AI War 2 v0.503 Released! "Nanocaust + Test And Revise Pass #2"

[BadgerBadger]

Well, the lore exists to serve gameplay, not gameplay the lore. What I originally implemented was the Lore version of the Unstoppable Nanocaust and it didn't make the game fun (I mean, it was fun to watch the AI be crushed like a ripe melon, but it wasn't a game anymore, it was a "watch the AI and humanity die" video). If you try to play with the Nanocaust right now then you'll eventually observe similar behaviour.

I think the Nanocaust should add some interesting strategic depth, but it also shouldn't completely dominate the game (though if we have a Faction Strength Modifier a la AIWC, I'll certainly make it capable of dominating the game). I think it is critical that the human player be able to defeat the default Nanocaust for "Fun Gameplay" purposes, and I think it's more interesting if the nanobots and the AI are "at war, with the outcome undecided" when the humans show up. If all you have to do is turtle until the Nanobots destroy the AI, well, that's boring.

@Oralordos: At the moment there isn't really a "Victory" or "Defeat"; the game simulation continues on after the Ark or Master Controller dies. So I don't know the answer to that. I could certainly add "Destroy the Nanocaust" to the objectives screen as a start, I suppose.

[WolfWhiteFire]

Well, that is why I suggested the building that pretty much shields the planet the AI's controller or whatever is on from the nanobots and that you could either make it have hp so humans could try a raid late-game to take it down so the nanobots will move in or just make it invincible or part of the controller or whatever it is called.

[BadgerBadger]

The intended mechanism for having the Nanobots really try to kill the AI is hacking; you'll be able to Hack the initial Nanobot Controller (which will have a unique name so you can spot it); this will let you reprogram the Nanobots to have a new behaviour: "Help the humans kill the AI".

The primary reason I'm hesitant to let the Nanobots get too strong is I think it will take Agency away from the player. In a multi-faction game (at least, with the type of factions I have in mind), the galaxy should be in a state of conflict between the various factions. And they will stay in some state of tenuous balance until you, the player, take action to tilt the balance. You might be weak by comparison to other entities, but your actions and decisions matter, and your hand can tilt the scales. You might also just get killed in the crossfire though.

The previous paragraph is a basic design goal of mine, and one that I hope is in keeping with Arcen's philosophies; this is their garden and I'm just trying to help grow some new deadly flowers. See http://arcengames.com/designing-emergent-ai-part-6-the-tempo-and-ai-vs-player-agency/

[WolfWhiteFire]

That seems like a cool mechanism for having the nanobots fight the AI, though I am guessing that the Nanobots REALLY won't like you trying to hack them. (I would say the AI too, but I am not sure how they would really know what you are doing, initially at least). It might be cool to have the opposite be possible with a far longer time and with the player being given plenty of notice to try to stop it, where the AI can hack the nanobots to have them focus on the humans. Kind of like a dyson sphere except it starts out hostile to everyone and rather than depending on whose if anyone's command station is on the planet it depends on who hacked them last. Just an idea though. Maybe something that can happen when the difficulty of the faction or of the AI is set at or above a certain threshold?

[BadgerBadger]

Keith's awesome C# interfaces let me design a custom nanobot response to hacking, so it's going to be cool!

Also, just because the nanobots dislike the AI doesn't mean they like you. I'm going to have some nasty anti-human interactions built in as well, never fear.

[WolfWhiteFire]

I had the impression that they start out hostile to everybody, and hacking them will make them hostile only to the AI and capable of actually killing the AI, did I have something wrong then?

[BadgerBadger]

No, you are correct, they start hostile to everyone. My comment was to say "Even though at the moment the AI can't make them just target you, they will target you in some worrisome ways without interference from the AI".