Stellaris Dev Diary #142 - Sectors

[Bearjuden]

Not entirely sold on this since it prevents sectors being disputed between empires, but...I'll keep an open mind. We'll see. I'm interested about this world designation, that's cool, and maybe the sector capital designation can grow into something more...

 

[Bad Bill]

Glad to see this, like most everyone else. Two comments/ideas: First, could sector size (the X in the statement "Sector range simply means all systems within X jumps from the sector capital") be settable via slider in the Settings available at the beginning of the game? Seems like that's a simple way to accommodate different playing styles, and it shouldn't be too hard to program (famous last words...).

Second, maybe I'm weird, but I personally don't mind the micromanagement of sectors NEARLY as much as I dislike the hassle of dealing with trade-route piracy. Might I humbly request that when we reach a point where sectors are working to everyone's (relative) satisfaction, some thought be given to having trade routes connect sectors rather than planets? Probably premature to do that until sectors are in better shape, but the micromanagement of patrols and other piracy counters drives me bonkers.

 

[Duuk]

I am not at all being mean when I say that I cannot wait for the future DLC that expands sectors even further.

I want internal factions to be led by sector governors, have sector events, have sector revolts, etc.

As a long-time CK2 player, this will make Stellaris SO MUCH more fun for me.

 

[Yagami913]

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.

Exaclty, i very like it. Its somewhat similar to mine. I mean in my version this "staggered build queue" saved in the ai from a previous planet and keep watching even after its finished his queue. But if the queue visible and i can modify is even better. What is your opinion? Is it good if u can build manually this "staggered build queue" or fill it with another planet building queue and modify from that.

 

[Galactic_Economist]

Hello there! I am very glad you decided to use our collective brain a bit more. I actually thought about the problem sector and I have a suggestion that, I believe, might be interesting.

Give us the ability to create planets blue prints. More specifically, let us design some planet templates, including the order of the buildings. Of course, there is a need to have "incomplete blueprints" as the specific design of districts depends on the features of the planet. To me, this idea have a few potential upsides.

1) It give the players who likes efficiency the ability to delegate and stop micro managing like hell. With a blue print system, it will be on me when the sector is not doing well, so I won't complain.

2) We could share and exchange blue prints in the community like we share and exchange about fleet and ship design. It would be even better if we could "import" from game to game or from other players blue prints. (BTW that should be a must for ship design so experienced player could stop repeating themselves about the "current ship design meta" and just share templates.)

3) You could offer us some starting blue prints so that the new players have access to what is the "standard" design of a planet.

4) To me, the potentially most important point: data collecting. With blue prints you could observe not only the actual actions of the players but also their plans and partially their strategies. To be informal but still a bit technical, let us define define a plan as an anticipated serie of state-contingent actions, for instance "build a forge (1st action) at 10 pops (1st state) and then build a mine (2nd action) at 15 pops (2nd state, conditional on the first)". A strategy is therefore a plan designed to achieve a specific objective. In the current design of the game, your learning algorithm for AI can only observed what is decided by the player at the moment he needs to do the stuff, but it cannot see how the players anticipate the future and build accordingly. (To be technical: observing an action at time t is not necessarily informative about the action at t+1, and so observing a serie of actions (the unfolding of a gameplay) is probably not a "sufficient statistics" for determining the player's strategy. If you did a bit of decision theory or game theory you'll understand that I define a plan as a mapping from a set of states of the world to a set of actions, and a strategy is a plan that maximizes an objective function. You cannot perfectly deduce strategies from observing only actions as there can be many plans that effectively yields the same serie of actions as the states are revealed.)

Anyway, hope it helps, and keep up the good work.

 

[Alexu]

Would be cool if you could designate a series of non owned planets as a "neutral zone" sector between yourself and other empires. Get Casus Belli if it's broken. Probably outside the scope of a sector rework, but it'd be nice if the system could support that down the line.

 

[Jorlem]

Space stations
We have discussed adding an auto-build function for construction ships, similar to auto-explore, which should hopefully solve this problem better.

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

If I may ask, have you seen my suggestion thread on construction ships? There is some discussion of automation in the thread, and my initial suggestion includes a way to "semi-automate" building outposts, that keeps the actual choice of where to expand entirely in the player's hands, but would reduce the micro involved in ordering the construction of said outposts.

 

[Kelben]

Looks like the goals mentioned need a better than now AI

That is the real issue I believe. Let the AI handle the development of your sectors to get a first hand experience of how the AI manage entire empires.

I still do not understand all this energy being spent on having an automatic and dynamic sectors creation/evolution. The manual layout before was just fine. Don't want people to use a single mega sector with all their planets in? Then limit the number of jumps one can draw the sector from said sector capital planet.

I hope I'll be proven wrong but I cannot imagine a dynamic enough sectors evolution that will be neat enough to satisfy everyone and react properly to edge cases scenario.

I would also be satisfied with sectors created with the map as mentioned by others people.

 

[Reilus]

I would love to be able to make some templates for starbase's modules, defensive stations and buildings, where I could just click on outpost, and then queue some template to build itself, because it becomes very tedious (especially in late game where you have to build a lot of starbases) very fast. The problem with current system is that you have to click on this starbase at least 4-5 times to queue next starbase upgrade, then click 6 times for modules and 4 time for buildings, and after that 10+ times for defense platforms, which is a lot of unmeaningfull clicks, which i would love to automate (it becomes so MUCH WORSE with mods like NSC where you have 12 modules and 12 buildings and 60+ defense platforms per starbase).
For example: in the late game i would love to have 4 types of spacebases, anchorage (for naval cap), fortpost (for defending choke points and my empire borders), trade station, and shipyards.
The same in my opinion is needed for planets (but to a lesser extent): mining worlds, agri worlds, tech worlds etc (also i'm fine with giving the decisionmaking to competent AI on what building to build on planets, but not with starbases). Also i would love to be able to change the planet type to whichever i want (maybe for some influence).

 

[Xrona]

add auto construction mining / research stations.

 

[Don_Quigleone]

With constructing mining/research stations, it would be nice if, when you tell a ship to build everything in a system it only subtracted resources for the next base in the queue, and if there's insufficient resources, it waits until there is sufficient resources.

This would eliminate a lot of pointless early game micro.

 

[phanboi]

Glad this is getting tackled Coule you also look at the UI for sectors and planets? As it stands by the late game the info panel gets full of sectors, most of which are closed all the time but just take up space. Sometimes though a player might need to have a particular sector’s info close at hand, like if it has an important fortress world which would be even more relevant in this case.

Perhaps an ability to designate planets “important” so that they appear in a separate “important planets” category could help with this, bypassing the sector info clutter.

yes I would love to have a planet window that you could sort different ways and goto with a clik

 

[Chaincat]

Please just let us draw our own sectors like we used to.

 

[Argen Tum]

I would again like to suggest a system like in my suggestion in the linked thread, where you can set how large defict a planet can run, and where jobs are assinged more on purpose, as opposed to random, as (I think) it is now.

https://forum.paradoxplaza.com/foru...nomic-overhaul-from-long-time-player.1160542/

 

[Strater]

Have you thought of adding a template system for planets, similar to what is described in this reddit thread?
https://www.reddit.com/r/Stellaris/...ates/?utm_medium=android_app&utm_source=share

 

[Dalos]

@grekulf Since you mentioned automation of construction ships would it be possible to expand context menu for science ships with something like "Research project(s) and continue surveying"?

Also if it is not already possible(maybe I missed it...) - would it be possible to add filter to outliner (list of planets, ships...) to include only those items that required action - idle ships, planets with unemployment...? Would help a lot.

Thank you and good luck!

 

[permeakra]

• Sector geography can feel wrong
• There are too many sectors in late-game
• No manual control of sector area

Hm.

I guess, that the design intention is to create a nudge for players to move systems into sectors and to provide some sort of cap to their size, potentially growing later into the game.

I believe, this can be achieved adapting administrative capacity mechanics. Consider following scheme

1. Penalty for going over Empire-wide Admin Cap is greatly increased
2. Each sector has its own Admin Capacity
   
3. When a system is in a sector, 3/4 of empire sprawl it produces are counted against Sector's admin cap instead of the empire-wide admin cap. So, a system in a sector produces 1/4 of Empire-Wide Sprawl.
   
4. Sector gains moderate penalty for going against its sector-wide admin cap, but it should be significantly less than for going above empire-wide admin cap.

Adoption of a mechanic of this type will provide a motivation to keep sectors small or moderate and keep systems in sectors. With addition of tech increasing sector-wide admin cap, size of late-game sectors can be adjusted.

 

[jady]

Why not both? That is the intention, at least.

Why both? If i can set a planet designation for example to agri world, why should i put this planet in a sector on top of that?

I also see other AI problems coming with two systems, where a planet designation AI and a sector AI working against each other.

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?

You should only implement some general planet templates that will work for most cases. But made them full modable, players can made the perfect templates for every situation.

 

[jady]

Sectors would not have their own economy, so things would affect you directly. The budget would simply assign them resources to build with.

Why do we need then a sector resource pool and shared sector resource pool?

I think we only need one resource pool and then simply assign resources for the sectors. For example, i can set that sectors can use 20% of the mineral income and 100% of the food (because i want to share the food with all planets).

 

[Antaro]

I do not fully agree with the sector rework, I think too much emphasis is being given to sectors, and I think the emphasis should be given to decrease micromanagement... Sectors is just a way to solve the micromanagement problem, but it is not the only one. Overfocusing in sectors might blind us to see other options.

I think most fo the people like taking high-level decisions (such as... I want to decrease the number of mining districts, and to increase the number of energy districts, or I want to create 5 silos), but the details on how to do it (go to several planets and change things in them) is tedious most of the time. I would like to propose something different that sectors...

I think it would be very nice to have a screen to macro-manage your empire. This could have information of the overall state of your planets, such as:
* You have 200/250 districts available (50 nexus, 50 energy, 50 minerals, 50 agricultural).
* You have 500 population (20 unemployed, 40 clerks, 20 scientists, 60 whatever...).
* You have 600 population space (10 population in overcrowded conditions in 3 different planets)
* You have 300/360 buildings.
* You have 3 Planets with crime above x.
* You have 4 Planets with amenities below x.

And that you could take decisions like:
* Build 5 energy districts (and the system should automatically try to find the best places to build it considering several factors, like planetary bonus, unemployment, population space...)
* Replace 5 mineral districts by agriculture ones (same rules as before).
* Build 5 silos (or research centers, or whatever...).
* Migrate pops to a particular planet (and they are taken automatically from other planets following some optimization rules).
* Move populations from one job to another (again, in which particular planet that would be executed is to the optimization process behind).

I think this will also create the possibility to create better AI-empires, because this optimization system would be separated from the decision-making, and the IA should only focus in high level decisions for the economy, and use the same micro-management optimization process than human players.

When I talk about the optimization of the high-level decision players should take, this does not need to be a fully fleged optimization system, but it could be something simplified like a set of rules with some weights. We do not need to be perfect, just reduce micromanagement!

Nice thing about this is that it could escalate to basically any size of galaxy. It can work from a few planets (maybe 10 or so) up to whatever number.

Thanks for reading it!