AI Defense System: Moving Forward From Core Shield Generators

[Sizzle]

I'll just toss my previous idea in the ring again.  http://www.arcengames.com/forums/index.php/topic,7794.msg65935.html#msg65935

That being said, I don't mind the AIP per minute while deep striking.

[HitmanN]

Dropdown box with even more options sounds interesting, and once again allows the player to affect what kind of game to build. Plus further expansions and DLC can add new options to that list, rather than become completely new mechanics on top of the old.

The poll does make me want to vote for multiple things though. I mean, let's assume that 30 people vote. 15 vote for A and 15 vote for B as primary choice. All 30 would choose C as a second vote. This means C would get more votes despite not being the primarily preferred choice. It can still be considered something people would be happy to use, and thus fills in the same criteria as being a primary vote, in the end. That statistic is never known though, because secondary votes aren't possible. :<

Anyways, even though I suggested the 'hold X number of planets' scenario, for a primary vote I'm leaning towards Wingflier's new AIP increase penalty for deep raiding, or the following suggestion from the old thread. Original suggestion by jordot42, based on something by WinterBorn, with some random pointers from me:

A takeoff on an idea from WinterBorn:  Each AI homeworld has a "human firepower inhibitor" that starts at 99% efficiency.  What this thing does is reduce all human weapon effects and ship abilities (damage, EMP length, cloak radius, weapon boosting radii, etc) by 99%.

How to counter the inhibitor you ask?  Destroy the inhibitor energy reactors of course!  A small reactor reduces inhibitor efficiency by 2%, medium by 7%, and large by 15%, for example.  Each AI would have its own set of reactors.  Can only be destroyed on planet ownership change.  The larger the reactor, the farther from the human worlds it is.

So for those who want to rush the AI, go ahead.  Good luck doing piddly damage per attack.  But for those who want an even fight, the conquering of several planets would be mandatory.  

So now you have a choice on how you want to proceed while (hopefully) encouraging more planet capturing, as per the objectives above.

As a story side-effect, having the AI use energy reactors does go with the AI technology being based on the human technology (since humans created the AI in the first place).

I was just about to suggest something similar, although mine was that the ships at AI home have a permanent personal shielding that starts at 100%, and goes down by destroying CSG's. However, the idea of adding various levels of reactors all over the map is a marvelous idea, and limiting the player firepower could work too.

Take twenty planets with 5% reactors, or if you want to try to play with a low planet count, you'll be nicely challenged as to where you'll find ten 10% reactors or similar.

I wholly support something of this sort. With the amount of choices as to what planets to take, lack of freedom would be non-existent , and expanding would be rewarding if you want high level reactors at far away planets, and you'd get the reward straight away as the AI immediately becomes a step weaker at its home defense, but aeach planet you take also makes the AI stronger with AI Progress. It's up to the player to decide what point is the best to strike. High AIP and weaker defense, or try going in earlier, with low AIP but against higher extra shielding.

Challenging, encourages expanding, somewhat penalizes turtling, offers instantly noticeable rewards, contributes towards endgame, doesn't force any captures... You'd be constantly building and adjusting your setup for the final attack, choosing how to weaken the AI's, how to avoid making their offense stronger at the same time, and when to make the decisive strike. Best idea so far, IMO.

Essentially, it's a choice to go against high defense with low AIP, or against low defense with high AIP. Also, a correction to what what I mentioned there; regarding the percentage of the personal shielding for the AI Home ships, I meant that the percentage implies how much of the damage gets negated. If you drop the Shielding to 90%, your weapons do 10% of their normal damage to the ships on that planet. It's essentially the same as jordot42's suggestion, except the effect is applied to the AI ships, and not the player ships. Both sound valid options though.


I'll hold off my vote for the moment until I know what the full setup of choices ends up being.

Just, please, for the love of god, find a better acronym than "AIDS" for this. DX

Yeah, forcing the AI to have AIDS is pretty nasty.

[Yuugi]

Just, please, for the love of god, find a better acronym than "AIDS" for this. DX

Yeah, forcing the AI to have AIDS is pretty nasty.

I just had to imagine me trying to, for example, get my little bro, who still lives at my mom's place, into the game, and drop a line like "Just click on the AIDS selector right there", with my mom overhearing that.

I wasn't sure wether I wanted to laugh or cry right there.

[Lancefighter]

Just, please, for the love of god, find a better acronym than "AIDS" for this. DX

Yeah, forcing the AI to have AIDS is pretty nasty.

I agree, I dont believe the AI needs AIDS personally.

[Kemeno]

Actually the notion of a drop down menu for this is the first thing that's come up that I really dislike. That level of fragmentation in the game design is going to make it very hard for you guys going forward, I would think. Every design deciision will impact 5 different play styles built around 5 different mechanics, isn't that something of a nightmare?

I disagree. As long as the mechanics accomplish the same thing (in terms of the four goals Chris put on the previous thread) and there aren't a *huge* number of them, balance *should't* be a big problem. Progress gating and straight planet capture don't address the opening multiple fronts thing directly and so I don't think they're ideal candidates. In fact, the core shield generators really don't either (at least not any more than gating or straight planet capture), since you don't HAVE to hold the planets after you capture them - we have to hope that the incentive not to lose an advanced factory or fabricator is enough to have people hold these planets.

Wingflier's solution is really great; it really seems to address all of the major goals of these AI Defense Systems, assuming that you have to control the planets in order to avoid the AIP increase. You'll need to take and hold territory in order to avoid the AIP increase while attacking homeworlds. Balance here is going to be really tricky initally, but for people who want the most open ended experience, this addresses everything in a very subtle way, which I think is awesome. It forces you to hold a planet, at least temporarily, within a few hops of the AI home, or you're going to have a bad time AIP wise - especially if you lose it!

I had mentioned the shield supression nodes (or something similar) but not as a discrete AI Defense System of their own - that was all you, Chris (unless someone mentioned them)

On the topic of incentives to get ARS's: maybe they should give a tiny amount of bonus knowedge per planet (say, 250 or so), per ARS. That would really add up, but it would also be pretty tricky to get, since you have to do a long range raid to get the ARS and then you have to bring it back to your "core" to get the k-bonus! (Of course, that could be solved by making them perma-cloaked or tough as nails or something like that...)

I'd like to re-lobby my idea of turning the shield generators into a mini-scripted campaign of sorts, where instead of blowing shield generators, you're just capturing ARS to figure out how the shield works and then figuring out how to make a virus to disable the AI shield (by capturing adv. factory + adv. fabricator), and then deploying the virus to disable the shield. This is probably too much work and does the same thing as the shield generators - but it eliminates the extra structures, and it's essentially what the shield generator mechanic reduces to anyways (take a bunch of ARS's, advanced factory, advanced fab, and one of x other random planets. Need to figure out how to work the counter-attack guard post in there... maybe the act of disabling the shield triggers a CPA or AI deep strike counterattack) That said, I find nothing *wrong* with the shield generator mechanic; I just think this is a little more interesting flavor-wise. Also, I find the idea of dumping a virus into the AI to disable the shield really cool.

Of the ideas on the list now, the ones I like best are shield generators (ideally with a modification that makes *forces* the player to hold a small bit of territory) and supression nodes. Wingflier's idea needs a name but I could see playing with that too.

[Yuugi]

I'd just like to note that I dislike those hard caps, like the "need 500 AI progress to enter" or "need x systems to enter", but I wouldn't be against them being added as an optional choice. As long as they remain like that, optional.

This doesn't include the shield-inhibitor idea, which'd essentially be a normal-game counterpart to the spire plot, in a way (similar rules of inhibitor placement compared to spire hubs). But I'm not sure if inhibitors or spire hubs overlapping in role like that (with the inhibitors being much less attractive options compared to the hubs since they don't offer the other benefits) really is good gameplay-wise. I wonder if it'd end up with a lot of people stuffing their inhibitors into the shields of their spirehubs to laugh as the AI tosses the anti-inhibitor waves (which'd presumably be less fierce than the anti-spire-waves) against the hubs with little effect.


I think I like the Core Shield Generators (as is) and the Core Planet Booster ideas most. The latter being that suggestion that you find and kill nodes strewn through the galaxy that weaken the home defenses without being a binary "defeatable/undefeatable" switch. I'm not sure who brought it up initially.

[Echo35]

Granted I voted on the Shield Suppressor option, but the anti-deep strike waves does bring up something. There is already an AI Plot that hunts you down when you kill one of the AI homeworlds, why not add a plot for anti-deeptrike hunter killers too?

[Hypha]

I guess the hunter-killers were more my idea, so I should clarify them. I got the idea from a suggestion once floating around about the AI having some sort of mega starship that patrolled around and enter into battles on the AI territory that swung the engagement completely in the AI's favour. In my mind, they function in spawning terms to the Devourer, in that they are present right at the start and can only move within AI controlled space. There are a finite number of them and they don't respawn. They don't like being too close to human controlled worlds and will generally stay away from human controlled systems but they will hover closer to alerted worlds than not. If a human fleet is deeper in AI territory, a hunter-killer will gravitate towards the attacked planet and the longer you stay in the area, the more likely you will have to fight through one of these uber guardians. You don't have to fight them however if you stay close to your own territory. As your territory get closer to the homeworlds, you will be more likely to face these ships as they defend access to the core worlds, so you can still attack the homeworld and are not shut off completely from ever accessing it but you got to go through quite a fight to kill the hunter-killers. The strength of the hunter-killers though is external, locked up in the ARS and other capturables and the more of them you are able to get, the weaker they become. In this way, they are reduced in strength from a devourer golem or an avenger to something much easier. I hope this will make deepstriking more difficult and at the same time, more fun, as you try to sneak under the detection of these units and have a measure of risk in entering AI territory even if you neuter a lot of them. Who knows, maybe let the ARS be allowed to do crazier things with the hunter-killers, like give the AI negative penalties on planets it goes through or let one of them go berserk. In this way, you don't have to capture most of the ARS to be able to access the homeworld but you can unleash a lot of "accidents" in the AI territory the more you do have. This is similar to the virus idea for the core shields but I feel it is more entertaining to unleash a haywire deathmachine rather than just power down some shields. Sometimes I just like watching the AI have to fight something that isn't me. Make the AI regret the day it trusted finicky human technology.

In lore terms, I would like to think of them like Death Stars that were once used in the Civil War but got turned over by the AI and now serve as its guardians. The problem with them is that they have a ton of manual overrides and there is much within the devices which the AI doesn't understand. There is little said about the Civil war and I would like the think that the hunter-killers were a measure which the humans hoped they could use to eliminate the AI if it got out of hand and were the flagships of the ancient navies. The AI was able to develop a countermeasure that turned them over to itself and incapacitate the crews but it is still afraid that if they are placed too close to human worlds, some hidden override it hasn't found yet will kick in and the original mission will resume. The AI doesn't know that knowledge was lost with much of everything else but it can be rediscovered in the ARS, resulting in discoveries that can power down weapons systems and shields and restore some functions to humans, ending up with a paradox ship which has its weapons split between human and AI targets. They are symbols of the Civil War and the past that defined humanity, a past which also haunts the AI.

[tigersfan]

Personally, I love the AIP/minute idea. I've read every side of this argument for days, and to me, this is the best idea I've seen.

[Winter Born]

First, another salute to the developers bending over backwards to accomodate the fan base.

Second, the ARS value discussion just points out the dificulty of balancing the number of ships in the game. Every new ship creates an ever growing number of degrees of freedom to balance.

[lovekawakawaii]

I like Hypha's idea. Make cool new defeatable monster guardians that implement these gameplay shifts instead of 'forced collectibles' and the pill will be easier to swallow

[Hypha]

It also buffs up the ARS since the extra ship usually isn't useful. You can't build your strategy around them cause you get something random and you need knowledge to get them to tech 2, where you are already generally starved for knowledge anyway. I get a better bonus out of upgrading starships than this new fleet ship and I generally already chose the ship that works best for the strategy I want to pursue. Having the ARS activate other stuff would be great, since the chance of getting something you want to really use and upgrade fully is being diluted with each expansion.

[Echo35]

It also buffs up the ARS since the extra ship usually isn't useful. You can't build your strategy around them cause you get something random and you need knowledge to get them to tech 2, where you are already generally starved for knowledge anyway. I get a better bonus out of upgrading starships than this new fleet ship and I generally already chose the ship that works best for the strategy I want to pursue. Having the ARS activate other stuff would be great, since the chance of getting something you want to really use and upgrade fully is being diluted with each expansion.

Doesn't matter that it's random, having an extra ship type with extra things you can make your fleet strong against and extra ships in general (Since ship caps are PER SHIP) is extremely useful.

Though I can agree that there are so many different kinds of ships these days, especially with the expansions, that only getting one of them at random is a very diluted bonus. But at least it's free, so to speak.

[Elukka]

I like both the gameplay and lore of Hypha's idea. It seems a lot less artificial than simply spawning something when the humans go too far. And I definitely wouldn't mind more hints and fragments of the galaxy that was, back in human times.

[Arcain_One]

In reading of all the comments and ideas for the last few days I thought of a simple idea: at first the AI either doesn't know of your existence or just barely and that you are of very little threat, this is why it doesn't come to annihilate you in the beginning. When you have taken 1/3 of the planets (this would be 3-40 planets) there will be no question of if your a threat, the humans should be destroyed post haste. This also goes the same for the vulnerability of the core AI, if it is now able to be destroyed then the threat should be squelched first. [This makes deep strikes more inline to game but that is not my point nor the direction the game is being developed]

My idea is to split the game into two parts:
The first part is where the AI has definitive impenetrable defense as well as a false sense of security and it doesn't take the human force seriously, the second is where the AI acknowledges the threat and it devoting more resources to deal with it. (Why focus on something outside of the galaxy when your about to be blasted away by something of 'little threat'?)
This is somewhat already done with the AI progress and also with Core Shield Generators now. The AI already has a impenetrable but compromisabe defense network and also steadily sends more ships to combat the human rebels (I'm running out of synonym here ) as the humans present more of a threat. I think it should be more so.

My idea of the defense network is based of an idea I've hear before: mini cores. These mini cores would be MK V worlds but would not be as strong as the core home world. How many mini cores would be based on the greater difficulty of the two AIs and based on the map size.
There are actually two ideas here:
1) these mini cores take the place of the Core Shield Generators: you must destroy all of the mini cores to attack the main two. After destroying the mini cores the AI will get much more aggressive and more potent to defend its now vulnerable cores.
2) the mini cores are buffers to the main core you could destroy the mini cores first or the main cores first but which ever is second will be that much more difficult in the same manner as #1.

Another thought is that humans can do many deep raids, why can't/wont the AI? The counter attack post is a great example of it doing just that, I would like to see more. (sadly I don't have any ideas to contribute at this time)

The last idea I have to tell of right now is human made Counter Defense Grids. The humans must construct these CDGs to undo the protection the AI has at it cores. The catch is the CDGs must be spaced out, thus forcing the players to either capture more planets or spacing the planets out and leaving each one more vulnerable. The number of CDGs needed would also be based on the greater difficulty of the two AIs and map size.