• 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 #385 - AI Benchmarks

Hi everyone!

The 4.0.13 update released today with the following changes:

Stellaris 4.0.13 Patch​

Improvements​

  • Behemoth Fury is now available to Wilderness Empires.
  • Improved tooltips for the following civics:
    • Functional Architecture/Constructobot
    • Environmentalist
    • Astro-mining Drones
    • Maintenance Protocols
    • Ascensionists
    • Augmentation Bazaars
    • Brand Loyalty
    • Death Cults
    • Dimensional Worship

Balance​

  • Mutagenic Habitability now counts all planet types as ideal for upgrading Gaiaseeders
  • Dramatically increased the draw chance for the Mineral Purification, Global Energy Management and Food Processing technologies
  • Rebalanced the Pleasure Seeker civics to transform Civilians into Hedonists
  • Logistic Drones are now Complex Drones not Menial Drones

Bugfix​

  • Fixed invaded pre-ftls not becoming biotrophies
  • People once more die when they are put in the Lathe
  • Bio-Swarmer missiles can now be used by all biological ships with medium weapon slots (including defensive platforms)
  • Pops that are being pampered will now be forcibly switched to the correct living standard
  • Replacing a district specialisation no longer destroys CyberCreed buildings that should be kept
  • Corrected a tooltip bug where a planet would display itself as a possible migration target.
  • Fixed capitalisation for resources in trade policies
  • Updated assorted modifiers that still referred to Clerks
  • A Trade deficit now causes Job Efficiency and Empire Size issues
  • Fixed the tooltip for the Polymelic trait
  • Armies now protect 200 pops from raiding, not 2
  • Blocked the Federation Code technology for some empires, for example homicidals. To draw the tech, the empire is also required to be in contact with someone they can form a federation with.
  • Blocked the Development focus task Form a Federation for some empires, for example homicidals
  • Added swaps for some empires, for example homicidals, for the Development focus rewards Federation Code, Xeno Diplomacy, and Xeno Relations
  • Updated the Colony view tab mentioned in the hint of the focus task Enact a Planetary Decision to say Management
  • CyberCreed pops with Ritualistic Implants can now colonise planets
  • Fixed Recycled and Luxurious traits not applying to Roboticists
  • Catalytic Processing Civic now lists correct information regarding job swap
  • Cost for repairing orbital rings when you use bioships is now correctly calculated
  • Gale Speed trait gained from Defeat no longer causes errors
  • Fixed scope for LeaderShipSurvivalReason
  • Fixed scope bug for ruler in leader_election_weight
  • Fixed Worker Coop gaining Elite strata jobs in too many places and tidied up the civic tooltip
  • Updated tooltip for Warrior Culture civic
  • Added a pre-list colon to the Feudal Society civic's tooltip
  • The everychanging stone can no longer cause artisans to have negative mineral upkeep
  • Gave the Neural Chorus advanced authority the pop growth speed modifier that had accidentally been assigned to Memory Aggregator
  • The Planetary Supercomputer no longer has an empire cap of 1
  • The Research Institute/Planetary Supercomputer no longer give scientist capacity
  • Added dashes to Traits tooltips and list items
  • Fixed trigger logic for criminal syndicates and federations
  • Fixed Offspring Bioships not being visible in game
  • Fixed Offspring Bioships not being labelled as non-offspring ships in the ship designer
  • The Machine Uprising will no longer spawn 100 machine pops for every 1 missing housing. However the pop-rework seems to have handled 6 million machine pops okay.
  • Stopped removing occupation armies for bombarded and invaded planets on savegame load
  • Repairing ruined buildings in zones is now always possible.

Performance​

  • Flattened pop job modifier node into planet one
  • Made clearing modifiers a fire and forget job

Stability​

  • Fixed a possible OOS when a player leaves the game.
  • Fix CTD when generating a Cosmic Storm mesh.
  • Fixed a random freeze when loading save with stations containing multiple defence platforms.

