Defense Supply Mechanic

[Minotaar]

While I was reading and thinking over this thread, a completely different idea appeared to me..

It is to significantly increase the radius of a turret's collision circle (or whatever it actually is), forcing a more staggered placement.

A couple things that led me to this:
1) We are discussing the "tower-defense" aspect of the game, as it's advertised. In all of the actual tower defense games, building space is a limited resource, and with good reason - why would you put your towers in different places if you could find a single one that's best? But in AI War, it's a non-factor, you can place truly ridiculous amounts of firepower on what amounts to a 25-square-foot-area (tm). I really think this is one of those cases where seemingly complete freedom just collapses in on itself, leaving an obvious best solution. (And AoE damage, which is the usual counterbalance to stacking stuff up in one place, is a very rare concern when defending)

2) Back when I was playing Starcraft II, I've seen a lot of discussion focused on pinpointing why it fells so different from the original Starcraft. One of the most interesting realizations was how much the difference in unit pathing impacted the feel of the game. In a nutshell, SC2 units use a better pathing algorithm that has a side effect of making the units form up in tightly packed balls. In SC1 the way unit pathing works is way less efficient, and the units tend to stagger much more. Because of this battles were larger in size and took more time to resolve despite armies actually being smaller on average.
It seems like this effect could be translated into AI War. Not for the mobile units, of course, we neither need nor can afford that. But for the turrets it makes sense.

Some of the implications:
a) Can't place a lot of turrets around a point without getting diminishing returns. You either have to sacrifice some of the range or put it out of range, which leads to..
b) It makes sense to have defense in depth - on a single planet! Positioning becomes both more interesting and more important. Munitions-Booster turrets anyone? Yeah, you can't have that with how things are right now, can you?
c) The battles can play out over a larger area, too, which is cool by me. Not sure if anyone honestly likes the 25-square-foot-area (tm) effect even if they like their single chokepoint (yeah, that's me)
d) The higher marks give you a higher defensive density because they take up the same amount of space, and it sure looks like they need some more incentives
e), no, E) It looks way more badass!
I mean, check this out

Taking a shot of Whipping Boy One at this point would be futile unless I showed you about 10 images at close-up.  Otherwise it's just a giant green blob of stuff.  I'm not sure even the close-ups would do it justice though.  Ignore the CPA.  That's literally just an annoyance at this point, utterly ignorable.

Giant green blob? Sounds exciting *yawn*. Now imagine actual spread-out defensive formations. Now imagine them firing all at once. There's people that do this already even with no restrictions,

I'm always trying to set my defences in a very orderly, symmetrical and aesthetically pleasing manner

therefore it matters! 

Alright, that's all I have for now, feel free to make holes in it! 

[chemical_art]

Bouncing off another summation:

1. Currently, all the chips and favor are toward the single defense world, and none toward multi defense worlds or a bend but dont break philosophy.
2. There currently is no combat or economic reason to not stack defenses on the edge of empire like a shell, except to counter raiders (really easy. Stick 6 snipers behind the border worlds.)
3. Turrets are not strong enough to be able to hold off waves or exo-waves if they are not stacked on a single world.
4. Turrets cannot win games, they prevent the losing of games
5. Turrets encourage stagnation.

To synthesize a bit on these points:
1. The only way to keep a fleet free is to have a turtle shell. Reclaiming worlds means tying of the fleet to recover, and multiple fronts means bouncing fleets around in a time consuming, nerve wrecking, and less likely chance to succeed way.
2. An economic nerf alone to the turrets will not stop turrets from being piled on border worlds, especially a single world. There needs a combat advantage to flexibility to make up for the inherent economic disadvantage of losing energy and M + C and time to recover a world.
3. Turrets cannot go on offense, and offense is needed to win games.
4. Turrets cause increasing threat walls. Spending K on turrets causes the dual effect of making the walls larger and decreasing the offensive ability to smash them.

[Hearteater]

How about an alternate, and fairly simple, solution to make defending multiple system viable without adversely affecting the current system at all:

1) Make Turrets immune to current attack boosts
2) Make a new Turret Attack Boost structure, which has a system cap of 1, and boost up to 100* turrets' attack damage by 100%*

Since you can only have one per system, to get the most bang out of your turrets, you want to spread them out with 100 per system so they are all double-strength.  You can still put them on a single system, and 100 of them still do +100% damage.  This unit would probably have a fairly small boost range allowing you to control which turrets get boosted better (think Tachyon Drone detection range) and would be at least cloaked, if not perma-cloaked.

* Randomly chosen example numbers, actual balanced numbers to be determined.

[keith.lamothe]

Well yes, that AIP + % increase would explode raid sizes eventually, but that was the point of my suggestion.  The game is thematically about avoiding the AI's notice. If the AI brings it's entire force against you, you get roflstomped.

I think that sort of "Sauron's Eye is Upon (Planet Name)" modifier would be one way to motivate some kind of defensive strategy that doesn't involve "stand here in broad daylight, and face them head-on" unless you really had just that much firepower.

It would still be fairly easy for the thoughtful player to adjust the current optimal (or at least most common) defensive pattern around, particularly through use of swapping the command station with a warp jammer.  Basically modify the current whipping boy pattern to look something like:

		(AI)
		|
(Squishy Insides)	--	(Whipping Boy)	--	(Decoy)	--	(AI)

Decoy would be a human planet with nothing important on it.  Either the whipping boy or the decoy would start with a warp jammer, and every other wave or so you would swap which one had the warp jammer.  That way the multiple-waves-to-same-planet bonus wouldn't stack up much or at all on either planet.  A cruder form would be to simply start with the whipping boy having a command station and the decoy not, and after a wave or two scrap the whipping boy command station (yeaaaah, not sounding so great, but warp jammers are expensive!), and alternate.

... so, actually, this sounds pretty good to me   Such a setup has more complex "map requirements" and the next logical step for the inventive player is to put some "preliminary" units on Decoy to help mess with the incoming attacks that start on it.  And either way, it gives the player a reason to think a bit more about their defense.  I'm sure there are better ways of dealing with it than the one I just outlined, and the exact map and situation would feed in to which one really fit best.  That sounds like a good thing to me.

My only concern, and I think this probably goes for anything I've talked about here too: would this cause more annoyance than it is worth?

I suppose the question has to be asked: What are we trying to accomplish here?

Primarily?  Make the game more fun.  If none of the ideas here will make the game more fun (because they take away choice without giving at least as much back, because they're just annoying, etc), then we do none of them.  There are plenty of other areas in the game where dev effort and your effort would be more likely to lead to improvement with less risk of messing things up.

But when the defensive side of the game is so marginalized (in terms of amount of thought that goes into it, and variations in actual practiced strategies, and "that was awesome!" moments) in an apparently large percentage of moderate-to-high-level games, then I really get the feeling that we're "leaving fun on the table".  In other words, there's a lot more potential fun experiences there that just don't happen because the optimal solutions are too obvious.

All that said, it may be that making a number of small, non-invasive changes (like this concentration multiplier for a series of waves finding only one target planet, making beachheads less of a binary choice between life and death) will be significantly better than trying to make a single more revolutionary change (like defense control caps).

[keith.lamothe]

How about an alternate, and fairly simple, solution to make defending multiple system viable without adversely affecting the current system at all:

1) Make Turrets immune to current attack boosts
2) Make a new Turret Attack Boost structure, which has a system cap of 1, and boost up to 100* turrets' attack damage by 100%*

Since you can only have one per system, to get the most bang out of your turrets, you want to spread them out with 100 per system so they are all double-strength.  You can still put them on a single system, and 100 of them still do +100% damage.  This unit would probably have a fairly small boost range allowing you to control which turrets get boosted better (think Tachyon Drone detection range) and would be at least cloaked, if not perma-cloaked.

* Randomly chosen example numbers, actual balanced numbers to be determined.

I like that.  I think it'd probably need to be an attack boost of something like 400% to really get the point across to the player.  That wouldn't affect things like tractors, gravs, and forcefields but I think the player would naturally want to combine those in due proportion with their offensive turrets (and if they really want to pile them all on the "shell", well, I don't see a problem with that; it will make the job of the backup defenses harder).

And this would provide a good hook for extension too: something like another turret-buffer that cloaks up to 100 turrets, but only if there are <= 100 turrets in the system, or something like that.  I don't know if cloaking would be a sufficient tactical advantage unless it was cloaked-while-firing, but that would probably be overpowered.  Or something that adds parasite damage to turrets, or paralysis (the recently requested "tazer turret"), etc.  Basically: you can still pile up your turrets if you want to, but here's tricks you can pull if you don't.

[keith.lamothe]

It is to significantly increase the radius of a turret's collision circle (or whatever it actually is), forcing a more staggered placement.

Hmm.  It couldn't be literally that, or ships (friendly or enemy) couldn't hold position within that area either.  But adding an extra distance for turret placement compared to other turrets would be doable.

I'm not sure how much it would add, but I'm interested to hear what others think on that.

[Hearteater]

Yeah, I think the main reason players are resistant to turrets-in-depth is you lose a system each wave and it is annoying to rebuild.  I was actually thinking cloaked firing turrets could work for such a system since they wouldn't need to be rebuilt ever raid.  Goes along with the Cloaked Command Station.  And if you tie in with the Turret Attack Booster, the Cloak Turrets + Attack Boost might give +300%, and the plain Attack Boost without Cloaking the turrets might give +400%.

For Whipping Boy + Decoy, why both with a Warp Jammer at all?  Just use normal Command Stations at both and since there are two Warp Gates to different systems, waves will naturally split between them, resetting the modifier automatically.  A Cloaking Command Station on the decoy would be ideal.

Turret Placement Radius: Would increasing this break existing games?

[rchaneberg]

Yeah, I think the main reason players are resistant to turrets-in-depth is you lose a system each wave and it is annoying to rebuild.  I was actually thinking cloaked firing turrets could work for such a system since they wouldn't need to be rebuilt ever raid.  Goes along with the Cloaked Command Station.  And if you tie in with the Turret Attack Booster, the Cloak Turrets + Attack Boost might give +300%, and the plain Attack Boost without Cloaking the turrets might give +400%.

I approve of this suggestion.
I don't think the game needs a new mechanic just to force a larger defensive surface area, I prefer suggestions like the 1-per-planet attack boost and cloaking stations to allow defensive worlds that don't have to stop an attack wholesale or be rebuilt.  As Hearteater said the main reason I build strong chokepoint systems is that if I build a defense in depth I have at least one, and probably 2-3 worlds worth of defenses to rebuild after an exo or cpa. With a chokepoint, I either don't lose a system and can just rely on remains rebuilders, or I lose most of my empire and typically either reload and build even more defenses or die to the next exo.

I also don't think that we need to cloak turrets while they are firing. Even when I do try and design deeper defenses my planets are going to be capable of stopping one wave at a time without fleet support, and exos tend to beeline toward your homeworlds so the only turrets I don't want the enemy to stop and engage are those defending my command station.
On the other hand, having turrets possibly have a firepower reduction while they are cloaked (similar to how forcefields work) seems like that would be best. Maybe similar to how flak and lightning turrets have a smaller reduction we could have snipers and missile turrets still doing almost full damage from cloak emphasizing that we want to build our defenses out of the line of fire and far away from any tachyon units.

[Toranth]

Yeah, I think the main reason players are resistant to turrets-in-depth is you lose a system each wave and it is annoying to rebuild.
(snip)

There are two other, very very large reasons: 
1)  Protecting valuable uniques.  That Factory IV?  That Blade Spawner Mk V Fabricator?  If I can only put light defenses into those systems, then to protect them I need to clear out all AI systems in 2+ hops.  This leads to the next major problem I see,
2)  AIP.  Defense in depth requires depth!  That means clearing out more planets, each of which is a major AIP jump.  The "Whipping Boy-Decoy" pair already requires one additional system to be cleared beyond what a single chokepoint system would require.


Lesser points that occur to me:
I like the idea of the "Eye of Sauron"-style "Attracting too much attention to one system is bad" mechanic.  It would certainly encourage having more than one defensive system, but provide a downside to the choice if the player wants to focus anyway.
The Warp Jammer is expensive, both as a command slot and in 5000K to unlock.  If having viable fixed defenses requires it and the twiddling required to avoid the "Eye of Sauron" effect, would it become cheaper/easier to use?
Exowaves.  Yes, I know they're optional, but I play every game with at least one exowave source active.  Usually two.  Exowaves are already painful, and that's with all the super-high-efficiency-slaughter you can get from stacking all of your turrets into a single system.
Limiting system defenses significantly reduces the purpose of the Black Hole Generator for human use, as well.  If a single system's defenses are limited, then the effect of the Black Hole Generator is greatly reduced, because there is much less to protect it. 
Finally, wrt annoyance:  time and cost.  A single Basic Turret Mk I costs 2000 M+C and 30 seconds to build.  Rebuilding a few hundred every 10 minutes, as the waves roll over a system or two each time, now imposes a significant penalty to the economy.  Which means more time waiting for fleet ships to rebuild, so we can resume the offense.

[Philature]

The current discussion raises a very good point: the games currently favour a unique whipping boy and having a single defensive choke point. Additionally, it was also suggested that the relative cost of turrets relative to fleet ships make them not as efficient unless you strongly focus them into a few or a single defensive point.

Even if I often have a single whipping boy and strongly focus my defense I do see the need to add diversity and have deeper guerrilla warfare tactic which would be fun. But as a matter of good game design, I strongly believe we should build system that reward alternative strategies and not simply punish curent strategies, after all the beauty of AI War lies in its depth of very interesting choices and strategies.

Hence, I really like the idea of a tower attack booster as proposed by Hearteater, I think they'll really bring the point accross that deepnening of defenses is good especially if the bonus is big and if other bonus can be latched on as well such as parasite, engine damage, armor damage and etc..

I also agree that taking the time to rebuild everything once the ennemies have passed your initial defenses is really annoying and is the main reason why I favour unique strong choke point. For guerilla tactic to be fun the mechanic to retake control of a given system need to be alleviated to become faster. Mechanic such as command station cloaking would not be bad but having simple thing, such as simplified or easier way to rebuild lost comand stations (such as having a destroyed command stations leave remains to rebuild instead of requiring a colony ship each time) would make the game smoother, faster and more enjoyable.

Finally, the "Eye of Sauron" idea is kind of interesting as well and really make sense, if you cap the bonus (let say 20% or 25%) its give enough incentive for the player to diversify their defenses but does not really force you into such a diversification. If the cap is too high players will try all srorts of shenanigans to avod the bonus and still use a single defense points, a lower cap give you just enough incentive to push you into trying something different and not neccessarily make your previous strategies useless while avoiding giving you too much incenctive to bypass the mechanic completly.

[Wanderer]

... so, actually, this sounds pretty good to me   Such a setup has more complex "map requirements" and the next logical step for the inventive player is to put some "preliminary" units on Decoy to help mess with the incoming attacks that start on it.  And either way, it gives the player a reason to think a bit more about their defense.  I'm sure there are better ways of dealing with it than the one I just outlined, and the exact map and situation would feed in to which one really fit best.  That sounds like a good thing to me.

My only concern, and I think this probably goes for anything I've talked about here too: would this cause more annoyance than it is worth?

I like thought and complexity in my defense.  I don't like constant micro there though.  Having to twiddle with my defenses after EVERY wave?  Nono, thanks.  Not at 5-10 minutes between each wave MAX.  That'll be as annoying as dropping back with my fleet constantly.

But when the defensive side of the game is so marginalized (in terms of amount of thought that goes into it, and variations in actual practiced strategies, and "that was awesome!" moments) in an apparently large percentage of moderate-to-high-level games, then I really get the feeling that we're "leaving fun on the table".  In other words, there's a lot more potential fun experiences there that just don't happen because the optimal solutions are too obvious.

Obvious isn't bad. Obvious isn't lacking in fun, particularly if you're still trying to attract new players.  Players who die too quickly never get to attack and just get run over.  The game, currently, at first glance looks like defending a lot of places could be a good idea, and that was part of my initial problems.  Right now, that's actually NOT obvious to a new player.

That said, I approve of the idea of 'encouragements' that will let you spread out more effectively.  Control Towers + Rebalance seemed one of the best ways to do this, even if it is invasive surgery.

[Toranth]

Random thought - Scale incoming wave strength by the strength of the defending system?  That would certainly give an incentive to use something other than one, single, super defended system to handle all incoming attacks.

Of course, actually finding numbers for that scale would be... difficult.

[Volatar]

I think the right direction to go is to encourage multiple fronts in various ways, rather than try to discourage a single front. Positive reinforcement vs punishment.

[CodeMichael]

You seem to just be dancing around a per-planet cap on turrets without calling it that.  It might be a soft cap, but it's a cap regardless.  The easiest thing would just be to make each turret beyond the soft cap reduce the system-wide effectiveness of that turret type and mark.  You can then fine tune the value so that more turrets cause the additional benefit to approach zero.
e.g 10 mkI basic turrets = all turrets at full power, but for each turret beyond 10 the overall effectiveness of turrets of that type is reduced by 3% (or something).  So at 30 turrets you've reduced the overall effectivness of mkI basic turrets by 60%.
I don't know what the correct values would be.  Ideally it would be enough so that a sane dispersal of turrest on a planets with lots of wormholes was not overly penalized (15 seems to be on the high end as these things go)
This lets people sub-optimize if they want to hard focus their defenses, which I think should be a viable option.

You can probably find some other "incentives" from keeping people reducing their front-line to a single planet as well.  Make the AI drop exo-wormholes every 4 hours or something.

[TechSY730]

From what I see,
To aid multiple chokepoints or multi-planet chokepoints (sort of like the setup Kieth showed, its a single chokepoint over several planets), turrets will need some sort of boost. Right now, you need near cap of them to effectively hold off mid to late game waves.
So some things that could help with that
1. Make them easier to get. Possibly increase cap or decrease K costs of higher marks (their current K costs often makes it counter-productive to unlock them). This way, you have more turrets to spread around.
2. Make each individual turret more powerful. This way, you don't need quite so many turrets to hold off attacks

However, while this would allow defensive setups not revolving around one planet to work better, it would also allow single-chokepoint setups to work better.
From what I can see, the concern is that a conventional boost to turrets, while would boost multi-planet defenses from iffy to effective, could also boost single-planet defense approaches from effective to overpowered.
I am assuming that this is why some sort of new mechanic or "limited effect boosting structure" or increased AI response to repeated attacks to the same planet or something is being proposed instead of the more conventional buffing techniques. Is this right?


I can think of three reasons why turrets don't work so great below near max cap.
1. DPS. As GUDare computed, laser turrets of all marks have a 5 million cap DPS, while bombers Mk. III have a 1 million cap DPS. Yet, laser turrets are still less useful. Yes, mobility is really that valuable, that 1/5th DPS is worth having mobility.
2. Durability. Turrets cannot run away when damaged. They cannot kite. Due to their fixed positioning, they are usually the first to take damage.
3. Limited range. If a unit flies outside of a turret's range, that turret is now useless at taking that unit out. A mobile ship could at least try to make it in range. Having more turrets means less likelihood of any one ship escaping all turret ranges

While the attack boost structure for only a limited number of turrets would help the sub-cap DPS issue without causing insane boosts to cap all on a single planet DPS, it doesn't do anything for durability or range. So what if my turrets can now deal 4x the damage if all of them still die in 4 seconds or their small "spread" means they are easy to kite around.

EDIT:
Another relevant question.
How much is mobility worth?
That is, if a unit where to lose its mobility, generally speaking how much increase to DPS and/or durability and/or range would that unit need to receive to still be considered worthwhile?
Or inversely, generally speaking, how much DPS and/or durability and/or range would you be willing to pay to give a stationary unit mobility?