Stellaris Dev Diary #153 - Empire Sprawl & Administrative Capacity

[methegrate]

While I like more modifiers to the various sources of empire sprawl, I'll add to the chorus and say I think it's a really bad idea to let pops cause sprawl. Currently empire sprawl is basically a tax on the sources of raw materials: systems and districts. If it becomes a tax on pops as well, it becomes just another upkeep, similar to consumer goods, crime, amenities, housing, food... We've already got enough of those.

I feel like most aspects of empire sprawl would be better calculated on a logarithmic scale. That's especially true if they use pops in the formula.

Right now the linear scale means that adding a second star system and adding a hundredth star system are, for all intents and purposes, the same thing. It shouldn't work that way though. Integrating the handful of stars right next to your capital should be essentially a trivial task. Ruling over stars on the far reaches of your empire should be an order of magnitude harder.

At least, that's how I'd do it. I'd make empire sprawl a curve where the more you grow, the harder it is to keep growing

 

[Methone]

On the whole machine empires vs hive minds: I think the issue goes beyond pure balance and wide or tall. Both empire types feel like Stellaris with less features.

One of many, many tiny things that make me feel these empires are basically just tacked on like an incomplete Mod: Many notifications make no sense. There is an anomaly that reveals a gas giant to be a barren planet, and the notification speaks about how my scientific community is embarassed by this oversight. Well, I am a hive mind. But apparently there is a scientific community? What? It is worse for machines, as they certainly have not even the potential for a community of individuals.

I love robots. My first Dawn of War was Dark Crusade with the Necrons, when I found out there is a synthetic ascension path, I was overjoyed. In the cinema I cheered for Skynet. The first empire I ever played was a gestalt consciousness. But now I just don't play hive minds/machine empires at all. They feel incomplete, and are frustratig to play, despite the power of machines.

Yeah, Gestalts have always lacked an enormous amount of flavor. I mean... bickering archaeologist even in Ancient Relics? For a Determined Exterminator?!

I'm not entirely unconvinced that Stellaris team sees Gestalt as a mistake they added once, and wish they could just make go away.

 

[Druplesnubb]

I definitely think that Administrators and the various ruler pops should help with Administrative Sprawl.

 

[Jobindo]

I'd say you have it in reverse... a Hive-mind is one singular entity, wherever it would be, it's one mind and one consciousness. Machine Intelligence could POSSIBLY have many voices that make it's own Chorus. Hell, look at the Geth: Platforms gain complexity and intelligence the more there are in a given network, and Legion is more than one thousand programs all in one shell.

Organics could also work like that. See Dominion Founders from DS9 Star Trek

 

[methegrate]

Also, while this is a little bit of a reach, I might change how sprawl affects an empire. I'd keep some things, like the cost on edicts and maybe traditions. But mostly I would make it affect systems and colonies directly. I would make admin cap essentially geographic, a measure of how far and how effectively you can govern. Based on empire sprawl and admin cap, the further a system is from the capital the more you:

- Reduce output and stability;
- Increase upkeep costs;
- Increase issues with deviant factions and ethics drift.

And I would make these penalties severe enough that players really do have to think about whether adding a new system is worth it (or if it might even harm them). Right now empire sprawl really isn't a factor. The penalties are real, but the benefits of expansion always outweigh the downside.

Basically, empire sprawl is all about not being able to govern the farthest reaches of your empire. So it should literally make those areas less governable. Instead of doubling the cost of tech, I'd cut in half the amount of research you get from far-flung colonies. Your frontier sectors and remote outposts should be hives of scum and villainy, reflecting the fact that the imperial writ carries very little weight here. There's a lot of ways to express this.

This would also open up lots of potential for when internal politics get expanded. Sprawl could be integrated into a better leader system or more interesting factions, letting them take advantage of worlds that are far past your administrative cap. Once we do something interesting with ethics drift, this could be a very big deal. Heck, this could even lead to a more interesting piracy and crime system.

Regardless, I wouldn't keep it as a generic, empire-wide sense of malaise. Trantor should still hum along. It's just that you've gotten too big to keep your boot on Tatooine.

 

[Frostyant]

Ok so this is a massive nerf to tall empires, at least using current numbers.
But the ideas aren't bad.
After all I *really* don't like that pop growth is the best thing in the game, which I think is what the devs are trying to fix here.



Actually, if numbers/civics are fixed then tall empires might become much better.

Some possibilities :

- Make systems and colonies just increase the cap more than pops (I am thinking maybe 2 for each system and 5 for each colony).
- Bureaucracy can only be built on your capital but have 3-5 levels of upgrades requiring you to heavily invest in your capital.
- Give megacorp -25% pop sprawl and give a civic for ALL empires (hive, machine, normal, megacorp) giving and additional -25% at the cost of +50% sprawl penalty.
- Bureaucracy can only be built on core worlds, and tall empires (ie megacorps, machines etc.) get a bigger core sector (so 6-7 jumps instead of 4).


Also give sprawl more teeth, like make it increase crime/deviancy and make those in turn increase pop upkeep (which is an indirect buff to hives).

Lastly, again for the poor hiveminds.
How about hiveworlds increase capacity in some way ?
This means that while hiveminds can increase their cap massively if they heavily invest (or else suffer the penalties even if reduced).

 

[BlackUmbrellas]

It's not that Machines HAVE to be wide and Hives HAVE to be tall, but if I want to play a science hive with Pooled Knowledge and Natural Neural Network, I should be able to without the game constantly going "Are you going wide yet? How about now? Now?". Same with wide machines - you know, the Exterminators and Assimilators come to mind?

Should you? Or should different types of empires behave differently? Hiveminds have ALWAYS been focused on expansion and "going wide", and the devs have made it clear they want Machine Intelligences to be highly efficient but have a smaller footprint.

I highly doubt that these Admin Cap changes will prevent an Exterminator or Assimilator from eating the galaxy, but "Hiveminds are for wide play and Machines are for tall" has pretty much always been the design plan and suddenly going "that makes no sense, it should be the opposite!" and trying to scrape together lore justifications is silly.

 

[Millbot]

I like how this change actually makes both the administrative capacity civic and the perk both worthwhile. The perk will scale with how much administrative capacity our empire has. Is the civic going to stay the same or will it be reworked similar to the perk and a percent? Or something else?

Also like the idea of hivemind being geared towards wide and machine wanting to be tall. In that vein, would it be possible to tweak some of the specialized worlds? For instance have hive worlds increase administrative capacity, while maybe having machine worlds reduce the sprawl pops on that world create? Also would mind seeing ecumenopoli get a tweak. I'm not a huge fan that people without the perk can get the full benefit of such worlds, but also less thrilled that both machine & hive empires, can create worlds that normal empires have to terraform (assuming non-synth acceded for the machine world). Could ecumenopoli be coded so that they create a sprawl malus for gestalt empire types? (ex. if a gestalt empire takes a ecumenopolis, the sprawl from pops they have on that world is tripled or quadrupled).

Also nice to see how sprawl and administrative capacity could be a good check on the whole "more pops always being better." If the numbers pan out right, this could be a great way to add some choice, where some of the time, when presented the choice to increase total population (either new pops, growth speed or assembly) and something else, that sometimes something else gets picked instead.

Only real concern is that this could make low resource systems really unappealing or feel punitive. Maybe systems where the player can only build one or two stations, should have a lower sprawl value.

Finally, will the tech for admin capacity be made available to empires that only have one inhabited world? Currently, people cannot boost admin capacity via research until they have colonized two worlds and I think one has to stick with two worlds until they hit the end game repeatable tech otherwise the admin boosts won't show up.

 

[Methone]

Lastly, again for the poor hiveminds.
How about hiveworlds increase capacity in some way ?
This means that while hiveminds can increase their cap massively if they heavily invest (or else suffer the penalties even if reduced).

Hive Minds have many problems with them and Admin Capacity is not one of them.

How about how poor they are at making Amenities, how their equivalent of City Districts provide NO jobs at all, compare to Machine Nexus Districts which give both Amenity Jobs and Energy Jobs?
How about how much food and energy their Synapse Drones eat for so little in return?
How about how their new colonies have to fill FIVE base jobs - two of which are those Synapse Drones I mentioned - before they can make basic resources to fuel said Synapse Drones?
How about how any empire that unlocks robots instantly eclipses Hives in pop growth, despite Hives being billed as being the kings of pop growth?
How about their possessing of ONE SINGLE interesting civic, Devouring Swarm, and a whopping TWO AI Personalities?
How about their Hive Worlds being literally worse than Machine Worlds? Not to even mention their Hive Worlds are just hand-me-down Prethoryn Worlds; at least Machine Worlds were added in alongside the Contingency's hubs.

Or should different types of empires behave differently?

They should behave differently when I build them differently. I'd quietly raised an eyebrow at Megacorps being pushed to tall, but this is just getting ridiculous. Like you click Authority and then that entirely decides everything about your empire?

Hiveminds have ALWAYS been focused on expansion and "going wide"

Only since 2.2, and pretty poorly at that.

and the devs have made it clear they want Machine Intelligences to be highly efficient but have a smaller footprint.

While also having map-painting civics?! So are they deliberately acknowledging that Exterminator is now meant to be the worst out of all the genocidals?

"Hiveminds are for wide play and Machines are for tall" has pretty much always been the design plan

Since when?! I don't remember a peep about this before 2.2

 

[BlackUmbrellas]

They should behave differently when I build them differently.

A Spiritualist empire should play differently than a Materialist one, that's a choice you make when you build them... and so is picking a Gestalt empire. Playing a Hivemind is a specific choice in how you built your empire. Playing a Machine Intelligence is a specific choice in how you built your empire. They're specialized builds that should ABSOLUTELY have specialized mechanics and play styles attached to them.

 

[BlackUmbrellas]

Only since 2.2, and pretty poorly at that.

Nope! Hivemind AIs have always been expansion-based to the exclusion of basically everything else. They expand and expand and expand and only really get hostile when they RUN OUT of space to expand otherwise. That's a special behavioural tag they've always had.

"Hiveminds are expansive" has ALWAYS been the intention, even when they were just a special sort of Collectivist empire.

 

[Herr der Erlen]

Machine empires will get a +100% malus to empire sprawl now.

Ahhh, thanks ;-)
Yeah, finally a step forward!

 

[BlackUmbrellas]

While also having map-painting civics?! So are they deliberately acknowledging that Exterminator is now meant to be the worst out of all the genocidals?

Why not... you know... exterminate organics without holding onto their territory? Go to war, claim a bunch of space, and once it's purged decolonize the planets and deconstruct the starbases. Actually WIPE OUT organics, don't just take over their territory. Simple.

 

[Kanzhuu]

I feel like most aspects of empire sprawl would be better calculated on a logarithmic scale. That's especially true if they use pops in the formula.

Right now the linear scale means that adding a second star system and adding a hundredth star system are, for all intents and purposes, the same thing. It shouldn't work that way though. Integrating the handful of stars right next to your capital should be essentially a trivial task. Ruling over stars on the far reaches of your empire should be an order of magnitude harder.

At least, that's how I'd do it. I'd make empire sprawl a curve where the more you grow, the harder it is to keep growing

The thing with logarithms in complex formulas is, that I sit in front of the game and have no way of quickly figuring out what the exact consequences of my next step will be. Logarithmic scale in combination with pops means that immigration or unchecked growth may let my empire implode like the "Empire" of Alexander the Great. (Not sure it counts as an empire.) And it would also mean, that your unstable, multi-species empire may return to stability (or indeed maintain stability) through genocide. That... Seems very wrong. What is the implication? They were a race of saboteurs? Blame the Foxfolk and their interplanetary finance schemes? I shudder.

Genocide against your own population usually does more harm than good, and I say 'usually' only because there may be some obscure example somewhere, but I doubt it.

 

[BlackUmbrellas]

Not to even mention their Hive Worlds are just hand-me-down Prethoryn Worlds; at least Machine Worlds were added in alongside the Contingency's hubs.

Machine Worlds are just hand-me-down AI Worlds from the pre-2.0 AI Rebellion Crisis.

You're wrong about the history of this game an AWFUL LOT for someone who makes arguments based on its history so much.

 

[Methone]

A Spiritualist empire should play differently than a Materialist one

Yeah, and different Materialist empires should also be able to play different. I should be able to go Technocracy Life-Seeded uber-tall, I should be able to go Mechanist Nationalistic Zeal wide. I should be able to go Materialist-Free Haven-Agrarian Idyll diplomatic.

I should be able to do more than pick "Wide, but with +1 Leader level cap" or "Wide, but with +20% army damage" or "Wide, but everyone doesn't wait for you to attack them to start hating you".

That's a special behavioural tag they've always had.

I'll give you this one. But I still maintain that Machines being pushed to Tall is stupid. I mean, half of their playstyles are all about going wide, Servitor encourages you to keep finding new organics to pamper, and if we want to count non-playables, Paperclip Maximizers and Gray Goo/Tempests are always portrayed as expanding across the universe.

Why not... you know... exterminate organics without holding onto their territory? Go to war, claim a bunch of space, and once it's purged decolonize the planets and deconstruct the starbases. Actually WIPE OUT organics, don't just take over their territory. Simple.

Oh boy, my favorite, clicking Deconstruct on nine HUNDRED starbases! Nevermind that Exterminators come with a Total War casus beli that gives it to you, not an Extinction CB that wipes it out entirely.

You're wrong about the history of this game an AWFUL LOT for someone who makes arguments based on its history so much.

I'm not going to rise to this.

 

[methegrate]

They should behave differently when I build them differently. I'd quietly raised an eyebrow at Megacorps being pushed to tall, but this is just getting ridiculous. Like you click Authority and then that entirely decides everything about your empire?

Hard disagree. Restrictions make the game more interesting.

Yes, an empire should behave differently when you build it differently, but that's not the same thing as saying every archetype should just be a blank slate. Hive minds are psionically linked, expansion-oriented empires. If you want to build and play that as a tall, one-planet empire then by all means, but that's part of the challenge.

I like to compare it to classes in a roleplaying game. Yes, your mage should behave differently if you build it differently. If you want to build a fighter/mage you can. But you've still started with a spellcaster, and the restrictions and challenges that come with that class are what make it interesting. The game wouldn't be fun if you could pick locks, cast fireballs, wear heavy armor and heal all with equal skill. Same in Stellaris. You can take your empire in any direction you want, but it's not fun if empires just start as empty vessels.

 

[methegrate]

The thing with logarithms in complex formulas is, that I sit in front of the game and have no way of quickly figuring out what the exact consequences of my next step will be. Logarithmic scale in combination with pops means that immigration or unchecked growth may let my empire implode like the "Empire" of Alexander the Great. (Not sure it counts as an empire.) And it would also mean, that your unstable, multi-species empire may return to stability (or indeed maintain stability) through genocide. That... Seems very wrong. Genocide against your own population usually does more harm than good, and I say 'usually' only because there may be some obscure example somewhere, but I doubt it.

Ha! Fair enough. I like the idea of sprawl happening on a curve, not a line, but you're right. It might be too much complexity without that much actual reward.

 

[Methone]

Hard disagree. Restrictions make the game more interesting.

Yes, an empire should behave differently when you build it differently, but that's not the same thing as saying every archetype should just be a blank slate. Hive minds are psionically linked, expansion-oriented empires. If you want to build and play that as a tall, one-planet empire then by all means, but that's part of the challenge.

I like to compare it to classes in a roleplaying game. Yes, your mage should behave differently if you build it differently. If you want to build a fighter/mage you can. But you've still started with a spellcaster, and the restrictions and challenges that come with that class are what make it interesting. The game wouldn't be fun if you could pick locks, cast fireballs, wear heavy armor and heal all with equal skill. Same in Stellaris. You can take your empire in any direction you want, but it's not fun if empires just start as empty vessels.

This isn't about starting as empty vessels. It's not 'you can pick locks, cast fireballs, and wear heavy armor'.

It's "You are a mage. You can only do heavy single-target damage. You can pick if you want fire or frost. Don't ever get any funny ideas about crowd-control spells or Area of Effect spells! You picked mage, you agreed to heavy single-target damage and ONLY that!"

 

[BlackUmbrellas]

@Methone IMO the problem here is that you're assuming "Hiveminds get bonuses that make them good at going wide and Machine Intelligences get penalties that discourage them from going wide" means "Hiveminds can ONLY go wide, Machine Intelligences can ONLY go tall".

People already regularly ignore Admin Cap to go wide and munch the galaxy. I doubt that these upcoming changes will make it impossible to do such as a Machine Intelligence, nor make it impossible to play a Hivemind as tall. But Hiveminds and Machine Intelligences are, BY THEIR NATURE, specialized build options, and them having baked-in aptitudes is fine and makes for more interesting gameplay.