Secondary Armor Bonus.

[SKW]

Is there any way to upgrade a unit's armor seperate of the standard upgrades?

Say, I wanted a upgrade for militia to gain additional armor (+1), how do I go about it?

 

[DaneTheBeast]

I guess creating a custom Upgrade would work. If not, then a custom ability that is hidden.

 

[SKW]

Problem is, there's no "Defense" upgrade. just attack and health.

And considering I'm upgrading Militia, seperating their hp from that of peasants is bound to be trouble.

Here's my current Version:
DISCIPLINE:
Increase Health and attackspeed by 20% for Human Infantry Units

 

[Wwallace_]

Problem is, there's no "Defense" upgrade. just attack and health.

So, how do you want to upgrade their defense?

You may use secondary triggers
ie
I have a upgrade that gives 6 increased stats
while normal upgrades only have 4

>Trigger detects when the upgrade is research
>Trigger auto researches the second upgrade, containg the extra 2 stats

 

[Death Adder]

The only upgrade type that increase armour use the "defence upgrade" data field for each individual unit.

If you need the unit to get different amount of armour from different upgrades then you probably have to make the armour increase lower and give extra levels of the upgrade which is supposed to give more armour.

Otherwise perhaps you can temporarily add an inventory ability and give a tome that increase armour to the unit(s).

 

[Xonok]

There is no tome that directly increases armor, but there are some item abilities that don't show and icon and do give armor.

 

[Arrr]

There is someway:
set the armor increase on the unit to 1,
then have two upgrades, one with only 1 effect (armor upgrade)
and the other with 2 effects, bot armor upgrade

this means that when researching the first upgrade, the unit will get +1 armor, while when researching the second one it will get 2.

 

[teozamait]

So, how do you want to upgrade their defense?

You may use secondary triggers
ie
I have a upgrade that gives 6 increased stats
while normal upgrades only have 4

>Trigger detects when the upgrade is research
>Trigger auto researches the second upgrade, containg the extra 2 stats

Wallace, the AMOUNT of armor a unit gains from an upgrade is controlled via the Unit's Object Editor Field - Armor gained from Upgrade. Even if an Abomination could benefit from two different armor upgrades, the armor it will gain per upgrade will always be +3, regardless of the upgrade.

A solution:

1. Make a custom ability based on the Item Ability +1 Armor.

2. Make a custom upgrade based on Flare research (or any upgrade that does nothing on its own - all of the fields are empty.)

3. Make the custom ability from point 1. have as a requirement the custom upgrade from point 2.

4. Give the ability from point 1. to the Militia. The armor will appear as a + 1 permanent addition to the unit's armor (in green, like the bonus from a Ring of Protection), but there will be no buff on the unit. It's not invisible, but it's the simplest way of doing what you want.

!Some Item abilities CANNOT have requirements! Even if the field is there, it won't do anything! But these are rare and from my knowledge +1 armor can have requirements

 

[Dr Super Good]

In worst case you can use triggers to apply the upgrade. When upgrade finishes you apply an armor bonus ability to all appropriate units. If a unit of the appropriate type is built and the upgrade is finished for the player then you can give them the ability as well.

You could change the unit type (similar to siege engines) with the chaos ability. Instead of being ordinary units, they become another copy of ordinary units with +X armor but otherwise identical stats.

By changing the unit to a hero and giving it agility and changing it back you can give it bonus armor and attack speed. However this is extremely likely to cause a fatal error so not recommended.

Finally you can say that all of this is too much work and instead just increment their armor a tier like normal armor upgrades do.

Be aware that armor raises equivalent hitpoints during attack calculations. By default each point of armor adds 6% bonus health for taking attack damage. Armor is also fractional, so even small sub-unit amounts gained from agility count. Adding 1 armor (initial armor does not matter as armor does not suffer diminishing returns) to a unit with 200 hitpoints will give it the ability to take 12 more attack damage in combat.

 

[teozamait]

Armor does suffer diminishing returns. Or at least it appears to do so when hovering over a units armor. A unit with 1 armor will reduce all damage taken by 6%, a unit with 2 armor will reduce damage taken by 11%.

Yes I have read the discussions regarding the fact that diminishing returns are not actually real and you do get the same HP per armor point at each level. The effect in-game might be the same but technically you get less % of armour reduction per point of armor.

Also it might actually still be diminishing when taking into consideration other types of damage reduction (eg. Defend).

 

[Dr Super Good]

Yes I have read the discussions regarding the fact that diminishing returns are not actually real and you do get the same HP per armor point at each level. The effect in-game might be the same but technically you get less % of armour reduction per point of armor.

Yes however % damage reduction suffers from increasing gain. Together the two cancel out to be linear.

Also it might actually still be diminishing when taking into consideration other types of damage reduction (eg. Defend).

It is only diminishing if the reduction amount is larger than the minimum damage of 1 since then an infinite increase in armor does not increase toughness. Defend and other abilities should multiply the damage like damage type does.

I do not know how hardened skin works so cannot comment how that interacts with armor. I would imagine it being applied before damage is dealt so is not affected by armor at all. If it is after then it will suffer some diminishing returns from armor, but seeing how it still helps lower damage you really cannot complain.

 

[SAUS]

Armor does suffer diminishing returns. Or at least it appears to do so when hovering over a units armor. A unit with 1 armor will reduce all damage taken by 6%, a unit with 2 armor will reduce damage taken by 11%.

Yes I have read the discussions regarding the fact that diminishing returns are not actually real and you do get the same HP per armor point at each level. The effect in-game might be the same but technically you get less % of armour reduction per point of armor.

Also it might actually still be diminishing when taking into consideration other types of damage reduction (eg. Defend).

If it went up linearly, like 6% per point of armour, you would notice once units get a decent amount of armour. They will just get ridiculous. The thing is, it makes armour when you have lots of armour more overpowered (until you reach a threshold - in this case, 100% physical damage reduction when you have 17 armour).

This is from the opposite end. I don't intend to show that wc3 increases HP linearly (though that is what it does), I intend to show you why linear % reduction is a super bad idea.

Consider a unit with 100 health. At 0 armour, it requires 100 physical damage to take them out.

Now add 1 armour. 94% damage taken. The amount of damage required is X, where:
X * 0.94 = 100
X = 100 / 0.94
X = ~106.4 (rounded to keep it clean)
You add 6.4% effective health by adding 1 armour to the unit when the unit started with 0.

Now let's look when the unit has 10 armour. 40% damage taken. The amount of damage required is Y, where:
Y * 0.4 = 100
Y = 100 / 0.4
Y = 250
You added 150% effective health with 10 armour. On average, you added 15% effective health per point of armour. The first point added only 6.4%.

Furthermore, let's add one more armour now. 11 armour. 34% damage taken. The amount of damage required is Z, where:
Z * 0.34 = 100
Z = 100 / 0.34
Z = ~294.1 (rounded again to keep it clean)
You added 194.1% effective health. Compared to the 10 armour, this single point increased effective health by 44.1%. That is almost 7 times as effective as the first point of armour that was added.

Add 5 more points of armour and the unit will have 25 times the effective health as their base health. One more point and the unit cannot die from physical damage. It is just purely bad design to have % reductions like this add up linearly.

 

[teozamait]

You could make it so that no unit would reach 15 + armor (which in melee games rarely happens).

I myself think that the armor system in basic Starcraft is the best, since it allows for some nice pros and cons to attack speed: fast attack units deal more damage to units with no armor but low damage to units with high armor (kind of how real life works, fast machine guns deal little damage to tanks).

But we are going off-topic here)

 

[SAUS]

You could make it so that no unit would reach 15 + armor (which in melee games rarely happens).

I myself think that the armor system in basic Starcraft is the best, since it allows for some nice pros and cons to attack speed: fast attack units deal more damage to units with no armor but low damage to units with high armor (kind of how real life works, fast machine guns deal little damage to tanks).

But we are going off-topic here)

Yes, but I like this topic haha. Capping the armour still doesn't prevent the 11th point of armour from being almost 7 times more valuable than the first point of armour.

Starcraft armour is a nice system for the reasons you say. It makes higher powered single attacks better against heavier armour, which makes sense.

In an RPG or more complex game like a moba, it is nice to have both. The shield block mechanic is not quite the same (unless you give it certain values), but it does have this effect as well. It is a shame it doesn't stack (fully, anyway). At least you can make it work with damage detection.

 

[Dr Super Good]

I myself think that the armor system in basic Starcraft is the best, since it allows for some nice pros and cons to attack speed: fast attack units deal more damage to units with no armor but low damage to units with high armor (kind of how real life works, fast machine guns deal little damage to tanks).

It has abysmal scaling though. Most RTS players pros agree such systems are bad.

Age of Empires 2 used the same armor scaling as StarCraft II. The result was where 1 research of arrow upgrade could double the damage a unit does to buildings (1->2) where as a research of armor upgrade could reduce the damage of other units by over 80% (5->1). This is why in Age of Mythology and Age of Empires 3 they changed to % damage reduction.

If you look at SC2 arcade you see many map, especially RPGs, suffering this problem. A tank hero becomes unkillable unless absolutely broken OHKO moves or impossible damage are added to bosses because their armor is so large everything does 0.5 damage (0.5 is the lowest damage from armor in SC2, not 1). On the other hand if you do not get armor everything melts your health in seconds dealing huge amounts to you which are impossible to regenerate. Most SC2 RPGs break around the 20 armor mark as a result since your HP vanish if you are not min-maxing armor otherwise you become pretty much un-killable.

WC3 armor scaling (also supported by SC2) scales far better. You can inflate enemy attack values to parallel your armor inflation so that the resulting damage you take is appropriate for your health.

It works in SC2 melee because armor scaling is purposely bad. As the game progresses they want units to kill faster. As such armor is more expensive than damage to research and damage grants more bonus than armor does. For example the Siege Tank gains 5 attack per damage research (15 max) where as the armor upgrade only add 1 each (3 max). This means fully maxed out siege tanks will kill faster. As such it is always better to invest in attack first and damage later. This applies to many units, not just the siege tank where the damage difference might be 2, 3 etc. Armor is only really useful in countering early game troops or units with low damage scaling which only gain +1 attack per upgrade since it keeps your troops equivalently as tough.

Two expensive units which do not really gain from upgrades are Carrier and Battle Cruiser since they gain +1 damage against the +1 armor.
Two expensive units which do gain damage from upgrades are Thor (+3 vs ground) and Ultralisk (+3 vs non structure) against +1 armor.

 

[SAUS]

It has abysmal scaling though. Most RTS players pros agree such systems are bad.

Age of Empires 2 used the same armor scaling as StarCraft II. The result was where 1 research of arrow upgrade could double the damage a unit does to buildings (1->2) where as a research of armor upgrade could reduce the damage of other units by over 80% (5->1). This is why in Age of Mythology and Age of Empires 3 they changed to % damage reduction.

If you look at SC2 arcade you see many map, especially RPGs, suffering this problem. A tank hero becomes unkillable unless absolutely broken OHKO moves or impossible damage are added to bosses because their armor is so large everything does 0.5 damage (0.5 is the lowest damage from armor in SC2, not 1). On the other hand if you do not get armor everything melts your health in seconds dealing huge amounts to you which are impossible to regenerate. Most SC2 RPGs break around the 20 armor mark as a result since your HP vanish if you are not min-maxing armor otherwise you become pretty much un-killable.

WC3 armor scaling (also supported by SC2) scales far better. You can inflate enemy attack values to parallel your armor inflation so that the resulting damage you take is appropriate for your health.

This is why the values in SC1 and SC2 are so low. The starcraft armour system does not play well with RPGs where all the numbers can scale for a long time. the point of that armour system is to create unit interactions that make sense. I agree entirely that it doesn't work in RPGs or other similar things, like MOBAs (LoL, Dota, etc.).

For an RTS, the starcraft armour system is perfectly fine. Imo, issues like the ones described for AOE are a result of poor use of the system.

 

[Xonok]

I'd say it's not really a problem. The problem is games that rely on insane scaling, like many MMORPGs. A more flat RPG that scales more into choices than power would not have this problem nearly as much. It also makes the game less top-heavy, which in turn reduces the grind-until-I-become-relevant aspect.

 

[teozamait]

Terraria could be considered as an RPG (well, depends if how you define the "role", but with some definitions even Diablo is not an RPG) and it uses a very similar system to Starcraft. Every 2 points of armor remove one point of damage. And weapons have different fire-rates, so high powered but long cooldown attack will be more effective against enemies with high armor, but overkill on smaller enemies.

I for one like this system more than a % one, since it allows for clean and clear differences between weapons and enemies. Enemies with high hit points but low armor are best killed with weapons that attack fast for a counter-example.

 

[Dr Super Good]

I for one like this system more than a % one, since it allows for clean and clear differences between weapons and enemies. Enemies with high hit points but low armor are best killed with weapons that attack fast for a counter-example.

Except this can be achieved with percentage based systems as seen in Age of Mythology and Age of Empires 3. You have your hard hitting units get massively reduced against some units while your soft hitting units are not and the opposite holds.

WC3 already has this. SC2 also has limited support for it.

 

[teozamait]

Except this can be achieved with percentage based systems as seen in Age of Mythology and Age of Empires 3. You have your hard hitting units get massively reduced against some units while your soft hitting units are not and the opposite holds.

WC3 already has this. SC2 also has limited support for it.

You can do it, but in a non-procentual reduction armor system this is can be done "naturally", there is no need to create unit types such as Heavy or Light, you just fiddle with their armor, attack speed, salvoes etc.

Like I said, the advantage is that its clean and clear.

Warcraft 3's current system is anything but clear, new players have the impression that armor reduction does have diminishing returns. Before I found out that this was not true my playstyle in regards to armor buffs was very different than it is today. If I had a high level devotion aura for example, I would often neglect researching additional armor upgrades, since I thought I get less with each consecutive upgrade.

 

[Dr Super Good]

You can do it, but in a non-procentual reduction armor system this is can be done "naturally", there is no need to create unit types such as Heavy or Light, you just fiddle with their armor, attack speed, salvoes etc.

Except it raises the nonsense "boarder cases" I mentioned earlier where a single upgrade suddenly doubles or triples the damage a unit does to something.

Like I said, the advantage is that its clean and clear.

It is less clear since a unit can still hit hard, but if it does not hit hard enough it deals practically no damage. All RTS veterans agree the percentage based armor system is better in RTS games.

The implementation varies. Most use flat percentage bonuses (+5% damage reduction) which usually means multi-tier upgrades improve in effectiveness. This was the case in Age of Mythology.

Warcraft 3's current system is anything but clear, new players have the impression that armor reduction does have diminishing returns. Before I found out that this was not true my playstyle in regards to armor buffs was very different than it is today. If I had a high level devotion aura for example, I would often neglect researching additional armor upgrades, since I thought I get less with each consecutive upgrade.

Easily solved by setting the constant to 0.01 instead of 0.06. Then armor means equvelant % more hp when taking weapon damage.

 

[teozamait]

"ALL" RTS veterans? I'm sorry, on what study or empyrical evidence is that statement based off? ALL
sounds like you are 100% sure, which no reasonable person can be.

Factually it simply is not true, since I have been playing RTS continously for the last 12 years (including all of Blizzard's RTS from WC1 to SC2) and I consider myself to be an RTS veteran and I prefer a non-percentage system. Therefore, since 0.01% of all RTS veterans do not prefer percentage based armor system in their RTS games, your statement is factually (strictly speaking) false.

I'm not claiming that most people prefer a non-percentage armor system or that a non-percentage armor system is the best. I am claiming to be speaking JUST FOR MYSELF because I have no access to studies with a wide range of participants (in the tens of thousands) and with a small factor of error.

The main issue with a percentage system is that armor upgrades can be just swapped with a health upgrade. Instead of plus 5% armor reduction you can have +5% maximum health. In a non-percentage system armor is a different layer of protection compared to health, which adds a new layer of complexity to the game. Look at how games like Fallout New Vegas do it - this non-procentual system allows for Hollow Point and Full Metal Jacket Bullets to exist.

 

[Dr Super Good]

"ALL" RTS veterans? I'm sorry, on what study or empyrical evidence is that statement based off? ALL

The core players of Age of Empires 1,2 and 3, Empire Earth / Empires Dawn of the Modern World, Rise of Nations and Age of Mythology. Basically the true old-school RTS group.

0 A.D., an open source RTS game made in the same RTS style as Age of Mythology mixed with Age of Empires, made a migration from fixed armor (Age of Empires 1/2 style) to percentage armor (Age of Mythology/Age of Empires 3 style). This decision was done after much debate. The final conclusion was the same reached by the Age of Mythology developers (which was made in response to feedback from Age of Empires 2 players), that the fixed damage reduction for armor creates too many annoying borderline cases where by the difference of a single unit can mean doubling a units damage or no change at all.

The example most commonly used is the Age of Empire 2 archer unit against tougher buildings. After some building upgrades it will deal only 1 damage due to the armor being larger than the attack. After upgrading the archer damage enough you will get a case where by instead of dealing only 1 damage, it will do 2 damage (doubling). Additionally some upgrades to the damage will not increase archer damage to the building as it might still be below the building armor (1 damage upgraded to 1 damage). Percentage based systems fixed this problem since then a damage upgrade always increased damage by an easily computer amount and armor upgrades always reduced damage by an easily computer amount.

The main issue with a percentage system is that armor upgrades can be just swapped with a health upgrade. Instead of plus 5% armor reduction you can have +5% maximum health. In a non-percentage system armor is a different layer of protection compared to health, which adds a new layer of complexity to the game.

No it cannot be swapped out with health because armor does not reduce all damage, only attack damage. In WC3 abilities ignore armor completely. This means that a unit with insane armor and low health (1000 armor, 100 health) might be near unkillable by weapon damage but die easily to damaging abilities.

Other RTS games incorporate armor with armor type. So that depending on what damage source is used the reduction will vary. As such a piercing attack might deal near full damage to a knight but a sword attack might do practically no damage to the same knight.

Look at how games like Fallout New Vegas do it - this non-procentual system allows for Hollow Point and Full Metal Jacket Bullets to exist.

You have no explained what those do exactly. If you give an example you should give an example, not reference an example which most people will not have played.

 

[teozamait]

What's the issue with borderline cases? You can balance around it, you increase the base attack value (like Starcraft does it, lowest is 5 not 1) and at the end of the day Marines with no attack upgrade will only deal 1 damage per attack to a fully upgraded Battlecruiser and that has won/lost hardly any game by itself. It simply is not an important factor, since most units have higher base attack values than 5 or 6. And it makes sense for machine guns to take extremely long to destroy a spaceship! Or for arrows to destroy a building.

And unit types such as heavy/light or damage types such as magical/physical are a nice extra layer, but they are "artificial". I agree that they are needed even with a non-procentual armor system.

In your example with abilities killing a unit with high armor you made a good point, procentual armor upgrades in WC3 are not the same as health upgrades, they are inferior to health upgrades. A unit with high health will resist longer to physical and magical damage, whereas a unit with just high armor will only resist longer against physical damage.