We expect the 4.0.14 release will be next week (probably on Tuesday), and is expected to include some fixes to a few infinite loops and some select balance changes (like splitting up Enforcers and Telepaths again). It will be a short work week here in Sweden, so it’s likely to be the only update of the week.

As I mentioned last week, with multiplayer stability largely handled, AI is one of our next focuses. Today I want to talk about AI benchmarks, and have a discussion with you about how we should measure “success”.

What Makes a Good AI?​

The AI in Stellaris has always been designed as very reactive, and AI personality has a massive impact on their behavior. Our goal is for our AI empires to feel like actors in the galactic play - acting in a manner consistent with their Origins, Authorities, Civics, and Ethics rather than always picking the “meta” play.

They do still need to put up a bit of a challenge though, especially at higher difficulties.

The first economic goal we make for our AI is “please don’t collapse in an economic death spiral”, and it’s actually far better at that in 4.0 than it was in 3.x. The current AI does NOT meet the second “provide an adequate challenge” goal though.

One of the fundamental tools we have for our AIs are resource targets in their economic plans. They’ll strive to reach those targets, and many of these are set as “scaling” - if they meet the target, they’ll raise the target the next month. This attempts to ensure that they’ll keep thirsting for ever larger research and alloy numbers (or food if they use bioships!) as is appropriate. This is one of the tools we also use to make them exhibit their ethics - Materialists scale their Research targets faster than other empires, so they’ll inherently be more likely to build more Research specializations, while Spiritualists are more likely to have a lot of Unity specializations.

Ironically, improving AI tends to consume any benefits we carved out through performance improvements. The stronger the AI, the more stuff they have - fleets, colonies, and so on.

Benchmarking​

One way to decide whether or not the AI is performing up to expectations is through benchmarking - what kind of fleet power, alloy generation, and research generation should they have by 2230, 2250, 2300, and so on? Around what year should they hit 10k fleet power?

Then there come questions around whether the benchmarks should differ based on personality type. Should it be different if they’re Democratic Crusaders vs. Peaceful Traders? Or does differentiating them there make the friendlier empires too weak?

I’ve got my own set of benchmarks that come from running 3.14 and from the multiplayer community, and in general, I’m okay with Grand Admiral being significantly harder than it was in 3.14. but I’m interested to hear what you all strive for.

How much research and alloy production do you try to have 10 years, 30 years, 100 years, and when the end-game crisis comes calling? (Include your preferred difficulty settings and galaxy sizes as well if you could, as well as if you change any other important settings like tech costs.)

What’s Next?​

We’re going to continue with 4.0 post release support.

Since the next two weeks are both short weeks in Sweden, our next Stellaris Dev Diary will be June 12th. (You’ll be hearing from me in patch notes in the meantime though.)
 
  • 76Like
  • 16
  • 5
Reactions:
By 2070, playing on scaling late game grand admiral, I'm ahead of every AI in tech in every game I've played since 4.0. Most are pathetic. a couple might be inferior. This was true even of my Wilderness run, where they are missing most of the synergy buildings. I would gently suggest that the AI is not optimizing their research enough, or possibly, at all.

I note also that I haven't been min-maxing research myself, except on any relic worlds I stumble across, so they're sucking wind against a non-optimizing player.
I noticed this as well in the game I've been playing on 4.0, albeit on a lower difficulty level. (I forget if it was mid-game scaling Admiral or one notch below that.) I started with Stargazers, too, which put me behind a little, and I did feel behind ... until something happened and everything popped off while I wasn't looking, and by 2070 everyone else is inferior or pathetic in all categories. I'm really not sure what the cause was, but I suspect it was that I finally researched and built the +1 to a specific researcher type buildings. (I did do a bit of a salvage slingshot for ship techs, too, but I don't think that was what did it.)

I'm also pretty sure that the specialized research districts giving more jobs than the generic ones is a thing that will further handicap the computers against humans: I might chase bonuses too much, but I can drop entirely or almost entirely bonused research facilities (depending on mood; I often do one specialized and an archives, at least as a hive mind) on planets where there is a research bonus and it has greater job density than building the mixed research district, on top of being able to better stack bonuses.

The AIs could probably chase numbers as well as I can, but I don't think they do it, and I suspect it just broadly puts them behind. But also, it has always been the case that you need to build a surprisingly large amount of research to stay on track, and I'm not sure the computers have ever done that reliably.
 
  • 1
Reactions:
Again bugs intrusive directive does not work
 

Attachments

  • Screenshot_105.png
    Screenshot_105.png
    3,3 MB · Views: 0
  • Screenshot_107.png
    Screenshot_107.png
    2,6 MB · Views: 0
  • ironman.sav
    1.008,9 KB · Views: 0
Speaking of AI, I'm enjoying seeing the toast notification of when empires are embracing new or fanatic ethics. Whether the behaviour is new I'm itself, I'm not sure.

It's old feature. Not sure about the current state, but in the past most of the galaxy tended to turn some form of xenophile, materialist or spiritualist due to the huge ethics attractivness bonuses they got from free aliens, diplomatic pacts and ascensions. It could prevent empire dropping non-supported ethics: the 2nd most popular ethic didn't get enough pops to reach required amount for embracing it because the biggest faction was getting all the pops. I believe the issue isn't as blatant anymore though I haven't really paid attention to it in any recent games I have played.

Old suggestion I made in 2021: https://forum.paradoxplaza.com/forum/threads/reduce-xenophile-attraction.1471138/
 
  • 1
Reactions:
My biggest issue is the scaling. I mostly play on Admiral, and at around 10 years in, i can get declared war on by an empire twice my fleet size. and then, 50 years in, every single empire is insignificant. do i turn off scaling so they arnt as ridiculously strong in the first few years? cool, every single empire is now insignificant in 25 years. do i turn up the difficulty? i now get stomped by an empire with 3x my power, but it takes 75 years to be insignificant. the whole scaling is off, they dont keep up and the raw bonuses just balloon their early game power but they cant use it to get a decent economy and they just hard fall off.
 
  • 3
  • 1Like
Reactions:
It's old feature. Not sure about the current state, but in the past most of the galaxy tended to turn some form of xenophile, materialist or spiritualist due to the huge ethics attractivness bonuses they got from free aliens, diplomatic pacts and ascensions. It could prevent empire dropping non-supported ethics: the 2nd most popular ethic didn't get enough pops to reach required amount for embracing it because the biggest faction was getting all the pops. I believe the issue isn't as blatant anymore though I haven't really paid attention to it in any recent games I have played.

Old suggestion I made in 2021: https://forum.paradoxplaza.com/forum/threads/reduce-xenophile-attraction.1471138/
Nice one. So it's just the notification that's been changed or at least the icon in recent times. Ty for the explanation
 
Interesting parallels between the CK3 and Stellaris team atm. Fans of both essentially giving feedback that the games' respective AI can't play the game (CK3's new difficulty mechanics introduction, Stellaris' update completely neutering the AI) and for those of us who've been around a while this will sound familiar. Because it is. Some instance of this conversation has been had multiple times for multiple games.

So, presumably, solutions like 'just teach your AI to play the game' aren't going to work. Either its too difficult/time intensive (especially for the pace of releases in Paradox games, imagine retraining the AI for every mechanical update [not saying we, as fans, shouldn't have that expectation, but clearly yah know... money]) or there's a different kind of issue, maybe it'll lead to too rigid builds that'll just collapse in a different way. To me the initial post reads as 'hey we're going on like a 10 week holiday mid June and I'd like to have at least something workable in place by then.' Band-aid fix, which fair enough. So probably fake/cheaty boosts to the AI's viability before then but I'm fine with that considering the alternative is me not playing till like the end of August.

With that in mind: I usually play on Commodore, not a min-maxxer (more or less the opposite honestly) with scaling on. At end game I expect at least a few AI empires to match me, or slightly exceed me, on all the metrics. I kind of need that because I need their help with the crisis (or crises) cuz I play with that on 5x and I don't want to deal with that slog on my own. And that wasn't a problem last time I played, which was a while ago admittedly. Like with the robot ascension DLC I think.

Anyway, that's what I expect from the AI under current circumstances.
 
  • 6
  • 2Like
Reactions:
  • Logistic Drones are now Complex Drones not Menial Drones
That's it? There's still no fix to maintenance drones? It's not even worth mentioning?

It's been talked about in detail in several threads here. Making the gestalt denizen the main amenities job renders gestalts a micro mess that's almost unplayable. Maintenance drones migrating away can crash your amenities, and there's zero way to have empty jobs on the same planet as maintenance drones, meaning you can't build jobs ahead of pop growth or as a target for automatic resettlement unless the planet doesn't need maintenance drones at all. So you can forget filling up a large world late game like a ring world or Rogue Servitor ecumenopolis (unless you're willing to manually resettle constantly).

Making logistic drones complex doesn't change that. They don't provide nearly enough amenities to rely on as the main amenities provider (also, they're trader jobs, not entertainers, so they certainly shouldn't be the main amenities job).

Plus, unless there's been changes not mentioned in patch notes, a fresh machine colony still has awful amenities issues due to zero proper amenity jobs, essentially requiring the immediate building of an amenities building just to be amenities-positive.

The fact that both building up brand new colonies and filling out late game large planets just sucks for gestalts, or at least for machine empires, has killed all desire I have to play this update.

Please, please, please prioritize a fix or at least acknowledge that it's an issue. Or if gestalts are supposed to play differently now, explain how gestalt amenities are supposed to work when maintenance drones can't be relied on.
 
  • 1
  • 1
Reactions:
I feel like we might wanna fix the lag problem before making the AI stronger. The fact that they're underperforming right now may well be the only reason the game is even playable. Unless its just me experiencing extreme lag before even hitting midgame.


As for the AI they just needs to learn to build their planets and not rely so much on their difficulty bonuses. Too often do i find the Ai with loads of undeveloped colonies just sitting there.

Its kind of funny to mention this now but it is maybe relevant so here it is; Why don't you train the AI to use the planetary automation system? Seriously why don't they just turn that thing on? In 3.x the planetary automation system was actually in its best state yet and though sometimes not entirely optimal it was dang close, you could literally turn it on and most of the time it would build a solid world, you wouldn't even need to look at it, you could trust that it was doing a good job. I myself used it all the time, so why doesn't the AI just do that? If they're so incapable of doing it themselves?

Edit: I forgot actually that planetary automation isn't working right now, whoops. Maybe just keep this in mind when fixing it then, factor in a way for the AI to make use of it.
 
Last edited:
  • 2Like
  • 2
Reactions:
Friendly empires should not go cheap on fleet cap, perhaps they should be less inclined to go over their own cap and instead invest those extra resources in research or unity, but they should never, ever, aim low on purpose, even fanatic pacifists need to understand the only way to keep their peaceful existence is through a strong military presence that discourages people from attacking.

Of course, they should also be more inclined to take defensive traditions like unyielding and create massive bastions around their borders while the more aggressive ones, expecting to fight less in their own territory, should give more priority for ship techs & investments.

As for what's expected from GA difficulty, let's be honest, there is nowhere else to go if you want a challenge, so don't be afraid to make it oppressive, GA difficulty should outpace any player doing any meta build (except literal bugs and exploits that need to be fixed, of course) and force the player to win through deeper strategy than simply outpacing the AI, the player should always feel like they have half as much fleet power than everyone else and if they want to win they need to play the ship designer very well.

