Cost Reduction Modifiers are sort of broken.

[Hellinfernel]

I wanna talk a bit about some of the more broken stuff in Stellaris in general (not just 4.0), and i thought Cost Reduction Modifiers are a good place to start.
For starters, "cost reduction" in this instance means basically anything that reduces upkeep or buying costs of anything. That might be ship upkeep or pop consumption or building cost or whatever. Those kinds of modifiers have a bit of history to lead to incredibly broken builds, like the -90% Ship build cost builds. Personally, I never went for this kind of stuff because while I like experimenting with stuff a lot, I usually dislike it if my game becomes completely broken. The thing here is, while those modifiers tend to be very modest on their own, if stacked upon eachother, they become very overpowered, because the value of each additional percent becomes higher and higher. It basically scales with itself. To give you an example:

Lets say, you have 1000 alloys and a corvet that costs 100 alloys (for the sake of argument). This means you can build 10 ships. Now you get an -25% cost reduction modifier.

1000 alloys/75 ship cost = 13.333 ships. Meaning this modifier is 3.333 ships worth.
Now let's say you get that modifier again, stacking additively.

1000 alloys/ 50 ship cost = 20 Ships. Meaning this second modifier is 6.666 ships worth.

Now let's repeat this process.

1000 alloys / 25 ship cost = 40 Ships. Meaning, this third modifier is 20 ships worth.

As you can see, the more you get of it, the more valuable it becomes, and it has potential to spiral very hard out of control. Thankfully, there is an easy fix for this.

In League of Legends, there was a stat called Cooldown Reduction, reducing the cooldown of your spells. It had this issue of being weak on its own, strong when stacked up, and so it had to be capped at 40 Percent. This in itself caused other issues, and later this whole stat got replaced with Ability haste. The way this stat is calculated goes as follows:

Effective cooldown = Original cooldown * 100 / (100 + Ability Haste)

If we apply this now on the ships, it would work something like that. Lets take the example from above and replace 25 percent cost reduction with... idk, "cost reduction points".

Effective ship cost = 100 * 100 / 125 = 80. 1000 Alloys / 80 Ship cost = 12.5 ships.

Now again for 50 and 75...

Effective ship cost = 100 * 100 / 150 = 66.7. 1000 Alloys / 66.7 Ship cost = 15 ships.

Effective ship cost = 100 * 100 / 175 = 57.14. 1000 Alloys / 57.14 Ship cost = 17.5 ships.

A voilà! Every point of this is now worth exactly the same and there are no possibilities more to stack those kinds of values together, so it isnt able to break the game.

The reason why this would be a neat addition to stellaris (and honestly also some other Paradox titles) is because of Modifier Creep, which is basically this games version of Power creep. The developers want to incentivize to play with the new content, and those almost always contain at least to a degree new modifiers. However, if certain modifiers cannot be added without breaking the game, then this limits the developers ability to give those to the player. This simple formula change would eliminate this danger entirely, and it also would get rid of some of the most broken combos in the game, of which I am honestly tired.

 

[Egodeus]

Yep, making sure that modifiers are multiplicative instead of additive is important. Additive multipliers in the negative are easily broken once you stack them enough, while additive modifiers in the positive are good in the beginning, but near meaningless by themselves once you have enough of them.

The point would be to do what you mentioned for the negative modifiers (or modifiers cheapening something) and do a similar multiplication to positive modifiers. After that you could do a generic balance pass on all positive modifiers as currently they are everywhere, and they are large, just the techs each give you 20% which is craxy, but now if you get a +5% modifier on energy production after the early game it feels like nothing, since you already have at least +100% modifiers on energy production, so it's effectively a 2,5% modifier. I'm currently running a +126 percent of modifiers to energy production (mind you, with just the first of the energy production techs) so getting the improved solar energy from the anomaly of 5% percent feels meh, because it barely changed my production, as I went from 2,26 times production, to a whopping 2,31 times production on energy, an actual increase of 2,2%. Not nothing, but close to it anyway. And as I get more techs and modifiers, the change by the improved solar energy is going to become even less significant.

But if we moved to a multiplicative bonus system for positive modifiers, a +2,5% modifier would always mean you get +2,5 percent more of the thing. One of the reasons the species production modifiers had been becoming less useful was this modifier creep. Now that they modify the job efficiency, meaning you have more effective workforce for the same job, they've become more useful again, but that only lasts until we get the same amount of power creep there. Because currently we have two sets of modifiers for the same thing that are multiplicative with everything from the other set, but not within the set. All modifiers for workforce efficiency are additive with each other, but multiplicative on the "X production" set, and vice versa. If you had a modifier for +50% job efficiency, and another of +50% production, the end result would be 2,25 times production, while if you had the same level of modifiers, but only in one category, you would just get 2 times production.

If you ask me, this makes no sense.

And yes, I know I got carried away, but modifier creep is a problem in this game. It is clearly evident once you get to add too many -% modifiers, which you commented on, but it is also a problem on the positive side, and I think both should be fixed. There have been threads about the issue many times in the past already, but I think your solution for the negative side is elegant and workable, while simple multiplicative change on the positive side would also make sense, along with a balance pass on the numbers, as otherwise the result would be nonsensical numbers.

 

[Muldeh]

I think OPs solution to cost reduction stacking is elegant, and wuld add space for stacking above 100% cost reduction but still getting benefits.

It isn't multiplicative like you suggest Egodeus, multiplicative would still lead to runaway scaling.. it's still very much additive. For example with 1000 alloys and a bse ship cost of 100, each 25% ship cost reduction with OPs solution would allow you to build 2.5 extra ships. While multiplicative nereduction would allow you to build 25% MORE ships each time.

Making all positive modifiers multiplicative runs into a big balance problem, where the difference between a sub optimal build and an optimal one will feel much greater than it currently does. To give an example, in a world where every modifier is 10% , a build with 20 of these will be over 5x as powerful as abuild with 10 if they are multiplicative, but is only 50% more powerful if they are additive as they are currently.

 

[Abdulijubjub]

A few thoughts:

• Cost reduction on something that isn't infinitely spammable isn't a big deal. It can go to zero and be fine.
  • ex. Mineral upkeep reduction for metallugists getting to 0 (or near 0) just moves a few pops from miner jobs into metallurgist jobs. It's just an output buff wearing a funny hat: 25 miners supporting 50 metallurgists with a +50% output buff make the same alloys as 75 metallurgists with -100% upkeep, and they both use the same pops. The pops (or job slots) are finite, so the upkeep reduction doesn't matter.
  • Also: Eternal Vigilance gives you free Defense Platforms (-100% cost). It's fun, but not game breaking, because there are a finite number of defense platform slots (and rapidly escalating starbase upkeep keeps you from scaling it).
• Cost reduction on things that are infinitely spammable have to be extremely tightly controlled to never go to 100%, and preferably never beyond 50% (where each additional bonus is functionally twice as powerful as it was to start).
  • ex. ship cost reductions could go to zero (or -90% after being capped) immediately after the release of Paragons. It was absolutely absurd: nothing could compete with building 10x as many ships for the same price.
• Cost reductions on things that technically aren't spammable, but can be heavily invested into, need a very close look.
  • ex. empire size reductions on pops increase/decrease your empire size by the same amount, if the number of pops is fixed. But reducing empire to near zero removes the rubber band that incentivized inward investment over mindless outward expansion through conquest (and also makes a few things like planetary ascension become spammable, making it worse).

So, some extremely problematic cost reductions now in Stellaris:

• Multiple ways to bring empire size to basically zero (including Cloning Ascension bringing you to -90% even without Sovereign Guardianship, which lets you bring systems/planets near zero too).
• Finding a ruined Mega-Art can bring you to -80% planetary ascension cost (Harmony, Covenant, Ascensionists, 2x Mega-Art). If you can cheese a vassal into making a third for you, you can hit -100%.

Can't think of any more at the moment.

You can make cost/upkeep reductions be multiplicative, but then you get into a funky case when they're in the same category as cost/upkeep increases. It becomes extremely opaque to the player what they're actually getting from various bonuses. Though you can solve that by just actually putting them in different categories. You can also make things more readable by swapping e.g. -20% to 0.8x (to make it clear that it's multiplicative with other things).

 

[A Cat]

Centena grants a -50% planetary ascension cost reduction.
Fun fact: Planetary Ascension costs can go negative.

 

[Abdulijubjub]

I think it's ok for things to be broken after you've already beaten the game.

 

[Twogs]

It isn't multiplicative like you suggest Egodeus, multiplicative would still lead to runaway scaling.. it's still very much additive. For example with 1000 alloys and a bse ship cost of 100, each 25% ship cost reduction with OPs solution would allow you to build 2.5 extra ships. While multiplicative nereduction would allow you to build 25% MORE ships each time.

A multiplicative system still keeps it under control though and has a fixed scaling instead of an escalating one like the additive we currently have has.

PS: Would be 33.33333% more ships each time

 

[Muldeh]

OH right, of course. Good catch.

 

[Critical Ethics]

A few thoughts:

• Cost reduction on something that isn't infinitely spammable isn't a big deal. It can go to zero and be fine.
  • ex. Mineral upkeep reduction for metallugists getting to 0 (or near 0) just moves a few pops from miner jobs into metallurgist jobs. It's just an output buff wearing a funny hat: 25 miners supporting 50 metallurgists with a +50% output buff make the same alloys as 75 metallurgists with -100% upkeep, and they both use the same pops. The pops (or job slots) are finite, so the upkeep reduction doesn't matter.
  • Also: Eternal Vigilance gives you free Defense Platforms (-100% cost). It's fun, but not game breaking, because there are a finite number of defense platform slots (and rapidly escalating starbase upkeep keeps you from scaling it).
• Cost reduction on things that are infinitely spammable have to be extremely tightly controlled to never go to 100%, and preferably never beyond 50% (where each additional bonus is functionally twice as powerful as it was to start).
  • ex. ship cost reductions could go to zero (or -90% after being capped) immediately after the release of Paragons. It was absolutely absurd: nothing could compete with building 10x as many ships for the same price.
• Cost reductions on things that technically aren't spammable, but can be heavily invested into, need a very close look.
  • ex. empire size reductions on pops increase/decrease your empire size by the same amount, if the number of pops is fixed. But reducing empire to near zero removes the rubber band that incentivized inward investment over mindless outward expansion through conquest (and also makes a few things like planetary ascension become spammable, making it worse).

So, some extremely problematic cost reductions now in Stellaris:

• Multiple ways to bring empire size to basically zero (including Cloning Ascension bringing you to -90% even without Sovereign Guardianship, which lets you bring systems/planets near zero too).
• Finding a ruined Mega-Art can bring you to -80% planetary ascension cost (Harmony, Covenant, Ascensionists, 2x Mega-Art). If you can cheese a vassal into making a third for you, you can hit -100%.

Can't think of any more at the moment.

You can make cost/upkeep reductions be multiplicative, but then you get into a funky case when they're in the same category as cost/upkeep increases. It becomes extremely opaque to the player what they're actually getting from various bonuses. Though you can solve that by just actually putting them in different categories. You can also make things more readable by swapping e.g. -20% to 0.8x (to make it clear that it's multiplicative with other things).

Drop to 0 is more powerful than +50% because, ironically, it's effectively multiplicative. If you've got +100% already then +50% is only a net +25%, but increasing your alloy producers by 50% is increasing your output by the full 50%.

More importantly dropping to just alloy production compounds the existing benefits to hyperspecialisation. There's no tradeoff on mineral sources when there's no need to source them.

(There's situational negatives to bumping a bunch of workers into specialist roles but that's another story)

Dropping to 80% upkeep is one thing and yeah is comparable to an output increaser but once the upkeep reducers bring the upkeep into trivially ignorable levels you start deleting mechanics. Which is fine if that's an intended game changer from a game changing civic or origin or planet modifier but it's not something that should happen just by accumulating mundane bonuses, except as a part of a deliberate design goal.

 

[Sweeze]

I don't abuse cost reduction modifiers, but I do abuse -orbital bombardment damage stacking because its almost free to get to high levels with virtually any build and it practically breaks the AI (and is probably equally unfun to deal with in MP). It should be similarly nerfed.

Drop to 0 is more powerful than +50% because, ironically, it's effectively multiplicative. If you've got +100% already then +50% is only a net +25%, but increasing your alloy producers by 50% is increasing your output by the full 50%.

More importantly dropping to just alloy production compounds the existing benefits to hyperspecialisation. There's no tradeoff on mineral sources when there's no need to source them.

(There's situational negatives to bumping a bunch of workers into specialist roles but that's another story)

Dropping to 80% upkeep is one thing and yeah is comparable to an output increaser but once the upkeep reducers bring the upkeep into trivially ignorable levels you start deleting mechanics. Which is fine if that's an intended game changer from a game changing civic or origin or planet modifier but it's not something that should happen just by accumulating mundane bonuses, except as a part of a deliberate design goal.

I don't agree that it pop or job upkeep bonuses need to be nerfed because they are OP, but I do agree because it breaks my suspension of disbelief in a game that is soft sci-fi where a combination of governmental policies and cultural beliefs can combine to make ships out of alloys out of literally nothing, or people can just exist without food. As some kind of super high tier tech/ascension ability it might make sense but not by stacking modifiers.

 

[Abdulijubjub]

Drop to 0 is more powerful than +50% because, ironically, it's effectively multiplicative. If you've got +100% already then +50% is only a net +25%, but increasing your alloy producers by 50% is increasing your output by the full 50%.

More importantly dropping to just alloy production compounds the existing benefits to hyperspecialisation. There's no tradeoff on mineral sources when there's no need to source them.

(There's situational negatives to bumping a bunch of workers into specialist roles but that's another story)

Dropping to 80% upkeep is one thing and yeah is comparable to an output increaser but once the upkeep reducers bring the upkeep into trivially ignorable levels you start deleting mechanics. Which is fine if that's an intended game changer from a game changing civic or origin or planet modifier but it's not something that should happen just by accumulating mundane bonuses, except as a part of a deliberate design goal.

It's a very powerful effect for some jobs, yes. But it doesn't explode to infinity like the others; it's balanceable.

A better way to describe it is that the cost of alloy output is pops. If you reduce the metallurgist upkeep to zero and only 1/3 of pops were previously miners in upkeep, it's actually a "-33% alloy cost reduction". It's inherently not that problematic because it's inherently capped.

The bigger the portion of the pops that upkeep represents, the more powerful it is (and it can wander back into problematic if that portion is above 50%). But by the time you can collect enough bonuses to make upkeep go to zero (for most jobs), the jobs at the bottom have become more powerful, so I think they're no more than half. I may need to redo the math with all the efficiency vs. output changes, though. This may not actually be true anymore.

Deleting mechanics, though, is a valid concern (it's the empire size mechanic example I gave). If alloy upkeep were meaningfully capped by, say, mining deposits, then that would be an issue (though they're not). It's more of a game design (apart from balance) issue when you reduce it below the point where you need miners at all, and can get everything from space deposits. It makes all miner things useless, which sucks. But it's not a balance concern (at least, not as directly).

 

[sandwich8]

I don't abuse cost reduction modifiers, but I do abuse -orbital bombardment damage stacking because its almost free to get to high levels with virtually any build and it practically breaks the AI (and is probably equally unfun to deal with in MP). It should be similarly nerfed.

Funnily, -orbital bombardment damage stacking is countered additively by +orbital bombardment stacking on the other side. For instance, even if you have -90% orbital bombardment, all it takes is the opponent to run the Aerospace Adaptation civic or the Annihilator admiral trait to totally counter your build

 

[across]

In a multiplicative system, values like -90%, -50%, or +100% always produce consistent effects, but in a reverse additive system, a -50% modifier loses its meaning when combined with a -90% one.