Knowledge raiding has become tedious and trivial. Change it.

[keith.lamothe]

Hmm, well, I'm reasonably convinced at this point that we should take the micro out of late-game k-raiding by some sort of research station like suggested before, without any untoward AI rage, and see if it really does make the game much easier or not.  If it does, then we can look at making it harder again

But I will need to run this one by Chris first, not the sort of thing I just do by myself

[RCIX]

Hmm, well, I'm reasonably convinced at this point that we should take the micro out of late-game k-raiding by some sort of research station like suggested before, without any untoward AI rage, and see if it really does make the game much easier or not.  If it does, then we can look at making it harder again

But I will need to run this one by Chris first, not the sort of thing I just do by myself

Do not worry young padawan, someday you will become an Jedi AI War master....

[wyvern83]

Guess I took too long writing this, but I'll contribute my thoughts anyway. I'm glad the untoward AI rage is being passed over for the time being. (I was not a fan.)

One idea that comes to mind is that if an AI command station is present on a planet, science ships can only generate knowledge on that planet if they are within, say, 2000 range of that station.  It would solve the k-raiding-off-in-the-middle-of-nowhere thing.

Or, somewhat simpler, that if there's an AI command station present, the science ships have to be within X range of the planet center.

Thematically I'd buy a close range data hack if you needed a reason to get that close.

My concerns would be the proximity of defensive mine fields and tachyon emitters at that range. I think there are a couple combat vessels with tachyon emitters as well. Patrolling scouts could also be a problem. Some systems are very packed and any number of these in close proximity of the command center could all but wall it off from a cloaked approach.

If you have to fight your way that close than it's not a raid, it's an invasion. I have strong opinions on this.

X range would have to far enough away to allow choice of method even in fairly dense systems.

Well, how about this: a command station module (or even special command station) that gets knowledge from all adjacent planets at 2/s or something, but causes the AI to aggressively spawn ships against that station (and if it's destroyed it's +40 AIP or something like that). Would probably be based on similar factors as my earlier suggestion of the super-terminal-like behavior.

I remember talk of modular command centers a while back, a module like this would be nice. Atm k-raiding must be done in enemy space, being able to have them come to you instead would be a nice change of pace. Having the effect be on par with the super-terminal-like behavior may be a bit much, would it key off of number of planet's being siphoned? I think it should.

Why not a new capturable AI object that blocks all player knowledge gain/supply in all adjacent planets while the AI holds it? Have it cost at least 20+ or so AIP if destroyed. If captured it could cost X resources per second and use Y energy (for each adjacent planet) but it will also steal 1 knowledge per second for each surrounding planet, maybe add a wave modifier as well.

Like an AI listening/observation post where you hack its data streams on capture to analyze footage of AI vessels in action perhaps? I like variety and we don't have any purely knowledge focused captureables atm.

It could also provide scouting intel for the area network it was connected to. (would simulate a previously scouted location based on the time of capture for systems that also had this structure in its network, say 3-6 planets depending on map size)

[Aegis]

I am surprised this topic got as much discussion as it did.

[keith.lamothe]

But Knowledge Raging would be such fun

Anyway, what I'm thinking is that the humans would have to build up some sort of research network, so in order to actually get knowledge from one you'd have to have, say, 4 of these special stations in the galaxy, and they'd have to be at least 2 hops apart from each other.  That's a couple new mechanics but I don't think terribly difficult considering that command stations already have special placement-eligibility logic.  Also, I think the knowledge gather rate would be 1/s, so if you wanted the k fast you'd have to do something in addition to that.  Probably a moderate wave multiplier on the stations too, but nothing precipitous.

The goal behind the minimum-station-count requirement being to make sure you can't do it super-early game; the numbers being subject to change of course.

Anyway, thoughts? I'm sure someone can shoot it full of holes

[Buttons840]

Thematically I'd buy a close range data hack if you needed a reason to get that close.

One last idea, although I still prefer my earlier proposal.  What if you could "hack the AI command station" and gain all 2,000 knowledge from the planet instantly.  This way you could focus on knowledge gathering for a brief condensed time and then be free to continue with your end game objectives without bother.

Basically, you could get knowledge raiding over and done with instead of being forced to thinly spread it over the entire game.

Oh, and when I say one last idea, that does not imply this is the last idea.

[Buttons840]

But Knowledge Raging would be such fun

Anyway, what I'm thinking is that the humans would have to build up some sort of research network, so in order to actually get knowledge from one you'd have to have, say, 4 of these special stations in the galaxy, and they'd have to be at least 2 hops apart from each other.  That's a couple new mechanics but I don't think terribly difficult considering that command stations already have special placement-eligibility logic.  Also, I think the knowledge gather rate would be 1/s, so if you wanted the k fast you'd have to do something in addition to that.  Probably a moderate wave multiplier on the stations too, but nothing precipitous.

The goal behind the minimum-station-count requirement being to make sure you can't do it super-early game; the numbers being subject to change of course.

Anyway, thoughts? I'm sure someone can shoot it full of holes

I like this.  Although if your going to consign them to the late game though this mechanic, then they no longer need a high resource cost or even a knowledge cost.  But yes, I find myself like this more and more as I think about it, because knowledge raiding can't be done away with in the early game by any means.  You must do your own knowledge raiding while you expand but once you've expanded you no longer have to worry about raiding planet after planet.

"get knowledge from one"

What is "one?"  Are you thinking of a command station or a stand alone structure?  You could make the research network command station (or whatever) give it a cap of 4, and, like you said, force them to be built at least 2 hops apart from each other.  Once they're all working you get 1k/s.  You could scrap and replace them once you have all the knowledge from all supplied planets, but at 1k/s this might never happen, and if it does I see nothing wrong with switching out the command station.

By 2 hops you mean if there is planet A - B - C  you can build one on planet A and planet C?

I'm excited, good ideas.

[rubikscube]

I am surprised this topic got as much discussion as it did.

well yah, last time i had a way better idea than this with nobody having a real argument to counter me and it got flamed, this has dozens of flaws but got more replies than this entire forum ever gets in a week in less than a day.

Spending money to reduce micro? how about using a fix to reduce micro, it's not like it gives you raw stats and micro is usually substitued by mods

[keith.lamothe]

I like this.  Although if your going to consign them to the late game though this mechanic, then they no longer need a high resource cost or even a knowledge cost.

Yea, I think it would still have an above-average metal/crystal (and thus time, minus engineers) cost, but it doesn't need to be a super-structure to construct.  And the goal isn't to make them exclusively late-game, though I guess it depends on definitions; I was thinking the latter part of mid-game, ideally right at that point where you can easily spare a knowledge-rambo fleet to do it manually; before that you can do it but it's a main-fleet activity.

[quote"get knowledge from one"

What is "one?"  Are you thinking of a command station or a stand alone structure?  You could make the research network command station (or whatever) give it a cap of 4, and, like you said, force them to be built at least 2 hops apart from each other.  Once they're all working you get 1k/s.[/quote]Yea, I think you guessed right; I mean they would not provide any knowledge until the whole network was up (though the moderate wave multiplier would still apply), and when they're all up you get 1k/s from each planet that they are on or adjacent to (that has supply).

You could scrap and replace them once you have all the knowledge from all supplied planets, but at 1k/s this might never happen, and if it does I see nothing wrong with switching out the command station.

Yea, not sure right now but I don't see a huge problem with that; something to worry about later anyway.

By 2 hops you mean if there is planet A - B - C  you can build one on planet A and planet C?

Yes, though as I said the numbers are subject to change.  One issue is that different map types make for very different levels of difficulty in achieving this, but that might be ok.

I'm excited, good ideas.

Yea, everything's fine when I start agreeing with you (just kidding)

[wyvern83]

I like it as well.

In a way it gives your middle-late game empire a sense of structure that links it together. Save for the flow of rear area ship production that some players have we don't have anything that does this. I think it's a nice touch.

Edit: Nvm I see my question was answered.

So adjacency would still matter then? I was hoping it would extend as far as your uninterrupted supply went. 

[superking]

'Warning - while you were typing 6 new replies have been posted. You may wish to review your post.'
no way, this tl dr is leaving the nest no matter what  :

I think the main problem is that a knowlege raid dosn't take the form of a raid per se

'raid' suggests at a swift, sustained attack that is completed before a superior force can intervene.

this sounds interesting, and requring skill.

however, as is, there is no aspect of a superior force that intervenes. when the planet has a large defence force,you either destroy it first or just infiltrate using a cloaker starship and not fight at all. In either situation, once the science vessel is in system it harvests resources. given that vessels are disposable and quick to replace, this makes aqquirng knowlege in the face of resistance a simple case of attrition. there is never any fear of failure- nothing is lost but time- and consequently any losses incurred during K gathering just provoke frustration with the resulting delay.

thus, the problems I perceive with the mechanic:

• In the event of a forceful K-raid, the risk is removed, making K gathering mundane
• In the event of a stealth K-raid, the risk is entirely evaded, likewise removing the interest
• Science starships are replaceable; losing them only costs time. This makes losing them no great deal.
• Knowlege gathering is continuous, meaning an interupted or prevented K-raid is still effective; in the event of losses, they are quickly reproduced and back K gaining. This makes AI action only capable of delaying, never parrying K raids
• K resource is gained, but nothing of value is risked; lack of risk or perceived threat causes boredom.

The mechanic can be broadly simplified as an exchange in which the player must pay time and attention (and losses if brute forcing and/or knowlege stealthing) in return for K resource.
risks of this activity: none
potential losses: time

a proposed solution

reform K-harvesting

instead of a steady income, implement a timer that begins to tick down when a science ship enters an enemy system eligable for supply raiding, maybe 200 seconds. When the timer expires, the science ship then harvests K very quickly.

On player held or neutral planets, the timer is skipped and K gain is very fast. This would make attritional attacks more diffcult while speeding up K gain on friendly planets. The time could also be longer on higher MK worlds to scale difficulty.

Timer length relative to AIP (higher AIP lower times): in the first 100, the whole process works out a similar speed as current on AI worlds (much faster on friendly ones with no timer ofc). by eg. 600, it is considerably faster. this is counterbalanced by increased difficulty of K-raiding during high AIP reinforcement.

this could be further enhanced by, for example, having science ships produced on low power mode, in which they behave normally but do not harvest knowlege. however, when placed on normal knowlege they uncloak if cloaked, lose the ability to move and the timer begins (perhaps the higher tech vessel can move..) this means cloaked ships can be used to move the science ships in system with gaurds, and move them out of the way- but once the K raid begins there is a window of danger, and if the raid is foiled before completion it fails without any gains.

these mechanics would make K raiding less related to attrition; a failed raid, one in which a ship dies without harvesting any K becomes possible. cloaking is useful but does not make the K-raid unassialiable. the addition of risk will create the sense of a time window of danger with a profitable result. Risk entails the possible loss of somthing vauable, so:

to create a resource at stake: I propose the loss of a science vessel to incur the cost of 1 AIP (the AI is concerned by evidence of humans gaining sophistication..) This entails a certain degree of anxiety with K-raids and should abolish that 'make 30 stations, put them in a transport, dump on planet, repeat' behavior. it also gives tougher, less efficent science vessels new signficance.

to create a risk: when a science vessel timer begins, an AI behavior is triggered. The behavior causes the AI to threaten the science vessels. idea as to what form this takes pending...

[RCIX]

That's..... not a bad idea! Though someone was bringing up the point that if you're going to just use a fleet and keep it in-system that long, you might as well take the system (and thus round up the knowledge real quick).

[wyvern83]

That's..... not a bad idea! Though someone was bringing up the point that if you're going to just use a fleet and keep it in-system that long, you might as well take the system (and thus round up the knowledge real quick).

That would be me.

[rubikscube]

well, neuter the planet then

[Awod]

But Knowledge Raging would be such fun

Anyway, what I'm thinking is that the humans would have to build up some sort of research network, so in order to actually get knowledge from one you'd have to have, say, 4 of these special stations in the galaxy, and they'd have to be at least 2 hops apart from each other.  That's a couple new mechanics but I don't think terribly difficult considering that command stations already have special placement-eligibility logic.  Also, I think the knowledge gather rate would be 1/s, so if you wanted the k fast you'd have to do something in addition to that.  Probably a moderate wave multiplier on the stations too, but nothing precipitous.

The goal behind the minimum-station-count requirement being to make sure you can't do it super-early game; the numbers being subject to change of course.

Anyway, thoughts? I'm sure someone can shoot it full of holes

I like the idea.

You could make them similar to colonies, or whatever those cloaked bases are that require saving.

To go further with what you were saying there would be 2 research stations in 2-3 hop jumps from your home world(then, so on and so forth from those stations), once you've explored a planet that has one there's a "chance" for it's cloaking to end. Which would add some spice into it.

Also, I am HORRIBLE at knowledge raiding, so this might actually work better.