I'd also like to see the end of the random ship & fleet templates, AI should not be doing nothing but mixed ships with mixed fleets, it should have templates of what a torpedo fleet should look like, a template for what an artillery fleet should look like, a template for what a missile cruiser fleet should look like, etc...

Also, military doctrines should be improved, right now it's hard to justify ever going for any other than the +speed +range one, after all, it perfectly fits every kind of fleet in the game, I specially think the defensive bonuses for fighting in your own territory are far too weak and that hinders the defensive/friendlier empires greatly.
 
  • 6
  • 1
Reactions:
That's it? There's still no fix to maintenance drones? It's not even worth mentioning?

It's been talked about in detail in several threads here. Making the gestalt denizen the main amenities job renders gestalts a micro mess that's almost unplayable. Maintenance drones migrating away can crash your amenities, and there's zero way to have empty jobs on the same planet as maintenance drones, meaning you can't build jobs ahead of pop growth or as a target for automatic resettlement unless the planet doesn't need maintenance drones at all. So you can forget filling up a large world late game like a ring world or Rogue Servitor ecumenopolis (unless you're willing to manually resettle constantly).

Making logistic drones complex doesn't change that. They don't provide nearly enough amenities to rely on as the main amenities provider (also, they're trader jobs, not entertainers, so they certainly shouldn't be the main amenities job).

Plus, unless there's been changes not mentioned in patch notes, a fresh machine colony still has awful amenities issues due to zero proper amenity jobs, essentially requiring the immediate building of an amenities building just to be amenities-positive.

The fact that both building up brand new colonies and filling out late game large planets just sucks for gestalts, or at least for machine empires, has killed all desire I have to play this update.

Please, please, please prioritize a fix or at least acknowledge that it's an issue. Or if gestalts are supposed to play differently now, explain how gestalt amenities are supposed to work when maintenance drones can't be relied on.
Lets not forget slaves either which are still not functioning whatsoever... two major playstyles left in an unplayable state for nearly 3 weeks now. I'm trying to be patient but my goodness... When will the game be in a playable state?
 
  • 3
  • 1
Reactions:
Hi everyone!

The 4.0.13 update released today with the following changes:

Stellaris 4.0.13 Patch​

Improvements​

  • Behemoth Fury is now available to Wilderness Empires.
  • Improved tooltips for the following civics:
    • Functional Architecture/Constructobot
    • Environmentalist
    • Astro-mining Drones
    • Maintenance Protocols
    • Ascensionists
    • Augmentation Bazaars
    • Brand Loyalty
    • Death Cults
    • Dimensional Worship

Balance​

  • Mutagenic Habitability now counts all planet types as ideal for upgrading Gaiaseeders
  • Dramatically increased the draw chance for the Mineral Purification, Global Energy Management and Food Processing technologies
  • Rebalanced the Pleasure Seeker civics to transform Civilians into Hedonists
  • Logistic Drones are now Complex Drones not Menial Drones

