• 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.
Sorry for the double post:

Since automation can build with energy credits that are converted 1:1 it can sometimes be a bit exploitable If you are short on minerals.

Could it be made to convert at market price and also adjust the market price as if the player bought minerals on the market?

More scriptable controll over the conversion values (e.g. allow modders to use script values to calculate their own conversion rates) would also be great as well as more scriptable controll over the system in general.
 
  • 1
Reactions:
Hey,

Thank you for designing this new feature!

Instead of starting with a detailed list, I would like to ask a question:

Why does the automation feature seem to have a much smaller priority compared to other empire management features?

It almost feels like automation is treated as an "extra feature" only for "certain people" and not at all as an integrated and natural part of the game flow. I can imagine explanations such as "we want gamers to micro-manage planets", "developing a good automation system is technically too much effort" or "it is just simply not feasible to implement what players expect, reality vs. expectations", etc. but I'm interested in what was the real design decision or blocker behind automation?

When I look at the players' journey throughout a game session, I can't think of anything else but that automation should be a more integrated and highlighted option. Sooner or later, the game naturally flows in the direction where you have to perform too many repetitive tasks and it keeps you away from the next fun part.

A typical game flow:

You have only a few planets

  • Managing is fun, easy, and even satisfying: "I want to do it myself"
  • Automation is irrelevant for experienced players
  • Automation would be actually a good introduction for new players to keep the learning curve less overwhelming
Having 2 sectors with 8+ planets
  • "Okay, I can still manage BUT it starts to feel too much"
  • Already thinking about automation but min-maxing is more important
Having a lot more sectors with even more planets
  • "Oh my... my fingers... my brain... GIVE ME AUTOMATION"
  • The player motivations are shifting: what was fun during the early game becomes tedious later. Not because players like micro-management less but because their focus is now elsewhere e.g. waging wars feel more fun and requires the same amount of attention and clicks as empire management. Doing both at the same time can feel overwhelming from mid-game
To sum it up:
  • Probably different players have different limits on how much they can micro-manage but surely there is a point for everyone that it becomes just too much.
  • To reach this point, you don't even have to be far in the current session
High-level problems:
The above-mentioned factors show that the game naturally evolves from manual management towards automation but automation is still...

Basically a hidden feature
  • Hidden in the colony management panel, a very hard to notice option
  • It is not clear that the colony designation isn't just a one-time setup for the general bonuses but it is for automation as well
  • Possible solutions:
    • More highlighted section on the planetary management panel about Automated Construction
    • More highlighted / intuitive settings on the Sector overview (F4)
Not enough intuitive
  • A tiny cogwheel icon that doesn't even look clickable at first sight - it is so easy to miss this feature
  • ..and even easier to miss the new (awesome) customization options... Having to right-click on the same icon is just anti-pattern as basically no other panels open to right-clicks, so it is not natural that I can find extra options this way (in my case, it was an accidental misclick that revealed the feature)
  • I have to click on every single planet in the Outliner to validate if automation is enabled or not
  • Possible solutions:
    • Using more standard UI elements + icon combination that makes it clear for users that there is something important and useful to configure
      • e.g. "Automation enabled / disabled" is actually written on the UI on a button and not on a tooltip
      • the cogwheel / wrench / etc. icon is used to open the customization panel by left-click just like other panels
    • Avoid using right-clicks, SHIFT clicks, CTRL clicks... it is not intuitive, not a typical everyday interaction pattern, and hard to remember
      • A different UI element that reassures the players if they enabled / disabled something globally.
      • Also, there is already an existing pattern for this already in Stellaris, setting species rights globally and individually.
        • Same here, setting automation options globally and individually
        • If there is an existing pattern for a similar problem what the user learned already, why introduce a completely new pattern?
    • Planet icons in the Outliner should show a tiny difference that indicates "automation is enabled"
Doesn't educate enough, not enough transparent
  • Minimal or blurry information about what happens exactly if I enable the automation in general or enable a filter option
  • The rules mentioned in Montu's video are critical, not knowing them is the source of frustration and misconceptions about automation in general "the sector AI doesn't work", "the AI is terrible"... this could be easily fixed by providing more info
  • Possible solutions:
    • Highlighting the core rules on the Customization panel, at least the critical ones, similar to an oversimplified FAQ
      • e.g. "Why the AI doesn't build anything? >>> resource deficit, no extra pop, etc."
Doesn't feel like an integrated part of the game session, automation is not offered for the already busy player
  • Players have to keep it in mind and specifically look for the automation feature every single time
    • New players don't even know what to look for, if they can have such a cool and click saver feature
  • The game doesn't try to remind, help, or educate about this
  • Possible solutions:
    • Highlighted in the tutorial for new players
    • A reminder for those who have unmanaged planets to turn on automation while e.g. they are having extra resources (e.g. long wars on multiple fronts, planets will be easily neglected)
Maybe I wasn't fair as some of the mentioned issues are addressed in the form of tooltips
  • This is awesome, as a lot of things are contextually explained in Stellaris and this is so much better than redirecting the players to an external site and breaking their experience
  • Still, if everything is equally hidden behind tooltips, hard to know what is important and what is just nice to have

Here is a rough sketch what I meant actually (only about the planet summary view but automation should be highlighted at other places as well)

stellaris-automation.png


Conclusion:

Automation in this game helps to keep the game fun for the entire session by taking over the repetitive work from the user.


I would consider it as a core feature and not just an "extra" as it keeps the players in the loop and makes it less likely that they abandon a game half-way because it feels too much. Instead, they can keep on going and focus on the next fun part, new challenges, wars, or crisis.

The new feature is awesome, just it feels like it is floating in the air. It could be more fluid how it is offered for the players as they will all need it eventually – hardcore and casual players too.


The awesome:

I might sound negative but I'm just too hyped to further ideate on the matter and just want to make sure everyone finds this helpful feature, which actually works great compared to the legacy versions. I really appreciate that this part now gets recognition and special attention. It is even more appreciated that players are involved and not just "we added this feature you wanted now shut up" but "how could we make this even better".

As a designer, not sure why is it worth for Paradox to still improve the game on this level, but as a player, I'm just happy about it. :)

Also, I just love that even small details were considered during the first iteration e.g. auto-disabling amenity related jobs, people can set something globally and per planet as well, and similar things.

One last consideration: I feel automation should never be perfect or 100% automated. It can also kill the mood if everything comes complete and nothing left to do. Planetary management is a fun part of the game. Later in the game, players could be offered more global options, toggles, and sliders to optimize their automated empires, somewhat a mix of the "new" and the "legacy" automation settings. Maybe something like the vassal settings screen, just somewhat more fine tuning options. Also, if automation will be so reliable that it is almost too good, maybe it could have some negative consequences to balance it, e.g. "-15% from every resource if you enable automation on a planet / sector" - it might feels less like cheating or easy mode.

----

Also, I second the Clone Vat problem, the AI crippled my Hive Mind economy / food production when I was too busy fighting some of my future livestock. =)

Thank you!
 
Last edited:
  • 6Like
  • 2
  • 1Love
  • 1
Reactions:
I'm 'hesitant' to use it because it's inherently flawed and not actually useful. You can mass build and ignore planets for a fairly low number of additional clicks, that are ultimately much more efficient to your economy. You want to solve all of these logic problems when all of it is reactive to your economy, and you shouldn't be reacting to the economy. You should be proactive. For that reason alone planetary automation is always going to be substandard. At best these sorts of changes only go to making the AI planets better, and considering that they still can't specialize a planet I don't see how these little things are at all a help.

If you want to stop people complaining about "tedium", try to make planet building feel more engaging/rewarding. In general, less things should just be button presses. Things like planetary decisions should be tied to buildings, not just available. The bonuses tied to setting a planets focus should be tied to a building instead. The flipping them every couple of years once you have built out the districts of the kind you need feels cheap. Similarly, even build up planets just feel like resource hubs because of how little investment you actually have in them. It's a big part of why things like Ecumenopolis and Ring Worlds feel great and powerful, and why under Dick people were annoyed that the payoff for them wasn't there.
 
  • 1
  • 1
Reactions:
New colonies disabling the colonist jobs because they get enough amenities from the capital, and then depopulating because the only pop is unemployed and auto-resettles, shouldn't happen. While the depopulation can be fixed by preventing the last pop from being able to auto-migrate, I think planetary automation should also always keep at least one job enabled so it never results in 100% unemployment.
That's what's happening! Had a few colonies depopulate and didn't know what was up.
 
It would be nice to have a system similar to the ship designer and fleet manager for planets and sectors. Basically let me tell the AI what buildings and districts I want built in what order and the AI just does it.
 
  • 1
Reactions:
Rural world designation will build energy -> mining -> farming -> energy -> mining -> farming
Currently the script does not take into account monthly income. What do we want here, construct the district with the lowest amount of monthly income?
Kind of. If any basic resource has 0 or less monthly income I'd want a rural world to construct the district with the lowest income, but if they're all positive I don't necessarily want to build more agriculture if food is lowest. Food usually has no uses besides upkeep and colony ships, and without any reason to save up a stockpile for large expenditures there's a limit to how much of a surplus I'm interested in having. There are basically a couple different tiers of 'how much food surplus do I want' depending on your empire:
  1. Catalytic empires want to keep food income at least as high as the other resources because they can have big swings in food use when switching planets between factory and foundry designations.
  2. Empires with the Engineered Evolution ascension perk want enough a monthly food surplus so they can afford to build Cloning Vats on colonies that don't have them yet. Increase the max monthly food threshold by 30 times (the number of colonies with pop assemby automation but no clone vats, fewer than 12 buildings, and excluding resort worlds + the number of habitats with pop assemby automation and no clone vats, and either an empty building slot or an unupgraded capitalbuilding).
  3. Primarily biological empires need enough of a surplus to keep up with pop growth, afford the occasional colony o refugee, and conquer a few planets - although empires that are conquering many planets will typically also be acquiring the infrastrure to feed the pops they conquer.
  4. Primarily lithoid or robotic empires that only use food for a small number of slaves and/or bio-reactors want to keep food income non-negative, but don't need a surplus for growth.
  5. Synthetic ascension empires don't care about having a negative monthly food income as long as they have a stockpile to absorb it, because the food upkeep will go away as pops are assimilated. Ideally the game would be able to predict the monthly food that will be needed after all assimilating pops become robots and build for that.
  6. Empires that don't use food don't want any.
A "used_resource_stockpile_percent = { type = <resource> percentage >=< <0-1>}" trigger would be nice to have.

Possibly if a rural world has booster buildings and districts for two of the three basic resources but nothing for the third, it should avoid starting to build for the 3rd resource unless absolutely necessary? I'm not sure exactly what criteria should be used to make that decision, though.

Maybe a 'months required to reach storage capacity' check would also make sense, if you account for the difference in storage capacity for energy? That would cause it to build the resource with the lowest stockpile if income is relatively balanced or a particular stockpile is significantly lower, even if that resource doesn't have the lowest income. This strategy would likely be of particular interest to empires regularly using the arcology project decision or doing a lot of terraforming.



Something separate to think about is the limits of the designation system. Unless/until I have a lot a colonies, my planets will often be built towards pseudo-hybrid designation plans, one for basic or advanced resources from district slots plus a booster building or two, and one for research, unity, or naval capacity from building slots. Planet automation can't understand this at the moment and can only be directed to build towards one or the other.

Also, the special decision designations can have different builds, but I don't know if planet automation can follow different plans for them (the automation test game I started isn't far enough in yet to unlock them):
  • Thrall-World
    • Any combination of basic resources
    • Fortress
    • Livestock
    • Pop breeder & exporter
  • Penal Colony: Preferably fortress since the regular fortress designation offers no benefits - it's more like a decision you use when a planet's being attacked or building a stronghold - but could be built for absolutely any specialization besides forge or factory since it has no building restrictions besides costing you the designation.
  • Resort World
    • Trade (But with no galactic stock exchange...)
    • Fortress
    • Unity. Should generally build a monument for the popless unity output regardless since you'll often have plenty of building slots without pops to employ in them.
 
Kind of. If any basic resource has 0 or less monthly income I'd want a rural world to construct the district with the lowest income, but if they're all positive I don't necessarily want to build more agriculture if food is lowest. Food usually has no uses besides upkeep and colony ships, and without any reason to save up a stockpile for large expenditures there's a limit to how much of a surplus I'm interested in having. There are basically a couple different tiers of 'how much food surplus do I want' depending on your empire:
...
I mean you are right, but we are going pretty far into very complex rules for what to build when. One of the biggest problems that we want to solve is that it is already too hard to understand and that we want to give the player as much control as possible.

I think an easier way of making the same concept would be to perhaps have additional checkboxes for fringe world designations such as:
Build Minerals [X]
Build Energy [ ]
Build Food [X]

I would prefer to have more "dumb" logic with more control left to the player if possible as it makes it much easier to understand
 
  • 4Like
Reactions:
An interesting idea for the automation would be If you could set resource income targets on a per planet, per sector and potentially even empire wide level. This way you'd have a finer level of control over what is being prioritized. Would also be neat flavor wise, as it could give you the feeling of actually commanding your empire with the click of a button and then seeing it react.
This is how enemy AI empire manages its economy. Setting empire wide targets for their economy. It is basically the opposite of planet automation, it's a planet vs empire type automation.
 
  • 1
  • 1Like
Reactions:
Why does the automation feature seem to have a much smaller priority compared to other empire management features?
I've only been around on Stellaris for less than a year so I weren't there when the feature was first made, if I would speculate I would guess that the feature was made as an afterthought to try and combat the huge amount of micromanagement required in mid and lategame (pure speculation).
Here is a rough sketch what I meant actually (only about the planet summary view but automation should be highlighted at other places as well)

View attachment 866673
I like your sketch, I agree with all your points regarding unintuative and hidden feature. Making big changes to UIs require 2D art and UX work/approval so it becomes a much more involved process where I can not simply work on it myself but have to engage the rest of the team, taking time from other tasks.
 
  • 2Like
Reactions:
The only other quick thing I'd suggest is allowing us to set a designation for our capital. I've never really understood why we can't do that and it prevents us from automating it in the way we can with other planets.
I wish this too, imo the capital designation modifier could be moved to a planetary modifier instead, like the prosperous unifcation modifier but permanent, and then you could choose the designation for your capital as well. As it is right now it is basically impossible to make an automation for the capital as it is impossible to know what the player wants.
 
  • 9
  • 2Like
Reactions:
I wish this too, imo the capital designation modifier could be moved to a planetary modifier instead, like the prosperous unifcation modifier but permanent, and then you could choose the designation for your capital as well. As it is right now it is basically impossible to make an automation for the capital as it is impossible to know what the player wants.
Would it be possible to, say, use decisions to set a planet flag/modifier for automation to look at in addition to the designation, in order to mimic having multiple designations?
 
Perhaps outside of the scope of this discussion, but one automation that would really help with the mid-to-late-game micromanagement would be the ability to automate starbase buildings and upgrades.

Because we can only queue starbase buildings/modules if there are slots available - even if an upgrade that would unlock slots is being built or is in the queue - building a handful of Star Fortresses with (say) 6 anchorages, a naval logistics office and a hydroponic farm becomes an exercise in repeatedly returning to individual starbases to queue up the small number of buildings/modules/upgrade available at that point.

EDIT:

This might not even require automation, simply the ability to:

(a) have multiple starbase upgrades in the queue at the same time.
(b) queue buildings/modules even when slots are not available, so long as upgrades that create those slots are already being built/are in the queue above them.
 
Last edited:
  • 1Like
Reactions:
Perhaps outside of the scope of this discussion, but one automation that would really help with the mid-to-late-game micromanagement would be the ability to automate starbase buildings and upgrades.

Because we can only queue starbase buildings/modules if there are slots available - even if an upgrade that would unlock slots is being built or is in the queue - building a handful of Star Fortresses with (say) 6 anchorages, a naval logistics office and a hydroponic farm becomes an exercise and In repeatedly returning to individual starbases to queue up the small number of buildings/modules/upgrade available at that point.
It's weird we have a ship designer and even a defense platform designer but no starbase designer. No planet designer I can understand given how varied planets can be, and habitats are sufficiently niche that having a whole mechanic for them alone would be weird, but everyone uses starbases and there's four sizes total.
 
I mean you are right, but we are going pretty far into very complex rules for what to build when. One of the biggest problems that we want to solve is that it is already too hard to understand and that we want to give the player as much control as possible.

I think an easier way of making the same concept would be to perhaps have additional checkboxes for fringe world designations such as:
Build Minerals [X]
Build Energy [ ]
Build Food [X]

I would prefer to have more "dumb" logic with more control left to the player if possible as it makes it much easier to understand
If you want "dumb logic" with high player control, a template system that works like the ship designer, except with planets, districts and buildings would be the cleanest solution imo. Someone would have to make the UI though, so I'd be a rather involved process so it's likely out of scope (?)...but maybe Eladrin could be convinced :p?
 
Solid suggestions so far:
- Checkbox setting for ignoring empire wide deficits when constructing new buildings, will build if it can afford regardless of monthly deficit
- Take into account empire wide construction queue when looking at monthly income, specifically a big issue when cloning vats is researched, can crash monthly food situation
- Ignore pop unemployment check when constructing some buildings like energy grids
- Enforcer job micro the same way as amenities
- Fringe world designation should build the district associated with the resource with the current lowest income per month

UX considerations:
- Extremely hard to understand why the automation system is NOT building something
- Likewise hard to know why it is building something or what will be built next
- Designation checkboxes can be ordered in automation priority order to make it much more clear in which priority the automation will construct something
- Suggestion: Orange symbol in outliner (and planet view) explaining what automation wanted to build but why it failed (too low monthly energy income etc)
Please give us a way to uncheck all boxes at once.

Also, give us the designation icon in the Outliner and Empire View, and Resettlement menus, alongside the planet render.

planetary%20designations.png


Also, the Empire View and Resettlement menus should show planet designation in text instead of planet climate.
 
Last edited:
  • 3Like
  • 1
Reactions:
  • Change colony color in outliner to indicate it is under automation
  • Slot on top bar to show resource balance and last month spend
 
  • 6Like
  • 1
Reactions:
What's the colony list to show me which colonies have automation turned on or off?

Does that list show sectors (and their automation toggle) too?
 
  • 1Like
Reactions:
I mean you are right, but we are going pretty far into very complex rules for what to build when. One of the biggest problems that we want to solve is that it is already too hard to understand and that we want to give the player as much control as possible.

I think an easier way of making the same concept would be to perhaps have additional checkboxes for fringe world designations such as:
Build Minerals [X]
Build Energy [ ]
Build Food [X]

I would prefer to have more "dumb" logic with more control left to the player if possible as it makes it much easier to understand
If adding more option toggles is on the table, then I agree that's a better solution. Once you hit your desired stockpile and income you can just shift click to disable food everywhere; you'll just need some logic for how an agri-world should behave if it's not allowed to build food.

I think ideally I'd like to see this option available for every district category that isn't already covered by another non-designation toggle - so in addition to the basic resources you could also direct it to build industrial (and specify alloy and/or CG on an ecumenopolis), trade, research, or unity depending on the district set used.

And in a perfect world I would love to have the choice between the current automatic prioritization based on designation and the ability to manually prioritize/allow/prevent construction for each type of output individually. Something similar to what we have for jobs, just a list of each resource plus trade and naval capacity. Having this visible but greyed out when designation auto-prioritization is enabled would also give some more transparency with how automation will behave when left to its own devices.

Some more thoughts:
  • Similar to rural worlds building the lowest income basic resource, it would be nice to apply the same logic to which strategic resource building refinery designations add; I think they currently build an equal number of them.
    • Also, similar logic for choosing which order to build resource processing buildings in.
  • A "Treat clerks, toilers, and servants as unemployed when deciding if you need to build more jobs" option would be nice. Just don't enable it on designations with trade automation plans.
  • A "Build slave processing centers on planets with slaves" option - they currently seem to only be included on the build lists for Penal Colonies and Thrall-Worlds.
  • The option to disable fortress worlds from building shield generators.
  • The option to disable Gene Clinics. (I typically don't want them on colonies that already have 100% habitability, and empires with Warrior Culture or Pleasure Seekers might prefer those.)
  • Options to enable the auto construction of special civic buildings like Noble Estates or Dread Encampments.
Looking through the scripts in common/colony_automation...
  • Farming automation shouldn't build Hydroponics farms if any agricultural districts are available (excluding habitation districts on habitats)
  • Generator automation should build the Waste Reprocessing Center for gestalts who unlock it.
  • Mining automation should usually be building strategic resource deposit buildings, unless you need minerals more at the moment.
  • Penal automation
    • Shouldn't be building precinct houses unless there's actually crime; usually there's none since adding an additional +100% off a base of 0 is still 0. Should probably just be handled by the regular crime management automation instead of attempting to preemptively build for crime that never materializes.
  • Research automation
    • Possibly shouldn't build a Research Institute until the planet has a couple researchers.
    • Should Reanimators build a Dread Encampment here? Necromancers are researchers.
  • Resort automation (which seems to be assuming a trade build)
    • Should be able to build Communal Housing for empires with Shared Burdens.
    • Should build a Numistic Shrine if available.
    • Should possibly have a failsafe in case the resort gets conquered by a gestalt.
  • Rural automation should probably be building strategic resource deposit buildings.
  • Slave automation
    • Shouldn't be prioritizing city districts, which are unavailable to the designation.
    • Should be able to build uncapped versions of districts.
    • Should possibly have a failsafe in case the thrall-world gets conquered by a gestalt.
    • Should build basic resource processing buildings.
    • Honestly I think this should just use the same automation plan as rural worlds as a base. Slave Huts can be added by housing management and overseers and Psi Corps can be handled by crime management. The only other potential difference is how highly you prioritize the Slave Processing Center, and maybe filling any remaining building slots with strongholds.
  • Trade automation
    • Should build a Numistic Shrine if available.
    • Shouldn't build commercial zones buildings until it runs out of available trade districts on habitats and ring worlds.
  • Unity automation
    • Shouldn't be building leisure districts on habitats. They produce entertainers, not culture workers.
    • Should probably be building the Ministry of Culture if available, so it has all three +15% unity buildings. Unless it's being deliberately omitted from automation since it's limited to one per empire?
Automation exceptions:
  • Crime management
    • Should build Overseer Residences on Thrall-Worlds.
    • Should probably still be able to build Psi Corps even if the option to build them everywhere is toggled off.
  • Housing management
    • Should be willing to construct housing buildings, particularly on colonies that can't build housing districts - i.e. Resort Worlds and Thrall-Worlds.
  • Amenity management
    • Should build leisure districts on habitats.
    • Should build Alien Zoos if they get their culture workers back or otherwise stop being worse than Holo-Theatres.
    • Possibly should build Numistic Shrines? They're lower amenities per job but still give +10 total (same as Gene Clinics), and Merchants are nice. You could apply the same reasoning to the Galactic Stock Exchange too, tbh.
    • Would be nice if when amenities dropped negative temporarily because the Luxuries Distributed modifier expired, it would send you an alert or something instead of silently enabling an entertainer that's just going to become an unemployed specialist as soon as you notice and refresh the decision.
  • Posthumous Employment Center
    • Is missing exclusions for col_bureau_spiritualist and col_habitat_bureau_spiritualist.
    • Should not be excluding col_resort, col_habitat_trade, or col_ring_trade - zombies make fine clerks!
    • Should probably also exclude specialist-based machine and hive world designations. Corporate empires might not be able to build them but they can acquire them through conquest just fine.
    • Penal automation builds worker jobs so it probably shouldn't be excluded either.

Possible bug, or at least QoL issue: If automation starts building something using the shared stockpile, and you cancel the construction, the resources go back into the local stockpile, preventing other sectors from using them.

In the attached save, automation is trying to build a mining district for the unemployed specialist on my capital. Possibly this is happening because the stockpile wasn't large enough for it to afford anything that adds specialist jobs, but it could afford a basic resource district? Feeding it another 200 minerals caused it to start a research lab instead.
 

Attachments

  • 2234.02.13 urban only.sav
    2,8 MB · Views: 0
Last edited:
  • 1Like
Reactions: