• 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 #306 - Habitat Experiments

Hi everyone!

I hope your summers have been going well! I got a bit sunburnt, but today we’re back and ready to talk about some of the promising experimentation we’ve been doing with the Habitat system.

We’ll be going through the entire development process in this dev diary, so there’s going to be a lot of ideas that were interesting but were subsequently discarded for various reasons.

Why Are You Looking at Habitats Again?​

The Stellaris Custodian team looks to three primary categories when deciding what to pursue:
  • Directives: Things dictated by me, the Game Director, usually for long term strategic reasons.
  • Community: Things you ask for. Pain points, quality of life improvements, bug fixing, and other good ideas from the community.
  • Passion: Things the individual developers really want to do.

Conveniently, a lot of times all of these align quite nicely.

Let’s start by looking at the history of Habitats in Stellaris.

Habitats were introduced way back in the 1.5 ‘Banks’ update in Utopia.

In 2.3 ‘Wolfe’, alongside Ancient Relics, we removed the Voidborne requirement to build Habitats, adjusted their habitability a little bit, and gave them varying districts based on what they were built over.

The 2.7 ‘Wells’ patch made the next major change to Habitats, adjusting their costs, requirements, and adding multiple tiers.

Since then, we’ve added a couple of special Habitat variants, and various other reworks have shifted their fortunes up and down in the overall balance of the game. Recently, there have been many requests from the community to review the tendency of AI empires building dozens of Habitats when they’re otherwise unable to expand.

Due to how production and population work in Stellaris, this led to an interesting quandary - it is theoretically “correct” for the AI to create many Habitats if it was blocked in, but it was tedious as a player to deal with invading up to a dozen Habitats per system. (The current interactions with population growth have also always been troublesome on the game balance side.)

Habitats were feeling far too common, were too good at certain things, and weren’t capturing the base fantasy that we were looking for. They’re the central pillar of a very popular playstyle that we wanted to preserve, though, so this made them a perfect target for “summer experimentation”.

Everything in this dev diary is considered experimental, and may or may not make it live.
All numbers are placeholders for prototyping purposes only. There is no set release date for any of these changes at this time, but we welcome community feedback.

Different Takes​

One of the most common requests from the Community was to add a Galaxy slider to restrict the use of Habitats. Options could have ranged from banning Habitats entirely, to “Nobody (except Void Dwellers) can create Habitats”, requiring the Ascension Perk to build them again, restricting only the AI, or placing (hard or soft) limits to the number of Habitats that could be built.

We also discussed “what if Habitats cost fractional Starbase Capacity to build” - with Void Dwellers and the Voidborne AP granting discounts to this value. This was more appealing, since the soft cap would control AI use of Habitats nicely without significantly hindering players that wanted to go all-in on them.

These discussions led to some questioning about whether Stellaris Habitats were satisfying the general fantasy well enough, and whether Habitats should be more “hard sci-fi”, with lower habitability bases or even ceilings for those accustomed to planetbound life, and whether we could make changes that would address balance challenges like Hive Void Dwellers.

A More Complex Take​

We made a list of some of the current challenges caused by the existing Habitat system, and this led to the idea of “what if all the Habitats in a system were linked?” We could retain the interesting expansion of Habitats across a system while reducing the burden when seizing the system, and potentially address some of the other problems introduced by an excess number of Habitats in the galaxy.

Alfray threw together an incredibly hacky and utterly unshippable version of this, and continued iterating on it during the Summer.

Under this variant, the first Habitat built within a system is the Central Habitat Complex. Additional Habitats are Support Habitats that add additional space and versatility to the Central Complex. A reminder, many values are grossly unbalanced placeholders in the following screenshots.

At this point I went on vacation, so I’ll turn this over to Alfray to talk about his investigations.

Once More Into the Alfray​

Keep in mind that the numbers shown in the below screenshots are never intended to be the final values, but were used purely for testing purposes of how the systems felt to use and play with.

Firstly, to counteract the expected changes that with minimal Support Habitats, the Central Complex would be small, cramped and overall not great to live on, I gave Void Dwellers extra districts and building slots as a unique modifier (This saw further refinement in a later prototype).

An early version of Void Dwellers

Support Habitats as Megastructures:​


The first iteration of these prototypes made use of Support Habitats as additional megastructures.

Habitat Central Complex, v1
Support Habitat v1
Expanded Support Habitat v1
Advanced Support Habitat v1

In this prototype, we had the maximum amount of each type of resource collection district (Energy, Minerals, Research) limited by the size of the deposits the habitats were constructed over, similar to how buildings for Strategic Resources are limited.

Support Habitats provided additional Districts, Building Slots, and Housing to the Habitat Central Complex, while reducing the Habitability (to reflect the civilian traffic between habitats) as they are upgraded. The final tier also allowed the Habitat Complex to use deposits on moons of their orbited planet.

On the surface, this prototype seems to satisfy our initial requirements and more:
  • Conquering systems with Habitat-spam was easier due to there only being one functional “planet” per system.
  • Constructing multiple Habitats per system felt rewarding as it upgraded your existing colony.
  • The removal of multiple starting colonies removed one of our main concerns for allowing Hive-Minds to have access to the Void Dweller origin - their high pop growth rate due to excessive numbers of spawning pools in the early game. (Iggy had some thoughts on this that he’ll be mentioning in a future Dev Diary).

A Void Dweller Habitat Complex with way too many districts

A Void Dweller Habitat Complex.

A very cluttered system

The rather cluttered system said Habitat Complex is in.

However, the Support Habitats couldn’t be interacted with outside being upgraded, which felt like a major downside. Enemy ships would happily fly past and ignore the Support Habitats, they couldn’t be specialised or downgraded.

All things considered, this prototype showed that making habitats into a single logical planet spread across many entities in a solar system felt good, but megastructures were not the path forward.

Support Habitats as “Starbases”:​


Keep in mind that the numbers shown in the below screenshots are never intended to be the final values, but were used purely for testing purposes of how the systems felt to use and play with.

The below screenshots feature placeholder art and the default art for starbases, their buildings and modules.


The second iteration of this prototype investigated treating Support Habitats as special Starbases (much like Orbital Rings).

In this prototype, the districts available to Habitat Central Complexes depend on the configuration of any Support Habitats in the same system. Thus construction of a Habitat Central Complex would automatically build a neighbouring Support Habitat in orbit of the same planet.

When built, a Support Habitat would start with a module that matches any deposits on the planet it orbits. Each <District> Module on a Support Habitat, gives +3 Max Districts of that type to the Habitat Central Complex.

Upgrading the Support Habitats, still provides the same modifiers as shown in Megastructure Prototype. Additionally each tier of the Support Habitat allows construction of an additional Support Habitat module and the second and third tiers allow construction of a Support Habitat building.

Expanded Support Complex v2

The starting Habitat Central Complex and its neighbouring Support Habitat for a Void Dweller empire.

New Habitat Complex v2

A newly constructed Habitat Central Complex, completely unspecialised.

Allowing the choice of which districts the Habitat Central Complex has access to via specialisation of the Support Habitats brings some interesting changes to how Habitat-dependent empires play.

Due to the nature of the prototype, the buildings for Support Habitats haven’t seen much investigation yet, but would likely include buildings much like those on an Orbital Ring, the lunar extraction support that Advanced Support Habitats experimented with in the Megastructure Prototype above and other such buildings.

Research Habitat Complex, v3 or so?

A Research Habitat Complex, using some of the district capacity to provide hydroponic districts.

Due to the nature of summer experiments, we can’t say if or when this prototype might make it into the live version of the game, but it’s something that we’re interested in exploring further.

…But the fourth one stayed up!​

Thanks, Alfray.