Bugfix​

  • Fixed invaded pre-ftls not becoming biotrophies
  • People once more die when they are put in the Lathe
  • Bio-Swarmer missiles can now be used by all biological ships with medium weapon slots (including defensive platforms)
  • Pops that are being pampered will now be forcibly switched to the correct living standard
  • Replacing a district specialisation no longer destroys CyberCreed buildings that should be kept
  • Corrected a tooltip bug where a planet would display itself as a possible migration target.
  • Fixed capitalisation for resources in trade policies
  • Updated assorted modifiers that still referred to Clerks
  • A Trade deficit now causes Job Efficiency and Empire Size issues
  • Fixed the tooltip for the Polymelic trait
  • Armies now protect 200 pops from raiding, not 2
  • Blocked the Federation Code technology for some empires, for example homicidals. To draw the tech, the empire is also required to be in contact with someone they can form a federation with.
  • Blocked the Development focus task Form a Federation for some empires, for example homicidals
  • Added swaps for some empires, for example homicidals, for the Development focus rewards Federation Code, Xeno Diplomacy, and Xeno Relations
  • Updated the Colony view tab mentioned in the hint of the focus task Enact a Planetary Decision to say Management
  • CyberCreed pops with Ritualistic Implants can now colonise planets
  • Fixed Recycled and Luxurious traits not applying to Roboticists
  • Catalytic Processing Civic now lists correct information regarding job swap
  • Cost for repairing orbital rings when you use bioships is now correctly calculated
  • Gale Speed trait gained from Defeat no longer causes errors
  • Fixed scope for LeaderShipSurvivalReason
  • Fixed scope bug for ruler in leader_election_weight
  • Fixed Worker Coop gaining Elite strata jobs in too many places and tidied up the civic tooltip
  • Updated tooltip for Warrior Culture civic
  • Added a pre-list colon to the Feudal Society civic's tooltip
  • The everychanging stone can no longer cause artisans to have negative mineral upkeep
  • Gave the Neural Chorus advanced authority the pop growth speed modifier that had accidentally been assigned to Memory Aggregator
  • The Planetary Supercomputer no longer has an empire cap of 1
  • The Research Institute/Planetary Supercomputer no longer give scientist capacity
  • Added dashes to Traits tooltips and list items
  • Fixed trigger logic for criminal syndicates and federations
  • Fixed Offspring Bioships not being visible in game
  • Fixed Offspring Bioships not being labelled as non-offspring ships in the ship designer
  • The Machine Uprising will no longer spawn 100 machine pops for every 1 missing housing. However the pop-rework seems to have handled 6 million machine pops okay.
  • Stopped removing occupation armies for bombarded and invaded planets on savegame load
  • Repairing ruined buildings in zones is now always possible.

Performance​

  • Flattened pop job modifier node into planet one
  • Made clearing modifiers a fire and forget job

Stability​

  • Fixed a possible OOS when a player leaves the game.
  • Fix CTD when generating a Cosmic Storm mesh.
  • Fixed a random freeze when loading save with stations containing multiple defence platforms.

We expect the 4.0.14 release will be next week (probably on Tuesday), and is expected to include some fixes to a few infinite loops and some select balance changes (like splitting up Enforcers and Telepaths again). It will be a short work week here in Sweden, so it’s likely to be the only update of the week.

As I mentioned last week, with multiplayer stability largely handled, AI is one of our next focuses. Today I want to talk about AI benchmarks, and have a discussion with you about how we should measure “success”.

What Makes a Good AI?​

The AI in Stellaris has always been designed as very reactive, and AI personality has a massive impact on their behavior. Our goal is for our AI empires to feel like actors in the galactic play - acting in a manner consistent with their Origins, Authorities, Civics, and Ethics rather than always picking the “meta” play.

They do still need to put up a bit of a challenge though, especially at higher difficulties.

The first economic goal we make for our AI is “please don’t collapse in an economic death spiral”, and it’s actually far better at that in 4.0 than it was in 3.x. The current AI does NOT meet the second “provide an adequate challenge” goal though.

One of the fundamental tools we have for our AIs are resource targets in their economic plans. They’ll strive to reach those targets, and many of these are set as “scaling” - if they meet the target, they’ll raise the target the next month. This attempts to ensure that they’ll keep thirsting for ever larger research and alloy numbers (or food if they use bioships!) as is appropriate. This is one of the tools we also use to make them exhibit their ethics - Materialists scale their Research targets faster than other empires, so they’ll inherently be more likely to build more Research specializations, while Spiritualists are more likely to have a lot of Unity specializations.

Ironically, improving AI tends to consume any benefits we carved out through performance improvements. The stronger the AI, the more stuff they have - fleets, colonies, and so on.

Benchmarking​

One way to decide whether or not the AI is performing up to expectations is through benchmarking - what kind of fleet power, alloy generation, and research generation should they have by 2230, 2250, 2300, and so on? Around what year should they hit 10k fleet power?

Then there come questions around whether the benchmarks should differ based on personality type. Should it be different if they’re Democratic Crusaders vs. Peaceful Traders? Or does differentiating them there make the friendlier empires too weak?

I’ve got my own set of benchmarks that come from running 3.14 and from the multiplayer community, and in general, I’m okay with Grand Admiral being significantly harder than it was in 3.14. but I’m interested to hear what you all strive for.

How much research and alloy production do you try to have 10 years, 30 years, 100 years, and when the end-game crisis comes calling? (Include your preferred difficulty settings and galaxy sizes as well if you could, as well as if you change any other important settings like tech costs.)

What’s Next?​

We’re going to continue with 4.0 post release support.

Since the next two weeks are both short weeks in Sweden, our next Stellaris Dev Diary will be June 12th. (You’ll be hearing from me in patch notes in the meantime though.)
I think a lot of problem that u place on ai are power creep based and i am not talking about bug builds like (cyber assention enforcer). The power creep got so bad in this update its hard to discribe.
50 per month at first day if u pick machine world start.
jobless(utopian ubandence - 75% upkeep triat ) build where u get 150+ reserch and unity day one if fire all your people outside form those that keep u form going into red.
Those outputs are day one. those thikns where not posible in previous updates. Not even grand admiral AI can beat those build and there is no point in making it so. Reserchers are one of worst ways to get tech in early game post this updates they cant fight new civics and those build.We should bring those builds back to normal lvl(and others like them) insed of trying to make ai be able to fight them while plaing empires with more traditional civicks.
 
  • 1
Reactions:
Now I was listening to montu plays latest video about the lag and the shape the game is in and I had a thought.

Why simulate the economy of AI players. No really. Why?

Why not create planet templates which create an ideal planet when the player gains or looks at the AI planet, populating pops based on species list from the empire? Instead of simulating the AI playing like a player why not create economy templates which have economy difficulty curves which are more inflexible, more set. We all already know the AI is cheating for resources when their planets look like jokes. So why not create templates of 'perfect' planets for when and IF the player interacts with them.

This way you will significantly reduce lag from all those calcs by them just not existing anymore. This way the AI just get resources based on the number of planets they own and it scales from there. From there they just raise the fleets needed to fill out the impression of a 'real' player. But behind it there are no pop calcs ergo no pops exist in the whole galaxy ONLY your own. The other pops exist in a state of potentiality, IE they exist if you access an AI planet or gain an AI planet but disappear when you leave.

You can have different sets of templates / resource curves based on the kind of civ and combination of ethics. No need to calculate huge sums pops contributing to an economy that doesn't work anyway!

Anyway this suggestion is made with the best possible intentions. Best of luck.
 
  • 5
  • 1
Reactions:
The AI is overall fairly bad at properly scaling their economy/research/unity in my experience. It's like they hit a brick wall around midgame, and don't work on scaling their resources more. I think this because of their exceptionally poor planet specialization, and unwillingness to build out and expand their planets. There have been many times I have gone to check out an AI economy, and they have unupgraded buildings, undeveloped planets with unemployment/civilians, and poorly developed fleets. They also need to be more willing to expand their fleet to the fleet cap more if they have a large surplus, as well as develop planets. I also want to see them build more resource storage on starbases, because they almost never do that in my experience.

The AI also should focus more on ascension and unity gain more than they do. Ascending and getting ascension perks should be a higher priority than it is, especially for spiritualist empires. They should also take APs that relate to their personality/situation more, and they should take the megastructre/arcology APs more too.

I can't give any specific benchmarks, but I usually have 1k+ alloys, 100k fleet power, and have ascended by 2300, but I would consider myself an above average player.
 
  • 1Haha
Reactions:
Personally, disappointed in both performance "improvement" and pop changes. I loved watching how genetically modified pops perfectly for certain jobs work on certain jobs outputing max amount of resources. Stellaris basically is a excel spreadsheet and you watch numbers grow, all that covered in my favourite Sci-fi genre, but now we have pops amalgamted in one big pod, you have little to none control on how and where those pops work. I hope pride does not blind you and you will revert or even better change pops so that they can be managed. Perfect world would be where you could geneticaly modify pops based on jobs, but i guess that is impossible to do with current engine/code.
I wish you all a peaceful life <3
 
  • 1Like
  • 1
Reactions:
Potentially nonviable idea but what if the benchmarks were dynamic to the player? On ensign the AI would attempt to meet 80% of the player's metrics, whereas on grand admiral it might be 800%.
That would be great actually.
(Not specific numbers but the idea)
Since even in case of specific player this vary.
For example, I'm trying to build my fleet depending on situation. At least 1 influence from power projection but heading towards max.
But on the other hand If I'm surrounded by befrienders (or I managed to make friendly neighbours) I concentrate on everything except fleet power.
If I see too many empires become stronger I starts to invest heavily.
So benchmarks should be dynamic?
This would make AI be more real. But that could be hard to implement, and it's difficult to guess what would happen in case of player also trying to match AI - power spiral that never ends?
 
  • 1Like
Reactions:
In an optimal world, the AI would not have to cheat to be powerful (ex. Grand Admiral = free resources for AI), but instead it would make better strategic choices, which would result in better fleets / economies overall.

lower difficulties is fine for the AI to do less optimal decisions, but anything higher would have to mean that the AI is actually competent and not what it is today. A competent player 10/10 times will win against an AI opponent even with maximum difficulty, because the AI is not inherently that intelligent ast it stands and has been an issue for the wider Stellaris community regarding their actual performance and decision making.

What I am hoping for is that the AI will be able to provide an actual challenge, which can for example be statistically measured to be better via benchmarking and I look forward to seeing the solutions brought fourth by the Devs to address the issue of AI competency.

Thank you Paradox for finally addressing the AI issue.
One person told me something i to this day remember. If strategy games had "good" AI that makes good decisions on high difficulties, then they would probably develop sapient computer and to run it we ourself would need a supercomputer :D otherwise that is no AI it is just a computer completing tasks, example empire has negative energy income it builds energy district. That is why most strategy games use that golden rule, high difficulty AI gets extreme boosts to everything, that especially applies to Paradox Games. Take Age of Wonders for example, on high difficulties you can outsmart enemy with 10x bigger and stronger army, if you can call outsmarting stupid an achievement.
 
Honestly the thing I was looking forward to the most was 4.0 improving the performance of the game. You released it and not only is it barely playable but the performance has reduced since 3.14 and now you're suggesting that it's only going to get worse from this point.

I get that you wanted to get it out before your holiday or for the Paradox sale or whatever but I think you would've been better off releasing BioGenesis without the pop changes and just delay the pop changes until it was in a better state.

It's kinda clear to see that the state of the game is probably going to be awful for a long time which is really sad considering the amount I played out of hype leading up to this DLC/patch release.
 
  • 11Like
  • 2
Reactions:
My biggest issue is the scaling. I mostly play on Admiral, and at around 10 years in, i can get declared war on by an empire twice my fleet size. and then, 50 years in, every single empire is insignificant. do i turn off scaling so they arnt as ridiculously strong in the first few years? cool, every single empire is now insignificant in 25 years. do i turn up the difficulty? i now get stomped by an empire with 3x my power, but it takes 75 years to be insignificant. the whole scaling is off, they dont keep up and the raw bonuses just balloon their early game power but they cant use it to get a decent economy and they just hard fall off.
GA, mid game scaling. Mid game date 2250, reduce crisis bonus strength to 0.5-0.75. And if you changing crisis stregth then move late game to 2325 for example.
 
Last edited: