Logistic Growth is Unredeemable

[Mewnine]

I suppose the main reason why I haven't had a lot of trouble with the new s curve mechanic is because clerks have become so crappy, I've turned all their job slots off, and automigration is taking care of the unemployed pops well enough. The empire wide malus means whatever pop growth you can get, you should get.

Edit: Neither system works, but due to the horrible design of the empire malus, the s curve gets a pass for now, because it would be even worse without it.

 

[Less2]

Keep in mind, having dozens of planets capped at 25 pops past the early game isn't really the fault of logistic growth, but of the empire pop malus. The empire pop malus is what is nuking pop growth to the point where it becomes a challenge to get 25 pops per planet. Growth isn't being capped because the choice is between 40 pop planets and 25 pop planets, but because the choice is between 25 pop planets and 5 pop planets that will literally never reach usefulness due to the empire malus.

 

[The Founder]

Most of the discussion on the forum since 3.0 launched has been centered around the empire-wide pop growth malus.

Yes, it really is reaching the point where they should make a venting thread for this. Like they did with FTL. And Sectors.
Because nothing much is comming from these discussion other then "me no like, remove."

However unlike FTL and Sectors, this is something that can be removed via modding. If enough people agree, then you should have no issues finding multiplayer games with the same mod. So you should still be able to play the game you want, without these rules.

1) It can be easily cheesed by making sure that there is no more than ~25 jobs on any given planet (exact number varies slightly based on capacity), allowing players willing to micro to double organic growth more or less forever. Most of the other issues with logistic growth follow from this one.

2) It increases population counts dramatically. Logistic growth effectively doubles organic growth. Right now, the empire-wide malus "hides" this to an extent, but if that is removed or reworked (and given the extreme backlash against it, there's a good change it will be. If not, mods removing it already have tens of thousands of subscribers), it will make lategame 2.8 performance issues come back worse then ever.

5) The AI has no idea how to handle it. While players can get x2 organic growth forever, the AI has absolutely no idea how to do this, and just plays as if it didn't exist. The game is already full of powerful mechanics that the AI has no idea how to use. Adding another one, and one that is incredibly powerful, makes the single player experience much worse. The AI already struggles (to put it generously) to deal with the other major economic changes in 3.0 (industrial districts and the empire-wide malus in particular); adding another one is a bad idea.

6) It incentivizes bizarre micro/gameplay incredibly strongly. Getting such a massive boost for leaving all of your planets half-developed feels silly and requires quite a bit of micro (having to do a new math problem for each planet). It also makes prebuilding non-city districts a terrible idea.

Two simple fixes:
1. Capacity is the lesser of Jobs or current capacity
2. As long as there are places to migrate too, this growth reduction is fed into the migiraiton system. However it must be substracted even if there is no place to migrate too - that is why trying to solve it via migration always failed.

4) LG is poorly explained (not really explained at all) in-game

It is a Logistics Curve. Not even the name has been explained by the original inventor. Like, where is that "logistic"?
However it is a very natural thing. In fact, it was developed to model the natural growth. So it is not a "unnatural" concept as man claim.

 

[mial42]

Minor counter-point: Machines are way too good otherwise. Higher per-pop efficiency of machine jobs and lack of habitability penalties means that machines should have slower growth for the game to be somewhat balanced. I think in most previous patches whenever I checked out multiplayer games people often banned machines for being way ahead of the curve, a mechanic helping organic empires is needed (but perhaps not this particular mechanic).

I agree with your other points though.

In my experience, machines were quite a bit weaker then normal organics in 2.8 due to lacking ascensions, ecumenepoli, and slavery, meaning they couldn't eco-boom as well as organics mid-lategame. They were stronger early on thanks to better metallurgists and habitability (although migration treaties make this a near-nonfactor very quickly), but since they couldn't conquer efficiently*, they couldn't turn this into a lasting advantage. Compared to something like techno-slavers => synth ascension or militarist slaver conquest snowballing, they got left in the dust very quickly and couldn't catch up.

Regardless, IMO the ideal nerf to machines (if one is necessary) would be removing their replicators at the lvl 4 capitals, which has the added benefits of reducing potential micro and lategame pop growth.

*Except assimilators, which were probably the strongest non-glitched empire in 2.8 and are definitely the strongest in 3.0.

 

[mial42]

Well, I actually studied population models in university, and logistic growth is definitely used to model it.

4) Realism. The very premise of Stellaris involves FTL communications and travel; realism should not be a reason to add a whole new system to the foundation of the Stellaris gameplay loop: pops. Not to mention that it isn't realistic, since (a) the only industrial population we actually have any data on, humans, does not grow like this, and (b) LG applies to ecosystems as a whole, not small parts of them. In Stellaris, the "ecosystem" is the empire, not the planet, since moving from planet to planet is completely trivial and most resources are empire-wide. But I need to emphasize that even if it were realistic, that would not be a reason to add it, since Stellaris is not a reality simulator. Systems should be added because they improve the game in some way.

 

[Jaxck]

Well, I actually studied population models in university, and logistic growth is definitely used to model it. Besides you really can't control what I like/don't like, I find logistic growth fun, and I dare say most other players do too, at least from what I've seen on the forum.

Dude. Logistic growth applies to rabbits, it doesn't apply to people.

 

[beleester]

Logistic growth empire-wide based on empire-wide planetary capacity seems the best way to go.

It would solve the problems with logistic growth, but bring back the original problem of "there are too many pops and the player is incentivized to spam habitats to get even more pops." Spam habitats/ringworlds -> Empire-wide capacity goes up -> more pops.

 

[Less2]

It would solve the problems with logistic growth, but bring back the original problem of "there are too many pops and the player is incentivized to spam habitats to get even more pops." Spam habitats/ringworlds -> Empire-wide capacity goes up -> more pops.

1. That has nothing to do with logistic growth, only the empire cap. If logistic growth was removed you'd be spamming habitats/ringworlds again, they were worth it before at normal growth rates and will still be worth it if we reverted to normal growth rates.
2. It's entirely possible to do all of this RIGHT NOW, the only catch is that you release the habitats as a vassal and re-integrate them every 10 or so years. But you absolutely get huge pop growth if you build an entire system of housing habitats, release as a vassal, and then integrate them at an (incredibly cheap) cost. In fact you potentially get twice as much pops as before thanks to the logistic growth system.

 

[beleester]

1. That has nothing to do with logistic growth, only the empire cap. If logistic growth was removed you'd be spamming habitats/ringworlds again, they were worth it before at normal growth rates and will still be worth it if we reverted to normal growth rates.
2. It's entirely possible to do all of this RIGHT NOW, the only catch is that you release the habitats as a vassal and re-integrate them every 10 or so years. But you absolutely get huge pop growth if you build an entire system of housing habitats, release as a vassal, and then integrate them at an (incredibly cheap) cost. In fact you potentially get twice as much pops as before thanks to the logistic growth system.

My point is that if you want the population cap to actually reduce the number of pops, you shouldn't allow the player to control what the cap is.

 

[Less2]

My point is that if you want the population cap to actually reduce the number of pops, you shouldn't allow the player to control what the cap is.

In that case you'd just have to delete habitats and ringworlds. Even before we had logistic growth and planetary capacity, we had living space (planetary cap by another name) and linear growth. Players controlled the capacity of the empire before and increased it by spamming habitats.

 

[currylambchop]

@mial42 I can't for some reason reply to your previous message, but the premise is wrong. It takes 3 months for my ships to fly from the centre to the edge of the solar system. So clearly there is no instantaneous travel between planets.

 

[mial42]

@mial42 I can't for some reason reply to your previous message, but the premise is wrong. It takes 3 months for my ships to fly from the centre to the edge of the solar system. So clearly there is no instantaneous travel between planets.

Instantaneous =/= trivial. It is not a major undertaking for pops or resources to travel between planets in Stellaris; look at how quick and easy resettlement (and auto-resettlement) are, or how planetside production can be instantly used for the entire empire.

 

[Oscot]

Dude. Logistic growth applies to rabbits, it doesn't apply to people.

Exactly which textbook are you getting your data for 23rd century interstellar population dynamics from?

We really should ban "realism" discussion here. Not so much because I think it's inherently pointless, but because 99% of posters think "reality" means "my extremely unsupported prejudices".
(Also, it is inherently pointless, for the reasons @mial42 mentions)

 

[Tisifoni12]

Hang on, logistic growth effects population ?
Just population or everything ?
If just population then why isn't it referred to as population growth ?
I assumed logistic growth applied to mining and industrial production . . .

 

[Gobla]

It is a Logistics Curve. Not even the name has been explained by the original inventor. Like, where is that "logistic"?
However it is a very natural thing. In fact, it was developed to model the natural growth. So it is not a "unnatural" concept as man claim.

Given just how constrained the Stellaris curve is, especially in 3.0.3, you really can't call it a logistic curve.

It's an asymmetrical plateau. Starting at 3, sharply rising to 4.5 and later sharply dropping to 0.3.

If you plot it there's literally nothing logistic about it. Just because there's a logistic curve buried in there somewhere doesn't mean the end result can still be called that.

 

[The Founder]

Dude. Logistic growth applies to rabbits, it doesn't apply to people.

Sure it does.
It applies to bacteria.
Rabbits.
Human colonists.
Humans living in a fully build up city.
And everything before, after and in-between.

Given just how constrained the Stellaris curve is, especially in 3.0.3, you really can't call it a logistic curve.

It's an asymmetrical plateau. Starting at 3, sharply rising to 4.5 and later sharply dropping to 0.3.

If you plot it there's literally nothing logistic about it. Just because there's a logistic curve buried in there somewhere doesn't mean the end result can still be called that.

So it a first goes up, then goes down until it Plateaus?
Because that is the literaly definition of a logistics curve.
Not looking like the contrived textbook examples does not disqualify it from being what it is.

 

[Gobla]

So it a first goes up, then goes down until it Plateaus?
Because that is the literaly definition of a logistics curve.
Not looking like the contrived textbook examples does not disqualify it from being what it is.

A logistic curve never plateaus. It approaches but never becomes a flat line.
A logistic curve has symmetry.

The Stellaris growth curve literally meets 0 of the mathematical properties of a logistic curve.

 

[The Founder]

A logistic curve never plateaus. It approaches but never becomes a flat line.
A logistic curve has symmetry.

The Stellaris growth curve literally meets 0 of the mathematical properties of a logistic curve.

Please provide a source for these definitions of a Logistics Curve.

 

[Gobla]

Please provide a source for these definitions of a Logistics Curve.

Logistic function - Wikipedia

en.m.wikipedia.org

Logistic curve | statistics

Other articles where logistic curve is discussed: population ecology: Logistic population growth: …population growth known as the logistic curve. It is determined by the equation

www.britannica.com

Logistic Equation -- from Wolfram MathWorld

The logistic equation (sometimes called the Verhulst model or logistic growth curve) is a model of population growth first published by Pierre Verhulst (1845, 1847). The model is continuous in time, but a modification of the continuous equation to a discrete quadratic recurrence equation known...

mathworld.wolfram.com

Keywords being that the curve is smooth, the Stellaris one is not, it abruptly changes it's derivative to 0 when it reaches the ceiling.

The presence of a logistic equation inside a function does not automatically make the function logistic. Just like the presence of an exponential equation does not automatically make a function exponential.

 

[The Founder]

Logistic function - Wikipedia

en.m.wikipedia.org

Logistic curve | statistics

Other articles where logistic curve is discussed: population ecology: Logistic population growth: …population growth known as the logistic curve. It is determined by the equation

www.britannica.com

Logistic Equation -- from Wolfram MathWorld

The logistic equation (sometimes called the Verhulst model or logistic growth curve) is a model of population growth first published by Pierre Verhulst (1845, 1847). The model is continuous in time, but a modification of the continuous equation to a discrete quadratic recurrence equation known...

mathworld.wolfram.com

Keywords being that the curve is smooth, the Stellaris one is not, it abruptly changes it's derivative to 0 when it reaches the ceiling.

The presence of a logistic equation inside a function does not automatically make the function logistic. Just like the presence of an exponential equation does not automatically make a function exponential.

The only times those sources mention the term "Smooth" was as a optional property.
The 2nd one only mentioned it for Exponential growth, making it very specifically not a Logistics Curve property.

I am also not sure what hte scientific definiton of "smooth curve" would even look like. Sounds like a human evaluation, not something we use in science.