause is something like the district limit being reached, or there being a lack of resources.

Queue preconditions could include many things: population, happiness, crime level, stability, available housing, available jobs, planet resource input, planet resource output, technologies, other buildings, etc. Each condition you set should have a "not" checkbox that lets you invert it.

The final idea I have for an option is the ability to alert the player of an item from a queue can't be built. My reasoning is that you might have a queue that requires a certain crime level to exist before it can build a building to mitigate the crime, but perhaps there is not enough space on the planet for this building. You want to be notified so you can remove a building and queue up something that can deal with the crime immediately.

I think combining pre-made templates with custom-made player templates will provide choices desirable for all types of players.

I also realize I might not be doing a good job of describing what I'm imagining in my head. In that case I suggest checking out my inspiration for this which is the AI behavior system in Pillars of Eternity 2. I'm basically trying to translate that concept of casting spells and using abilities into building things in Stellaris. Here's a guide for the system in PoE2: https://steamcommunity.com/sharedfiles/filedetails/?id=1392162466

 

[Avalon8782]

Well there is already the Holy World mechanic. So it would probably need to be more a culture world that all empires use.

Doesn't Consecrate World need to be used on uncolonized worlds?

 

[Orctavius]

I strongly agree with people who have already posted that automation should be implemented at the penitentiary level. A given sector is likely to include worlds of varied development potentials.

 

[methegrate]

I strongly agree with people who have already posted that automation should be implemented at the penitentiary level. A given sector is likely to include worlds of varied development potentials.

Ha! I assume autocorrect has struck again?

 

[Walt (oc.ca.us)]

One additional thing I'd like to see in the mix is the rare resources. In my current game, I'm fighting to get enough of them. But, I can't have sectors focus development on them, so it's into micro. And I ran into what the AI will run into: early, basic commodities were produced inefficiently; now, with various enhancement buildings, the basics are in some cases over-producing, and I have trouble producing enough rare resources.

Maybe the best thing would be for the AI to "look" at the top of the screen and trade: what's being over-produced; what's being under-produced? Make what's being under-produced. Consider replacing buildings that are over-producing.

 

[Beliskner504]

Sectors definitely needed a touch up. Glad to see you guys trying new things.

My only concern is that sectors would become necessary, and with the current governance AI, would leave your planets significantly less optimised than you could do manually. In past updates, this has been the case, and it has been significantly more beneficial to grow and develop a planet yourself, before passing it off to a sector. I'd also like to see how this plays into Administrative capacity.

Hopefully we can get a sector system which works in a balanced way, across all playstyles

 

[RaptorCommander]

One thing that always stopped me using sectors in 2.2 is that you can only set specialization for a whole sector, but different planets lend themselves to different specializations. I'd like to see the ability to set specialization on a per-planet basis.

Could not ageee more. I would much rather be able to group planents by specilisation then geography.

Prehaps we can have 'ministers' which control how types of worlds develop. These ministres could then be expanded to fall under control of a faction. Giving bounese or penilties.

So there would be one for each resource type

Ministry of extraction (minerials)
Ministry of agriculture (food)
Ministry of energy (energy)
Ministry for entertainment (amenities and unity)
Ministry for industry (consumer goods and unique)
Ministry for war (alloys)
Ministry for sceience (sceience)

 

[Kayden_II]

Sectors should help to alleviate the player’s need to micromanage everything

Some additional thoughts in regards to this:
01. You should consider to rethink this strange "X building-slots for Y POPs"-relation since this makes it necessary to observe all colonies in a continous manner to look for (opened) building-slots. I don't really know, why you've changed it, but in the pre-release-phase, there was this "infrastructure"-feature, which had given the player the ability to build up a colony at once ...
02. point 01 could lead to a lot of "unmanned" districts / buildings, which cost (nevertheless) a lot of upkeep-resources. At first, I had thought about an (automatic) feature, which disables such "unmanned" districts / buildings until there're unemployed POPs to "man" them, but I think, that it's a more "elegant" solution to pass the district-/building-upkeep-resources to the "manned" POP-jobs instead ...
03. point 02 could lead to a lot of chaos in which order "unmanned" districts / buildings should be "manned", so that there's a need for a "proper" (aka not the current) POP-job-priority-system ...
04. By the way, point 01 and point 02 should also eliminate the need to observe all colonies in a continous manner to look / wait for unemployed POPs and more than that, if there're unemployed POPs on a colony then you should know exactly, that they're (probably) in a demote-process.

 

[Kinkness]

Planet designations
We really like the planet designations, i.e. “Mining World, Agri World, Forge World”, but we also want the player to have more control over them. We want to add the ability to manually set a planet designation, in addition to the automatic setting. If you designate a planet to be a Mining World, it should perhaps also be quicker to build mining districts there. It should generally feel cooler to colonize a world, and based on its features, immediately be able to decide it should be an Agri World – and designate it accordingly!

We also hope this will make it easier for the AI to specialize their planets a bit more in certain cases.

This I love. While the automated version is nice, sometimes a planet could be.. literally anything, and just because I build a energy grid or something on it first because my empire is suffering in htat, doesn't mean that's what I want the planets main focus to be. Really sucks when it automatically decides "Ahh this large mining district planet.. you want it to be a generator world because you built a generator first! gotchya!" with no way for me to tell it what I really want.

If the Sector AI is hooked into it as well, then its a great way for the player to macro manage the sectors, having a sector option be "By Planet type", the player can set the planet type and the AI Sector does its best to follow that.

Governors
Although governors will remain mostly the same as to how they are now, we will try to remake the governor traits to be a bit more generic and applicable to a sector as a whole, as opposed to being so planet-specific with their bonuses.

Another great one imo. The completely specific governors really didn't work well, same with the very specific sector settings. Just because you have a majority of planets with good food, doesn't mean EVERY planet in that sector is good for food.. It's the same for Governors.

Having passive bonus's that cover a more wider range of planets in 1 bonus would be nice, or governors with multiple different types of bonus's, like Unity/food/science bonus in 1, allowing governors to cover a wider range.