i fear that "helping" things alleviate might be too late at this point tbh
- 2
Finally uninstalling this game after many years and no longer buying dlc.
- Thread starter ChillB
- Start date
-
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.
- Status
i fear that "helping" things alleviate might be too late at this point tbh
Well, the game still has 15-20k daily according to steam charts, about the same as EU4, and higher than CK3 which just came out. It's still holding steady despite the terrible state the game is in currently. Certainly not too late to try and fix the game. If the AI wasn't in such a sorry state the game would likely catch HoI4 in terms of daily player base.
But they realllllly need to stop focusing on big new additions because it just adds more layers that the AI has to navigate, and it can't even deal with the features we have now. Adding new major gameplay additions is paradoxically making the game worse at this point, as each expansion further exposes the flaws in the AI. The first thing they need to focus on is getting the sector AI to understand how strategic resources, resource stockpiles, and building levels play with each other. Until this happens, you can never turn over systems to a sector to manage because the sector AI will crater your economy in less than a month without fail. The inability to utilize sectors is what has caused the massive micromanagement headache, and also what spurred the band aid system of empire admin/bureaucracy that allows to manage as many systems as you could conquer; this was basically an admission that they could not get sector AI to work and that the only way to allow players to actually win a game was letting them control every aspect of every planet. Of course, on anything but the smallest galaxies this means micromanagement hell in the mid and late games. This should be priority #1- because fixing this not only means that the game isn't micro hell for the player, but it also means the opposing AI empires economies will actually be functional and they might pose some sort of challenge beyond the first 20 years of the game.
Last edited:
- 5
- 3
This shows the difference in play strength. For me, even the default 2.8.1 AI isn't helpless after 20 years, although it eventually becomes so. As for 1.9.1, I find even its "fair", no-AI-bonuses level to be genuinely quite challenging.
- 9
What confuses me about the AI is that Stellaris' economy is actually the kind of problem that a computer should be quite good at.
There are only two end-goal resources in Stellaris: alloys and research. Everything else in the game is just a means to those two ends. From food to amenities to strategic resources, each other resource just feeds into the production chain that ends with either new technology or new ships. (To a limited degree this is also true of unity, but tradition trees honestly feel like a demo of the actual system they plan on implementing someday. That mechanic is too short to matter much.)
So the only choice the AI has to make is what balance to strike between alloys (shipbuilding and starbases) vs. research (new technology). Or, put another way, do they build more ships or better ships? And that seems like something that you could honestly just code in for each AI personality. Erudite Explorers? You're 40% alloy/60% tech. Honorbound warriors? 80/20.
From there it's just an optimization problem. What is the minimum amount of every other resource you need to produce the absolute maximum number of alloys and research. That's a formula. There's no judgment involved, it's just math. The computer should be great at that.
Ultimately that's what confuses me. I completely understand the AI being bad at the parts of the game that require judgment. Diplomacy, for example, makes perfect sense as a sinkhole for the AI. But while the 2.2 economy is very complex, it is just a massive optimization system with only one choice involved. Once you decide how to split your focus between alloys and research, the rest is just math with a whole bunch of variables.
- 6
- 6
- 1
The AI is not designed to be good at refining other materials, or at handling upkeep in general. It used to be that this wasn't an issue - minerals replaced alloys, science did its own thing, and IIRC strategic resources mostly just granted buffs to very specific things without being a necessity for upkeep. The AI could handle 'make lots of XYZ, spend lots of XYZ'. It can't handle 'make X but retain a stockpile but also turn some of X into Y and some other X into Z to fuel making more of Y'.
- 3
- 1
I definitely get that. I'm just curious what makes it so hard to update the AI for this new system. Upkeep and refining new materials are all just variables. Like the example you give. I would disagree that an AI can't handle that. In fact that's exactly the type of formula a computer is best at solving.
Making Z costs materials X and Y. Making Y costs materials X and W. For any given month, solve to maximize Z and minimize W, X and Y. That's just math. The more steps and variables you add, the more complex it gets. But it's still just math, and fairly straightforward math at that.
The 2.2 economy is super high complexity, with layers of supply chains and lots of variables, but it's a low depth optimization problem at its heart. Of course I don't disagree that the AI hasn't been programmed to do this. I'm just curious what the problem is.
Last edited:
- 9
- 3
2.2 Sure as heck killed the AI, i hope some have seen their builds on planets. Conquered a capital, -30 amenites and missing building slots, i conquer colonies with like 5 pops, all districts built on them jumping my admin cap extremely high. Yeah ai with the 2.2 update are a huuuuuge joke
I wish Paradox could revise their design guideline. I want to complain about complexity in Stellaris. Ideally, a game should be easy to learn, hard to master but Stellaris is the opposite. There are a lot of shallow mechanisms in Stellaris and players have to take sometime to figure out how they work but none of these mechanism is deep enough that player usually find optimize solution quickly. Best example is tradition system. It has potential to synergy with empire ethic and ascension perk. It could have combo bonus like EU4 idea group policy or influence diplomacy like CIV5 ideology. However, tradition system ended up stopping at irrelevant flat bonus, wasted all of its potential. Game designer of Stellaris should decide what are the core mechanisms and deepen core mechanisms over time with extensible framework. They should stop adding more mechanisms and focus more on the interaction between each core mechanism.
- 16
- 2
- 1
To be fair, it's up to the Dev team to know what is a priority to fix/improve, and how to do it in a relatively balanced and efficient way.
Administrators are abysmal, they are just a way to totally ignore empire sprawl for a very little investment (something like 2 or 3 dedicated worlds and a bunch of crystals) They should have been nerfed long ago, but Pdx keep ignoring g everything related to balance, sadly.
Espionnage is a nice idea, for sure, but it's really not the time to add such a feature in the current state of the game. The devs shoot themselves in the feet thinking that more features is a solution to the game problems. Even more if they release it without beta test before, and just let it half finished like every time.
- 11
- 1
Stellaris is a good game, good value for money, and will give you many hours of pleasure. Quite easy to sink dozens or hundreds of hours into it.
The killer though is that it has potential to be a GREAT game. It is so nearly there. If the issues around player experience, micro and dysfunctional AI can be addressed, the game will be so much more than it is now. That is the tragedy.
Nobody expects an 'intelligent' AI (lol), but we do expect a functional one that can at least handle the game mechanics. If planetary and sector governors could be made to work as intended at game launch, and the player was reduced to controlling the equivalent of one sector (a demesne), without feeling like they shot themselves in the head, that would be amazing.
From a player experience/UI perspective, the biggest missing design element is that the game never transitions the player from a planetary governor to a sector governor to an empire governor. You are always a planetary governor, for every planet you gain.
The killer though is that it has potential to be a GREAT game. It is so nearly there. If the issues around player experience, micro and dysfunctional AI can be addressed, the game will be so much more than it is now. That is the tragedy.
Nobody expects an 'intelligent' AI (lol), but we do expect a functional one that can at least handle the game mechanics. If planetary and sector governors could be made to work as intended at game launch, and the player was reduced to controlling the equivalent of one sector (a demesne), without feeling like they shot themselves in the head, that would be amazing.
From a player experience/UI perspective, the biggest missing design element is that the game never transitions the player from a planetary governor to a sector governor to an empire governor. You are always a planetary governor, for every planet you gain.
- 8
- 1
Shallow is the word that always comes to mind for me. for all paradox games so far. and i remember often enough a lead said they laid the groundwork for great stuff they could add later.... They never do. and i still think it is because of the DLC model, nothing can be to deep because every DLC has to be optional.
saying that i still somewhat enjoy stellaris but i stop playing every game long before crisis or awakened empire because i know long before that if i would crush them or would have an even fight and the fun stops often enough 50 to 75 years before that and even if i force myself to "the end"it would be just a game of whack a mole with the fleets.
- 9
- 1
Victoria II is perhaps the deepest PDX game - it is so deep that for some nations World Conquest is genuinely impossible (horror!) - although, mind you, it’s possible even for Greece or Luxembourg.
It's worth noting, however, that Paradox announcement of espionage was met with a very positive response in the forums. We grumblers are a minority.
Here we come to the question of profitability (nerfing this would bring no profit while annoying worse players - in fact, there was a thread soon after the expansion got released accusing Paradox of making conquest and wide play completely unviable with this "bureaucrat upkeep" nonsense) although it’s not like nerfing Bureaucrats or Enforcers would be that costly. Unless internal PDX mechanisms are a Byzantine Bureaucracy.
It's worth noting, however, that Paradox announcement of espionage was met with a very positive response in the forums. We grumblers are a minority.
Last edited:
- 2
- 1
Like I said earlier. Paradox chases what gives them money, because that's what shareholders want. Making a functional AI doesn't get them as much money as making a shiny new DLC does. Thus, Stellaris AI is continually piled with new features it doesn't understand how to use, instead of the devs taking a six month period to overhaul the AI to deal with the new economic system.
You say it's easy, and if it was as simple as 'build x building, get y' you'd be right. But you also have to train the AI to handle jobs and their priorities, move around pops, specialize planets (with their own modifiers that might make them more useful in certain ways than others), split alloy production between starbases and outposts and ships of all types... you get the idea. The system ends up too complex for the AI to handle because everything is randomly generated and they're not guaranteed anything (well, outside of difficulty modifiers but that's different).
- 2
I remember when the forum was almost in rebellion a few months ago regarding the state of the game. Many players were upset and the movement was picking up steam (no pun intended). In swooped the new community liason and a few developers chimed in and hinted at some fixes coming down the pipe for the shiny new species pack.
The necro fluff was released, there was some minor improved performance and some pathing may or may not have been resolved with the AI (not entirely sure?).
People ate it up. The forum angst subsided and the Stellaris development team could safely take a couple months off for the Holidays. Now we fast-forward to today and I visit the forums and see that many of these problems still persist and that the majority of people that were up in arms that would not have purchased the DLC "Unless the game was fixed" have the Necro pack in their forum badges.
Friends & colleagues....this game is not 'fixed'. I am not even sure I see a roadmap of how to get to that point from the development team? This is an important step and something the community needs moving forward, not platitudes or empty assurances. An actual plan to get this game to where it is going...it has literally been years at this point.
What is the plan?
The necro fluff was released, there was some minor improved performance and some pathing may or may not have been resolved with the AI (not entirely sure?).
People ate it up. The forum angst subsided and the Stellaris development team could safely take a couple months off for the Holidays. Now we fast-forward to today and I visit the forums and see that many of these problems still persist and that the majority of people that were up in arms that would not have purchased the DLC "Unless the game was fixed" have the Necro pack in their forum badges.
Friends & colleagues....this game is not 'fixed'. I am not even sure I see a roadmap of how to get to that point from the development team? This is an important step and something the community needs moving forward, not platitudes or empty assurances. An actual plan to get this game to where it is going...it has literally been years at this point.
What is the plan?
- 17
- 2
There was some improvement with the Prethoryn, some clearly irrational behavior the AI started to do after the "rework" was no longer present (although the AI are still too timid), and the team asked for feedback, with saves and descriptions. Plenty was given, but not reacted on so far. Presumably, people are willing to wait whether the feedback would be acted on in the text patch - for all complaints about how the people here are so whiny, they are quite patient.
I haven't purchased Necroids, for one, although I was impressed enough with them even asking for feedback I purchased Federations with Steam Wallet money. It appears that I was too charitable, but only the next patch will prove it. If the Unbidden/Contingency are still nonfunctional, it will be a clear proof that the new stance was a fake.
I haven't purchased Necroids, for one, although I was impressed enough with them even asking for feedback I purchased Federations with Steam Wallet money. It appears that I was too charitable, but only the next patch will prove it. If the Unbidden/Contingency are still nonfunctional, it will be a clear proof that the new stance was a fake.
- 3
I remember getting buried in "Respectfully Disagrees" from people who didn't even comment (some of whom didn't have any posts to their account at all) when I complained about the parlous state of the game a few months after 2.2. I feel satisfied that my attitude toward the game at that time was the correct one, and still is today.
- 7
- 2
- 1
I really wish 2.2 hadn't made Sprawl so visible. Its always been a part of the game, its amazing how being visible made people freak out about arguably its weakest incarnation.
- 4
Another topic: how presentation affects player attitude. Civ5 is famous among "analytic" players for ending up with an overly punishing escalating tech cost per city (5% additional tech cost per city), but it's well hidden. Now, it's not a good thing - mechanics need to be clearly presented. But it certainly prevents player panic.
Last edited:
- 1
- Status