In Fallout, there are bullets that amplify the damage a unit takes but also the armor - Hallow Point bullets (so they deal extra damage to units with no armor, but if a unit has high armor they would deal less damage than even a standard bullet) and bullets that reduce the damage a unit takes but also its armor - Armor Piercing bullets. This can be achieved with a procentual armor system, but its very easy to calculate just how much damage you will and properly plan ahead what bullets to use when you just have simple, flat numbers.

 

[vonsteyr]

Why not have an armor system that incorporates both ideas? A small negation in actual damage and a larger percentage negation for scaling purposes? It makes sense logically that a unit with sufficient armor would be effectively immune to very weak weapons and would likewise take less total damage from even stronger attacks, and it would allow for a better mixture of counter-tactics (such as armor piercing attacks that negate the base armor but not the percentage armor). I mean, you can stab a inch (2.5cm) thick plate of steel with a dagger as many times as you want, it's not going to leave a hole.

WC3 incorporated this in a way with the Mountain Giant's Hardened Skin upgrade, which is a direct reduction in damage rather than a percentage one.

Personally, (and having played many, many RTSes) I feel that while a strictly percentage based armor system is perhaps slightly better for game balance it takes away from the nuance and character of the game as a whole. I've also played a lot of RPGs, and there is a feeling of real accomplishment and strength when you get a character who is excellent at tanking, and a 1.5% further damage reduction as you equip that godly plate mail just doesn't cut it for me. A game like WC3 that draws so heavily on RPGs for its overall feel might benefit from a gameplay (obviously important) standpoint but not necessarily from a "role-play" stand point which, after having played WC3 for so long, has also become one of its big draws.

Diminishing returns is rarely a fun thing to experience in games.

edit: also poor SKW I wonder if he ever figured out his problem

 

[Dr Super Good]

and a 1.5% further damage reduction as you equip that godly plate mail just doesn't cut it for me.

Damage reduction by itself is actually a exponential value. At 100% you become infinitely tough. People already gave examples to prove this.

This is why Diablo III uses "Toughness" to quantify gear differences. That new Legendary Ancient armor in Diablo III Might only raise your damage reduction by 1% here or there, but it still adds +500,000 toughness or more.

 

[SAUS]

Diminishing returns is rarely a fun thing to experience in games.

The point is that it isn't diminishing returns - it is the opposite, actually. If you add 6% with that godly plate when you have 94% damage reduction already, you become invincible. Maybe the best way to fix the problem is to change what is displayed. Instead of displaying % reduction, display the % effective health / health modifier % (it also applies to every other form of health, such as health regeneration and various other healing effects). I don't think this can be properly done in WC3, though.

A mix of both armour types is good - and it is in WC3 as you state. Shield blocking via the hardened skin ability also ties into this. You can fully block an attack. Being a % chance just makes it so you can't be completely invulnerable to certain attacks. This system is used in DOTA and it allows people to become tanky in multiple different ways but also be vulnerable in different ways as well (since numbers are balanced knowing the countermeasures that are in place).

 

[teozamait]

Yeah, I think there is value in using both systems. This is what I'm doing in my Enhanced Campaigns, I'm applying the Hardened Skin ability to some units, and I like how it is turning out

I have no issue with units that are invulnerable to low-damage attacks or immune to a small category of units, we already have units that are completely immune to Magic attacks in WC3.

And I would also agree that the current WC3 system could be more honest, showing and explaining that it increases your health by 6%.

 

[Dr Super Good]

And I would also agree that the current WC3 system could be more honest, showing and explaining that it increases your health by 6%.

The system was engineered for RPGs. Hence why it is in Diablo III as both Armor and Resistance formula.

If you read the support site it does clearly mention how armor works. The problem is there is no in game UI to explain this to the user so without taking a minute to look it up you will have no clue.

Why 0.06 was chosen for the % per armor is also a mystery. It is far more clear if it was 0.01 and all armor values multiplied by 6.