With Upcoming Banks....will game be too easy?

[Te'Ryn Tec]

First off let me state i am super stoked and excited for the banks upgrade so this is not a hate thread.

Secondly, i play exclusively insane difficulty and while it is difficult to get your empire up and on its feet, once you do there isnt much of a challenge once you get your first doomstack to a respectable size.

My question is, with all the advancements being made and the sheer awesomeness that im seeing with the teasers for banks, is any improvements being made to the AI empires?

How well will the AI be at utilizing the new features, will the game still basically be the "i have the biggest best doomstack, all thats left is 1000 war dec's needed to control all systems" game we currently have?

The game just seems so underwhelming after you reach that certain point. With all the new upgrades and features it seems like A) ill reach that point faster or B) Im just going to play to see how the new features pan out.

Ive still yet to officially "win" a single game. One reason is due to the massive lag and late game performance and two because i always reach a point where im unstoppable yet i have a hundred of RL hours to just war dec over and over again claiming 2-3 systems at a time cuz war score cost yolo's me into a stupid grind of war dec's and researching repeatable research with nothing left to discover or explore.

Im hoping banks properly allows the AI to utilize the features to make a much better end game experience. Not only making wars more enjoyable(challenging) but also in terms of expanse of late game enjoyment(events RP'ing, interactions with AI empires etc etc)

 

[Fourthspartan56]

My question is, with all the advancements being made and the sheer awesomeness that im seeing with the teasers for banks, is any improvements being made to the AI empires?

I know that they're improving the Sector AI in Banks but I would be shocked if they don't improve the general AI. Since it is an incremental process that has been going on since Stellaris was released and will continue for the rest of the game's development lifespan.

How well will the AI be at utilizing the new features, will the game still basically be the "i have the biggest best doomstack, all thats left is 1000 war dec's needed to control all systems" game we currently have?

I highly doubt that they would introduce new features that the AI could not at-least slightly handle.

The game just seems so underwhelming after you reach that certain point. With all the new upgrades and features it seems like A) ill reach that point faster or B) Im just going to play to see how the new features pan out.

Yeah this is somewhat of a problem.

Ive still yet to officially "win" a single game. One reason is due to the massive lag and late game performance and two because i always reach a point where im unstoppable yet i have a hundred of RL hours to just war dec over and over again claiming 2-3 systems at a time cuz war score cost yolo's me into a stupid grind of war dec's and researching repeatable research with nothing left to discover or explore.

Like AI development optimization is an incremental process that they have done and they will continue to do.

Im hoping banks properly allows the AI to utilize the features to make a much better end game experience. Not only making wars more enjoyable(challenging) but also in terms of expanse of late game enjoyment(events RP'ing, interactions with AI empires etc etc)

You may be disappointed then, the AI will almost certainly be improved but making an AI for a game like Stellaris is hard and what you want is not going to just happen in one update.

 

[gajop]

I haven't been following the streams, but judging by the dev updates, combat will be mainly unchanged. This doesn't mean combat AIs won't improve, they're run by some pretty simplistic rules right now. I don't think Banks will make anything drastically easier, and in any case you can always mod in higher difficulty (Improved Space Battles in particular can make late game crisis and fallen empires very hard).

 

[Me_]

I sure hope they improve the AI. AI fleets being caught in infinite loops where they can't decide which system to go to. AI running back and forth. AI having starvation in most systems. They must make sure that the AI has at least some understanding of the new stuff, but also fix the old stuff.

 

[Fourthspartan56]

I sure hope they improve the AI. AI fleets being caught in infinite loops where they can't decide which system to go to. AI running back and forth. AI having starvation in most systems. They must make sure that the AI has at least some understanding of the new stuff, but also fix the old stuff.

They will, the idea that they would add features without the proper programming for the AI to use it is just silly.

 

[Me_]

They will, the idea that they would add features without the proper programming for the AI to use it is just silly.

And yet they did so a few times with other games. CK2 and EU4 had a few features that were, at least when first introduced, player-only. Usually those were diplomatic options.

 

[Alblaka]

I highly doubt that they would introduce new features that the AI could not at-least slightly handle.

In before AI building Dyson Spheres in border worlds only under it's control due to a single nearby Frontier Outpost.

But yeah: The more complex (thus awesome) a system introduces is, the more the AI is gonna suffer from it.
I wouldn't be surprised to hear that the AI will have no faction system whatsoever (aka cheating).
Then again, I will prefer a barebone-simulated cheating AI over one that just fails at everything.

 

[Fourthspartan56]

In before AI building Dyson Spheres in border worlds only under it's control due to a single nearby Frontier Outpost.

Ok.

But yeah: The more complex (thus awesome) a system introduces is, the more the AI is gonna suffer from it.

I agree, but suffering from it is not the same thing as not being able to handle it.

I wouldn't be surprised to hear that the AI will have no faction system whatsoever (aka cheating).
Then again, I will prefer a barebone-simulated cheating AI over one that just fails at everything.

I find this hard to believe, the AI does not cheat currently on Normal and I cannot see Wiz adding cheating.

 

[Alblaka]

I find this hard to believe, the AI does not cheat currently on Normal and I cannot see Wiz adding cheating.

...
You DO realize the AI already cheats in several aspects of the game, i.e. having unlimited core sector systems?
AI's in ALL Paradox games have cheats, simply because there's no point in simulating certain aspects of gameplay for all AI entitys, or because it's too complex a task to handle.
(Other examples: No naval attrition for EUIV AI fleets, no leader cap for HoI IV AI nations...)

The question was never whether the AI cheats, but how much cheats it needs to compete with the player in any new gameplay mechanic implemented.

 

[Diezy]

Sometimes I have some worries that countries can grow to be too self-sufficient in their little borders. Civilization 5 leaps to mind, where AI's that went on a good world conquest spree weren't that many. Conqueror AI's are in minority, or they get stomped like Fanatical Purifiers.

Buut in Banks, ethics can change! Which means, AI personality can change during the game... At least I hope so.

Here we can see... Super wide, peaceful, Harmonious Collective, friends with everyone...
*flips switch*
Hegemonic Imperialists on galaxy conquest spree! MUAHAHAHA! And they have allies too, the horror

Right now, everyone's alignment is known from the start of the game, but, afterwards, with shifting AI personalities, Fanatical Purifier might be much more of a threat, if a strong empire that does well becomes the Fanatical Purifier in mid-game!

 

[Tigey]

...
You DO realize the AI already cheats in several aspects of the game, i.e. having unlimited core sector systems?

Really? This doesn't match my experience, as I see the AI make use of sectors all the time (I occasionally load up as the AI in games where I want a challenge to gift them ships/research/buildings/planets, etc). In my experience its very rare for the AI to go over its core systems cap, and it usually addresses it within a few turns when it does happen.

 

[Alblaka]

Really? This doesn't match my experience, as I see the AI make use of sectors all the time (I occasionally load up as the AI in games where I want a challenge to gift them ships/research/buildings/planets, etc). In my experience its very rare for the AI to go over its core systems cap, and it usually addresses it within a few turns when it does happen.

So that was added after all.
Sorry then, last I've known the AI did not use sectors (and I admit that was discussed back around 1.1ish), and I don't usually play non-ironman (thus, no watching AI with console commands).

Kinda surprises me though, since it's a mechanic which limits the effectiveness of large empires and reduces player micro. Former can be done in a less CPU-intensive way (aka, slapping a -% produced modifier on some planets, if it's even necessary to give the AI that 'nerf') and latter is irrelevant for the AI, thus I wouldn't have expected Paradox to actually go through the effort and making the AI use sectors proper.

Right now, everyone's alignment is known from the start of the game, but, afterwards, with shifting AI personalities, Fanatical Purifier might be much more of a threat, if a strong empire that does well becomes the Fanatical Purifier in mid-game!

This would actually be hilarious.
Rare, but hilarious.

It's possible though, if it's implemented the way Wiz mentioned, i.e. that purging a species makes it's pops grow xenophobe.
Aka, find a huge peaceful megablob, purge a planet, watch the blob turn into a fanatic purifier in range, stand back and let it purge every small xeno filth empire nearby to save you the trouble.

 

[Fourthspartan56]

...
You DO realize the AI already cheats in several aspects of the game, i.e. having unlimited core sector systems?
AI's in ALL Paradox games have cheats, simply because there's no point in simulating certain aspects of gameplay for all AI entitys, or because it's too complex a task to handle.
(Other examples: No naval attrition for EUIV AI fleets, no leader cap for HoI IV AI nations...)

The question was never whether the AI cheats, but how much cheats it needs to compete with the player in any new gameplay mechanic implemented.

False, Wiz has very clearly stated in the past that the AI does not cheat on Normal difficulty.