Stellaris Dev Diary #385 - AI Benchmarks

[TilloKirov]

The AI must provide a sufficient challenge to be interesting. (Speaking of which, I see a lot of players using mid game scaling - until we get it in a better state, you may want scaling to be off.)

The sufficient challenge consists of two parts: the AI needs to be strong, and player stats shouldn't be excessively overpowered.

As for PVE

Right now, the AI is weak because it hasn't adapted to the new building mechanics. Players are overpowered because "Civilians" are too strong, making researchers a joke. Additionally, the population growth in version 4.0 is completely broken, likely due to genetic ascension being too rapid. Some examples show it's possible to begin ascension within just 3 years.

Trade value is too easy to acquire and the numbers are inflated.

Last, the civic "Genesis Guides" and Individual Machines have been top-tier since its introduction. If you don't balance this stuff, what's the damn point?

LOL, got any actual counterarguments for the balance issues I pointed out? Don't just silently downvote.

 

[Seridor]

I know many won't agree with this, but the other way to fix the game is to seriously reduce the power creep which is introduced all this time with the many DLCs.

Teaching the AIs using the same paths and combos as the player is hard, and probably wouldn't be preferable. so instead we should reduce the benefits of some of those paths and combos.
Some examples:

-leaders effects, I'm sure the AI will not use them as effectively as we do
-anomaly/special event bonuses some of those are so powerful that when i get them I rather restart the game, because large part of the challenge is gone. And most of those are not available to the AI.
-planet specialization-ok this might be controversal- specialization should offer great benefits, but less than it is offering now, because this too adds to the power creep, especially against the AI
-research cost, and novadays mainly tradition costs should be rebalanced. I can and do play with the cost slider, but default should be changed when necessary. Some player chosen traits and combos can lead to very fast ascension which obviously the AI won't do (and shouldn't do).

 

[TilloKirov]

I know many won't agree with this, but the other way to fix the game is to seriously reduce the power creep which is introduced all this time with the many DLCs.

Teaching the AIs using the same paths and combos as the player is hard, and probably wouldn't be preferable. so instead we should reduce the benefits of some of those paths and combos.
Some examples:

-leaders effects, I'm sure the AI will not use them as effectively as we do
-anomaly/special event bonuses some of those are so powerful that when i get them I rather restart the game, because large part of the challenge is gone. And most of those are not available to the AI.
-planet specialization-ok this might be controversal- specialization should offer great benefits, but less than it is offering now, because this too adds to the power creep, especially against the AI
-research cost, and novadays mainly tradition costs should be rebalanced. I can and do play with the cost slider, but default should be changed when necessary. Some player chosen traits and combos can lead to very fast ascension which obviously the AI won't do (and shouldn't do).

Leader effects had been consistently nerfed until version 3.14, but I think they've reached a decent balance state in 4.0. Investing 50 years to nurture a leader is one of the core player enjoyments in the game.

 

[dawedqafdAD]

I know many won't agree with this, but the other way to fix the game is to seriously reduce the power creep which is introduced all this time with the many DLCs.

Teaching the AIs using the same paths and combos as the player is hard, and probably wouldn't be preferable. so instead we should reduce the benefits of some of those paths and combos.
Some examples:

-leaders effects, I'm sure the AI will not use them as effectively as we do
-anomaly/special event bonuses some of those are so powerful that when i get them I rather restart the game, because large part of the challenge is gone. And most of those are not available to the AI.
-planet specialization-ok this might be controversal- specialization should offer great benefits, but less than it is offering now, because this too adds to the power creep, especially against the AI
-research cost, and novadays mainly tradition costs should be rebalanced. I can and do play with the cost slider, but default should be changed when necessary. Some player chosen traits and combos can lead to very fast ascension which obviously the AI won't do (and shouldn't do).

Power creep, the endless argument from every grand strategy game. Experienced players complain about it because it makes the game too easy for them but what they fail to realize often times is that if you were to reduce it significantly, new players would struggle and the AI would remain as they are, incompetent. In other words experienced players will find the result of reducing power creep to negligible while inexperienced players will feel the effects and likely have less fun thanks to the game suddenly becoming harder without their input. Its always better to do it in reverse through optional buffs to the AI, i.e. opt in difficulty rather than forced difficulty.

Reducing excessive power creep is sometimes necessary for game balance, but its best to try to keep nerfs to a minimum when possible, imo of course.

 

[Sotek]

I generally play with mods - especially ones like Guilli's Planet Modifiers that make the AI problems worse - so I'm not a great resource for this in terms of actual numbers.

However, I can mention some things about strategy that can be phrased as an algorithm that I believe the AI does not do, that it should be willing to do, that would make it a lot stronger, so:

1) I specialize worlds based on what I expect to need in the medium-term, and what the world is good at. If it has beneficial modifiers and I have any desire for that resource? I specialize in it, even if I'm doing okay on that. Sometimes this means I end up producing extra minerals! Sometimes it means I produce extra energy. Sometimes it means I've got an insane heavy-duty research planet early because I grabbed a relic world early on. I am going to specialize at least one world for industry and at least one world for research early, at some point I will get a specialized world for unity but often that's not till world 6 or 7 or 10, my capital tends to be a bit of a mish-mash but pre-4.0 it tended to focus on industry and/or science and I will absolutely respecialize it later in the game.
2) I am absolutely willing to run nominal deficits or surpluses in a resource, because I then set up ongoing trade deals to smooth that out. This often means I end up having to buy consumer goods early, or have to buy food, or in 4.0 has meant I have to buy energy, but as long as I am trade-positive (in pre-4.0 it was energy-positive, obviously), I don't care; it means I start prioritizing making up for the deficit a bit more, but when I am developing a research-specialized planet, I will still develop that research-specialized planet even when I am in consumer goods deficit as long as I can make up for it with trade.
3) Likewise, for food, minerals, and consumer goods (and in 4.0, I think this will also apply to energy but I've yet to actually get a real energy surplus in 4.0) I set up monthly trades if I find myself too positive, or if my trade (pre-4.0 this was energy) balance is getting concerning. Ditto for strategic resources; I typically only want to accumulate a couple a month, of those, so when I get more once I get a few hundred or thousand in the bank I set up a monthly sale on the grounds that 'money' is better than a resource I'm not making heavy use of - and I relax this as my production goes up. Minerals I just set my threshold to avoid capping out since I spend them on development a lot, but consumer goods and food I often will aim to have literal single-digit profit (though I don't tune it too much - an AI probably wants to tune it in part based on stockpile size - it should sell less when stockpiles are lower and more when stockpiles start getting higher). The other way an AI could do this is to set monthly buy orders to maintain a small positive balance in anything in production deficit and then adjust the monthly sell orders to keep itself trade-positive and if something is within, say, 90% of the cap sell 100% of the surplus per month.
4) Early on, I will also buy minerals as required to keep my construction going. Later on, when that stops happening and my trade balance starts getting higher, I buy alloys periodically just to use my trade income productively; this might slow down once an AI hits naval cap, but that also presumes it's at starbase cap and has fully developed its starbases as anchorages.

So, TL;DR for my suggestions for AI in this game: They should be less afraid of deficits, as long as they can patch them up with monthly purchases. Specialization is too valuable, stacking positive modifiers is really, really strong, and it usually seems better to go into deficit in one or more resources early on and use the market to mitigate shortages than it is to avoid having shortages at all. Likewise, some resources are only valuable beyond your basic needs as trade goods, but often have relatively easy production; if a planet has bonuses to a basic resource, it can absolutely be worth investing in that basic resource even if you're already selling your excess.

 

[Seridor]

Power creep....if you were to reduce it significantly, new players would struggle and the AI would remain as they are, incompetent.

The main focus should be to teach the AI how to play at least on an acceptable level, right now there must be bugs preventing this.
About power creep: I know there are many like me who prefer slower paced, very balanced games, while others prefer to be powergaming, both are valid. New players could adjust the difficulty settings, but eventually they will learn the game, and once they start to optimize they will need to seriously increase the AI bonuses just to keep the challenge.. and when it gets to 50-100% bonuses the game is already a rush and breaks at several points.. imo..

 

[Carl_Bar]

There are a number of detail problems with the AI. But the biggest problem that i see is fundamentally structural. The current way the AI functions at its core has some major disconnects from how a player actually plays the game.

The core problem is this idea of chasing a target, then when that target is hit, chasing a new higher target. This has several problems.

Problem One: The single biggest problem is that infinite growth in all resources simply isn't desirable.

Everything except Unity, Research, and Alloys exists purely to pay some form of upkeep. That could be pop upkeep, job upkeep, building, station, ship, e.t.c. upkeep. but its all paying for upkeep, and the major sources of upkeep are generally ships, jobs, and pops, (this is a little less true early on as stations and buildings can consume a surprising amount). So the various upkeeps will only scale at the rate your pops and ship count is scaling.

In a perfect world you'd produce exactly as much of the upkeep resources as you need and scale your production of them to exactly match your rate of pop and ship numbers growth. In reality this is completely impractical, there are too many things involved in determining that for players to keep track of, and i doubt even an AI could really do that effectively.

My standards method, (and i imagine most other players approach), is to instead build enough production to meet my needs plus a small excess, (the absolute size of this excess grows as the total absolute upkeep value does).

So long as i've got that buffer and have enough pops to work the jobs i try to build either research or alloys production


Problem Two: An AI focused on ever increasing growth has an active incentive against redeveloping.

Others have raised the problem of the AI doing a terrible job of specialising or redeveloping worlds. Doing so typically comes with a downside of causing your income in one or more other resources to either stagnate, or even get worse, in exchange for better longterm production of another resource. It's a short term negetive in exchange for a longterm gain. But an AI thats permanently focussed on "get bigger now" is incentivised to pursue short-term gains even if it hurts their long-term growth.


My thoughts on how the AI should function

The AI for food, minerals, energy credits, and consumer goods NEEDS to be operating on the basis of "attempt to produce between X and Y times my current consumption". i don't know exactly what X or Y should be, but i'd start aroudn 1.1 for X and 1.25 for Y. Meaning if is below 10% more than upkeep it will prioritise growing that resource on an emergency basis, and if it's at or above 25% it will stop growing that resource. Alloys since they are used for upkeep should have the same formulae and priorities in place, but with the stop on reaching Y part skipped if all other upkeep resources are at or above their Y value. Research should only be grown when all upkeep resources, (including alloys), are at or above their Y value.

As noted the specific values given are just my best guesses at good starting points, the actual values will probably need some experimentation.


Note: For empires using bioships food would have to be treated as a non-upkeep resource. As it's also used to build ships.


You should also (naturally), include some personality bias on which of the two out of alloys or research it focuses when all resources are at or above their Y values.

Unity is the only resource IMO where you can functionally get away with a fixed scaling with time only, as it just doesn;t have the same level of benefits in infinite amounts as research and alloys.


And yes i'm aware "Alloys or Research" is a pretty limited pool of items to run the personality slider along. But thats the nature of stellaris in its current form. The only resources where infinite amounts are useful to any significant degree are alloys and research. So those are your two options to choose a balance between when defining your approach to how you play the game on an economic level.


Naval Cap and Trade needs a bit more complication for their setup, (for naval cap i'd say it should only attempt to grow if all upkeep resources are above their Y values and the unused naval cap is equal to or less than the fleet command limit), but again infinite amounts of either aren't useful.



I haven't hit every problem with this breakdown of how the Ai currently works, (it's late and i want to post the key points before i get distracted), but the two i mentioned are the big ones that don't involve getting down into tiny details in the weeds.

 

[Koopatin]

btw.


while i have all of you here, did not read all patch notes but


is it WAD or BUG that we just integrate gestalt machines instantly into our empire when we conquer them?

Yesterday they wanted to end all life in the universe. Today they just work in my empire like nothing ever happened.

 

[reberuappu]

not setting AI for failure could be a start, for example:

MSI starts with gene clinics
they have no gas production
to "fix" it they decolonize two of their three planets
so they just default in 1,5 years and have massive debuff for 10 years

 

[Jashkar]

I had to take a bit think about what to type here after reading the posts here on the forum. I am not being rude nor disrespectful. I love this game, its unique and there is nothing else out there like it. This game gotten me into modding and learning how to code things. It has been fun to play with friends over the years and playing fun games solo yet with 4.0 I think that fire has diminished. I felt like I been lied to, and I feel like I paid to be a beta tester.

Thinking hard about the content released with Biogenesis leaves to feel that it is unfinished in areas and makes me question some choices on things or feels outright lazy in other areas.

The Deep Space Citadel is virtually the removed Fortress mechanic from the earlier version of Stellaris but instead of being freely built it counts as a ‘megastructure’ with a limit per system. I am fine with those changes but what I don’t agree with is the idea of removing a mechanic only to add it back at a later date behind a expansion without little to no content directly tied to it kind puts a bad taste in my mouth. Because I bought the game when it was a thing and I practically paid twice for it when considering the expansion. You can polish something from before but don’t say its new when it already existed but with a new coat of paint.

The DSC also doesn’t seem to apply skins to it from the different mechanical shipsets, just the stark white hull no matter the shipset. Which goes against the norm for all the megastructures prior if I recall. I’m not expecting a unique model for each shipset given prior megastructures only applied a texture but not even doing skins for them for the mechanical shipsets just screams lazy to me.

The DSC not having a bioship equivalent model also feels extremely lazy. When A bioship empire builds it looks so out of place. Bunch of flesh platforms around a white smooth metallic surface. Now I can understand it if another empire builds them and I built the defense platforms, were just using the structure that already existed, but building the DSC it should look like my empire built it. Also couple of megastructures such as the Arc Furnace don’t even have a bioship skin applied to them It made me stop playing with bioships after I seen that. Visually I like it when things meld together.

The new Fallen Empire Is a wonderful addition to the game but didn’t anyone consider adjusting bioship empires that go down the route of cosmogenesis to use the Bioships of the Hive Fallen Empire instead of the normal Fallen Empire ships?

On a side note, it feels strange seeing the Fallen Empires not using the advance government system that best showcases their ascension paths. It’s a flavor thing but small details do add up over time. I think this was lightly touched on the machine age if I recall right?

The 4.0 update itself. It has felt like I paid to be part of a beta test, this mixed with the stuff above makes it feel tad worse. We are 13 patches in and 4.0 doesn’t feel marginally better. Performance is all over the place. Sometimes are fine other times it randomly gets bad with seemingly no reason why. but with the AI being absolutely braindead and the idea of them being fixed with the possibility of performance to worsen, it makes me question what the point of all the changes was if performance is gonna get worse. 4.0 was said to be a good performance update and right now it doesn’t seem to be one at all. Is it better in some places yeah but I cannot say for certain until the AI is actually functional.

When referring to the update overall, granted this update was going to adjust content from all areas of the game. It was major changes and I’m not expecting the world, I expected some bugs or some exploit because it’s hard to nail down everything in development because were all human. (I hope). But to keep some changes like the reworked leader system for the sake of making leaders simple to use after everyone in the beta hated it was a poor decision. I am glad that system was ripped out and the old one put back in. I don’t know if the system that affected traits was reverted as well, the one whatever position your leader is in effects the outcome of traits of on level up. I do thing that works well thematically.

I really like how with the new planet system I can make very specialized planets and I like the idea of districts swaps or basic resources districts having more usage. Just a lot of things to mess around with. I like to see this expanded upon in the future.

For things addressing the AI. I like to see AI interacting with another Ais. I would like to know what the AI’s goals. Are they wanting to take over X space or want to build a federation with Y. I like to help or hinder these kinds of goals. Allows a bit of roleplay and story creation, but with how it stands now the AI did seemingly random stuff from an outside observer.

I like to have my galaxy be filled with AI to make it feel alive as much as possible. I will typically play on the largest galaxy (1k stars) with max AI. It fills up fast and the AI and I either come into conflict or diplomacy blooms. Usually conflict.

I like to see them actually build megastructures and go down their paths of ascension, in the 3.14 I would see them at least have 3 or 4 get to their ascension paths before the end of the game, and rarely ever see them build megastructures like the nexus. I like to see them do more of that.

I think we who criticize certain things do so because we all love Stellaris...it is an amazing game! Which is why it is SO PAINFUL to see it take certain "bad paths" with confusing innovations that, instead of the refined simulation that has been for years, gamify it and on top of it are sold as being done to "improve performance" that instead worsen...and then we get talks about "changing other stuff" while the game is painfully "on life support" with a huge amount of bugs and problems still on-going and on top we get told, straight in the face, that performances had been improved when most people here can easily find videos proving with a screen-to-screen comparison that the performances have worsened from 3.14 to 4.0!

 

[Leycer]

There is a need to lower the number of ships in the late game.
I practically changed the file to increase the damage, defenses of the ships, stations; in general, I also increased the mant costs to double or triple.
Following with the lowering of the feet size and fleet command to half, there is a moderate lag decrease in the late game.
Should consider a "size" amount of the feet, like 18 Frigate-Corvettes, 10 Destroyers, 10 Cruisers, 7 Battleships, and the titan ships.

 

[OldyGuard]

I think in terms of general feel of the ai is that no matter what difficulty or build they should have a good decent fleet by 2400. If no disastrous context, the ai should be able to build and maintain a fleet size appropriate to difficulty. When I play on normal or captain the ai almost never has a good fleet, and on normal I never see them get above teir 3 Armour or shields nor do they get battleships. I get new players need a crutch but the ai never having a good fleet on what should be the intended standard difficulty of the game (normal) is kind of ridiculous. Now what I define as a good fleet is just a good amount of battleships, cruisers, etc. You have more data than I do so as the developer its up to you to find the right fleet power target for the ai on each difficulty. I also agree tho that the ai should try to "rp", but what is "role play". It should already be meta fundamentally to play into your build so the ai not playing optimally at least a little bit is kind of insane to me. If I play auth militarist then yah by logic sense I should have a bigger fleet than if im not. That doesn't mean play like an idiot and build fortresses and precincts on every planet lol. or if I am spiritualist yah in theory I would have a lot of priest or unity workers than other empires but why would compromise my empire by only building temples and not anything else?

Love the game but the normal empire AI has never been in a good state it seems. I get it. Lots of different variables for the AI to work with but the AI should at least be smart enough to maintain an empire that I can tag over, look at their planets and say "okay this makes sense" rather than be confused on whatever its trying to atm. The ai building good planets can also give new players in game examples on how to build planets.

 

[Basti1981]

The main focus should be to teach the AI how to play at least on an acceptable level, right now there must be bugs preventing this.
About power creep: I know there are many like me who prefer slower paced, very balanced games, while others prefer to be powergaming, both are valid. New players could adjust the difficulty settings, but eventually they will learn the game, and once they start to optimize they will need to seriously increase the AI bonuses just to keep the challenge.. and when it gets to 50-100% bonuses the game is already a rush and breaks at several points.. imo..

This is my major problem with this game atm. The AI is not able to play the new system in any reasonable way.

After 2 games on admiral, my experience was that after the early game, where they AI gets buffs, the challenge I normally had before 4.0.X just imploded.
The AI did not build big fleets, they fell short in research and economy, this lead to an underwhelming diplomatic power and me feeling like a cheater in a galaxy without tension.

The grey tempest or the crisis, can only be stopped by me, everybody else is completely helpless. I am alone and not member of a mighty community anymore.

And, as mentioned above, in 3.14 all this was better and somehow working. Without the challenge my motivation to play it is down to 0.

Since I am a stellaris player since the first hours I remember how long it took to get the game on the level of 3.14 (with a reasonable performance btw).

Now I feel like a kid with a broken christmas toy:
You were looking forward for weeks to unpack it at christmas day.
The dissapointment set in after you saw its not working
Your parents tell you that it will be fine in some days when the shops open again

The good thing is:
Now I am grown up, I have more patience.
The bad thing is:
From the development history of stellaris I expect it not days but month, until my toy works like expected...

 

[Troyen]

Power creep, the endless argument from every grand strategy game. Experienced players complain about it because it makes the game too easy for them but what they fail to realize often times is that if you were to reduce it significantly, new players would struggle and the AI would remain as they are, incompetent. In other words experienced players will find the result of reducing power creep to negligible while inexperienced players will feel the effects and likely have less fun thanks to the game suddenly becoming harder without their input. Its always better to do it in reverse through optional buffs to the AI, i.e. opt in difficulty rather than forced difficulty.

Reducing excessive power creep is sometimes necessary for game balance, but its best to try to keep nerfs to a minimum when possible, imo of course.

New players are mostly not playing with all the DLC. They might have a subset (if any) or whatever is newest, but they won't have nearly as much access to civic and origin combos or pure power buff mechanics (like astral threads's physics boost) as experienced players. The exception would be players on the subscription, but the very steady stream of DLC recommendation threads on reddit or the forums suggest quite a few players aren't using the subscription. And those that are aren't generally aware of the most powerful combos you can make (outside of being told them by experienced players).

Also, most of the time when power creep is introduced, the AI is not updated to fully take advantage of it (AI is one of the last things worked on in a patch cycle and almost always doesn't get enough time). The other systems in the game are also not updated to account for the new power swing. Things like the mid and endgame crises. Remember the 3.X days, especially pre-3.11 tech rework, when a medium difficulty AI could stomp the crisis on their own using default settings (2300/2400/2500 1X crisis)? Even these days, Fallen Empires are considered to be pushovers for most experienced players by the time they're scripted to awaken, with the exception of the brand new one.

If you take the approach of always buffing and never nerfing, then you're either setting yourself up for a mountain of balance adjustments on the other game systems (like the crises, space fauna, leviathans, etc.) or more likely slowly rendering them obsolete and meaningless over time. The "always buff" approach always leads to a weaker game overall as all consequences get watered down, which makes aspects of your game's systems become meaningless. For example, take a look at the current role of the three base strategic resources in 4.0. At the moment you can produce so many that they're pretty undervalued, especially compared to prior iterations of the game. Finding a space deposit of one used to be a cool bonus in 3.14 but now is as meh as finding a low-energy star.

Iirc, the 3.11 technology nerfs were relatively popular. I think one of the DDs said 70% of the people who played the beta and took the survey liked the changes, even though it came with a big set of nerfs to research, because it corrected the power crept pace of the game and made the middle part more relevant again. Of course, Machine Age came and promptly stomped on that balance again, but for a while those mid-tier ship upgrades mattered and you weren't skipping T2 or T3 components because the next one was ready to be researched in a few months.

Finally, I think the default difficulty for new players is Cadet, which already gives new players bonuses. At some level, new players do have to learn the game. The new players asking for advice on how to handle their 10K fleet vs an AE in 2420 aren't going to be saved by any amount of power creep that most experienced players would find reasonable, they're just going to have to learn the game expects them to be stronger than they thought (or that they can adjust their galaxy settings for a slower-paced, more peaceful game).

 

[Koopatin]

There is a need to lower the number of ships in the late game.
I practically changed the file to increase the damage, defenses of the ships, stations; in general, I also increased the mant costs to double or triple.
Following with the lowering of the feet size and fleet command to half, there is a moderate lag decrease in the late game.
Should consider a "size" amount of the feet, like 18 Frigate-Corvettes, 10 Destroyers, 10 Cruisers, 7 Battleships, and the titan ships.

it could be the number of fleets, not ships in general. When I moded the fleet size max, I also had good effects. The AI seems to make a lot of micro fleets when the first fleet is full. Also, when they can not replenish fleets (same as the player bug), they have a lot of single-ship fleets, and they never seem to merge fleets but have dozens of fleets with their own target.

There is also still a bug/problem when the AI decides to attack a station and changes its mind every few days to months. should also be lag fodder.

Also, when I changed into an AI to see something, there were fleets en route to a system that flickered the AI target on and off; no clue if that means it does change its mind, but the system was not connected to the fleet system, so it was never able to reach it and just stood there.

Also, also, there is the fact that a fleet that decides to attack a system and changes its mind because an enemy fleet arrives just stands there, not fleeing, just waiting and not knowing what to do. I could guess that still costs calculation.

@Eladrin, Do you think that could be worth looking into for lag purposes? i mean it should be looked into sometime anyways because those mini fleets that never merge and do not know what to do are bad for many reasons.

 

[Sneaky_minon]

it could be the number of fleets, not ships in general. When I moded the fleet size max, I also had good effects. The AI seems to make a lot of micro fleets when the first fleet is full. Also, when they can not replenish fleets (same as the player bug), they have a lot of single-ship fleets, and they never seem to merge fleets but have dozens of fleets with their own target.

There is also still a bug/problem when the AI decides to attack a station and changes its mind every few days to months. should also be lag fodder.

Also, when I changed into an AI to see something, there were fleets en route to a system that flickered the AI target on and off; no clue if that means it does change its mind, but the system was not connected to the fleet system, so it was never able to reach it and just stood there.

Also, also, there is the fact that a fleet that decides to attack a system and changes its mind because an enemy fleet arrives just stands there, not fleeing, just waiting and not knowing what to do. I could guess that still costs calculation.

@Eladrin, Do you think that could be worth looking into for lag purposes? i mean it should be looked into sometime anyways because those mini fleets that never merge and do not know what to do are bad for many reasons.

I'd like to note that I've seen AI-behavior that backs up the notion that the lagging might be in sigificant part caused by the AI being unable to make up its mind on where it's fleets need to go.

turning around, moving to/from the edge of a system, etc.

so both laggy and a detriment to the AI's functioning.

 

[Dagorad]

Fewer fleets and starbases would have so many benefits that you could make a whole list:
- less lag resulting from fewer things on the map
- probably less lag resulting from the fact that the AI has to control fewer things and take fewer things into account
- greater strategic aspect (building "anchorage number 45" is not the peak of strategy, nor is mass production of fleets)
- better visual effects (battles of several fleets or several fleets overlapping each other do not look good)
- reducing the number of things for the player to click, at some point it is certainly tiring for many to expand another starbase or control another fleet
- some ascension perks would gain in value

Each subsequent fleet, each subsequent expanded starbase should be more expensive, burden the player's economy more, etc. (I hope that "trade" in its current form is supposed to stimulate it, but it certainly requires some balancing). Then there would not even be a need for artificial limits (the number of ships could be tied only to "soldiers" jobs), because the growth curve would flatten itself. Meanwhile, I have the impression that it is quite the opposite, at some point you fall into a spiral where you have more and more and more of everything (even "+5%" technologies boost this), growth curve simply shoots up. This would also have other benefits, it would slightly reduce the gap between those who do better and those who are left behind.

Of course, there are also problems, such as federations and vassals as a way to bypass the flattening growth curve (as is the case with the growth of pop(s)). The problem is even bigger because vassals and federations are things from DLC (this is a mental shortcut, but you know what I mean), so simply nerfing them is out of the question (I am writing from the perspective of sales and players' complaints). But I think it could be partially balanced by other bonuses for players who are developing their empires and not looking for mechanics to bypass the flattening growth curve (as long as they are not too big bonuses, it would not be about creating a new meta or making it not worth having vassals at all)

 

[fourteenfour]

Hi everyone!



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.

​

Gas lighting 101.

Sorry, but performance was a key element promised in this new version of Stellaris and frankly besides being tired of the all the bugs, some of my favorite empire types still not truly working, the one thing I can say is, the AI does not feel any better than before.

Really comes down to the fact you have created an economic model that has too many dependencies for your team to model properly. Resources just seem to exist to have them and a few exist just because they always have. Worse is that many empire types require use of the market to properly function and players can easily exploit the market in ways the AI code cannot. Using your own example of supply thresholds, players will have similar goals but also at the same time willingly take on deficits to reach another goal to gain an advantage now.

I would love a resource pass and normalization of all empires to similar sets of resources. My one big annoyance has always been consumer goods which just add a layer of complexity to the economics with no real payoff. Now that trade is far more important in game between trade and amenities all the needs consumer goods held can be managed through them. Like special resources which tend to trap the AI when it does upgrades, do we even need them anymore? I do like the end game resources that can be used for non generic purposes but motes/gases/crystals are noise now

Anyway, I am down to a game start once a week if that now as I am just waiting for the flow of patches to slow to the point it feels like the game is back to the relative state where it was fun to play.

 

[Turk Pasha]

Gas lighting 101.

Sorry, but performance was a key element promised in this new version of Stellaris and frankly besides being tired of the all the bugs, some of my favorite empire types still not truly working, the one thing I can say is, the AI does not feel any better than before.

Really comes down to the fact you have created an economic model that has too many dependencies for your team to model properly. Resources just seem to exist to have them and a few exist just because they always have. Worse is that many empire types require use of the market to properly function and players can easily exploit the market in ways the AI code cannot. Using your own example of supply thresholds, players will have similar goals but also at the same time willingly take on deficits to reach another goal to gain an advantage now.

I would love a resource pass and normalization of all empires to similar sets of resources. My one big annoyance has always been consumer goods which just add a layer of complexity to the economics with no real payoff. Now that trade is far more important in game between trade and amenities all the needs consumer goods held can be managed through them. Like special resources which tend to trap the AI when it does upgrades, do we even need them anymore? I do like the end game resources that can be used for non generic purposes but motes/gases/crystals are noise now

Anyway, I am down to a game start once a week if that now as I am just waiting for the flow of patches to slow to the point it feels like the game is back to the relative state where it was fun to play.

I agree that they did promote performance as a key takeaway from these updates, now the performance is worse, they must fix it... a player with decent computer should able to run a small-medium size galaxy until the late game without lags. I am lagging by first 50 years...

 

[fourteenfour]

You know, a person doesn't have to be a developer to have every right to disdain both the Paradox testers and their management. Just look at the release of version 4.0 and their lies on the eve of this release, when many people on the forum were screaming about the extremely sad state of the beta.

They deliberately released an unfinished product, which means the consumer has every right to disdain.

Specifically the testers, if they even exist, because it seems to me that testers are only us, maybe incompetent or not having enough people. But the result of their joint work is a shameful 4.0.

That's why your attempt to defend them looks ridiculous.

I actually agree with you and ask the community, how do they defend this release given we have THREE HUNDRED AND SIXTY TWO bug fixes since release. Not one, not ten, not a hundred, but we are now over three hundred. This does not include balance changes, changes listed as improvements, AI, or UI/UX

For convenience I attached the link to the consolidate page of patch notes, its actually to large by any means to cut and paste just the bug fixes into a posting.

Patch 4.0 - Stellaris Wiki

stellaris.paradoxwikis.com