That variant listed achieved a lot of the goals we were looking for, but was cobbled together out of the scripting equivalent of sticks and twine as a quick and dirty implementation. It also required a lot of back and forth clicking that we really weren’t too fond of. So after that one burned down, fell over, and sank into the swamp, we came up with another iteration.

My feedback: Simplify things.

The latest variant we’ve been playing with has been especially promising. In this one, we turned the “Starbase” style Support Habitats into single tiered “pre-specialized” units (renamed to “Orbitals” for UX purposes) rather than requiring Modules to be built on them - so you could build a Mining Orbital, Research Orbital, and so on.

This dramatically simplified the flow of building out Habitats while simultaneously improving the implementation.

It's an Orbital!

Pre-Specialized Research Orbital.

Habitat Transit Hub. Hey wait, Maintenance DRONES?

Unique buildings on the primary habitat complex can increase the effects of the orbitals.

Upgraded Habitat Transit building.

We’re still doing some experimentation with this model, but so far we’re liking what we’re seeing. Technologies can add special Orbital types or buildings that can modify the primary Habitat Complex, and it’s very easy for us (or modders) to add new types.

We've been looking at jobs per districts too - the Complexes have different challenges from the older Habitat system, and further updated the Voidborne Ascension Perk. Void Dwellers will start with its effects (similar to how Teachers of the Shroud empires effectively start with Mind over Matter).

Void Dwellers Final Text
Voidborne Ascension Perk v3.final.final(2)

Void Dwellers get Habitat Build Cost reductions in Traditions.

What’s Next?​

For now, I’d like to get some of your thoughts on what you’ve seen today, which we’ll bring into our internal design discussions. It would also be great to get feedback on whether you like this sort of diary, where we go through the overall process (including the failures).

Next week I'd like to talk about a Summer Experiment relating to leaders that didn’t pan out quite so well, and our plans on how to proceed with that.

See you then!
 
  • 125Like
  • 44Love
  • 16
  • 7
  • 6
  • 1Haha
Reactions:
It really doesn't, what hurts performance is the ai spamming them everywhere.
I mean I have played with the void dwellers before, it certainly does affect performance when you have 10 or more habitats in a small space.
The logistic growth cap (and floor) are the cause of this problem, not anything that alleviated it.
Yea, the original mod that logistic growth was based on had the base growth as 1 or 2 or something. It was more balanced than having the base growth as 3.

If the base growth was 1 or 0.5, then they could do away with the stupidity of 'empire wide pop growth scaling'.
 
  • 1
  • 1Like
Reactions:
Well, I respect that you feel that way, but I think you're in the minority of the player base on this. Many players (self included) along with the Devs are of the mind that the fun of reducing tedious management is more valuable than the fun of having multiple 'independent' habitats.
I get where that's coming from, but it's very likely these changes are hardcoded, so I won't be able to fix this with a mod.
 
  • 1Like
Reactions:
I mean I have played with the void dwellers before, it certainly does affect performance when you have 10 or more habitats in a small space.

Yea, the original mod that logistic growth was based on had the base growth as 1 or 2 or something. It was more balanced than having the base growth as 3.

If the base growth was 1 or 0.5, then they could do away with the stupidity of 'empire wide pop growth scaling'.

Why is empire wide pop growth scaling bad?

Personally, I think they should move all pop growth to the species tab, and make it more explicityly an empire wide mechanic - the main issue I see is that colonies cause pop growth, when it should be the opposite (pop growth should cause the need for more colonies)
 
  • 4Like
Reactions:
I mean I have played with the void dwellers before, it certainly does affect performance when you have 10 or more habitats in a small space.
In my experience, it's not that much of a difference in performance to have ten habitats in one system versus ten habitats in ten systems, the slowdown usually comes when the ai always builds several habitats in every system it owns.
 
  • 1Like
Reactions:
In my experience, it's not that much of a difference in performance to have ten habitats in one system versus ten habitats in ten systems, the slowdown usually comes when the ai always builds several habitats in every system it owns.

