Jazehiah wrote:

Well, I know that some people won't like war automation; that's why autowars have to be limited. It's an exponential growth factor; it doesn't really matter how many resources you have, all you have to do is research and design and plant a seed, and so long as someone out there didn't plant a seed before you, your numbers will be far superior,

...unless, by starting early, they gimped themselves by using older, slower construction models. All you, or anyone else, has to do is build a faster construction module.

and you're almost guaranteed to win. Especially if you automate strategy and don't have to be there every second of the day. My point is that if the AI does this, and a player wants to defeat the AI, that player, who doesn't want to do this, will have to, and thus there have to be limits so that that player doesn't have to do this, and so that you can't easily beat the game with your automaton fleet. It has to be balanced. All paths have to have equal benefits and drawbacks, more or less. That's why there has to be limits, or at least major weaknesses.

Unless you take my suggestion to limit the evolution of AI fleets to only what they manage to salvage. If technology evolves, the AI has to work to catch up. The AI will always be behind. With machine learning, you can sabotage its growth and tactics for later on. All you have to do is pretend to evolve slowly by poking with old ships only to do a clean sweep with a modern fleet.

Also, unless the AI cannibalizes its old ships to make new ones, it will always have a bunch of older ships in the fleet. Those ships would make for easy pickings, wouldn't they? But let's say your AI fleet does cannibalize most of the outdated ships. Then, the fleet may be strong, but relatively small in number. It also can't be any more advanced than whatever you're throwing at it.

If you're out of money and the AI got your only really good ship, yeah. You might have a problem. But seriously, that's the case with any race more advanced than you. If you really want fewer aggressive races, pick a different seed. Also, one of the trade-offs of having robots control your ships, is that they have to be replaced to see improvements in pilot performance.

TL;DR:
1) Slow construction builds slowly. Just build faster.
2) Make AI such that if they only see week fleets, they'll be weak.
3) If you're that worried, pick a different seed.

That's what I mean by limits. All of your suggestions are very well thought-out and reflect what I mean. Except for the robots being unable to improve. I mean, it's well thought-out, it just doesn't reflect what I mean. But that's a small point that really doesn't have to be argued.
So... yeah. I guess. We agree. *awkward silence*

Limits break Limit Theory. This is supposed to be a universe of no limits.

HKY09 wrote:Limits break Limit Theory. This is supposed to be a universe of no limits.

-_-
We've discussed this already, and while we could discuss it again potentially, you aren't adding anything except a statement that has already been made. If you have anything new to say, like a way that we could make sure that Autowars aren't overpowered that no one's thought of, or how they aren't, even without limits, or how self-replicating survey ships (probes) would affect gameplay, please do.

Why is having an unwinnable situation a bad thing? I can imagine some pretty epic stories evolving from that. Plus, as the universe is generated as you go along, you could always just leave the area of the super robofleet and try and amass a fleet of allies to take them on. I get that some people may not like that, but then they can just use a different seed or play with the options to get the scenario they would want to play in. I imagine LT being similar to Dwarf Fortress in that, at some point, fun is generated from watching to see how your impenetrable fortress of an empire is destroyed by leaving the gate open. Just trying to survive and stay alive in a area where there is an autowar going on would be a fun challenge in itself. In the end, none of it really matters, because the player can always just run away to an area of the galaxy far enough away from the crazy robot army where you can do your own thing without worrying about them. I am of the opinion that such galactic disasters should be allowed to happen, because what kind of scify game would this be if there weren't massive robot armies to fight with?

In the spirit of debate however, what if sections of the fleet were controlled by executives, while most of the robots were workers, much as human armies work. Then, no matter how big of an opposing force, you can use stealth and tactics to find the leader of that portion of the robot army and disable it or something. That would have the effect of shutting down the rest of the army in the system, or maybe rendering them so ineffective that a player force would be able to feasibly take them on even when vastly outnumbered.

Another option would be to allow the existence of anomalies in space which would allow a clever player or AI to even the playing field against a seemingly invincible force. Perhaps nebula which scramble robotic sensors, or the use of fancy tactics to use an asteroid field as a way to funnel enemies in specific paths to so they can be easily dispatched by defensive turrets, would be examples. If you give the player/executive the tools they need to get leverage over a superior force, that allows the existence of massive robot fleets, but also provides tools to deal with them. If the AI uses those tools against the player... well, that's the player's problem.

its bad because the player would always start in an universe where some variation of von neumann war machine won already.

the universe doesnt start when the player enters, it gets simulated quite a while before that.

so any "i win" strategy would already have taken over all the area the player could ever fly to.

and trying to fight such a force would be as futile as a single group of 5 people going headfirst against the empire from warhammer 40k, everyone knows how it will end.

and any "regional manager" would be as dispensable as the minions he is commandeering.

also do my previous statements not include any orders of magnitude of production capacity, regardless of how fast you could destroy them, they can build the stuff faster than you can cycle your weapons.


all in all it could only be anything but futile and over in 1 digit minute timeframes if the AI is artificially gimped to enable you to win.


as an extra game mode you have to specify to enable it it could be fun, with an adapted AI

if the AI plays straight.. you lost.
without any fun.