So, this whole armor thing

[keith.lamothe]

Could that flat reduction value simply vary with scale then, just as pretty much every other value varies? Or am I misunderstanding how scales work?

It could scale with that, but then the Armor Ship's durability vs other scaled units (i.e. fleet ships, etc) would vary massively with scale.

[RCIX]

Or hell, all armor could be % reductions. Flat reduction sometimes just doesn't work (like in the case where you have to account for mass variety in number of shots vs shot damage).

[keith.lamothe]

Or hell, all armor could be % reductions. Flat reduction sometimes just doesn't work (like in the case where you have to account for mass variety in number of shots vs shot damage).

I don't actually mind that a flat reduction nails high-rof attackers.  That's one of the things I like about the current mechanic, except that swarmers are already getting nailed by several balance dynamics.

The problem is that you can't have flat reduction, scaled units, and non-scaled units without scale making a massive difference in the effectiveness of said flat reduction.

[Toranth]

My current idea, which both fixes the situation and gives a role to (the majority of) armor-related units now:

Remove armor as a general mechanic, then the few units which are armor-dependent still manipulate damage. For instance, armor ships would have armor as their "gimmick" and say "This unit reduces all incoming damage by a flat amount per shot" or something. Armor boosters and inhibitors would increase and decrease damage taken by a % to all enemies/allied units respectively. Armor Rotters would apply a stacking % damage increase on the target, up to some very high cap (yay anti huge units!!). etc.

Would there still be an "Armor Piercing" ability?  If so, would it be flat (ignores X armor, or X minimum damage) or percentage (Ignore X% armor, does x% minimum damage)?
If Armor Piercing is replaced by your +/- damage modifier, what happens to the "Armor Damage" ability?

Not criticizing here, just curious as to your thoughts.

[Mánagarmr]

Or hell, all armor could be % reductions. Flat reduction sometimes just doesn't work (like in the case where you have to account for mass variety in number of shots vs shot damage).

I don't actually mind that a flat reduction nails high-rof attackers.  That's one of the things I like about the current mechanic, except that swarmers are already getting nailed by several balance dynamics.

The problem is that you can't have flat reduction, scaled units, and non-scaled units without scale making a massive difference in the effectiveness of said flat reduction.

Oh. There are non-scaled units of course. I never thought of that. Whoops, yeah that complicates things!

[keith.lamothe]

If Armor Piercing is replaced by your +/- damage modifier, what happens to the "Armor Damage" ability?

I believe that's what he was addressing with:

Armor Rotters would apply a stacking % damage increase on the target, up to some very high cap (yay anti huge units!!).

Which is actually similar to what I've thought of replacing armor damage with in the past, if armor went the way of the dodo.

Something like minipods might just lose the armor damage altogether and get something else to compensate, but the armor rotter itself could get this.

[RCIX]

My current idea, which both fixes the situation and gives a role to (the majority of) armor-related units now:

Remove armor as a general mechanic, then the few units which are armor-dependent still manipulate damage. For instance, armor ships would have armor as their "gimmick" and say "This unit reduces all incoming damage by a flat amount per shot" or something. Armor boosters and inhibitors would increase and decrease damage taken by a % to all enemies/allied units respectively. Armor Rotters would apply a stacking % damage increase on the target, up to some very high cap (yay anti huge units!!). etc.

Would there still be an "Armor Piercing" ability?  If so, would it be flat (ignores X armor, or X minimum damage) or percentage (Ignore X% armor, does x% minimum damage)?
If Armor Piercing is replaced by your +/- damage modifier, what happens to the "Armor Damage" ability?

Not criticizing here, just curious as to your thoughts.

No armor piercing, armor is a lot stronger when all the important stuff doesn't just ignore armor anyway.

[keith.lamothe]

No armor piercing, armor is a lot stronger when all the important stuff doesn't just ignore armor anyway.

Well, if you have no armor piercing, and armor is just a % reduction, then isn't that just a new hull type?

[Chthon]

Just a thought here, why does armor have to be a flat % anyways.  The way I understand it, currently:
RealDamage = max(.2 * Damage, Damage - Armor)

This results in a minimum 20% of damage dealt when damage is less than 1.2 * Armor, but armor quickly becomes negligible to larger shots.

Not a flat % system.  However what about something like this:
RealDamage = Sqr(Damage) / (Damage + Armor)

In this example:
When you have lower damage than armor, than armor your damage drops off.
When you have equal damage to the armor, damage dealt is 50%
When you have higher damage than armor, you deal an increasingly more amount of damage with little lost.  Eventually reaches a point where it approaches the current system of Actual Damage = Damage - Armor.

What you get from this formula is: Damage * ratio of damage to armor.  Armor piercing effects will shift the spectrum by simply reducing armor like it currently does allowing for more damage to go through rather than being simply being absorbed by the armor.  Then you just normalize HP to about 50% of it's current value, or you could double damage.

This way ships within a tier can be evenly classed, but not have their armor completely be ignored by a higher tier or two (one tier higher 34% more damage, two is 50% more).  After a certain point though it becomes a landslide.

Some datapoints to help you see the curve this would create on damage to armor ratio.  All work is done in % of armor the incoming damage represents vs. % of damage adjustment
10% = 9%
20% = 16%
30% = 23%
40% = 29%
50% = 33%
60% = 38%
70% = 41%
80% = 44%
90% = 47%
100% = 50%
110% = 52%
120% = 55%
130% = 57%
140% = 58%
150% = 60%
160% = 62%
170% = 63%
180% = 64%
190% = 66%
200% = 67%
300% = 75%
400% = 80%
500% = 83%
600% = 86%
700% = 88%
800% = 89%
900% = 90%
1000% = 91%
2000% = 95%
3000% = 97%
4000% = 98%
5000% = 98%

Edit:  Just thought I'd throw in a few examples here of how the new armor system would work.

Bomber Mk 1 vs Bomber Mk 1
9760 damage, 1200 armor.  Actual damage = 8691 (11% damage absorbed)
Bomber Mk 1 vs Bomber Mk 1 with armor booster
9760 damage, 1200 x 3 armor.  Actual damage = 7130 (27% damage absorbed)
Bomber Mk 2 vs Bomber Mk 1
19560 damage vs 1200 armor.  Actual Damage = 18429 (9% damage absorbed)
Bomber Mk 2 vs Bomber Mk 1 with armor booster
19560 damage vs 1200 x 3 armor.  Actual Damage = 16519 (16% damage absorbed)
Bomber Mk 1 vs Bomber Mk 2
9760 damage, 2400 armor.  Actual damage = 7833 (20% damage absorbed)
Bomber Mk 1 vs Bomber Mk 2 with armor booster
9760 damage, 2400 x 3 armor.  Actual damage = 5617 (43% damage absorbed)
Bomber Mk 1 vs Bomber Mk 5
9760 damage, 6000 armor.  Actual damage = 6044 (38% damage absorbed)
Bomber Mk 5 vs Bomber Mk 1
48800 damage, 1200 armor.  Actual Damage = 47628 (2.4% damage absorbed)
Bomber Mk 1 vs Missile Frigate Mk 5
9760 x 6 damage, 1500 armor.  Actual Damage = 57097 (2.5% damage absorbed)
Bomber Mk 1 vs Missile Frigate Mk 5 with armor booster
9760 x 6 damage, 1500 x 3 armor.  Actual Damage = 54381 (7.2% damage absorbed)

From these examples you can see that armor boosters can become more relavent without ruining damage bonuses.  Actual final armor values may need tweaking.

[keith.lamothe]

RealDamage = Sqr(Damage) / (Damage + Armor)

(...)

When you have equal damage to the armor, damage dealt is 50%

The idea sounds interesting but I'm not sure how these two lines go together:

If firing a 100 damage shot vs a target with 100 armor, wouldn't that give;

Sqrt(100) / (100 + 100) = 10 / 200 = less than 1 HP worth of damage.

Unless "Sqr" means "Square of" instead of "Square Root of" but that would (edit: yep, I shouldn't be trusted with math, that's what it meant, as that way it yields 50)

[Cinth]

RealDamage = Sqr(Damage) / (Damage + Armor)

(...)

When you have equal damage to the armor, damage dealt is 50%

The idea sounds interesting but I'm not sure how these two lines go together:

If firing a 100 damage shot vs a target with 100 armor, wouldn't that give;

Sqrt(100) / (100 + 100) = 10 / 200 = less than 1 HP worth of damage.

Unless "Sqr" means "Square of" instead of "Square Root of" but that would (edit: yep, I shouldn't be trusted with math, that's what it meant, as that way it yields 50)

(dam)^2 would have been better nomenclature for using the square function  

It looks confusing to me too.

[keith.lamothe]

Damage*Damage causes less CPU nervousness too

But (Damage*Damage)/(Damage+Armor) has the same underlying implementation problem as the current system.  I had typed out an explanation but my browser ate it, so real quick:

On epic, on high caps, single mkI laser gatling firing on mkI zenith starship (ignoring armor piercing) is 150 damage vs 1000 armor or (150*150)/(150+1000) = ~19.6 or about 13%

Same thing on ultra-low is 1200 damage vs 1000 armor or (1200*1200)/(1200+1000) =  ~655 damage or about 55%

If you're curious, factoring in the gatling's 600 armor piercing gives (150*150)/(150+400) = ~41 or about 27%, and (1200*1200)/(1200+400) = 900 or 75%.

Very significant balance implications of different scales.

The current system has this problem too and hides it by some nasty ju-ju involving having two sets of armor and armor-pen numbers for each unit: one for against scaled units, one for against non-scaled units.  Makes it very difficult to derive any (correct) general tactical theory from the mechanic.

Which is probably the main reason I want to move to a % reduction or just get rid of armor altogether.

[Hearteater]

As my previous favorite Armor idea isn't winning any votes, what about this very simple option:

"Armored" ships do not take bonus damage regardless of what the hull type match says.  Or rather, they take MIN(bonus, 1.0).  Unless the enemy has Armor Piercing, which allows them to deal their bonus damage against Armored targets.  If that's too harsh, maybe instead half hull bonus, so MIN(bonus/2, 1.0).

[Arcain_One]

I've argued in the past that the ultra-light, light, medium, heavy, and ultra-heavy armor types are somewhat flawed.
In my mind, the hull type should represent the "material", the armor stat should reflect how "quality" that material is (how good it is at doing its job of deflecting damage), and HP should reflect, well, how much damage it could take (this last mechanic is probably in a good state right now ).
Yes, the names themselves are just costmetic, but it would help avoid the "temptation" to lump stuff together based off of stats not related to what hull type is supposed to model, which is what happened with the 5 hull types mentioned above (and to a lesser extent, the "role" hull types like artillery)

Instead of "material", the hull type stat could be tied into the "role" instead (like sat rename polycrystal to bomber, instead of say, renaming artillery to, IDK, magnesium). However, that does seem like it would cause a bit too much "homogenization" in roles (aka, make the whole thing of "role clones" like the "fighter clone" or the "bomber clone" an even worse problem, now that damage bonuses would, more or less, treat all of them the same)


Yea, I know that this thread is about the armor stat, but armor, hull type (and thus damage bonuses), and HP are the three "pillars" of durability both overall and per match-up. All three of them are linked intricately when balancing.

While I don't agree with it, I do see why the "just remove an armor mechanic entirely, hull type and HP are enough" idea has gotten some support. The amount of effort to competitively balance N factors is (asymptotically) exponential over N. Thus, even going from 3 to 2 factors can make balance much more achievable and more stable, especially given the number of ship types in the game. (Anyone know how the "effort" scales over number of entities types to balance, as opposed to the number properties of those types? Is it linear, or polynomial, or NlogN?)

I'm reluctant to say this but maybe the problem here is the bigger problem with the whole mess. The armor mechanic + the hull mechanic + the 'other stats' mechanic don't mesh together. This could be resolved with taking out armor but that leads to rebalancing and repurposing (not much of a problem when your tweaking the game anyways). I view that armor can not be seen as a side idea that can be added on to fit a purpose (don't duck tape wings onto a bus and expect it fly).

The point of my ramblings is I view the hull mechanic needs more of a overhaul than the armor mechanic, or both at the same time.

[keith.lamothe]

@Hearteater: Not super wild about that idea, but it would be better than what we have, yea.

The point of my ramblings is I view the hull mechanic needs more of a overhaul than the armor mechanic, or both at the same time.

Seems a bit vague so I don't really understand the problem you're seeing with it (not saying I don't see problems, but that doesn't mean I know which ones you see).  To clarify, what changes would you want to see happen?