The difference is that ten habitats in 10 systems is a vastly LARGER proportion of your overall empire size in this new version - and would be way more expensive.

10 habitatis in the current version might only support 30 to 50 pops without void dweller or max upgrades, whereas the same number in the new system could potentially cover your ENTIRE empire.

The idea is that one habitat in the new system is modular, and can supply housing, jobs, and buildings equal to half a dozen old habitats, with only one window to manage - but the same amount of output.

Thus, building 10 habitats in 10 new systems would be a much longer task, and inefficient as you would need several hundred pops to fill them all.
 
  • 5
  • 1Like
Reactions:
Fantastic to see this getting looked at! I've seen lots of complaints of people going "aaaargh why are there so many habitats, make the AI not do it" and me replying "but they're actually good, if perhaps somewhat tedious", so I'm glad to see the exploration of solutions that can make both camps happy. ... even if I usually just play 2.1, 2.8, or 3.2 nowadays.
 
  • 1Like
Reactions:
Why is empire wide pop growth scaling bad?
It doesn't make sense from a lore standpoint, and also encourages vassal spam.

Realistically, pop growth is determined per-pop rather than through an entire country. It also makes even less sense that having more pops would slow pop growth down in the way that the empire wide pop growth scaling works currently. It's very immersion breaking and incentivises cheese like releasing vassals and then reintegrating them to avoid the penalty.
 
  • 5
  • 2Like
Reactions:
I wish I could build habitats on asteroids. I know it would be really hard because sometimes they are very close to each other but the role play makes sense when you build mining habitats near really rich mineral deposits etc.
 
  • 2Like
Reactions:
The difference is that ten habitats in 10 systems is a vastly LARGER proportion of your overall empire size in this new version - and would be way more expensive.

10 habitatis in the current version might only support 30 to 50 pops without void dweller or max upgrades, whereas the same number in the new system could potentially cover your ENTIRE empire.

The idea is that one habitat in the new system is modular, and can supply housing, jobs, and buildings equal to half a dozen old habitats, with only one window to manage - but the same amount of output.

Thus, building 10 habitats in 10 new systems would be a much longer task, and inefficient as you would need several hundred pops to fill them all.
The issue is that I don't want to have habitats be one massive world I put all my pops on, I want them to be small, but plentiful.
 
  • 2
  • 1
Reactions:
I realy want the option to disable all habitats in the game start options, is the only mod i actualy use -k

All the ideas are interesting lets see where they will go
 
It doesn't make sense from a lore standpoint, and also encourages vassal spam.

I think you're mis-understanding my question.

What I'm imagining is that instead of seeing the pop growth ticker on a planet, you see it on the species screen for example, for each species. When a pop is created, its then placed on a colony based on some automatic weight calculations.

But critically, this means that having more or less colonies doesn't drive your need for pops - instead, having more pops drives your need for colonies; which seems more accurate to how nations expand in real life.

Its also unclear how such a system would encourage vassal spam.
 
  • 2
  • 1Like
Reactions:
The issue is that I don't want to have habitats be one massive world I put all my pops on, I want them to be small, but plentiful.

Well, you can probably still build one habitat per system if you want to spread them out over the empire, and they will just be very small. But if you want a bunch of habitats in ONE system, you now just just make the complex bigger.

Shrugs - But, you can't make everyone happy all the time.
 
  • 2Like
Reactions:
With this, they should add new habitat exclusive buildings to counter some of the big downsides for expanding the habitat. Like a life support building or a monorail building that improves habitability. The monorail building could be upgraded to like a hyper rail building for higher levels. Since the size is reducing habitability from hurting movement and not just life support issues, they could have it reduce max habitability (because species won't make much difference to moving) than also reduce regular habitability as the life support is more stressed. So adding a life support building improves habitability. Building an monorail/infrastructure building increases the max habitability but not regular habitability.
 
  • 1Like
Reactions:
It doesn't make sense from a lore standpoint, and also encourages vassal spam.

Realistically, pop growth is determined per-pop rather than through an entire country. It also makes even less sense that having more pops would slow pop growth down in the way that the empire wide pop growth scaling works currently. It's very immersion breaking and incentivises cheese like releasing vassals and then reintegrating them to avoid the penalty.
I think the issue is that the pop system itself is unrealistic, so any changes would also seem unrealistic.
 
The big reason I haven't been using habitats recently is that orbital rings are just so powerful. The extra resources from miners, technicians, metallurgists, etc. have ended up leaving habitats kind of in the dust. This looks like there'd be a good way to implement those buildings - is there any plan for that, or are they just going to be added as another tier of upgrade on the habitat?
 
  • 2Like
Reactions:
With this, they should add new habitat exclusive buildings to counter some of the big downsides for expanding the habitat. Like a life support building or a monorail building that improves habitability. The monorail building could be upgraded to like a hyper rail building for higher levels. Since the size is reducing habitability from hurting movement and not just life support issues, they could have it reduce max habitability (because species won't make much difference to moving) than also reduce regular habitability as the life support is more stressed. So adding a life support building improves habitability. Building an monorail/infrastructure building increases the max habitability but not regular habitability.

Instead of luxary apartments, you can have private moon houses!
 
Its also unclear how such a system would encourage vassal spam.
In the current game, it does already encourage vassal spam. Because when you have 1000 pops for example, it takes 350 points to grow a pop. If you split those pops over multiple vassals, it would only take 125 points or so to grow a pop.

What I'm imagining is that instead of seeing the pop growth ticker on a planet, you see it on the species screen for example, for each species. When a pop is created, its then placed on a colony based on some automatic weight calculations.
It might be better than the current system of empire wide pop growth scaling in terms of game balance, but also doesn't make sense from an in-universe perspective. It would mean that 10 empires with the same number of pops and space as a single empire would grow 10 times as fast, which again unfairly buffs vassal-heavy playstyles and also isn't how reproduction works in real life. It would be better than the 2.8 pop growth or the current empire-wide scaling, but I would prefer a properly implemented logistic growth system where pops produce pop growth, and carrying capacity stops it.

P.S. Essentially, I'm saying that pop growth should directly come from the number of pops, rather than from colonies or the empire itself.
 
  • 4Like
  • 1
Reactions:
Well, you can probably still build one habitat per system if you want to spread them out over the empire, and they will just be very small. But if you want a bunch of habitats in ONE system, you now just just make the complex bigger.

Shrugs - But, you can't make everyone happy all the time.
Habitats shouldn't be massive planets that you build, we have ringworlds for that. My question is what was wrong with the current system that required changing everything about it? Ai spam could be fixed by changing the ai to not build as many, which would improve performance, and micromanagement could be fixed by the player by turning on planetary automation if you don't want to deal with it.
 
  • 2Like
Reactions:
It would mean that 10 empires with the same number of pops and space as a single empire would grow 10 times as fast, which again unfairly buffs vassal-heavy playstyles and also isn't how reproduction works in real life.

Ah, ok, I see the confusion. That's now how I imagine it would work. The base rate that your empire produces pops is not a static number that never changes - the modifier would be based on your current pops, your empire size, your available housing, and modifiers from things like planet stability.

In short, you can adjust the math so that an empire with 100 pops and an empire with 1000 pops grow at different rates based on policies and resources of the empires.
 
Habitats shouldn't be massive planets that you build, we have ringworlds for that. My question is what was wrong with the current system that required changing everything about it? Ai spam could be fixed by changing the ai to not build as many, which would improve performance, and micromanagement could be fixed by the player by turning on planetary automation if you don't want to deal with it.

Agree to disagree. There have already been many posts in this thread and previous ones as to peoples complains with habitats.
 
  • 1Like
  • 1
Reactions: