Stellaris 3.0.3 AI Feedback Megathread

[pmchem]

and here is a run that went out to year 2300 but I only looked at the save from year 2253 because there were already several problems apparent. some may have been partially caused by wars, but, war is certainly not the source of all the problems (most of these empires are not at war).
bug report here:

Stellaris - [3.0.3] [bfcc] AIs doing bad economic things in the beta (3/many)

Description [3.0.3] [bfcc] AIs doing bad economic things in the beta (3/many) Game Version 3.0.3 beta What version do you use? Steam What expansions do you have installed? Synthetic Dawn, Utopia, Leviathans Story Pack, Apocalypse...

forum.paradoxplaza.com

The attached screenshots detail particular problematic empires. The naming convention for the PNG screenshots is (empire #)-(year).png.
For example, 9-2253.png is the empire via "play 9" at year 2253.

I only examined year 2253 for problems. I found enough to stop there. I am uploading many saves to help you investigate but have NOT examined them all.

1-2253: general mess. at war, but not upgraded fleets. zero minerals. many habitable planets uncolonized. employment issues, intel issues.
2-2253: out of food.
3-2253: never colonized the planets in its space??
5-2253: total fail cascade. out of energy, minerals, food, goods. planets at low stability.
9-2253: out of food, planets at low stability

and of course the typical "not upgrading fleets" problems for various empires

EDIT: I just realized the OP asks people to only post once. my bad. that seems hard to make work with ongoing testing, though?? anyway, I'm done with tests for now.

 

[hart30]

It is the year 2429 now in my game. Huge Galaxy. The 2 best AI Empires (one fanatic purifier and one devouring swarm) have both over 30 planets. Their fleet capacity is around 1400, yet they only fill their fleet up to ca. 400 naval capacity and was for some unknown reason most of their alloys in Habitats, Stargates and megastructures. The Crisis soon will appear and the best Empires only have 90k on fleetpower, while having the potential for 400 - 500k (their eco is strong, but they somehow dont care About their fleetsize. All the other smaller Empires barely reach 50k.

Now to their eco. They have Admiral difficulty, so each AI empire has currently plenty ressources, but for some unknown reason they start accumulating more and more unemployment. They have the district and Building space, as well as the ressources to employ every Pop, but for some reason they dont do it, every planet Ends up with around 10 unemployed Pops galaxywide. An incredibly waste of ressource potential.

Due to this the AI is now inferior. I have the Feeling that it behaves worse than in 3.0.2. It does not collapse anymore after losing planets in wars, but the Prize was too high. Before the AI had way bigger fleet and massed up a lot of alloys and ressource, but by losing or gaining planets during wars, the AIs eco often broke down, because suddenly they did not have enough raw ressources anymore for all the Industry districts they suddenly gained / raw ressource districts they lost.
Now the AI just wastes most of its potential. If someone simply does not use his ressources, his eco cannot break through planet loss/gain during war. This makes it much weaker from the start, while before it grew weaker after several wars and even then it was not guaranteed that it breaks completely, if it was on the planet gain side.

This could be a possible better solution, than wasting all its potential:
Maybe if an AI conquers a planet, some of the planets Industry districts (2-3 per conquered planet) have to be destroyed, so the AI gets the ability to build new districts again. That would be a better help against Industry heavy ressource sucking after a new conquest. But this would not help against the AIs eco Breaking from planet loss. Maybe if an AI has Major losses during a war it should be allowed to redevelop some of its districts for a certain amount of time. Then it would have a Chance to Rescue itself.

 

[cblue22]

I've noticed that the AI decides to have the syncretic species when playing with syncretic evolution be residen status when playing fanatic xenophobe. Idk if this is how its supposed to be but I was thinking if they are Fanatic Xenophobe or Xenophobe they would enslave them and if they are Authoritarian maybe not. Also would it be possible to get a Q and A stream for feed back on changes every once and a while?

 

[slv]

While this post is not about my experiences in playing 3.0.3, I hope it will be helpful in finding the culprit of the issues.

