For the Devs: Non-Gestalt Machines

  • 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.
all I want personally is.... machine empires NOT to be gestalts, and acutally jsut be a single ai controlling everything by itself (which isnt a gestalt)
 
In Stellaris, independently sapient machines are an advanced technology far beyond the reach of starting nations. Asking to change that is like asking to start with the ability to build battleships - even if it didn't provide any mechanical advantage, it wouldn't fit with the lore of the game.
Terraforming Gaia worlds are also beyond the reach of starting nations, but you can start on one anyway.

Habitats are also beyond the reach of starting nations, but you can start on some anyway.

Building a Ring World is also beyond the reach of starting nations, but you can start on one anyway.

All of those fit into "the lore of the game" somehow. There's no reason this request couldn't also fit.


Can you please make it possible to make non-gestalt machines? I can't be the only one who wants that.
From what I can tell, you want:
- Robots (immortal leaders, robot traits, assembled pops, energy upkeep, etc.)
- Robot pops and jobs require Consumer Goods, unlike a Gestalt MI
- Ethics like a normal empire
- Civics like ... MI or normal empire?
- Normal empire buildings and jobs


As an Origin, maybe you're playing the people brought back to life by the In Limbo event chain -- finishing that chain unlocks this Origin for future games.
 
  • 5
Reactions:
And how would that be different, mechanically, from how machine intelligences are currently implemented?
from what i can tell they want to be able to have democratic sentient bots at year 2200 without any of the synth bonuses
 
  • 1
Reactions:
from what i can tell they want to be able to have democratic sentient bots at year 2200 without any of the synth bonuses
Or imperial sentient bots! With sentient bot heirs for when your sentient bot leader suffers an "unfortunate accident".

I think the gist of it is, "can I have a normal empire, with machine pops instead of bio pops?"
 
  • 2
  • 1Like
Reactions:
Or imperial sentient bots! With sentient bot heirs for when your sentient bot leader suffers an "unfortunate accident".

I think the gist of it is, "can I have a normal empire, with machine pops instead of bio pops?"
pretty much, and while i don't dislike the idea it would require some heavy balancing to make the empire in question not brokenly strong. although then again i don't remember the last time balance was a significant concern in stellaris.
 
Habitats are also beyond the reach of starting nations, but you can start on some anyway.
That's a good point - Void Dwellers start out with an advanced tech that they are able to replicate. It is basically the same situation as a hypothetical day-one synth origin, so why do I object to that but not to Void Dwellers? I think it's because I consider habitats to be weak and synths to be strong. So if a day-one synths origin was weak enough, perhaps I'd be OK with it? Something for me to ponder, I suppose.
 
all I want personally is.... machine empires NOT to be gestalts, and acutally jsut be a single ai controlling everything by itself (which isnt a gestalt)
But... that is a gestalt. Hive Minds, for instance, are literally described as "these drones are like the fingers of a body" and even autonomous drones are more akin to "that knee-jerk reaction your spine can send out because the brain's too far away" than separate people.
 
  • 6
Reactions:
As an Origin, maybe you're playing the people brought back to life by the In Limbo event chain -- finishing that chain unlocks this Origin for future games.
I like this, but Stellaris has no precedent for that kind of unlockable content, from what I understand... If that's ultimately how they implemented this sort of origin, it'd be the first piece of content in the game's history to do so.

Unless I'm mistaken, and some content used to be gated in this manner in very early versions.
 
That's a good point - Void Dwellers start out with an advanced tech that they are able to replicate. It is basically the same situation as a hypothetical day-one synth origin, so why do I object to that but not to Void Dwellers? I think it's because I consider habitats to be weak and synths to be strong. So if a day-one synths origin was weak enough, perhaps I'd be OK with it? Something for me to ponder, I suppose.
Synths *are* strong, but they're mostly strong because of all the buffs they get through techs and APs.

The part left over when you remove the techs and the APs looks pretty reasonable to me.
 
  • 3
Reactions:
But... that is a gestalt. Hive Minds, for instance, are literally described as "these drones are like the fingers of a body" and even autonomous drones are more akin to "that knee-jerk reaction your spine can send out because the brain's too far away" than separate people.
Their name, "Gestalt", also heavily hints at that. I don't think there's an exact English translation for the concept, but roughly speaking a Gestalt is a shape or form, not quite identifiable as a human, but unambiguously an entity of sorts.

Problem is, many events seemingly contradict this notion and paint a very muddy picture of drones as individuals, not parts of the whole that act autonomously. That's likely just a result of events largely not taking Gestalt into account properly.
 
  • 3
  • 3
Reactions:
So basically a species that have invented space travel without inventing the Internet?
They invented the Internet... and decided they didn't want it inside their bodies.
 
  • 2
  • 2
Reactions:
This thread is filled with terrible arguments that are just trying to find excuses for why we shouldn't be able to have non-gestalt machine empires.

Fact: There is no valid excuse for that. It just happens to not have been done yet. It has nothing to do with technology or lore or balance - balance would be easy to adjust so Synth Ascension still has a strong advantage over starting as robots, and the lore/technology arguments are even worse because Stellaris galaxies are filled with ancient hyper-advanced civilizations that have technology far beyond this from the start, and the concept of a gestalt machine empire is more technologically advanced than individual AI robots anyway.

If you ask me though, we shouldn't have been able to start as machines AT ALL, and becoming a machine empire - whether gestalt or not - should be something that happens during the game. That would make the technology argument more sound by placing all starting empires on a more technologically even ground.
 
  • 4
  • 1
Reactions:
They invented the Internet... and decided they didn't want it inside their bodies.
Wait, what? How do you even distinguish between artificial machine consciousness (which on a certain level is just an algorithms of data transfer, analysis and process) and any kind of data transfer technology?

If we accept the assumption that data transfer technology like the Internet is a technological prerequisite for the development of any kind of machine consciousness, then surely they are indistinguishable. Suggesting otherwise, to me, is a totally incomprehensible and immersion-breaking idea.

IGranted, it wouldn't hurt my gameplay since you can always switch the immersion-breaking species off, just like I do with "corporate death cults" and "sapient rocks". However, with so many other issues with the game, I'd rather the devs spend their time on things that actually matter before working on a feature I will switch off by default anyway.
 
And how would that be different, mechanically, from how machine intelligences are currently implemented?
for one, there would be no leaders like scientists, generals, admirals, there would only be 1 leader
for second, texts would change from "we" to "i"
for third, alot of things that you could do with this concept for example unique takes on admin cap, influence, traditions/ascention perks, etc
 
  • 1
Reactions:
But... that is a gestalt. Hive Minds, for instance, are literally described as "these drones are like the fingers of a body" and even autonomous drones are more akin to "that knee-jerk reaction your spine can send out because the brain's too far away" than separate people.
a singular ai controlling empty machines is not "machines forming a network which then creates the controlling ai"

its a single thing controlling many, not many forming a single being
 
  • 2
  • 1Like
Reactions:
i could see an origin that gives droid tech as starting tech and gives you a planet full of droids. the originwould also allow this civ to use droids as leaders and in all strata/job types, not just the ones they can normally do. perhaps with a perfomance penalty for research and ruler tasks since they are just a special kind of droids and not fully sapient.

so they'd be somewhere between the machine intelligence and organics. they'd still have to make it to late game to become proper synths with the associated production bonuses etc.

could work, imo.
 
  • 1
Reactions:
the problem is the concept ... thats all .

Starting with ringworld , habitats , gaia worlds , or others like hiveworld or machineworld ... has a story behind and a meccanic , as reason and direction .

Starting as a race of indipendent droids would need a major justification , and the best one would be to have the tech of an ascension at day 0 .
The event of the lost data that can be put into droids required another advanced species to create them , and they are not "awakened" , no happiness , no political ideology .

The concept is too close to an ascension , or mechanist start or a gestal empire .
 
  • 1
Reactions:
The part left over when you remove the techs and the APs looks pretty reasonable to me.
Well, when you first make them, robots can't take ruler or specialist jobs. That might be a bit of a handicap in the early game.
Problem is, many events seemingly contradict this notion and paint a very muddy picture of drones as individuals, not parts of the whole that act autonomously. That's likely just a result of events largely not taking Gestalt into account properly.
Drones are supposed to be nothing more than cells in the body that is the gestalt, but the reality of producing millions or billions of drones and deploying them on an interplanetary scale is, some of them go wrong. No manufacturing process is perfect, of course. Deviant drones are a kind of mutation or cancer in the gestalt's "body", parts that are acting in an unexpected and potentially harmful manner.
Wait, what? How do you even distinguish between artificial machine consciousness (which on a certain level is just an algorithms of data transfer, analysis and process) and any kind of data transfer technology?
Programming, of course. If your machines are designed for a highly interlinked, always-online existence, they might consider whatever arrives from the network to be their source of truth. But if they're designed for use out in the real world where communications can be spotty and transmission lag is a significant factor, they may need to subordinate the information they get from the network to what they get from their internal sensors.

(Take a self-driving car, for instance. The last thing you want is for it to hit a clearly-visible obstacle that it identified, but couldn't get an answer from the consensus about.)
 
  • 1
Reactions: