• 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.

Stellaris Dev Diary #30 - Late Game Crises

Hi folks!

We’re getting close to release and there is not much left to talk about that we haven’t already covered. The only remaining major feature is, I believe, the “Late Game Crises” events, and I really don’t want to spoil them, so bear with me if I’m being slightly vague this time…

stellaris_dev_diary_30_02_20160418_message.jpg


Now, last week I talked about how large empires will have to worry about keeping all manner of political Factions in check. This is one of the ways we try to keep the game interesting and challenging past that crucial point when you often tend to lose interest in most strategy games and feel that you’ve already won. It’s not much fun to spend hours of your life mopping up the final resistance just so you’ll get to see that sweet acknowledgement saying “Victory!”. Another way to keep a game interesting is through random occurrences that can upset your plans even at a very late stage. This is where dangerous technologies and late game crises enter the picture.

stellaris_dev_diary_30_01_20160418_dangerous_tech.jpg


Some technologies are clearly marked as being “risky”, for example Robot Workers. Now, you might not always risk having your victory snatched out of your grasp, but in this case at least, you really are gambling with the fate of the galaxy. Just researching such a technology is safe; it’s the actual use of it that carries the danger. For example, the more sentient Robot Pops there are in the galaxy, the higher the risk is that they will come to deem organic life unfit to exist and rise up in a well-planned revolt. Unless crushed quickly and with overwhelming force, such a Machine Empire will quickly get out of hand and threaten all the remaining empires in the galaxy. Sentient robots will out-research and outproduce everyone. If the revolt is centered in a powerful rival empire, you’ll need to think carefully about when you want to intervene; a savvy player might time it just right and be able to mop up both the robots and the remnants of the rival empire. Leave it too long, however, and the robots will overwhelm you.

stellaris_dev_diary_30_02_20160418_diplomacy.jpg


The idea is that you will usually see one of the possible late game crises every time you play, but the chances increase the longer it takes you to win. However, it’s very rare to see more than one in the same game. The different threats vary in nature and behaviour, and can offer opportunities as well as posing an enormous danger to your survival. For example, it might be possible to reverse engineer some really unique technologies from these galactic threats, but the geography of the galaxy might also change in your favor…

That’s it for now my friends! Next week, we’ll change tack completely, and do a two-part, in-depth guide for modders.
 
Last edited by a moderator:
  • 213
  • 99
  • 1
Reactions:
Watch and despair! all life will be absorbed by the Prethoryn Scourge! HAK HAK! :eek:

I have to say I didn't think it was possible to get any more excited about the release of Stellaris but once again you've proved me wrong.

Roll on the 9th May, my new favourite day of the year :p
 
So you guys think that is possible to create AI and remplace your pops by synthetic for having an AI goverment played by yourself?

I think we know 5 late game crisis for now:
-AI revolt
-Zeg/tyrannid comming from dark space
-Energy beings comming from another dimension if you research too much FLT
-Sanctuary world
-Fallen empire resurgence
 
I think his point was that the people in them didn't think they could be dangerous until they were. Likewise you shouldn't get warnings about "dangerous tech".

There are even today many people who think advanced AI research may become dangerous sometime.
There are other sciences that are deemed too dangerous or to have shifty ethic complications like genetic modification, nano-technology or the cloning of humans.
 
  • 2
Reactions:
Next time I neeed to laugh condescendingly at someone in a shitpost, I think I might have to go with "[USERNAME]? HAK HAK HAK!" Not here, mind, somewhere where it will be more out-of-left-field.
 
  • 6
Reactions:
How likely is the AI to trigger AI revolts, etc. Is it in their programming to research robotics, and if so how cautiously. I'm aware that the AI doesn't think, is it determined by their ethos? As an example, are ruthless capitalists more prone to try and exploit as much as possible and hence they build robot slaves to harvest ALL THE MINERALS. While in contrast a spiritualist empire might deem non-organic beings as heresy and not use them?

Will the Computer-AI trigger the AI revolt crisis? arghh, too much inception.

What if the computer AI gains sentience? Will Wiz be able to stop it?
 
Ah I just remembered when Wiz mentioned about Fallen empire's getting involved if you disturb Tomb worlds. I reckon awakening some old super advanced threat will definitely be one of the crises.
 
I don't think giving equal rights to AI should give the player anything close to total immunity against robocalypse. "Clearly the filthy meatbags gave you rights solely because they were afraid of your might, and why should you be chained to equality, when you deserve so much more?" A part of your robot population staying loyal makes more sense than all of them attacking you whatever you do, though.

Also, while some of the insurgent AI's suddenly deciding they like peace and harmony is a rather overt transformation, a sort of Ethos Drift in their empire would have interesting potential. While they're still locked in intense struggle, all kinds of divisions can lie dormant, and come to fore when they're decisively winning. Besides, a lot of players would just quit at the point AI is curbstomping the galaxy, but if they might split into, for example, genocidal faction and overlord faction that can live with organics as vassals, and start to fight each other, the game would stay interesting longer.
 
  • 5
Reactions:
@Doomdar
I just want to make certain it is correct, in the picture it says SENTIENT AI, that is self-aware machines that distinguishes themselves as seperate entities from others and that they know that they are a specific individual and can pass the mirror test. Does that mean they already have SAPIENCE? That is a research prior to the one in the picture has already granted the AI the capacity to reason on a human level? Without both it's not dangerous
 
  • 4
Reactions:
I don't think giving equal rights to AI should give the player anything close to total immunity against robocalypse. "Clearly the filthy meatbags gave you rights solely because they were afraid of your might, and why should you be chained to equality, when you deserve so much more?" A part of your robot population staying loyal makes more sense than all of them attacking you whatever you do, though.

Also, while some of the insurgent AI's suddenly deciding they like peace and harmony is a rather overt transformation, a sort of Ethos Drift in their empire would have interesting potential. While they're still locked in intense struggle, all kinds of divisions can lie dormant, and come to fore when they're decisively winning. Besides, a lot of players would just quit at the point AI is curbstomping the galaxy, but if they might split into, for example, genocidal faction and overlord faction that can live with organics as vassals, and start to fight each other, the game would stay interesting longer.
AI loyalists should be a thing.
 
  • 9
Reactions:
Shouldn't dangerous technologies also give good rewards? +5% research speed does not sound like much.
 
  • 10
Reactions:
There are even today many people who think advanced AI research may become dangerous sometime.
There are other sciences that are deemed too dangerous or to have shifty ethic complications like genetic modification, nano-technology or the cloning of humans.

Agreed, and besides, there's only a finite number of technologies. Sooner or later, people will figure out which technologies are dangerous, and it will become like Dwarf Fortress and the dark secret of adamantine. There's some newbies who find it out the hard way, but despite some vague gestures in the direction of trying not to spoil it, it's an open secret to anyone who lingers around the forums or wiki for any length of time, even if they've never seen so much as a sliver of adamantine.
 
  • 5
Reactions:
Freespace Descent and Freespace 2 are some of the greatest games ever made! :D

Shivans are awesome, but are not the first galactic invaders in S-F. The Berserker is probably book with "prototypes" of Shivan and Reapers.
And yes, Shivans > Reapers, becouse instead of talking bullshit about being undefeatable and incomprehensible, they act like one. So-called "menacing" reapers are the usual cartoony villains. I was surprised that Suvereign didnt made Evil Overlord Laugh at the end of his bullshiting.

Thank You, Thank You very much ! At last somebody who knows what I am talking about and how awsome it is, glad to see you here :)
 
I don't think giving equal rights to AI should give the player anything close to total immunity against robocalypse. "Clearly the filthy meatbags gave you rights solely because they were afraid of your might, and why should you be chained to equality, when you deserve so much more?" A part of your robot population staying loyal makes more sense than all of them attacking you whatever you do, though.

Also, while some of the insurgent AI's suddenly deciding they like peace and harmony is a rather overt transformation, a sort of Ethos Drift in their empire would have interesting potential. While they're still locked in intense struggle, all kinds of divisions can lie dormant, and come to fore when they're decisively winning. Besides, a lot of players would just quit at the point AI is curbstomping the galaxy, but if they might split into, for example, genocidal faction and overlord faction that can live with organics as vassals, and start to fight each other, the game would stay interesting longer.
I'm not suggesting empires with equal rights for AI get total immunity from the crisis. I'm suggesting that each population of AI get their own ethos based on how they've been treated. So that such an empire would be a lot more resistant to it than one with sentient robot slaves.

This would manifest as only parts of your robot population going rogue instead of all of them. This would simulate robots making their own decisions instead of being forced by scripting to suddenly try to murder everyone when the day before they were perfectly content.
 
  • 5
Reactions:
Shouldn't dangerous technologies also give good rewards? +5% research speed does not sound like much.
Depends on just how risky they are. We know there are a line of tech's related to AI, this is just one of them. Presumably there are other tech lines related to robotics as well. Just one or two of these risky techs probably aren't very risky on their own. If they were their benefit would have to be higher as you say.
 
I just hope it isnt going to be predictable and annoying. Oh boy, im arriving at late game, will it be robots again? I liked in other strategy games how there are medium size random events and small ones too as well as potentially game ending ones.
 
  • 6
Reactions: