Stellaris Dev Diary #141 - Exploring the Future

[SeekingEtermity]

Archaeology is too cool an idea to not be in the game, but honestly, at this point I'm severely tempted to just mod the relevant Scientist trait out of existance due to its sheer terribleness. I've written about it before, but I'll collect my thoughts here, in a subsequent post.

I really want to make archaeology a major part of Stellaris, something that actually deserves its own scientist specialization and so on. It's largely a game about exploring an old and life-rich galaxy, after all! The exploration in Stellaris is one of its key differentiators from practically every other 4X/GSG out there, be they space-based or not.

With that said, here are my thoughts on the current problems with the Archaeologist trait:

• It currently only applies to Precursors instead of applying to all the other "some old civilization was once here" anomalies.
• It only applies to anomalies, not even to the special projects they reveal.
• There are only six anomalies per percursor, so the trait is literally only ever useful six times.
• The bonus isn't very big. "Carefree" is 70% as effective at precursor anomalies (and is outside of them, too), but more importantly, amonaly research time for hard anomalies scales roughly exponentially with level difference (doubling ever two levels between -3 and -8) so the +50% research speed (multiplies the research time by 2/3) is only slightly more effective than gaining one additional level on those six anomalies. For example, a -5 without any modification to anomaly research takes 1080 days, while a -6 with Archaeologist takes 960 days. That's awfully minor, and the high level of the precursor anomalies means you *need* to level up your Achaeologist (doing non-archaeology things) before tackling the precursors or it'll take so long you'd be better off with a random level 3 scientist.
• While it's technically beneficial (unlike negative traits), it's possible to start with it, polluting your leader pool or even your starting explorer with a nigh-useless trait in their sole slot guaranteed to have something beneficial. In most cases I'd rather the scientist have nothing and be allowed to gain two traits from level-ups, as that has a better chance of producing something useful. In other words, it has a huge opportunity cost.

Things that would make it less-awful (pick any combination):

• Repeatable research projects to study precursors once you find their homeworlds, providing a continuous stream of minor but non-trivial rewards. Archaeologists could complete these projects much faster, or simply be necessary to research them at all.
• Far greater numbers of precursor anomalies, either because there are more precursors and/or because such a massive and important empire somehow managed to leave more than seven (counting the homeworld) signs of its existence (each)…
• Give Archaeologists some substantial utility related to ruined megastructures.
  • Gain general research from them
  • Gain specific tech progress from them
  • Make it take less time to repair them
  • There are lots of options!
• Give Archaeologists some substantial utility related to tomb worlds.
  • Find cultural remnants (Unity!)
  • Research the society that destroyed itself (Society!)
  • Study the lessons of the dangers of existential threats (Influence!)
  • etc.
• Give Archaeologists provide a bigger boost when assisting research on any homeworld (which will at least give them some special utility while waiting until they have the levels to tackle level 6 - level 10 anomalies). Optionally, make it extra-good for precursor homeworlds
• Give Archaeologists a minor boost to anomaly research on non-precursor anomalies, like a mini-Carefree. Alternatively, give them their full boost on any archaeological-themed anomaly, regardless of whether it involves precursors.
• Make Archaeologists faster at precursor-related special projects, in addition to anomalies.
• Make the bonus much, much bigger (like, +200% or simply "as though three levels higher") so that even if they're only useful six times, they're actually meaningfully more useful those six times than any random scientist who is merely slightly more experienced.

Archaeology really seems like something that could be useful throughout the game. After you've exhausted all there is to learn about your homeworld (because that's ever going to happen), and your local precursors (because people who could build planet-wide cities and ringworlds and such aren't going to be interesting enough to take more than, oh, a decade or so of research), and random other events and worlds scattered across space (seems like there ought to be a lot to learn from tomb worlds...), maybe they could cease to be specially useful... but until then, they should have some real impact.

 

[endymon]

My personal gripe is that ship designs need to be more consistent over time. It really breaks immersion for me when I can upgrade missile frigates into rail-guns and then later on upgrade them again to fire energy weapons.
When the only thing consistent about a ship is its class name, that just destroys it for me. There is nothing more to "upgrading" my ships than opening the designer, swapping EVERY part out at my whim into whatever my military needs happen to be at that exact moment, or whatever technology happens to be the most advanced at the time. Then press the upgrade button, and its done shortly. Far quicker than building brand new ships.

For all its faults, I really thought the system in Sword of the Stars 2 made a lot of sense. Maybe not the prototyping aspect, but the upgrading a ship's systems within the same weapon type was fantastic.

 

[KingAlamar]

None of those things are relevant to the problem outlined by the previous poster. It wasn’t a general complaint about micro. The complaint was that you must frequently resettle pops to avoid having your economy crash. That is not a problem that actually exists. It’s entirley possible to play a game from start to finish without ever resettling a single pop.

If it matters my first game under 2.2.0 was as an Egalitarian. For RP purposes I chose not to resettle. Seeing as I was playing against vanilla AI though it didn't really matter. In my next playthrough [vanilla machine empire] is where I felt the power of resettling

 

[JosunUrashima]

@grekulf

For sectors we want to give players better control and automation, so that sectors can better help alleviate the need to micromanage very large empires. We haven’t decided on the details yet, but the ability to shift planets between sectors seems like something we would want, as would the ability to give a sector monthly subsidies for its development. As always, your thoughts are very valuable to us, so we would love to hear what it is that you want the sectors to do for you.

.....

wasn't that sort of how the OLD system for sectors worked?

Backgrounds: Split up some civics into backgrounds, and add more backgrounds.

• Civic flavour: Spend more time on making the civics feel more unique and fun

• Institutions: Define which institutions make up your empire’s internal departments (such as Diplomatic Corps, Xenology Bureau etc.), and their funding, size and power.

• Espionage: Intel to determine how much you know about another empire, spy actions, cloaking, sabotage & general mischief.

• Religion & Cults: Similar to factions, cults could appear in your empire during certain circumstances. Spiritualist empires would most likely have “imperial” cults. Worship of powerful entities etc.

• Archaeology: Explore the ruins of ancient civilizations

• Subject contracts: Allow overlords to better customize what type of subjects they have, tribute levels, benefits to subject etc.

• Federation depth: Allow federations to level up, have different election types, taxes etc.

• Galactic Council: Create a sort of a ‘space UN’ with galactic politics and diplomacy

• Primitives: Allow for more interactions with primitive pre-FTL species

• 
  
  ALL OF MY YES FOR ALL OF THESE!

Now just make sure that when you do a big diplomacy update/expansion, you make sure that espionage, subject contracts, federation depth, and galactic council are all part of it. Might as well do it all at once.

 

[Keizaal]

About the Galactic Council. Would it be possible that there could be more than one Council in certain situations or that its decisions are not so absolute? Because in Civ games I have always liked the World Council mechanics, but I've also always hated how ridiculously absolute the Councils decisions are despite the fact that even in real life there are those who ignore such decisions or have just outright left UN in protest to some decisions.
So could it be possible to ignore Galactic Councils decisions (naturally resulting in opinion malus) or even leave the Galactic Council and possibly form another one with other nations that are discontent with the current Galactic Council and its decisions?

 

[Kryndude]

* Fixed pre-sapient policies disappearing if you purge all of them from your space, making fulfilment of certain faction demands impossible, because what are the screams of the innocent next to that sweet, sweet influence?

You can't purge pre-sapients due to bug, what is this talking about?

 

[jady]

Kick sectors out. They are in the game since 1.0 and never really worked.
Give as instead planetary focus.
When i can give a planet a science focus with one click, why i need a sector anymore? Sectors are total unnecessary and a burden.
Make this build orders modable and the players will optimize them for you so no dev ressources needed later.
Win Win for everyone!

 

[Col. W. T. Philmore]

Kick sectors out. They are in the game since 1.0 and never really worked.

in 1.9 they were pretty good

 

[jady]

No they were not good because of the
mechanics. Sector has its own ressources so they can starve for example. And they give you only 75% ressources back.

When my planets are really my planets, all ressources come and go in one pool. This would be much better, because you know they are my planets and should give me all ressources not 75% or whatever.

And with planetary focus seting the management will be easier.

 

[Weyird]

I like everything being considered for the future!
The only thing that's missing as far as what I want is a more interesting and fleshed out first contact process.
"We have discovered intelligent alien life! This is the greatest discovery of our race's entire history!"
You click the notification away, maybe look to see what and where they are, and continue on.

 

[Aevums]

Small idea for pop growth migration management.

Problem is too much micro management, but it doesn't make sense if pops can just magically transport to different planets for no cost.

New edict: Migration Subsidies. Costs a bunch of energy, makes pops able to migrate from planets with unemployment to planets with jobs entirely on their own, as if you moved them manually.

 

[xeleth12]

I hate to be like this but: It's all nice and everything, love the new features and how you fix the problems but what about the performances and the game being absolutely unplayable mid/late-game for a lot, a lot of people?
The lag/stuttering is absolutely massive, even when playing on the smaller scales and having a good machine.
At this point I feel actually bad for buying out most of the expansions (example the War in Heaven features) since a lot of those features that come in the late game...well I can't play the game anymore at this point. It takes up to 10 seconds in real life for one second in the game. This is a problem, and a big one.

I know it is a complicated issue and comes from a lot of sources and would probably take up a lot to fix/improve but I would much like seeing it adressed rather than implimenting new features (that are mostly broken when coming out and stay that way for long) and especially ones that add to the problem. I'm thinking about the enormous amount of patrolling fleets to protect trade by example (wich is an absolutely ridicule system in my opinion and should be completely revamped, it shouldn't be tedious in anyways and having so many fleets moving around obviously add to the performance issue)

It seems like the performances issues will never be properly adressed. Unfortunately I wont be playing this game/buying stuff from it anymore despite loving it, not until I can actually play the god damn game.
With love~

 

[shandou]

loving the proposed changes but please please dont forget about eye candy. more background images for systems especially for those in nebulae

these go a long way in terms of flavor and dirt cheap to make!

 

[Serenity84]

An obvious one, especially without Xenocompatibility, is for species to grow mostly proportionately to the number of them on that world.

Some disproportionate growth for minority pops is fine. Otherwise immigration becomes less useful because the new pops get drowned in an ocean of your own. The issue was that minority pops were all that grew. Instead of doing some catching up and then slowing down

 

[Ikael]

Wow, there's a lot to ponder about. First of all, thank you to the devs, you seem to be aware of the game's most pressing needs, so best luck to Grekulf and the team!

And now, for the rest:

Regarding short term goals

- I do think that automation could work better at a planetary level. Building queues, planet templates and the likes could help to offset a lot of micromanagement problems
- The Sector true potential lies on the political game and the sense of a "galactic geography", rather than being an effective tool of automation, me thinks
- I really have no idea of how to solve the inmigration / pop growth condruum. Perhaps a decision to determine if a planet growth should be focused towards migration towards another certain planet or for pops to stay in their homeworld?
- I say HELL YES to ascension perks and civics being more empire-defining and unique rather than providing a set of straight % bonuses. Why not give unique techs to technological ascendacy? Unique edicts for "executive vigor"? Unique leader traits for "trascendent learning"? There's a ton of things that could be made in order to further differenciate empires with each player decision (hint: Mutually exclusive Traditions, for example).

As for the long term goals:

• Institutions: Define which institutions make up your empire’s internal departments (such as Diplomatic Corps, Xenology Bureau etc.), and their funding, size and power

It sounds... intriguing. Intra-empire politics surely needs some fleshing out, and so do each different empire's culture. Institutions could be a good alternative way of expending unity, perhaps?

• Espionage: Intel to determine how much you know about another empire, spy actions, cloaking, sabotage & general mischief.

Yes! Give us way to mess around with other empires outside open war, please. Also, spec-ops, guerillas and false flag attacks! This could also be a golden opportunity for fleshing out leaders, which are one of Stellaris more under-developed and under-utilized aspects.

• Religion & Cults: Similar to factions, cults could appear in your empire during certain circumstances. Spiritualist empires would most likely have “imperial” cults. Worship of powerful entities etc.

That would be awesome. Empire's influencing each other via spies, religion and culture would be something to behold. Also, it would be great to see religions and "imperial cults" differenciating from each other. I really would love if my pacifist cult of mother Earth would have different bonuses from Death cultists. Also, this is a really great way of expanding the Shroud's reach and interactions.

• Archaeology: Explore the ruins of ancient civilizations

The game would vastly benefit from a mid and late game exploration phase, with anomalies changing your empire in dramatic ways. Also, xenology, archeology and the study of alien races should really be tied into diplomacy somehow, and require a resource investment, so you would have some well-known neighbour races while other ones would remain shrouded in mystery until the end of the game (unless you devote enough resources in order to unlock their secrets...)

• Subject contracts: Allow overlords to better customize what type of subjects they have, tribute levels, benefits to subject etc.
• Federation depth: Allow federations to level up, have different election types, taxes etc.
• Galactic Council: Create a sort of a ‘space UN’ with galactic politics and diplomacy
• Primitives: Allow for more interactions with primitive pre-FTL species

Empires having a hidden agenda (aka: winning conditions) would help a lot with all of that. Also, Federations should be a kind of a "high risk, high reward" type of investment, rather than a no-brainer for non-genocidal empires: Join a federation in order to enjoy its bonuses (perhaps, create different bonuses for each different type of federation) but expose yourself to the risk of being politically absorbed by said federtion if any member musters enough influence / favour / votes / resources to bid for unification. It could really offer some interesting dynamics that would translate very well into multiplayer as well!

 

[Infin-Vec]

Any possibilities of robotics pop xenos/non-xenos identification?
I can't assimilate or purge, or even stop building non-ascension robotic pops as a materialist-xenophobic-egalitarian synthetic ascension empire.

 

[Infin-Vec]

* Fixed robots/synths being treated as organic when conquered by hive mind empires, and getting locked in the livestock job

LOL.
Seems the hive mind empire pops were having some good teeth and good stomach that can eat robots.

 

[Taritu]

I hope the military AI changes make more of the controls available to modders. As it is right now, modders haven't been able to fix the military AI.

And, while i know i'm a bit of a broken record, I feel the game needs a LOT of AI improvements. The way the current building AI works makes it harder to mod/fix than it should be.

 

[aantia]

What I'd *really* like to see is a "generate random flag" button I usually waste ages agonising over how to make my flag as distinctive as possible. Thanks!

 

[Todie]

These are some exciting notes and future plans!

As self proclaimed ambasador for all things related to economy-AI i would for now just like to frame/reiterate concerns on this subject, already voiced here.

There may be more being worked on than was shared here - but not much was shared on this issue - an issue that is arguably quite devisive in the community.

Sounds like an excellent list for future development, however it's missing some separately needed improvements to the research system. First, the AI needs to get better at keeping up with the player in researching new techs. the current default strategy to win is to just turtle and tech up for a century until the AI's left in the dust, which should not be the case. Either the AI needs to be rebalance to prioritize tech better to keep up with the player, or some sort of tech diffusion mechanism be introduced to boost empires that have fallen behind. Second, I'd love to see more variety to the tech tree. Currently every game feels like you go through the same old tech tree with only the ascension path affecting the few techs you get... it'd be great if you could go down different tech branches depending on civics, traditions, your actions, or even random chance.

As for a couple other nitpics, i agree that Ecumenopoli and ringworlds need to be rebalanced... maybe require a hundred or so pops to enact the ecumenopoli decision so that they feel more like a natural development of your city worlds? As for ringworlds, the main benefit from them aside from the size is easy access to space from the rim... maybe give them a custom "spaceport" district that generates trade, naval capacity, and could even give a boost to ship production?

Also, please, please, PLEASE give us an option to swap leaders so I dont have to go manually assign my scientists to ships when swapping one out for their research bonus!

In my experience, unmodded AI falling behind in tech is a symptom of rather deep issues across different categories ai decision-weighting. Especially lately, as the tech speed penalties are quite easily outpaced by high volumes of scientist jobs.

@grekulf Thank you, that was a good insight. Two things that jumped out for me: first, this:



Come on, spill it. What happened? This is what we had before 2.2, and it strikes me as really weird how yourself here, and some other devs in a few other cases, seem to have just forgotten that you already went through this. Why wasn't the pre 2.1 sector stuff ported and adapted, instead of ripped out and replaced with the current rubbish?

Second thing: The AI. Glad you're targeting specific issues with planetary development and giving the combat an overhaul. However, still very concerned that a more thorough overhaul of AI budgets/economy/planetary development is needed.

What about the Economic AI? The mil AI is nothing withoutt an economy to support it?


Imho these two should be part of post Update support and be incoporated in a 2.2.7 Patch before moving on. For me the game is not enjoy able at all in a wide empire at the moment.
And the Pop growth mechanic feels so extremly wonky i dont want to play biological ascension or egalitarian empires at all.
This means i am severly limited in my play style options right now.

Ive been thinking about this recently; about how to best/simplest illustrate how AI military capability is underpinned by a fee economic parameters - that in turn depend on many moving parts.

Very exciting but I hope to see the braindead AI doing a better job in this patch or the next.

In general I love the update. It clarifies where the devs want to take the game. My only reservation is that I would have liked more of a focus on resolution of short-term (?) issues like:

• Economic AI
• Performance & Stutter
• [Additional near-term] QoL changes
• Micro

I will give credit though. I was asking for more communication and we got a decent chunk of it today.

Not building a trio of precincts without a whiff of crime to counter, which they are fixing in 2.2.6, is a good start.

Also, probably should want to aim for an alloy income of higher than 20 (I think?) when deciding to build more refineries, at least once it's unlocked new ship hulls and bigger starbases.

Especially since it just kinds sits on the mountain of minerals it is bringing in instead, selling off surplus and lowering the market price until I could make more minerals building energy districts to buy off marking than with building mining districts...

While i happen to agree, and these examples only scrach the surface - this discussion and the questions you try to answer is subjective in nature.

In terms of this Expected AI capability We may be able to agree on a fluffy term like ”illusion of competence” - but as for what that should mean and how (/when? / if?) to achieve this in Stellaris, is an open question for @grekulf @Jamor and the team.

... as is, a player that understands the economic mechanics in stellaris can wield them to create exponential growth of power with relative ease - even with just a handfull of planets, even without significant min/maxig in terms of ethics/covics/traits. - by contrast, an unmodded AI is very far from being able to do this in a comparable way... this can completely brake immersion for players who are looking for a strategic challenge.

EXACTLY my point ... I assume you think it's working properly [or close enough]. I tend to disagree.

Some things that I consider "not right" would be:

• I would think things like "upgrading buildings" would be a good start. I've never seen it properly manage to do that under 2.2.4 or lower.
• Understanding what buildings [in general] are good at what point in the game seems like another important factor.
• Building Archology Projects or otherwise specializing their planets to leverage their economy to the next level would be nice.
• Leveraging Caravaneer ships for early boosts in power would be nice to see [maybe shouldn't blame AI for that???]

I could list a ton of other things [mostly 2.0 and greater features] that the AI hasn't been properly taught how to leverage.

FYI: I used to play 1.9.x at Captain difficulty if memory serves. I'm not any better player now but even Admiral with AI mods isn't presenting a proper challenge. Getting the AI to be at least as competitive [which isn't much] as 1.9 was wouldn't be a bad goal

EDIT: Not intending to start a flame war. Just want to point out that some people [due to performance, stutter, economic AI, etc.] may justifiably feel that the game hasn't been "right" since 2.2.0 released.

I agree with you somewhat.

For multi-player I would assume the game is pretty good ... barring some micro, QoL, and similar issues. Not "broken" IMHO but may qualify as annoying.

For single-player folks that value "competitive AI" as their primary goal then, for that person, I wouldn't mind if they said the game was broken.

As for AI goals I don't need it to do as well as a human ... even a bad human. If we got the 2.2.X AI up to the level of the 1.9 AI then I would consider that 'good enough'.

Edit: My favorite phrase recently when referring to what I'd like the AI to be .... "Illusion of Competence" ... Basically don't be so brain dead that it is immersion breaking.

I hope the military AI changes make more of the controls available to modders. As it is right now, modders haven't been able to fix the military AI.

And, while i know i'm a bit of a broken record, I feel the game needs a LOT of AI improvements. The way the current building AI works makes it harder to mod/fix than it should be.