I've spectated a lot of games in 3.0.*, both modded and unmodded and, I think, I can pinpoint many reasons for possible failures of the ai. This is a list of things that (according to my impression) are responsible for 70%+ of the causes of the breakdown. Also it's important not to believe me without checking, I do not have access to inner workings of the code and I don't know how most of the systems work in detail, my assessment might be wrong.

/common/ai_budget

I think mineral_budget spending on planets should have large static_min past the early game (at least 400). So whenever AI wants to build mining district it will always be able to, AI sometimes keeps spending veeery low mineral income on stations instead of increasing said mineral income.

/common/buildings

Main problem here is usage of two scripted triggers, "buildings_simple_allow" and "building_relaxed_basic_income_check". Both of them seems to be written pre-3.0 (they are located in /common/scripted_triggers) so they fit pre 3.0 economy better, each of them assumes that districts should be prioritised before buildings. In pre 3.0 economy buildings were responsible for specialist jobs and construction of them could crash the economy, this is why these triggers exist.

This is fine for some buildings, but is absolutely harmful for amenity buildings. Each of them requires buildings_simple_allow and this trigger will wait till the districts are built. This stops the construction of holo-theaters, low amenities crash the economy which further stops the trigger from evaluating to yes. AI keeps building more districts, amenities go lower and lower and economy never recovers. At the very least this trigger should be removed from the holo-theater, but it would be better if they were removed altogether (and replaced by a building-specific triggers)

/common/districts

Right now the field (ai_resource_production) makes AI think that an industry district will always provide consumer_goods. This isn't true on forge planets, right now all industry districts show as if they give 1 alloy and 1 consumer_good. It would probably be better if instead industry districts on the planet with forge designation had ai_resource_production of 2 alloys, industry districts on planets with factory designation showed ai_resource_production of 2 consumer_goods and planets without designation showed 1/1 split.

Second, industry districts probably need a trigger in allow = {} similar to "building_relaxed_basic_income_check". While it doesn't happen in 3.0.3 (more on that later, in /economic_plans section), in 3.0.2. AI was regularly overbuilding those trying to follow the current economic_plan. If there was a trigger checking that energy,food. minerals income are greater than 0,0,50 respectively this wouldn't have happened.

Also from my experience if you increase ai_resource_production of rural districts (mining, farming, generator), AI is more likely to take serious care of raw resource deficits.

/common/decisions

Population control and discouraging growth shouldn't be used, same with "halt robot assembly" (if AI truly wants to halt robot assembly, it's better to have a destroy ={} field in the robot assembly instead).

/common/edicts

Somewhat of a surprising suggestion, but low difficulty AI really can't afford nutritional plentitude. The only thing that comes out of this edict is starvation. The edict should have much lower weight (or, ideally, weight = 0).

/common/colony_types

Honestly, even though visually everything looks terrible when observing (none of the designation make sense, the horror), I don't think think this is major. Overall I think bureaucratic designation is overused and the best way to make AI do "correct" designation is to have a scaling weight. If the weight of each designation (mining, farming, bureaucratic, research) was increased by 3 for 1 building/district, by 6 for for 2 buildings/districts, by 9 for 3 buildings district the resulting distribution would be better and more pleasant for humans to see (more mining districts = proportionally higher weight). This way as a result of a weight comparison designation will be chosen depending on what designation has the largest amount of relevant buildings/districts.

/common/economic_plans

As I see, 3.0.3 economic_plans were a big change for 3.0.2 in that alloys and research income/focuses were removed. This does indeed help with 3.0.2 problem of AI overbuilding industry, but it has a side-effect of AI not prioritising science and alloys nearly enough (it's frequent to see AIs with 15 alloys per month in 2400 according to some of the screenshots in this thread). To fix it is possible to keep 3.0.3 economic_plans, but add a subplan that in case of good raw resource production (potential = {minerals, energy, food > 100,20,5 } for organics and 100,20,0 for non-organics) adds some research and alloys targets. Lack of those targets may be the cause of unemployment and empty planets, but I am not entirely sure if it's the only reason.

Alternatively it's possible to instead come back closer to 3.0.2 plans and write a trigger that disallows construction of new industry districts if economy situation is dire (or, actually it's possible to do both. Extra redundancy doesn't hurt the chaotic system).

/common/starbase_modules; /common/starbase_types

Hydroponics is a very very good module, it can really help low difficulty AIs if it was allowed to build them everywhere (and had high weight of it). While it sounds awfully specific, but it indeed is a large change in how ensign AI can manage its early food needs.

/common/species_rights

Slavers will rebel. I don't think anything can be done with it (AI can't resettle, even human slaving empires can't be managed without resettling), apart from disabling AI slavery which isn't a good solution. Some things can help though, for example I think any slaver should prioritise identured servitude over any other slavery type (except for serviles, serviles can have caste). I know, it's not #optimal in human case, but this would lower risk of rebellions massively.

/common/technology

Some technologies need to be almost autopicked by AI, namely economy technologies (I suggest to increase the weight of these technologies at least in 10 times, preferably 100). I am talking about technologies which give 20% boosts to miners/technicians/farmers, technology that unlock hydroponics and technologies that unlock resource edicts (capacity_overload, mining_subsidies, etc.).

Frankly, I don't know whether these technologies are "optimal" or popular among players, but they immensely help low difficulty AI manage their upkeep and production.

 

[kgptzac]

I'm seeing glaring problems in early game.

• Hiveminds for some unknown reasons spam Synaptic Nodes buildings, causing Tree of Life hivemind never have enough food for their 1st colony ship.
• Galactic Doorstep hivemind somehow doesn't want to colonize nearby habitable planets.
• Doomsday AI somehow don't get habitable planets spawned nearby and still generally do nothing to stop its doom that leads to rebellion on their one and only capital planet.
• Generally the AI really likes to go on excess with building industrial districts while going into deficit with other resources such as food.

And these are only the really obvious problems that can be easily found by having max amount of AI in huge galaxy and turn on observe. Around year 2225 and go find the worst performers and you'll almost guarantee to see failing AI to have these symptoms.

 

[Salv94]

I have already made a bug report on this issue but I might as well mention it, since it can be a big deal. Right now, federations may hinder the AI due to the fact that it can choose allied civilian ships and civilian stations as targets to follow. Here's the report:

Stellaris - AI fleets follow allied civilian ships and stations ---- 3.0.3 1d63 & 3.0.2 91c1

Description AI fleets follow allied civilian ships and stations ---- 3.0.3 1d63 & 3.0.2 91c1 Game Version 3.0.3 & 3.0.2 What version do you use? Steam What expansions do you have installed? Synthetic Dawn, Utopia, Leviathans Story...

forum.paradoxplaza.com

It's from the previous version but the issue persists in the new patch.

There's another problem I've observed, but given how many potential allied targets the fleets now have I can't be sure of it. When selecting targets to follow, the AI chooses those with weak military power. There was a target with high priority but they chose a low priority starbase in the system next to it. I think the AI uses the same logic as with selecting hostile targets. Allied fleets will not consider targets stronger than themselves. I saw this happen with two fleets in the same system as well. They chose the smaller one with lower priority.

As for economic performance, I think the other posts have mentioned similar observations.

 

[Jman5]

Something I have noticed looking at the AI is that I am 40+ years into the game and I don't think a single AI has built even 1 research lab. I am pumping out 500+ research while the rest are stumbling along at about 1/5 that pace (I haven't conquered anyone). The only 3 non-district buildings they seem to most prioritize are Alloy, Consumer Goods, and Bureaucrats. Even worlds that are designated for tech don't have any.

Instead of having the AI build those alloy/CG buildings everywhere, they should be building industrial district. Then in those open spaces they should be putting more Research labs and Unity buildings.

 

[grommile]

Doomsday AI somehow don't get habitable planets spawned nearby

that's how Doomsday works: not only is your planet going to blow up, you don't get your guaranteed habitables.

 

[Baalor204]

The pop growth mechanic is still flawed. I'm just going to cross post what I gave on steam, to make things easier.

Opted in to try the re-work, quickly looked over the patch notes. Its still a problem, and its clear that Paradox either is unaware of the problem, doesn't know how to address it, or doesn't want to. All three are an issue, so hopefully someone from their design team will stumble upon my post here.

The amount of growth needed should be based on planet, not empire wide! At the beginning of the game, it takes 100 growth for a new pop. Under baseline 3.0, the formula I believe is 100 + Total Empire Population / 2. The beta makes this 100 + Total Empire Population / 4.
That's not addressing the issue! It doesn't matter what number you put as the denominator. It will always run up against the same problem(s) eventually. That's the real issue.

My suggestion; get rid of the increase or make it specific to planets, so the formula on a planet would be 100 + Total Planet Population. Could even have a multiplying factor, making it 100 + Total Planet Population * C.

Disclaimer; I have over 1000 hours and 4 years of playing Stellaris. I know how to play, I care about the game. I will mod this ridiculous mechanic out, because its clear its not a system meant to balance the gameplay, but rather address lag. There are better ways to do that, from a software engineering perspective I deal with that at work as well. We don't push systems that dissatisfy the customers because it would be better performance.

 

[grommile]

My suggestion; get rid of the increase or make it specific to planets, so the formula on a planet would be 100 + Total Planet Population. Could even have a multiplying factor, making it 100 + Total Planet Population * C.

This doubles down on the "you have to do goofy shenanigans for best pop growth" aspect introduced by the pseudologistic curve for planetary growth points per month.

 

[T5_k]

Playing as a machine empire, iron man, tech/tradition cost @ x0.25, difficulty @ Grand Admiral, growth required scaling @ x0.05, logistic growth ceiling @ x1.5

All empires have a relative power of pathetic in all fields, safe for FEs. However, I did get a really lucky start.

I cannot comment if AIs perform better or worse now, as they practically never posed a threat to my beautiful machines
I do remember AIs posing a greater threat before the 2.0 update, however, that was a long time ago.

And now, taking the risk of commenting off the AI topic - I would like to suggest changing the maximum size of habitable planets to 20, as this would fit great with the new changes to primary building progression and help keep the pop numbers under control, I believe.
Oh, and, could you maby, perhaps, pretty please lower the upkeep cost of special resources for special districts of Ecumenopolis and Ringworlds by half? Currently, they are an ungodly drain


Thank you devs for this awesome game, for listening to us, and for making the game better with each update! <3
Attached is the savegame ironman.sav (game year ~2260) and ironman1.sav (game year ~2300).

EDIT: Added opinion/feelings about upkeep cost of specialty districts on Ecumenopolis and Ringworlds, added save

 

[HistoricalScore1]

This has been my experience so far with 3.0.3. I've been playing rather than spectating. I won't play on Ensign unless I'm using StarNet AI mod. I prefer a smart AI than one that cheats and gets a boost bonus resources/free modifiers for fleets. I'm thinking of normal players or new players with these suggestions.

TLDR: The pop growth slider is confusing, for me and I'm certain new players. The building changes seem to make it luck-focused (too RNG-based to be fun) for normal non-overpowered origin builds to succeed. Other origins can have places with many, many jobs, while normal empires have to hope they roll a high resource district planet to keep up with resource buildings. Please focus on more immediate balancing needs. Civics, traits, origins. There are very good suggestions for all of these in the Suggestions section of the forum.

My Settings changed from default:

• AI Aggressiveness: High
• AI Empires: 16
• Difficulty: Captain

Pop Growth: Please explain/clarify to this lowly human, what the pop growth sliders do, specifically. I understand logistic growth, finally, and noticed the default was lowered to 1.5 from 3. Before I come up with a determination, I'd like to understand the second slider thingy. It was explained to me. This buffs those that go wide (and live), doesn't reward players for making tough choices to optimize carrying capacity on planets versus build another district. Unlike 3.0.2. where I had to decide, "do I build another resource district and lose some of that capacity? Or do I build a city district to keep pop growth near the ceiling?" Gone. Economy is one aspect of a strategy game, and pops are a resource. I think planets shouldn't just grow pops. But pops + planets together would make things more challenging/rewarding and allow many different playstyles. Instead of carrying capacity, perhaps on a planet/pop/housing available/amemities/happiness/stability basis. More work? Yep. Better in the longrun? I don't know. I just have read as you have that this isn't working. For example, in my latest game, I have Rapid Breeders, and adaptable. So my guaranteed are at 90%. If I go Expansion Tradition and take the extra 10% pop growth, I get an underwhelming 1 less month before a new pop is made. On my main planet from 59 months to 58 and on colonies from 19 to 18 with 3 pops on one and 4 on another. Meh^meh.

Crime Lord Deals: Good! Was too easy to exploit for free stability. Now there's a cost for non-criminal builds. Very, very, good.

Clerks: Yay! Clerks are less crappy filler roles. I have to make a choice about locking out clerk jobs to force resettlement of unemployed pops vs get equivalent of 1 Technician output! Thank you! Most I think will, unless they are Trade Federation/builds, which is fine. I've kept some clerks to keep amenities above 0 while building Holo Theatres.

Resource Buildings: I fear this really, really buffs builds that have, say a ring, Ecu, Life-seeded, over normal playstyles/origins unless RNG is on the player's side. The aforementioned have guaranteed high number of jobs for those worlds/places, so they get maximum benefit from using one building slot to boost all those jobs. Whereas normal empire has to pray for an okay roll on a planet with 9+ one resource district slot. "Oh, I have only planets with industrial slots and only 2 agriculture districts or 5 mining districts? I guess I lose/restart." That's not a fun mechanic, to me. I've had several games where the most non-industrial districts I've rolled was 8. I went just Explorer Trait/Build to see.

Medical workers: Less useless. The habitability is ooo-kay. Everything else is worthless until you get cloning (which takes way too long) compared to robots, making this super inferior and frustrating for those who don't want to just go robo/synth.

AI: (see screenshots)

The Great
I've experienced the Non-Ally AI doing things I only did with StarNet AI mod for a long time. They actually skipped a neighboring system with their Construction Ship to secure checkpoints. Now I have to watch all my science vessels, like a hawk (I love it!) to make sure while I'm surveying the AI isn't almost done building a Starport to force my scientist out. LOVE IT! LOVE IT! In a war, I saw 2 different fleets of 1.XK corvettes coming toward my 800 power Starbase. One stayed back while the other attacked. For a few seconds. Then the other one joined and together they crushed. Brilliant!

The Bad
AI still seems to need to cheat for getting resources/fleet etc. instead of playing better. At least that's my perception given not debugging the game. In that screenshot, the bottom empire is Materialist and Fanatic Xenophobe. I've mostly militarized my economy (I didn't colonize that one other planet yet) I'm Fanatic Militarist and Barbaric Despoiler. And Ally AI...it's bad. Like, really, really terrible. I've seen my ally, I'm in a Defensive Pact with suicide their fleet of 500 into a Starbase of 540 while a fleet of 800 waited. Then went in when the enemy brought a fleet of 1.4k to said Starbase and joined the suicide. They don't follow when encouraged, when in great opinion, and in a white peace with other neighbors so there's literally ZERO threat to them except the current enemy.

Bottom Line
Thank you everyone for your hard work. I'd really hope your next focus is on what I consider more immediate. Civics, Origins, Traits, and Ascension Perks. Most take Technology first because it's just THAT much better than others. Most don't play certain Civics/Origins because they are just that hindered. I have an empire I made, where I can tech rush and have 1K+ research pretty fast, but that's one playstyle. Where's my marauder/barbaric despoiler option? I play a non-overpowered style and the AI just cheats to take away my advantage I grinded for. And yeah, I'll say it on every post from now on--->Please explain to us lower beings why you haven't added customizable namelists so we don't ruin the checksum when putting in a text file. That's been asked for since day 1 for us role-players. I even paused, changed colony to industrial to make districts for less cost then changed back to colony then unpaused nearly every time. Do I get an edge? Nope. And again, all these tech empires need is 2 hangar bays in a Starport and that's 2.1K fleet power right there. Holding off an attack.

War3.sav is where I declared on the empire to my south to see if they could be punished for so much expansion. Nooope.
war.sav is where I'm about to get wrecked by the top empire who I decided to give back a coffin for +20 relation a few years ago...LMAO.

Cheers and thanks for all your hard work!

EDIT: Added "--->"

 

[JohnnyBravo40]

I just played a game for several hours on ironman at ensign difficulty in the beta. This was my first time playing since before the necro stuff. I like the addition of espionage, and eco changes, it spiced up the game more than I expected. I liked having the sliders for Pop cap and the bonus modifiers. I will say though that The AI was still very underwhelming. I havent even had a midgame crisis and they pale in comparison to all parameters of my empire. I could wipe them (all other empires) out combined. Well except for the fallen empires. That to me is the biggest downside to the game fun wise, I don't like having the difficulty mechanic only affect resource income by giving empires a boost. I just starts to make it impossible to keep up with their production and I would get stomped in the early game. And as far as using scaling difficulty, it really just still does not address the issue, Ill have way too easy of a time early on, and then when the bonuses start kicking in its kinda too late. I would like to see the AI have variable decision parameters that define the difficulty. (in other words, they play better at higher difficulty, not simply an economic boost). Some things they could do is take advantage of things like "metas" or optimal builds/game mechanics that might be unknown or even impractical for a human player to utilize for them to be competitive. As far as population growth and empire building goes, I really like the changes, I always used to play on 0.25% planets because it was so hard to manage them and also made owning a planet actually valuable. But now I was playing on 0.75% planets and everything felt great, I did not feel overwhelmed trying to manage the planets. I did however start to feel way too much pop growth from increasing planet numbers. That base growth for a planet of 3 is just WAY to strong. I think it should be 0 (zero). Or at the minimum allow a slider to modify that value at game creation from like 0-3 or maybe even more? idk its a game someone might have fun playing with 5 or 6. This would solve 2 issues with one stone, first being the massive incentive to colonize colonize colonize. (that was not a typo). second being to give the feel of the curve being less steep, it felt like the planets accelerated in growth too much (went from too little growth early on, to too much at mid capacity) but it could be that it was simply a "feeling" due to every planet getting an auto +3 growth per month, it just felt like I was having to pause too much because I had unemployed pops on some planets, and others with too few workers. Anyways I was playing on 2x Logistic Growth Cap and 0.05 Empire Growth Scaling (I did have to google how these actually affected gameplay so it would be nice to add a more specific description of them, the ones provided were so vague.) Going forward I feel like combat needs some love, corvettes have too much evasion, cruisers have too little. I also think the upgrades for weapon damage, evasion, armor/shield, ext. should have linear progression (or at least more linear). Same with upgrades that offset each other, if a sensor upgrade gives for example +5 tracking, then thruster upgrade should offset it evenly with +5 evasion. I also don't like that the values given to ships for evasion aren't clear, for example afterburners will say it gives +5 evasion, corvette will get +5, but then a cruiser only gets like 0.75. Like c'mon make it do the same for all ships. If +5 is too much for large ships then lower it to like +2 or +1, we don't want or need OP corvettes. and Lastly the fights are just a mess visually, cant tell what the heck is going on because all the ships just swarm each other and go all which ways. slow them down, have there actually be an approach portion to the fight before they get all tangled. I hope the feedback helps and leads to more content in the future. Thanks!

Edit: I almost forgot It would be nice to have a way to diplomatically get systems from allies or place claims on systems that they also have claims on during war. When I was playing I had a planned system I was going to take from an enemy, my ally ended up starting a war with him before I had finished all my preparations. He ended up placing a claim ahead of me and then I was completely unable to place my own claim, It ended up going to him after the war and made all 3 empires basically dotted up with random system grabs. It would be nice to be able to simply demand a system/planet for both enemies and allies as well, so I don't HAVE to wage war just to get a system. Lastly It would be cool to allow claims to affect uncontrolled systems so I could deter other empires from taking it in the first place, basically taking the system would anger me/ or AI empire with a claim (if I take the system), making hostilites more likely.

Edit 2: Also research exploads into mid game, I was already on level IV tech items with option to get lvl 5 before mid game crisis. The amount of research required for each tier needs to increase exponentially not linearly. since pops grow exponentially it doesnt take long for me to pop out new tech every couple of months.

 

[Valanoir]

Just started a game on admiral difficulty, by 2250 the machine purifiers next to me had settled at least 6+ planets wheras the hive mind on the otherside of me had settled exactly 0.

 

[King Harkinian]

While work is obviously needed on the economic AI, I'm quite pleased with how the military AI has turned out. I am no longer seeing fleets stuck moving back and forth, wars are generally active and eventful and the AI attacks on multiple fronts and uses armies properly. This makes the game a lot more fun to play - after all, the economic AI problems can be mitigated by simply playing on Grand Admiral.

One economic issue I have to point out is that there are still bugs in pop job priorities - for example, slaves that have bonuses to Energy Credit production are instead AVOIDING Energy Credit jobs. Here's a thread about it, with at least 4 specific examples of coding errors: https://forum.paradoxplaza.com/forum/threads/three-job-weight-corrections-code-included.1470975/

I'd also like to point out that the current "Random" Crisis setting is problematic. With tech going as fast as it does, players will generally have Jump Drives by endgame start even if it's set to much earlier than default. This causes the Unbidden to spawn early in most games. Please bring back a more balanced weighting for Crises, where the other Crises can also be triggered by players. Surely the Contingency should react to a player trying to blow up the galaxy, for example? It's what it exists to prevent, after all. Just make "Random Crisis" a separate setting from the standard, player-influenced setting.

 

[Tsavong]

Played one game now up to 2350. Ensign difficulty, default growth settings, large galaxy , 0,5 ,planets. Some thoughts about the ai in that game:

1. The ai empires focus heavily on corvettes, alloy production no matter what ethics they have. That leads to the ai exterminators, purifiers etc being dangerous to me early game and makes early conquest more difficult, but leads to serverall problems:

- all ai empires do so , so there is not much expansion by any ai since they neutralize each other
- ais dont build reasearch labs and so are soon far behind in sience and no threat anymore. 20 years into this game i could just roll over them

2. I did not see any 1 planet rebellions anymore
3. About 2300 the great khan events started. Great khan ai expanded good, did not suicide himself into an Fe. Ai at this point about 2k fleet strength.

4.2350 and the best ai empire still has a fleet strength of 6 k, while i already have about 1,5 million. Reason : The ai like said above does not build labs...

Overall : I believe the ai improved as its now stable and does not collapse anymore. But its to focussed on alloy production .It should in my opinion especially when its not militaristic or an exterminator focus more on science

 

[HistoricalScore1]

Update on some AI decision making. Seems like the Ally AI makes worse choices than enemy AI. I read that after an AI is vassalified by the player they lose certain bonsues. Wonder if the same is for Ally AI. The enemy was smart and ran away from my 1.4K fleet with their 900 power. A few different times of loading the game, I had the Ally AI suicide into Starports or especially moving to system with and engaging an enemy fleet 2x its power while other fleets were sitting back a few systems away.

The opponent AI (neutral then) shortly after first contact. Was doing what a human would do and going for choke points with construction ships. Even skipping a system or two or lock it in! Love it!

EDIT: Adding more decision making screenshots before I retire this playthrough. I lost anyway to the top empire before turning back time. Played through to see how AI and I suppose now, Ally AI acted. Ally AI is very, very disappointing.

 

[Gilbert95]

I did not get a chance to try the 3.0.3 open beta until that past weekend after the most recent updates were made. The original feedback thread is now locked and I want to provide my feedback. This is not generally about the AI, but my overall experience with pops and the economy on 3.0.3. I hope it is okay to post this here.

One of the key updates in this was changing the clerk trade value to 4, so I wanted to see what could be accomplished by focusing on TV. I played a xenophile megacorp with the civic that boosts TV and species trait that boosts TV. At game start I had a 55% boost to TV. I turned on consumer goods trade policy and quickly realized I needed neither technicians nor artisans to keep my income good in EC and CG. I turned those jobs off as soon as I could build replacement build jobs for them and did not turn any on before 2300. With mineral-free CG production, I also found that space based mining was sufficient to supply my foundries and so have almost no mining districts. Thus most of my workers are clerks or farmers.

To keep the focus on TV, I built a fair number of corporate culture buildings (one or two on most planets) and between that and the trade league policy added with my fed was able to sail through all traditions by early 2300s (note, 1.25x costs slider).

I was only able to make it to mid-game but by then I was swimming in all resources except food because I decided to take the genetic ascension path and cloning vats. But running a food deficit didn't matter when I could easily buy food on the market to cover.

I got the Cybrex precursor so I pushed to finish and get the first section rebuilt quickly to see how pop growth changes did on filling this up (note, default 1.5 and 0.25 on the sliders). It still feels slow, even with pumping immigration buffs.

I didn't get to late game yet, so no comment on how overall empire growth or performance are impacted by pop growth changes. I also don't know how the lack of late game scaling on TV output bonus affects the late game economy, but knowing normal production buffs don't affect, I hesitated to add any, like the new production modifiers that get added onto the capital building with certain techs.

Overall, for the early to mid-game I enjoyed being able to run a different kind of economy, but once I grasped how it worked this way, it almost seemed too easy. However, that may be due to other issues (like tech cost scaling).

 

[hangry]

What we are looking for:

• Is it too easy to out-tech the AI now?

The answer is yes. Playing on grand admiral with no scaling all AIs are inferior in tech to me. It's only 2260, I produce 569 tech and the AI can't keep up. I saw that in every game I played after the last update of the beta patch and it's not like I'm playing with an heavy emphasize on tech in my current game or the games before that.

 

[rubert]

The answer is yes. Playing on grand admiral with no scaling all AIs are inferior in tech to me

This was my impression as I well. I started with F. Xenophile/Militarist MegaCorp but after I had finished Psionic Ascension I decided to trigger AI uprising and switch to the Machine Empire (Driven Assimilators) because AI empires at that point were not a real challenge. Resource bonuses from the higher difficulties help to prop up the AI empires.

Quick observations around 2400 when I decided to call it quit after defeating Unbidden:

• AI is not redeveloping the planets to replace the basic resource districts with Industry districts even with a big excess of minerals/food/energy
• Same applies to replacing housing districts in the planets with excess housing and lots of unemployed pops with most of the other planets full. This was issue mostly with the smaller empires which got stuck with fewer colonies.
• AI not building anything to fill empty districts or building slots even with 10+ unemployed pops and plenty of resources
• AI not upgrading fleets even with lots of stockpiled alloys, at peace and free shipyards. They are upgrading star base defense stations which are pretty much waste of alloys.
• Fleet behavior of AI during wars seems better now than pre-3.0: several fleets attack at the same time even if individual fleet would not be able to take down enemy fleet or starbase.
• During my machine rebellion the AI which took over my old empire split the two battleship fleets into tiny fleets which I was easily able to defeat even after disbanding most of the autogenerated ships because of lack of energy.

/common/technology

Some technologies need to be almost autopicked by AI, namely economy technologies (I suggest to increase the weight of these technologies at least in 10 times, preferably 100). I am talking about technologies which give 20% boosts to miners/technicians/farmers, technology that unlock hydroponics and technologies that unlock resource edicts (capacity_overload, mining_subsidies, etc.).

Quoting this as this is the part I best understood: this would be a huge improvement over the current weights were most of the stuff seems to have fairly equal weights so AI essentially picks stuff at random with slight bias based on the traits of the researcher. Even if in theory that might make empires more distinct it doesn't work in practice if the empire fails because they do not research basic essentially required technology.

As a general feedback I feel like that the Fallen Empires should get buffed based on the difficulty (or if they already receive buffs they should get more) as even vanilla AI on Grand Admiral starts bullying them fairly early and they run out of ships quickly. More ships, more frequent and larger replenishment fleets rather than just when they have barely any left.