Armor values?

[FergardStratoavis]

Wasn't sure if this thread belongs here, but it's been something that's recently hit me.

Assuming you make a standard RPG, armor is going to add points to your Defense/MDF. There are tiers of armor, corresponding to the increasing threat of enemies. Makes sense so far, but what if the progression isn't linear? What if your characters walk around to and fro for whatever reason? Sure, one could probably strike some kind of balance if they tinkered with numbers enough, but I wonder if there's perhaps some more elegant way of solving it?

Why not simply have your armor reduce the Physical rate (assuming that's the most common attack around)? I realize this is nothing groundbreaking and may even be entirely obvious, but if all armors simply reduce appropriate element rates rather than add flat numbers to your defensive values, perhaps that might help maintain a healthy amount of usability and threat for earlier armors and enemies respectively? Some special gear could still raise DEF/MDF by flat values, but that would be rare to come by or reserved to accessories.

What is your approach to the notion? Perhaps I'm just being stupid, but I can't help but wonder about the effectiveness of such plan.

 

[TheoAllen]

You know, I had once in the same boat as you, but I thought differently. But instead of trying so hard to put a defense stat, I thought of "why having armor at all? can the equipment just adds more HP, because defense point is basically just more effective hp?".

The advantage thing about this design is that you can ignore defense stat in the damage formula (faster development), and if you somehow managed to show damage dealt in the skill info window, it will remain consistent. Say, in your skill information window, it says "damage 100", it will damage more or less with that number, factoring the variance. Another good thing from it if you want to make battle longer, you can just up enemy's HP without having an extra headache of "what number I should put in this enemy's DEF stat?". If I want a buff/debuff, I'd just put a state that increases/decreases damage taken by a certain percentage instead of reducting the DEF stat, sounds like a win-win solution.

I tried the default subtraction formula (atk - def), as well as armor scaling, and nothing worked so far. So I left my game with no defense stat at all because I couldn't make it "work". However, I did want to have a defense stat. The idea that came up was I'd just give every battler a flat percentage reduction. Say, enemy A will have 20% damage reduction, enemy B will have 30%. However, the twist is, you can reduce their defense to zero damage reduction when the condition meets. Because the armor has its own HP, calculated separately when dealing damage.

Result? It worked well so far. I like the result.

 

[Doktor_Q]

The general point of armor is just "better armor = harder to kill." I'd say there's probably... ~four common ways you make someone harder to kill in an RPG, and most of the time, you're using a combination of at least two of them:

1. Additive health, like @TheoAllen mentioned. It's incredibly simple, and the math often boils down to "this armor means I can survive another 2.5 hits." The downside is that you can't really make different types of armor exactly, because another 10 HP stops another 10 damage, regardless of what that damage comes from. Sometimes, you actually have a separate "armor" or "shield" pool with extra rules, like armor not healing during a fight, but refilling completely between fights.
2. Resistance, where the armor makes you take 25% less damage. About the same as percent-based health bonus, but with the added advantage of being able to resist different sources by different amounts- physical/magical, per-element, etc. So now we can make things that ignore defense, make different attack types good or bad against different people, and even adjust how hard something is to heal (more resistance = less damage = easier to heal).
3. Subtractive defense, which is the typical "atk - def" formula RPGMaker likes to toss at you. It's easy to think up, the math is simple, but balancing is very delicate- if attack gets too high defense basically doesn't matter (100 - 3 is not much change), if armor gets too high things get impossible to kill (100 - 97 is much big deal). Basically, if you aren't careful it's really easy to make defense so weak it's useless, or so high it's overpowered, or possibly both at different times. As of the recent versions, RPGMaker gives you separate Def and Mdef stats so you can easily have two resistances, but probably not one per element.
4. Evasion, which comes from a more D&D approach of armor making it harder for someone to actually hurt you at all. In video games this seems to be the staple of light armors and shields. Because it's RNG-based it's innately hard to balance, and the all-or-nothing nature means you probably don't want this as your primary defense, unless your game uses large numbers of low-damage hits, at which point statistics take over and it's similar to resistance. Like subtractive defense, RPGMaker gives you separate physical / magical evasion, but you also get critical evasion and status resist which are arguably similar.

Personally, I'm a fan of combining additive health and subtractive defense: Defense is powerful but hard to get, so most of your armor just gives more health. One particular advantage is that the two types have very different effects on the math: Defense is dramatically more effective as the attack gets weaker, so multi-hit moves or being swarmed is less dangerous, but powerful moves, attack buffs, etc. can break through. You definitely need to keep track of how much attack and defense players and enemies can get at any point in the game, though.

I'm not really a fan of evasion, and typically remove the accuracy system from my projects outside of specific optional abilities, usually in the player's favor- maybe an accessory that sometimes lets you dodge, or a really powerful skill with an explicit chance of failing.

Another common one is "(atk - def) * mod", basically crossing resistance and subtractive defense. Mod is often complex, but the end result is still to make several "types" of durability. An example is a damage formula like (weapon atk - armor def) * (user str / target res)- this models a case where your weapon has to be good enough to pierce the enemy's armor, but you can also boost that damage by having more strength, or reduce the damage by having more resistance. Implementing this probably means doing more math and tracking more values separately.

Lastly, some games go all-in on resistance- you just get a different damage rate per damage type, and it's about hitting the enemy's weakest spot, while having them hit your strongest.

 

[Kes]

I think this is probably better in Game Mechanics Design, the way the discussion is shaping up.

[mod]Moving this[/mod]

 

[mathmaster74]

@FergardStratoavis I think Doktor_Q has a good breakdown. I would say low level armor should maybe just use additive health. As the player progresses and enemies get tougher, keep using additive health with new armors, but add resistance to curb some of that villain o.p. Finally, at high levels, I would provide armors that add in higher DEF and MDF as well as (optionally) some evasion rate increases as icing. The main thing is figuring out when the additive hp for equipped armor doesn't seem realistic. Since resistance is based on percent, it can easily ramp as needed from there until another 1% damage resistance gets to be too much of a leap. When that happens, add in the rest to balance...but remember that armor is not the player's only defense. They can heal too, right?

 

[Wavelength]

Why not simply have your armor reduce the Physical rate (assuming that's the most common attack around)? I realize this is nothing groundbreaking and may even be entirely obvious, but if all armors simply reduce appropriate element rates rather than add flat numbers to your defensive values, perhaps that might help maintain a healthy amount of usability and threat for earlier armors and enemies respectively? Some special gear could still raise DEF/MDF by flat values, but that would be rare to come by or reserved to accessories.

The main problem with using the Physical (or Magical) Damage Rate - rather than stats like DEF and MDF - to have armor reduce incoming damage, is that the effectiveness of said armor is automatically scaling throughout the game as you take higher and higher amounts of damage. If you are also increasing the Percentage Reduction of damage that armor provides as players buy higher and higher tiers of armor, you are creating a multiplication in your scaling (higher percentage of a higher amount of damage) that's hard to keep up with offensively.

Example: Level 1, battlers have 50 HP, deal 10 damage (before armor reduces it), and have 50% damage reduction from armor. Battlers take 5 damage from each hit, and survive 10 hits before falling. Level 30, battlers have 500 HP, deal 100 damage (before armor reduces it), and have 80% damage reduction from armor. Now battlers take 20 damage from each hit, and survive a whopping 25 hits before falling. If the battlers have innate stats like DEF and MDF that are growing, and you have this growing percentage reduction on top of it, then the disparity will be even worse.

The other negative idiosyncracy of such a system is that in long games with many tiers of equipment, the Physical (or Magical) rates will start getting close to 0, which makes for really weird balance.

All of this, taken together, is why Percentage Reduction tends to work well in short games where there is no progression through different "tiers" of armor throughout the game (perhaps there are several equally-good armors that offer different benefits), but works poorly in longer games where the player is expected to progress through better and better tiers of armor throughout the game.

However, with the right formula, you can use stats like DEF and MDF to provide a percentage reduction that works a lot better than using the Damage Rate features. For example, a formula I like a lot is (a.atk * sp) / (b.def + n) where sp is the Power of the skill and n is a number identical in every skill (I use 20 but if you have big numbers in your game you'll want something more like 100). Essentially, this causes a higher DEF stat to increase the denominator of the formula, therefore providing a percentage reduction to the base damage (a.atk * skillpower) in a smoother way. The one weakness with this formula is that if your game is long, and battlers' HP increases dramatically throughout the game, then you will need a third factor in the numerator in order to keep the battle pace consistent (I sometimes go with a.atk * sp * a.level).

 

[woootbm]

I'm actually going to be trying a % reduction system in my next game. To @Wavelength 's points, though, it will be a relatively short game.

The other reason I am confident in the system, though, is because I'm willing to trudge through all of the math in the game manually and make sure everything shapes up sensibly. On top of that, I have a lot of attention to balance. Basically, any time you vie for higher armor, you will be costing yourself your other stats severely.

I also want to emphasize something that others have brought up, but not exactly this point. And that is that 100% armor means infinite power. Because of that, you really have to be careful with how much armor a player can have, especially past 50%. Even taking into consideration what I said about balance above, generally speaking every point closer to 100% armor will be worth it no matter what.

I'll show you. See, 75% reduction is twice as good as 50% reduction. Even though it's only 25 percentiles more, that is the math. (50% is half damage, 75% is one fourth). And it keeps going. 99% reduction is ten times better than 90%, even though it's only 9 percentiles more (90% is one tenth, 99% is one one-hundredth).

For longer games, what you can potentially do is have the formula scale with level. Like Word of Warcraft, for example, your armor is actually a "rating". So, say at level 1 it might only take 10 rating to get 1% reduction, but at level 100 it probably takes 1000 to get 1%. That way, you can introduce armor with wayyyyy higher ratings than lower level armor, but not get crazy with how much benefit you are getting.

 

[Diretooth]

One thing I have wanted to do within my own games is to find a balance between strong weapons and strong armor, but when you simply escalate the damage and defense values as the game goes on, you're not really 'growing'. When the name of the game becomes 'spam move until enemy dies', it becomes difficult to balance that.
My attempts to help balance has been the implementation of elemental systems, where using the right element a la FFX or Lost Odyssey can make combat easy, but not using the right element makes it more difficult. But there also comes the issue with effectively meaningless stat raises, especially if you're going with the basic formula, which if you want to consistently have small battles be around 3 turns, midbosses to be around 5-10, bosses to be around 10-15, and the final boss to be much higher, you'll have to do a lot of math.
While FFX allows for modular equipment, to the effect that you could essentially stack defense boosts, it still does offer stat growth that makes the early game trivial, and the late game survivable.
In Lost Odyssey, stats remain largely the same, and the only real benefit from leveling up that I could determine was more HP and more skills, which a level cap per area was put into place to prevent massive level grinding, which is fine, since you will know eventually where you are at a good place to continue the story. The enemies do get progressively harder, but the extra health does help. Equipped armor doesn't change, but weapons do, which increases attack, and rings modify the effect attacking has as well.

That all being said, there's never any real risk/reward with armor in most games. You get stronger armor, you become more difficult to kill, but enemies also have the same. I think that one game that does offer at least a halfway decent system is Dark Souls, specifically how heavier, more protective armors offer good damage reduction at the cost of potential speed loss if your equip load is too high, but being lighter makes you faster, albeit easier to kill, and thus reliant on more dodging to survive. There is, of course, a way to minimize equipment load to maximize defense and speed, but it never gives you a one-way ticket to steamrolling the game, as it still remains very difficult.
For me, personally, I think a risk/reward system for a limited amount of armors could work, coupled with an elemental system that determines how much damage an elemental attack or weapon does, as well as armor scaling for monsters of different types, with methods to bypass their defenses with reasonably high cost skills, with the same methods being available. Granted, it's easier said than done, as evidenced by the relative disorganization of my post. I may post further to elaborate or offer further insight.

