Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Inventor Raccoon]

Unless you had 20-30 Beta/Alpha cores going at once, the chance of that happening is low. The average Beta core would require nearly 2 cycles before failure. The average Alpha a little less than a cycle. You can always tweak the numbers down or scratch the "need to be there" issue. Point being: there is risk involved in using AI.

It's not a very fun risk though. A chance to get a permanent debuff if you don't constantly check back to micromanage? A better idea would be something like making AI cores draw in Remnant attacks, and successful attacks from factions who hate AI cores (the Hegemony and Luddic Church specifically) would result in the destruction of those cores. That way there's a counter - you can use resources to build more defensive structures, so you lose resources in the short term but gain in the long term because of the core's efficiency.

[Bribe Guntails]

Qyestion: Why can't markets import from more than one other market?

Fluffy Answer: Markets don't have the trchnological capability to handle multiple imports. (But then how are they handling multiple exports?)

Crunchy Answer: It creates a vacuum better suppliers can fill. (But that favors a system in which only the big suppliers can exist, while small markets e.g. Asharu will die off)


I can't find an explanation for why single import is better than multi-import.

[FooF]

Unless you had 20-30 Beta/Alpha cores going at once, the chance of that happening is low. The average Beta core would require nearly 2 cycles before failure. The average Alpha a little less than a cycle. You can always tweak the numbers down or scratch the "need to be there" issue. Point being: there is risk involved in using AI.

It's not a very fun risk though. A chance to get a permanent debuff if you don't constantly check back to micromanage? A better idea would be something like making AI cores draw in Remnant attacks, and successful attacks from factions who hate AI cores (the Hegemony and Luddic Church specifically) would result in the destruction of those cores. That way there's a counter - you can use resources to build more defensive structures, so you lose resources in the short term but gain in the long term because of the core's efficiency.

I won't belabor defending the idea much more but I think it's more of a calculated risk to install AI cores than it is a "not fun" one. The benefit is that you get improved efficiency now, while possibly running into issues some time down the road. You could take that core idea and extrapolate it dozens of different ways.

I like the idea of an active risk but drawing in enemies would be suicide for an upstart outpost. I guess it depends on the mechanic by which the cores signal an attack. If it's automatic, then no thank you. If it's RNG, it's not much different than my idea except that you fight a fleet instead of uninstall a core. If cores have a certain radius of effect, you could ensure your own destruction if you're not careful or bypass the drawback completely.

[intrinsic_parity]

@Foof, I don't think anyone is arguing that having a risk associate with cores is a bad idea, I just think having that risk be arbitrary and annoying when it could be tied into other mechanics like combat is a bad idea. Combat is one of the best aspects of this game, having a risk associated with combat is the best way to go I think. You could also make it so that certain factions won't trade with you if you are using AI cores. I'm sure you could come up with many more ways of achieving the same effect, the point is that there are a lot of better ways to achieve this effect that tie into other mechanics in a more fun way rather than a random chance that you have to fly back and push a button. I'm all for some additional factors to consider when installing AI cores.

[Bribe Guntails]

Are the other factions so omnipotent that they know when you instill AI cores into your outpost? No NPCs other than administrative personnel in the outpost can know that.

Related: Administrative personnel are visible when you enter a market and open the Missions and Personnel page.
A market with lots of Sectors would require more people in charge, and a big market with fewer people may indicate AI cores are abound or are actually short staffed.
Additionally: AI cores have variable capability in mimicking a human; Gammas are bad at it, Betas are great if not scrutinized, and Alphas are indistinguishable. Putting cores in roles 'visible' in thr Personnel page will draw suspicion.

The reason why you'd want to hide your use of cores is because factions prohibiting them would get suspicious when traded with (slight rep loss per month). E.g. the Church would get suspicious fast and descend upon you in holy war, the Tri-Tachs would actually like you more for using AI.


There you have it, combat and campaign risks rolled into a pattern predictable by the player.

[Megas]

The reason why you'd want to hide your use of cores is because factions prohibiting them would get suspicious when traded with (slight rep loss per month). E.g. the Church would get suspicious fast and descend upon you in holy war, the Tri-Tachs would actually like you more for using AI.

If player does not like those factions and is already hostile to (or plans to fight with) them, and using cores has benefits and no other drawbacks, then using cores would be good.

If cores are played up to their treacherous lore, then no sane person should use them as cores are simply killer demons by another name and classic 'sealed evils in cans' that invokes all of the 'evil is not a toy' tropes.  Sooner or later, they will break free and kill people.  If core treachery is not strongly enforced by the game, cores are simply another game element to be munched (or avoided if the drawbacks are too great).

[Bribe Guntails]

If cores are played up to their treacherous lore, then no sane person should use them as cores are simply killer demons by another name and classic 'sealed evils in cans' that invokes all of the 'evil is not a toy' tropes.  Sooner or later, they will break free and kill people.  If core treachery is not strongly enforced by the game, cores are simply another game element to be munched (or avoided if the drawbacks are too great).

Tri-Tachyon's Strategic Modeler needs a word with you.

The way I see it AI can be very dangerous if left unchained with a vague directive and no restraints. Here I quote:

"A powerful AI system tasked with ensuring your safety might imprison you at home. If you asked for happiness, it might hook you up to a life support and ceaselessly stimulate your brain's pleasure centers. If you don't provide the AI with a very big library of preferred behaviors or an ironclad means for it to deduce what behavior you prefer, you'll be stuck with whatever it comes up with. And since it's a highly complex system, you may never understand it well enough to make sure you've got it right."

— James Barrat

AI is only as dangerous as the people responsible for its behavior are lethargic, or you're Tri-Tach and you manage to let them off the chain.

[Megas]

The lore hints that Tri-Tachyon leadership may be AI, so Tri-Tachyon using cores (covertly) is expected.

Amoral AI with intellect greater-than or equal to humanity should be able to corrupt any directive given to it.  Failing that, (the author can make) AI can lie or cheat and break directives if the game or show must go on.

[DatonKallandor]

I think you're overselling the AI is going to kill you idea somewhat. We regularily hand over gunnery control to AIs on ships. We've got AI drones being used for combat. It's probably just putting extremely high grade AIs into unsupervised positions that's risky. You know, pointing a brilliant AI at an automated factory and saying "go build combat ships and fight the Hegemony" and then walking away - the Tri-Tachyon way.

[RandomnessInc]

When it come to one of your outposts being attacked you should get an option to barf out any ship in the market/storage and have them refitted with whatever's available, maybe with a hull mod that's something on the line of "Ill-prepared crew" or "untrained crew" because the crew wasn't ready or just random people who know how to pilot/work on a ship rush aboard to man it causing the ship to have less combat readiness/stats.

(Assuming outpost defense combat is a thing)

Just thinking out loud.

[Megas]

@ DatonKallandor: I know there is automation that the game shows is safe enough, such as Terminator drones (which were retconned from beta to delta) and Wasps.  The lore is clear enough that there is a threshold where AI past that is too smart.

[Techhead]

Qyestion: Why can't markets import from more than one other market?

Fluffy Answer: Markets don't have the trchnological capability to handle multiple imports. (But then how are they handling multiple exports?)

Crunchy Answer: It creates a vacuum better suppliers can fill. (But that favors a system in which only the big suppliers can exist, while small markets e.g. Asharu will die off)


I can't find an explanation for why single import is better than multi-import.

I think it's more "one well-beaten trade route out-competes more expensive ones", and combined with the (rough) log scale, multiple smaller markets aren't numerous enough make up the demand without shortfall. (Or they are numerous enough, and shortfall comes in the form of overhead of managing that multiple commodity streams.) A market that (conceptually) is just short of meeting the demand falls on the same point of the scale, and naturally grows by that little bit to meet the demand.

[gruberscomplete]

I think you're overselling the AI is going to kill you idea somewhat. We regularily hand over gunnery control to AIs on ships. We've got AI drones being used for combat. It's probably just putting extremely high grade AIs into unsupervised positions that's risky. You know, pointing a brilliant AI at an automated factory and saying "go build combat ships and fight the Hegemony" and then walking away - the Tri-Tachyon way.

I mean, the Tri-Tachyon were extremely successful with that approach, given how the Hegemony was nearly crippled twice from the AI wars.

You could say that it might be more advantageous to an AI to follow orders than to disrespect them, and in this case it was given the freedom to do anything it wanted, yet we still have the outcome that the AI was true to it's commands and didn't go about conquering the Tri-Tachyon.

So we could say that your Alpha-level AI administration would be very happy to build you a stellar empire.

[xenoargh]

You know what would be fun to do with the AI ideas?

Let's set the history back a minute, and have the player experience the Second AI War.  That gives us:

1.  A Faction that wants to expand (and evaporate our outposts).
2.  A big-bad to fight, who gets stronger over time if not fought.
3.  Something that can destroy the Sector (finally, a credible threat).
4.  Possibility of working diplomatically with the Sector's other players to solve the issue (or fail, and everybody dies).
5.  Excuses to have ultra-tech show up to push players outside the normal power-curve in late-game play.
6.  The events in question can happen whenever we want or even shut off in the options, so the game's difficulty ramp can get tuned better.

[ChaseBears]

late to the ball game here, but intrigued by the blog post having only just now read it. i like the simplification.

i am tbh not entirely certain of the purpose of shipping capacity, other than insuring there are markets with permanently unsatisfied import requirements; it seems like a stat that only matters if its modified by an event to create temporary shortages?

the topic of how to permanently effect markets and create interesting economic interactions concerned me most; temporary measures obviously wont do when one wants to create a permanent outpost. And you don't really want destroying actual markets to be something the player is engaged in, at least not as a casual thing.  The only solution I came up with was feeder outposts and way stations used by the AI.  The outposts would function as feeders to another market and not as markets of their own; for example, an asteroid mining facility or a hydrogen fuel-scoop that supplies a nearby market, increasing its production.  These would be interactable and destroyable, possibly even the subject of faction missions.  Destroying one (or otherwise neutralizing it, like with a hireable spy, skill interaction, or ancient AI virus you found in the boonies) would decrease production and cause a temporary stability hit.  Feeder outposts could even be semi-randomized, causing economic variations between gamestarts.  NPC Waystations would work similarly, providing connections between markets but able to be sabotaged or destroyed to cause economic chaos.

Now that I think about it, one great advantage of npc waystations would also thematic; the gradual destruction of trade infrastructure and the economic consequences thereby seems central to Starsector lore.