Ye Ol' Armor Debate

[keith.lamothe]

Oh yea, if we went with the 1000/(1000+Ar) the actual armor values would need to be made up basically from scratch   The current ones would do quite little indeed.

[Hearteater]

Thinking about the amount of actual DPS swing overkill gives has made me question how important 10% vs 15% damage reduction from armor would actually be.  So the question is, would numeric Armor values be superior to a small set of fixed percentage named armor types, as summarized in the below post?

I don't have a direct problem with that setup Tech, but one concern with such a system is no one wants to multiply out two values in play (and three would be right out).  The armor value system gets around this by being a relatively minor factor, a simple percent, which people are good at groking.  If I could get x6 armor Hull and x3 from armor, that's x18 which is a huge swing in damage.  So I think enumerated Armor types would need to end up translating to simple percents (20% reduction, 40% reduction, etc) that are cancelled by the attacker's Armor Piercing.  Think how AP works in Warhammer 40K (basically, Armor Piercing is all or nothing).

Say four armor types: 25% reduction, 50%, 75% and maybe 90 or 95%.  Nice solid numbers that are quick and easy for people to internalize easily.

EDIT: This system of armor could still support scaling the actual mitigation percentage by attack damage to make big hits less affected by armor, which I'd mentioned earlier here.

[Wingflier]

But by how much would attack damage mitigate each armor type?  Would it be the same for all types?

Also, would armor piercing simply be a tag that ignored armor altogether?

[keith.lamothe]

Having thought about it, right now I have two main goals for armor:

1) Be able to have some "armored" types that have a higher "effective" cap-health against types that lack sufficient armor-piercing.
2) Be able to have those same "armored" types have a higher "effective" cap-health against types that rely on high rate-of-fire.

To that end, my current inclination is towards the following (which is basically an amalgamation of previous suggestions from here) :

1) Change armor_piercing calculation from just a type-specific bonus to (some_function(log10(damage_per_shot))) + type_specific_armor_piercing_bonus, with the result being a value between 0 and 1 with 0 meaning no armor piercing and 1 meaning "ignore armor".  This number is precomputed for each type and would be based off the "mark I" damage per shot for that ship type, rather than scaling with mark.  All the player sees for this stat is a % of armor penetration, so they don't need to know how it got that.

2) Make armor and armor-piercing not scale up with mark level.

3) Change damage-done-through-armor calculation from:
- Max((damage_per_shot-(armor-armor_piercing)),damage_per_shot*0.2)
- => Max(
   (damage_per_shot*1000)
   / ( 1000 + armor*(1-armor_piercing) )
   , damage_per_shot*0.1
)
Which is more complex but that's fine.  Could also change the display of the armor stat to the % reduction given when the attacker has no armor piercing (max 90%).

4) Make hull multipliers come after armor is applied.

5) Obviously, pick new armor and armor piercing values for... well, everything   The general idea balance wise is that armor is "cheaper" than cap-health in the sense that 10% armor would cost less than 10% more health.  Similarly, a +10% armor piercing type-specific bonus would cost less than simply doing 10% more base damage.

Thoughts?

[Hearteater]

I like that system, and my last post was basically referring to step #3 and is it really worthwhile to have any arbitrary percentage reduction possible?  I'm fine with it either way, just thought the question bared thinking on.

@Wingflier
Armor Piercing would be listed as the armor you penetrate, and any armor of that type or weaker is ignored.  So if the armor types are: Light(25%), Heavy(50%), Fortified(75%), Exotic(90%) and you had Armor Piercing: Heavy, you would ignore Light and Heavy Armor, but suffer fully from Fortified and Exotic armor.  That creates much more significant match-up effects, especially when paired with hull types.

[keith.lamothe]

I like that system, and my last post was basically referring to step #3 and is it really worthwhile to have any arbitrary percentage reduction possible?  I'm fine with it either way, just thought the question bared thinking on.

Right, it was worth thinking about.  If I was only trying to achieve goal 1 then it could just be 4 possible levels of armor, 4 possible levels of armor piercing, done.

Goal 2 requires more granularity to work pleasantly, I think

[Hearteater]

You can still modify the percentage of the four fixed armors by the units attack value.  Effectively the Light(25%) = 333 Armor, Heavy(50%) = 1000 Armor, Fortified(75%) = 3000 Armor, Exotic(90%) = 9000 Armor.  The main two differences would be handling armor piercing and armor buffing effects would be odd with only four armor types.

But meaningful armor has me extremely happy, so I'll stop bothering you and let you get some actual work done

[keith.lamothe]

You can still modify the percentage of the four fixed armors by the units attack value.  Effectively the Light(25%) = 333 Armor, Heavy(50%) = 1000 Armor, Fortified(75%) = 3000 Armor, Exotic(90%) = 9000 Armor.  The main two differences would be handling armor piercing and armor buffing effects would be odd with only four armor types.

But meaningful armor has me extremely happy, so I'll stop bothering you and let you get some actual work done

Actual change will probably have to happen after the expansion (we're planning to release AS as 6.0, and then do another official about a month after as 6.1 or whatever with an upgrade to Unity 4, because that will break PowerPC support), because the rebalancing work will be fairly substantial.  Though I think we can pare it down to a reasonable amount of time by doing some math models first on what a given set of numbers and formulae will give in terms of change to effective-cap-health in the various scenarios.

On the other hand, there's more time between now and October than there will be between October and 6.1, probably, so maybe doing this sooner than later wouldn't be a bad idea.

I guess the main question there is: are there any strong objections to the approach I described in the post above?

[Wingflier]

No objections here.  Since we're already testing the new energy system which seems very beta (will probably need a lot of fine-tuning before it's finished), I say we throw in the new armor system and hull type redistribution as well and get it all out of the way at once (kill two birds with one stone).  Sure, it may not be pretty for awhile but that's why it's called beta ;p

Now that we've come to a conclusion about armor's purpose in a broader sense:

1) Be able to have some "armored" types that have a higher "effective" cap-health against types that lack sufficient armor-piercing.
2) Be able to have those same "armored" types have a higher "effective" cap-health against types that rely on high rate-of-fire.

I want to talk about how it will be specifically implemented into the game.

My understanding/vision of the new armor system is this:

1. Because armor will effectively become a new "hull type" (even if its not labeled as such), the current hull type multipliers can probably be reduced in many or most scenarios to reflect that.

2. Bombers will be the quintessential "anti-armor" unit.  I think this fits their category and role very well, and gives players an instant understanding of their purpose. The ships they were previously designed to be strong against may need to be armored to compensate.

3. Fighters will become the "death by 1,000 papercuts" example role.  They may have a small level of armor piercing to compensate.  While weak and easily destroyed, they should excel against most targets except the most heavily armored ones, which bombers are made for.

4. Frigates become an "escort" ship archetype, with no armor piercing but perhaps medium armor plating (making them vulnerable to bombers and stronger to fighters).  Their role is to take out the weaker Fighter-class ships from a distance before they can come in range and reap havoc.  The firing rate or missiles per salvo may need to be increased to accomplish this new role.

In other words, making the role of the 3 triangle ships (and their counterparts) more intuitive, and less "special damages" based could be easily accomplished with this new armor system.  It would be neat if the new system was simple enough for new players to easily understand, while deep enough to give a unique and meaningful role to all the current bonus ships.

Am I understanding this correctly or is my interpretation off?

[Wanderer]

I guess the main question there is: are there any strong objections to the approach I described in the post above?

In the end, from the 'on the fly' player perspective, I'd like to clarify what we'd see and what effect it would actually have (round numbers).

Ship A has 10,000 HP and 50% armor.
Ship B does 2,000 damage/shot and has 75% armor piercing.
Ship C does 2,000 damage/shot and has 25% armor piercing.
Ship D does 2,000 damage/shot and has 0% armor piercing.

Ship B takes a shot at ship A, Ship A ends up at 8,000 HP because the AP beats the armor.
Ship C takes a shot at ship A (repaired), and Ship A ends up at 8,500 HP because 25% of its armor remained 'unpierced'.
Ship D takes a shot at ship A (repaired), and Ship A ends up at 9,000 HP because its armor fully kicked in against the shot.

Is this correct?

[Diazo]

Hmmm, here are my thoughts on first glance.

1) Change armor_piercing calculation from just a type-specific bonus to (some_function(log10(damage_per_shot))) + type_specific_armor_piercing_bonus, with the result being a value between 0 and 1 with 0 meaning no armor piercing and 1 meaning "ignore armor".  This number is precomputed for each type and would be based off the "mark I" damage per shot for that ship type, rather than scaling with mark.  All the player sees for this stat is a % of armor penetration, so they don't need to know how it got that.

If this is precomputed, why bother with the formula? I can't see the in-game difference as the game is referencing a static % at run time.

Otherwise, no issue for me here.

2) Make armor and armor-piercing not scale up with mark level.

I realize you mean this in the sense that it does not get multiplied by Mark level as currently exists. I'd still like to see something that makes a higher Mark level more powerful then a simple doubling of its raw stats but leaving that out will not break this implementation.

3) Change damage-done-through-armor calculation from:
- Max((damage_per_shot-(armor-armor_piercing)),damage_per_shot*0.2)
- => Max(
   (damage_per_shot*1000)
   / ( 1000 + armor*(1-armor_piercing) )
   , damage_per_shot*0.1
)
Which is more complex but that's fine.  Could also change the display of the armor stat to the % reduction given when the attacker has no armor piercing (max 90%).

Pretty much what we have been talking about.  100 Armor = 91% Dmg done, 250 Armor = 80%, 500 armor = 67%, 1000 Armor = 50% and so on. You would have to get up to 9000 armor to hit the 10% minimum damage threshold, I suppose with armor boosters that might be hit?

4) Make hull multipliers come after armor is applied.

Oh wow, big change. With this attack multipliers no longer act as effective armor piercing, a ship with armor piercing could now do significantly more damage then a ship with attack multipliers under the right circumstances.

5) Obviously, pick new armor and armor piercing values for... well, everything   The general idea balance wise is that armor is "cheaper" than cap-health in the sense that 10% armor would cost less than 10% more health.  Similarly, a +10% armor piercing type-specific bonus would cost less than simply doing 10% more base damage.

Unavoidable really. Having said that, most units will be at 0 armor and 0 armor piercing I would think, units like the fighter would drop to 0 armor (from 300) in the new system?

My only objection comes back to the fact that Mark levels would become linearly straight. A Mk II is exactly twice as powerful as 2x Mk I ships. I'd still like to see a small bonus of some sort for upgrading but it looks like that is not in the cards.

Having said that, we'll see what I think tomorrow, I've already flip-floped on this topic once.

edit to add:

Wanderer: That is not quite how the system works, at least with the formula Keith has posted.

Ship A has 10,000 HP and 50% armor.
Ship B does 2,000 damage/shot and has 75% armor piercing.
Ship C does 2,000 damage/shot and has 25% armor piercing.
Ship D does 2,000 damage/shot and has 0% armor piercing.

First, 50% armor is 1000 armor so:

Ship B takes a shot at ship A and Ship B ignores 75% of Ship A's armor, so Ship A has 250 armor and takes:
2000* 1000/(1000+250) = 1600 damage (80%) taken.

Ship C takes a shot at ship A and ignores 25% of Ship A's armor, so Ship A has 750 armor and takes:
2000* 1000/(1000+750) = 1143 damage (57%) taken.

Ship D takes a shot at ship A and ignores no armor, so Ship A has 1000 armor and takes:
2000* 1000/(1000+1000) = 1000 damage (50%) taken.

D.

[Wanderer]

My only objection comes back to the fact that Mark levels would become linearly straight. A Mk II is exactly twice as powerful as 2x Mk I ships. I'd still like to see a small bonus of some sort for upgrading but it looks like that is not in the cards.

Roughly 4x, actually.  You've left out doubling of health which allows for longer survivability as well.

[TechSY730]

No objections here.  Since we're already testing the new energy system which seems very beta (will probably need a lot of fine-tuning before it's finished), I say we throw in the new armor system and hull type redistribution as well and get it all out of the way at once (kill two birds with one stone).  Sure, it may not be pretty for awhile but that's why it's called beta ;p

Now that we've come to a conclusion about armor's purpose in a broader sense:

1) Be able to have some "armored" types that have a higher "effective" cap-health against types that lack sufficient armor-piercing.
2) Be able to have those same "armored" types have a higher "effective" cap-health against types that rely on high rate-of-fire.

I want to talk about how it will be specifically implemented into the game.

My understanding/vision of the new armor system is this:

1. Because armor will effectively become a new "hull type" (even if its not labeled as such), the current hull type multipliers can probably be reduced in many or most scenarios to reflect that.

From what I understand, it won't be. The code seems to still allow armor (and armor peircing and armor rotting) to be an arbitrary integer. It may be displayed different (like the %reduction from a shot with no AP), but the internal armor value can still be all over the place. (presumably from 0 to 1000, though the damage equation would still work for values outside of that range, just a bit oddly))
Now, it may be that the values used will be a discreet subset of the possible range of armor values, but that wouldn't be a property of the armor, but rather of current balance.
Now it could be that Kieth will implement an discreet enum of armor ratings, and map those to armor values for the equation. IDK. But the damage equation allows for high granularity.

2. Bombers will be the quintessential "anti-armor" unit.  I think this fits their category and role very well, and gives players an instant understanding of their purpose. The ships they were previously designed to be strong against may need to be armored to compensate.

3. Fighters will become the "death by 1,000 papercuts" example role.  They may have a small level of armor piercing to compensate.  While weak and easily destroyed, they should excel against most targets except the most heavily armored ones, which bombers are made for.

4. Frigates become an "escort" ship archetype, with no armor piercing but perhaps medium armor plating (making them vulnerable to bombers and stronger to fighters).  Their role is to take out the weaker Fighter-class ships from a distance before they can come in range and reap havoc.  The firing rate or missiles per salvo may need to be increased to accomplish this new role.

In other words, making the role of the 3 triangle ships (and their counterparts) more intuitive, and less "special damages" based could be easily accomplished with this new armor system.  It would be neat if the new system was simple enough for new players to easily understand, while deep enough to give a unique and meaningful role to all the current bonus ships.

I would be fine with making their roles more obvious and focused (the missile frigate sort of needs some sort of a buff), but I think that the hull type multipliers need to stay, though perhaps in a lesser magnitude, to keep their "triangle" of counters within themselves. (Standard fighter beats bomber, bomber beats missile frigate, missile frigate beats fighter, though the "order" of the triangle can be adjusted). Also, if the standard fighter is to become the "death by a thousand cuts" style ship, should the AOE immunity be moved from the missile frigates to the fighters, or have the AOE immunity be removed all together from the triangle ships? (Personally, I would just remove it. AOE immunity seems to exotic for the standard triangle)

[TechSY730]

My only objection comes back to the fact that Mark levels would become linearly straight. A Mk II is exactly twice as powerful as 2x Mk I ships. I'd still like to see a small bonus of some sort for upgrading but it looks like that is not in the cards.

Roughly 4x, actually.  You've left out doubling of health which allows for longer survivability as well.

Not quite, two Mk. I ships would have double effective attack AND double effective health, which is captured by the Mk. II ship. So yes, a Mk. II is basically two Mk. Is

The whole linearly straight thing as Mk. goes up is an intended design goal. There is really supposed to be no advantage for going up in mark other than more ship cap. (Right now, despite the good intentions, the Mk. going up with armor basically violates this goal, even though that change was meant to satisfy it. Oops)
Now whether this is a design goal that could use tweaking (like allowing slightly more than linear growth in durability) may be a good idea.

[Diazo]

edit: bah, Tech the ninja beat me to it.

My only objection comes back to the fact that Mark levels would become linearly straight. A Mk II is exactly twice as powerful as 2x Mk I ships. I'd still like to see a small bonus of some sort for upgrading but it looks like that is not in the cards.

Roughly 4x, actually.  You've left out doubling of health which allows for longer survivability as well.

I don't follow. 1x Mk II and 2x Mk I have the same amount of health.

Yes, this means the Mk II keeps its full DPS for longer then the Mk I which losses half its DPS when the first Mk I dies, but I'm not sure how valuable that is, in my major fleet engagements I have ships dying a lot faster then the reload timers.

Yes, it's an effect but my feeling is that in game it is a lot smaller then it looks on paper.

D.