"Empire Sprawl Penalty" Needs to be reworked

[Hansatron]

Now we have ways to boost admin cap, but this has in effect turned the scaling penalty into what is basically a % tax to everyone. This very much favors high sprawl empires, since everyone is paying nearly the same % to not have penalties so more sprawl means more resources.

Having a certain % of the penalty always apply and/or limiting admin cap generation is needed for the penalty to be meaningful again.

It would be interesting if they could somehow take the federation laws system and construct an internal governance system that had elements like "centralization" and fleet contribution - so, for example, spread out empires would need to have laws to keep things running efficiently that would effectively increase the "tax" they are paying in administrative overhead.
Perhaps tied into a rework of sectors. But ultimately the idea would be a compact "merchant republic" might be running laws that allow it to be very efficient, while the "sprawling empire" has twice the size but is spending a lot of that margin on extra overhead to keep things together. It would be a good way to work in these "empire sprawl modifier" concepts and give a segue to stuff around actual internal politics, and makes room for feudal empires with vassal to make a return.

 

[Jarolleon]

The sprawl penalty, or its prior iterations, exist to limit snowballing. It basically makes the slope that snowball is rolling down less steep. This is a common thing in 4x games and shows up in mechanics like corruption and the like.

The very early version of this was around from the beginning of the game IIRC but was not very visible and was heavily weighted towards colonies and systems owned, had not mitigation, and was % based. This made small worlds and low quality systems unattractive since you could easily be at a net negative tech for expanding. Going wide got you resources but smaller entities could keep up tech wise.

The biggest example of the weakness of the early system was the one planet strategy, originally started as a meme build but turned out to be very powerful. By having a minimum of colonies and systems, you could basically break the tech curve by having lots of output and tiny penalties, capped off with science megastructure and repeatable 100-150 years into the game.

Megacorp made the penalty more visible while also taking most of the bite out of it, but people started to freak out over having a penalty.

Now we have ways to boost admin cap, but this has in effect turned the scaling penalty into what is basically a % tax to everyone. This very much favors high sprawl empires, since everyone is paying nearly the same % to not have penalties so more sprawl means more resources.

Having a certain % of the penalty always apply and/or limiting admin cap generation is needed for the penalty to be meaningful again.

You could buff the cap increases from tech and nerf the bureaucrats. The bureaucrats would then represent administrative overhead (which they kind of do already, they're just too competent for it to be significant), & the bigger you get the greater proportion of your population is consumed by the expanding bureaucracy. I can see why Paradox moved this way, with the Federation rework the old system where admin cap was only increased by tech, a single civic and an ascension perk would have severely gimped any non-federated playthrough.

 

[Archon87]

I personally think that the thing that finally broke sprawl as a useful mechanic with a real purpose (instead of just another resource to manage) was the introduction of bureaucrats. Being able to spam jobs that increase the cap make the entire point of sprawl meaningless. It was supposed to be a rubber band mechanic to prevent rapid expansion before you had the infrastructure to handle it. In previous versions of the game, you could definitely get around the penalty, but only by pumping out more research and unity to overcome it. Bureaucrats just bypass it entirely, while the old strategy of pushing out as much research (and to a lesser extent unity) can still be done simultaneously. This leads to ridiculous situations where you snowball into an unstoppable force within the first 100 years, with research into advanced techs only taking a year max. And this is doable without even really playing a build that specializes in either research or bureaucracy. Sprawl needs a balance pass, badly.

My personal fix for bureaucrats would be to limit the construction of new admin buildings to one per planet, and do away with bureaucrat worlds. This allows wide empires to address the admin cap in a limited fashion, but they will still be likely to go over it.

 

[leftkeh]

My personal fix for bureaucrats would be to limit the construction of new admin buildings to one per planet, and do away with bureaucrat worlds. This allows wide empires to address the admin cap in a limited fashion, but they will still be likely to go over it.

Getting rid of the planetary designation and the leader trait could IMHO be removed (at least at a beta stage), if only to see its effectiveness. On the other hand, limiting the bureaucracy building to 1 per planet would probably force wide empires to build one on every planet, just as they do with the robot building.

 

[Kahldris]

I agree that empire sprawl and capacity could use some tweaks. I always thought the bigger you get that the.more sprawl you get from everything, this making it harder to hold together your empire the larger it gets. The increase should be very small though. And of course you still need the proper tools to ensure you can have a large empire if that's what you really want. As if right now it is too easy to stay above the cap.

Honestly how it used to work was better for the game. I cant believe how fast I get tech and unity now. Even with empires that aren't supposed to be good at that stuff. That's not good for gameplay.

 

[Franton]

All empires have the same set of +5% admin cap techs. All empires have other bonuses like a +10% civic, etc. Since these are all uniform for everyone, I Ignored them.
Only gestalts get the 8 job upgrade; and only machines have a tradition that gives their coordinators +3 admin cap generation, which applies before any of the +% effects.
Even if only filthy organics have bureaucrat governors and bureaucracy worlds (can't recall ATM) that would still leave them at 50->60 admin cap from 5 jobs (plus all the other bonuses everyone can get) with machines still pulling 120 to 144 from one building.

I checked in my DA empire, and they dont have b'crat planet designations. The basic maintenance drones add 15 cap each, increased to 18 with the 'Integrated Preservation' tradition, for 90 total with an upgraded building with 5 jobs. I haven't seen a governor with b'crat trait either, not sure they exist for MEs, although it might be possible for a cyborg governor to have it, provided you allow your 'borgs to become leaders. Assuming they can have such governors, and including the +5% tech that I forgot about, we're talking about 103.5 cap per building. I have no idea how you get to 144. Organics can get to ~14 as I've seen in my game, although it looks like I'm missing a bonus in the calculation: governor bonus + planet designation + tech only adds aup to 35% bonus on a 10 cap base, but I've seen at least 70 per building!

I haven't considered civics, as I don't consider them to be worth it: it's so easy to handle sprawl anyway.

 

[Franton]

Because of how sprawl penalty scaling works (an empire wide % penalty based on the absolute overage) there is hard math at play. If we just consider research,, 1 bureaucrat = 10 points of cap = +4% tech costs. So an extra bureaucrat isn't worthwhile until you have 25 researchers. (Ignoring unity etc for a second, and pretending we can either employ researchers or bureaucrats)
But as you grow, and you numbers of researchers becomes large (say, 200) then the value of 1 bureaucrat to mitigate being over cap - still +4% research rate - is the same but one researcher only increases your tech rate 0.5%.
Thus, if you played optimally and plotted the overage, you would see a slowly narrowing function where the more pops you have, the less and less you allow yourself to get over before employing a bureaucrat. Eventually, at 250 researchers, just one point of overage (+0.4%) equals the value of a researcher (+0.4%), so it is better to never go over cap at this point. The real point where this happen would be lower because obviously you have other penalties, but it works as a simple example to show why there are strict limits on how much you can "ignore cap" without severely suffering.

Indeed, once you get past a certain amount of research, you gain more research by making sure to stay under cap than adding a researcher while over cap.

However, the way the math is handled means that the effect of 1 point of sprawl gets less the more you are over cap! This means there are two ways to handle sprawl: either 100% ignore it, or 100% stay under cap. There is no middle ground! I haven't done the math at which point the former strategy actually gives you better results than the latter, but I suspect that ignoring sprawl entirely always gives you a worse result - if only because at some point the cost of traditions, leaders and edicts become too painful.

 

[KubiG37]

I think the whole Empire Sprawl/"admin cap" mechanic has become an useless annoyance and lost its purpose completely.

It's severely punishing (especially for machines), yet so easily attainable that it is now nothing more than an "annoying thing you have to do after conquering other planets", something of an automatic chore, like clicking on "create core" in EU4. Sometimes I conquer a planet and gain +180 % tech penalty. I build an Uplink and boom, it drops to 0 again.

Honestly I have no idea why did they rework it like that, the old mechanic worked in a sense that it was punishing you for blobbing uncontrollably, and you couldn't do much about it, so no micromanaging, and the maluses weren't that bad. It completely killed any kind of "tall" gameplay, or Tech-oriented vs Blobbing, quantity oriented gameplay. The bigger your empire is, the faster you can research (and blob).

Playing assimilator Machine empire - Only thing stopping me from blobbing is my own will to not blob. I can just max. research and even with 3x Tech cost setting I research a new tech every 6 months. AIs are still using deflectors and chemical thrusters, while I think about researching Zero-point reactor, it's really dumb.

 

[Altruist]

The problem with sprawl is that its intention has become unclear.

Initially there was a rather clear role: wide/tall game strats, sprawl was a malus to wide and giving tall builds a chance.
Also slowing the good players (or experienced players) a bit down to give the AI a chance (for this role sprawl should had been, btw, only for players and not for the AI).

With the introduction of bureaucrats, as other posters have already mentioned, admin has become another resource to handle and it can be handled.

Results:

• Tall builds have vanished (or are rather useless for anything other than roleplay or when you are unlucky enough to get boxed in for long).
• Admin/sprawl has become a rather superfluous part because handling it has become a necessety (very much like consumer goods). Strategy games are about choices. If you havent got to make a choice but something just needs to be done, it is making a game only artificially bigger/complicated. Admin/sprawl and consumer gooods both fall into this category. As a counter example: Research and alloy production, while they are surely necessary, pose neverthless a true choice for the player to what extent you allocate your resources to the one or the other.

 

[Hansatron]

90 total with an upgraded building with 5 jobs.

Machines and hives both have an 8 job upgrade, the "system conflux" and "confluence of thought." The Regular empires originally had one that, according to the devs, was unintentionally left in the game upon release of 2.6 and they ended up created a new type of code to properly disable it. (I still think it was a coding error!)
8*18 = 144.

 

[Secret Master]

I've taken to just not worrying about it so much; I'll occasionally spend some time "catching up" to my cap, but largely I just let my automated administration worlds build up at whatever pace they choose to. Especially with the way Edicts work now, going over admin cap by running an extra edict is perfectly reasonable.

That's kind of how I deal with it.

Since I can't see any impact from sprawl on alloy output, I figure it's not that important in the early and mid game.

Hell, my first mega-corp with the release of the current DLC, I was just going over the limit willy nilly without even thinking about it. It wasn't like I couldn't afford edicts or ships or branch offices.

 

[Franton]

Machines and hives both have an 8 job upgrade, the "system conflux" and "confluence of thought." The Regular empires originally had one that, according to the devs, was unintentionally left in the game upon release of 2.6 and they ended up created a new type of code to properly disable it. (I still think it was a coding error!)
8*18 = 144.

Thanks for the clarification. My ME hasn't made it past strategic resource techs yet because I keep getting attacked by overwhelming neighbours and I just spent 20+ hours game time on an extended war to eliminate two of them while the third was eating 2/3 of my territory in the opposite direction ( I honestly don't know what this smilie is supposed to express - for me it's meant to express the utter exhaustion it was causing me after so many hours of risky and complex strategic moves to lure the way stronger enemy fleets into the wrong direction while my own took their inhabited systems - thankfully the AI is really bad in managing transports and never managed to take back those systems)

I assumed 5 jobs because that was what that posting that I referred to explicitely said. P.S.: I just noticed you said 5 jobs for the other empires, but "one building" for ME. Apparently I missed that subtle differennce.

 

[Janx14]

I was just pointing out that there seems to have been a conscious thought to give machine, hives, and regular empires different levels of efficacy on making admin cap because of their sprawl penalties. But because we can fully mitigate sprawl, we see that it plays out in the reverse, with machines having the cheapest admin cap generation.

PDX continues to make the same mistakes when it comes to hiveminds and machine empires. I feared this would be the case with 2.6 and it seems its exactly what happened.

In 2.2 PDX said hiveminds were to be 'high population, lower efficiency" empires, and that machines were to be the opposite. So they gave hiveminds good early game growth....but because they are biological and stuck to 1 habitability without genemodding, and can't even make robots they aren't terribly great at expansion. Meanwhile, machines can colonize everything and had many ways to eclipse biological pop growth until recently when it was brought more on par....but machines can still colonize everything.

And again, they give hiveminds superficial advantages that look great on paper, but then make what they're good at meaningless....and heck, make machines the most efficient at making it meaningless, once again, making hives inferior at the very thing they're suppose to excel.

 

[DerMoe]

Hi together,

Well catching up the sprawl ist not easy in the early game, because you need to decide whether you should build sprawl buildings or science and alloys. There is a certain limit of pops working in specific job before you actually need to take care about sprawl in the early game (I can recommend the videos of Stefan Annon).

As soon as you can dedicate a planet to sprawl it gets too easy. The only thing is, it has the characteristic of growing in the background, if you do not take care. Since 2.6. you can research so fast with sprawl at 1/1, that you get through tech after 125-140 years. It is the same every time nevertheless which empire you pick. Wars and less planets can only delay the trend a little. The current state makes the late game more about picking repeatables. I like sprawl in general, since you needed to decide what tech you can take, which makes every game/empire a little more unique. This idea is kind of lost since 2.6. Tech just got to cheap.

Therefor I suggest an idea:
First I would remove the dedication of sprawl planets and go back to the old penalty system of sprawl. I think it was growing constantly and the penalty was linear proportional. This would make tech and other stuff more expensive again. I would not remove the sprawl building in general, but change the way they need to be build. I would add them to certain situations, which increases sprawl additionally to the old mechanics. Then the placement of sprawl buildings based on special situatiolns and certain planets or sectors actually need more bureaucrats. These extra sprawl numbers then grow unproportional and are added to the main linear growth.

Here are some ideas for these situations for regular empires:
1. If a sector grows unproportional to its area in space vs. planets and pops you need to build sprawl buildings. It could be beneficial to build them on the sector capital.
2. You could measure the distance (number of jumps not direct) to the main capital of each non core sector capital and add sprawl penalties, which you can reduce by building bureaucrats in this sector.
3. Planets with high amount of pops (>80) need an additional building.

On hive minds, the additional penalty could be based on keeping the empire together.
1. If e.g. a planet is too far away from the capital or the next planet with low local sprawl, it causes an extra amount of sprawl. This way you can connect the outer perimeters via synapse drones or coordinators to the inner hive core.
2. The sprawl on high pop count planets could be used here, too.

Well introducing the mechanics could be performed by different planet modifier levels or a number which is planet based as e.g. trade, the amount of each planet is added together and added to the basic sprawl.

Cheers, Moe

 

[Archael90]

Currently, 3 authorities have this modifier:
Hive Minds: -25%
Megacorps: +50%
Machine Empires: +100%

However, with the bureaucrat system, not empire needs to ever go over their admin cap regardless of size. Producing admin cap is cheap, so everyone is very close to their cap at all times, making these 3 modifiers that are intended to shape empires, wholly useless.

It can likely be seen that Machine and Hive admin cap jobs reflect these intentions - machines have a high output coordinator and high sprawl penalty, hives have the reverse. But because the penalty is never relevant, hives end up with the most expensive admin cap and machines get by far the most efficient admin cap job / building, producing 144 cap from a single building late game. (Hives,meanwhile, can push out 40, while despicable regular empires push 50 out of 5 jobs in 1 building.)

I don't know what to do about this, but it seems like there's a few pieces of the puzzle still floating around to get this just right.
(Also, megacorps don't need the penalty anymore: just directly increase the sprawl from branch offices if you want to be concerned about that.)

IIRC devs sais somewhere that they are just testing new empire sprawl system, and it will be adjusted. At the moment penalties are not huge, and dealing with Empire Sprawl is not hard either, but it is going to change.
I dont remember where i red this, and not sure i understood it :x

PDX continues to make the same mistakes when it comes to hiveminds and machine empires. I feared this would be the case with 2.6 and it seems its exactly what happened.

In 2.2 PDX said hiveminds were to be 'high population, lower efficiency" empires, and that machines were to be the opposite. So they gave hiveminds good early game growth....but because they are biological and stuck to 1 habitability without genemodding, and can't even make robots they aren't terribly great at expansion. Meanwhile, machines can colonize everything and had many ways to eclipse biological pop growth until recently when it was brought more on par....but machines can still colonize everything.

And again, they give hiveminds superficial advantages that look great on paper, but then make what they're good at meaningless....and heck, make machines the most efficient at making it meaningless, once again, making hives inferior at the very thing they're suppose to excel.

So meybe they should change it - hives concentrate on efficiency and quality, and machines on number, and sprawl :V

 

[Khalael1155]

What if Admin Capacity works both ways. like "stability", with penalties and bonuses. You can work the math to make it more suitable for tall empires.

 

[Franton]

I think the key problems with sprawl and cap are:

1. The effects directly scale with research and unity output, while the counters hardly scale at all. This leads to the problem that at some point there is really no (economically sensible) alternative to always keep your cap above your sprawl.

2. The penalties have a very notable effect when you're just a few points over cap, but hardly noticable when you are a 1000 points over*. I feel it should be the other way round: make it so that going a few points over doesn't make a difference, but going 1000 points over will stop your research/unity generation dead in it's tracks.

3. Admin cap is a limitless resource. You just produce as much as you need to always stay within cap. It's a no-brainer. You could get the exact same effect by simply reducing all economic output by ~10%. I. e., the current mechanic has no effect on limiting empire size, because cap isn't limited.

The second point could be fixed with a slightly different formula. (provided you agree it needs fixing) The first as wll, but that will be trickier. And tbh I'm not sure in what direction a change should go.

The last is easy to fix: make admin buildings 1/planet, and make sure that each race (standard/hive/ME) can get their cap to a reasonable level with just that 1 building. (i. e. adjust #jobs and cap/job to arrive at a reasonable maximum)


* P.S.: I should elaborate: what I'm referring to is the effect of going one more point over cap than you already are - I didn't intend to state that the penalty for 1000 sprawl over cap is somehow less - that would be silly

 

[Ferrus Animus]

The last is easy to fix: make admin buildings 1/planet, and make sure that each race (standard/hive/ME) can get their cap to a reasonable level with just that 1 building. (i. e. adjust #jobs and cap/job to arrive at a reasonable maximum)

The problem is that a maximum would limit empire size beyond which certain elements would simply become too tedious to progress. There's plenty of gamplay styles (like exterminators) that are all about map painting, not to mention as a 4X that is one of the play styles people enjoy.

The issue is more which costs should scale with empire size and to what degree? Currently admin cap is a productivity tax for sprawling empires. which is the point. It slows the benefits of expansion and requires some infrastructure investments. The effects of harsher limits would be to either slow viable expansion even more (and the game is slow enough already) or making expansion a chore past a certain point (And the game has already e4nough aspects that feel like a chore).

 

[Franton]

The problem is that a maximum would limit empire size beyond which certain elements would simply become too tedious to progress. There's plenty of gamplay styles (like exterminators) that are all about map painting, not to mention as a 4X that is one of the play styles people enjoy.

The issue is more which costs should scale with empire size and to what degree? Currently admin cap is a productivity tax for sprawling empires. which is the point. It slows the benefits of expansion and requires some infrastructure investments. The effects of harsher limits would be to either slow viable expansion even more (and the game is slow enough already) or making expansion a chore past a certain point (And the game has already e4nough aspects that feel like a chore).

I do agree. The fact that sprawl currently affects research and tradition cost, and the odd way how this affects research speed is part of the reason of the second point I made. If they would affect research speed directly (e. g. -0.1% per point of sprawl over cap), then you could easily go 10 points over cap without noticing it too badly, but going 1000 points over cap would be a no-go. However, all that would mean nothing as long as it's trivial to stay under cap. Therefore we need a limit to b'crats.

Applying the effect to something else, such as pop growth could work as well. I actually like that suggestion. But, again, we need to limit the amount of cap, and tying it to the number of planets seems a reasonable proposition to me.

P.S.: limit per planets means that it is possible to get higher with technology: habitats and ring worlds. But if that is not desired, the buildings could be limited to natural planets instead. (or just don't allow them on habitats. Ring Worlds are so late in the game, it would be too restrictive)

 

[KubiG37]

Perhaps adding some mechanic like autonomy, together with making the sector AI less useless?

1. Empire Sprawl is still a soft cap that has slowly raising penalties for crossing it
2. Instead of the chore of building bureaucrats and effectively removing any penalties, make it possible to reduce your empire sprawl by giving planets (or preferably automatic sectors) autonomy - less empire sprawl, no penalties, but % of energy/or other raw resourcesof that planet/sector is lost (or kept locked in that sector)

You now make a choice between an 'absolutist' controlling everething (somewhat inefficiently), or delegating something to the planets themselves, at the cost of having less direct control/resources. This system could then be greatly expanded upon to create a small AI controlled 'states within a state', a loose federation of planets, if you will. Empire governments could have various effect on it as well.