 

[mathmaster74]

@woootbm Interesting point about damage reduction above 50. Yes, if you jump by 25% there's a huge change, and if from 90 you hop up 9% there's a huge change. Here's my thought for using damage reduction with armor: Start with a reasonably low % when you introduce these armors (say 10%), make slight changes available per level to aid balance, and cap it at a reasonable % and level before moving on to use of armor that adds DEF and (optionally) Evasion Rate benefits. Example: you were at 50% damage reduction with an armor. Now a 51% damage reducing armor becomes available at the next level. How much more effective is said new armor? 1 additional point of damage per 100 of the enemy's dealing will be defended against. Maybe an enemy at this point deals 5000 points of damage per blow. That means a measly additional 50 points of the 5000 will be defended per turn. You still take 2450 damage compared to 2500 with the old armor. Is it better? Yes. Is it much better? Not really. If this enemy can kill me with one, they most likely can with the other just as easily. Now say I survived and another armor with 52% damage reduction comes along at the next level. How much better? 1 point of every 100 gets ignored again, so if I grind the same enemy their 5000 point attack does 2400 damage. Is it getting better? Yes. Am I o.p? No. The danger with using percentiles comes in when you are too near 100% and enemies are not sufficiently damaging to still make it sting, which is why I suggest the cap on damage reduction and the shift to a new model at high levels. Aside from this, consider enemy damage and HP growth with enemy level. If I faced this enemy at a given level where 50% damage reduction was the best armor available and couldn't get the 52% damage reduction armor until 2 levels later, how much more damage and HP has the enemy gained as a 2 level higher enemy? (I like to think it adds balance and interest when enemies are monkeys, so I like to match player level to enemy level whenever and however possible.) This makes the new armor's effectiveness even more negligible, but also makes getting the new armor almost necessary to remain competitive against the enemy. In this case, damage reduction above 50% is no big deal. Dizzying.

I'm willing to trudge through all of the math in the game manually and make sure everything shapes up sensibly. On top of that, I have a lot of attention to balance.

That, ultimately, is the best policy. What works best by test? Practice will trump theory every time.

 

[woootbm]

Yep, I put in a hard cap on defense thanks to a mod. I might make that cap more severe with more testing, too (I already reduced the max once already).

This makes the new armor's effectiveness even more negligible, but also makes getting the new armor almost necessary to remain competitive against the enemy.

So the problem with enemies scaling to compete directly with high armor is that your low armor allies will suffer too much. I once worked on a game where the tank classes had INSANE mitigation levels, so the raid designers had to increase the damage of the enemies significantly in order to keep them a threat. What ended up happening was that the enemy damage was so high that they would absolutely obliterate anything that wasn't a tank in one hit. In fact, I think they generally did multiple times more damage needed to one-shot a clothie.

In the end, they had to drastically adjust the soft caps and hard caps of tanks so that they could nerf the damage of the enemies. It was... pretty frustrating for players that if a tank had even a microsecond of mismanaged threat or positioning that half the raid team would be killed

 

[Wavelength]

One thing I have wanted to do within my own games is to find a balance between strong weapons and strong armor, but when you simply escalate the damage and defense values as the game goes on, you're not really 'growing'. When the name of the game becomes 'spam move until enemy dies', it becomes difficult to balance that.

I liked the rest of your post a lot, but I wanted to focus in on this opening paragraph, because I really don't see the Equals Sign between 'stats treadmill' and 'spam moves until enemy dies'. I think they are two completely different areas of combat design that need to be addressed.

I don't have a major problem with the 'stats treadmill' where as you progress through an RPG, you gain stats from level-ups and better equipment, and the enemies you find in the new areas are stronger than in old areas so that the challenge remains consistent (or slightly grows) even with your higher stats. (Same with stronger weapons vs. stronger armor in PvP games.) It's a tried-and-true formula; it may not be the best mechanic ever devised but it does its job well. You get the pleasure of seeing your numbers go up (and can feel the power of them when you crush early enemies that used to be a struggle), and you still get to enjoy challenging battles.

At any level of stats for the player and/or enemies, "spam move until enemy dies" is a just a symptom of bad skill design, bad enemy design, and/or bad choices of mechanics to include in combat. It means that there is one move that is clearly better than the others, and that enemies don't provide enough power or diversity to encourage the player to do anything different (as well as no other mechanics in place to reward the player for doing anything different). There are a lot of different ways around this - the elemental system you mentioned is one; cooldowns are another; random selections of tools/resources in combat (Disgaea) are another; use of buffs and debuffs along with a challenging enough baseline (Epic Battle Fantasy) is another; simply creating enemies that are very different from each other in their behavior and strengths is yet another (Tales of series).

But I don't see any way in which creating the need to improve your stats in order to keep up with the growing power of enemies ever directly contributes to the stale dynamic of "spam move until enemy dies".

 

[bgillisp]

@Wavelength : I'm guessing you never played Betrayal at Krondor. They used a % based damage reduction in that your armor blocked x percent of the damage. The way they did it was beginning armor blocked 15% of all damage, and it went up about 10% per new armor you found, with the best armor blocking 70%.

Now, they did have the armor get worse the more damage it took too, so if you had an armor that blocked 15% damage but it was at 80%, you actually blocked 80% of 15%, which was 12%.

It seems to have worked for that game, and Betrayal at Krondor is anything but short.

 

[Wavelength]

@Wavelength : I'm guessing you never played Betrayal at Krondor

The only Betrayal I've ever played was "at House on the Hill".

They used a % based damage reduction in that your armor blocked x percent of the damage. The way they did it was beginning armor blocked 15% of all damage, and it went up about 10% per new armor you found, with the best armor blocking 70%....It seems to have worked for that game, and Betrayal at Krondor is anything but short.

So it sounds like there were a total of 5-7 "tiers" of armor throughout the game. A lot of RPGs (particularly JRPGs and MMORPGs, but also some WRPGs and P&P-RPGs) have more tiers than that. Since long games usually have more tiers than this, I noted the percentage reduction could be a poor choice for those kinds of games - but if the game happens to be long and only needs 5 tiers of armor throughout the course of play, then yep, it can fit just fine.

One slightly-negative idiosyncracy is that as raw percentage reduction increases, the effectiveness of each percentage point increases. It's really obvious when you have, say, 85% reduction vs. 95% reduction (you are taking just one-third of the damage in the latter case compared to the former), but even at 30 vs. 40, each point from 30 to 40 does more than each point from 20 to 30 in terms of increasing your "effective health" (the number of hits you can take before you fall). Exponential returns (as opposed to diminishing returns) can be hard to design around when you need balance to be really tight. When your target for balance is wider, and you're working with reasonable numbers (i.e. not 95%), you can still hit the target.

 

[bgillisp]

@Wavelength : Ha! I've played that board game too! As for the game I mentioned, it was like 6 tiers of armor, I think it went 15% 25% 35% 45% 55% and 70%. The way it balanced the numbers more though was weapon damage went up significantly throughout the game, to where the last weapon of the game could do 180+ damage to someone without armor.

Granted, this was a game with low HP values too, and max HP was hard to increase, which probably helped the balance as well. The follow up game Betrayal at Antara your max HP didn't increase at all in the entire game. Instead you got better by better weapons, armor and skills.

 

[Tai_MT]

Wasn't sure if this thread belongs here, but it's been something that's recently hit me.

Assuming you make a standard RPG, armor is going to add points to your Defense/MDF. There are tiers of armor, corresponding to the increasing threat of enemies. Makes sense so far, but what if the progression isn't linear? What if your characters walk around to and fro for whatever reason? Sure, one could probably strike some kind of balance if they tinkered with numbers enough, but I wonder if there's perhaps some more elegant way of solving it?

Why not simply have your armor reduce the Physical rate (assuming that's the most common attack around)? I realize this is nothing groundbreaking and may even be entirely obvious, but if all armors simply reduce appropriate element rates rather than add flat numbers to your defensive values, perhaps that might help maintain a healthy amount of usability and threat for earlier armors and enemies respectively? Some special gear could still raise DEF/MDF by flat values, but that would be rare to come by or reserved to accessories.

I've tinkered with this a little, but it really only seems to affect the dev side of things. Players will rarely know what your damage formulas are, so it doesn't matter much to them. An armor that reduces damage by an extra 10% over the other... the player is only going to know, "It's better, equip it". As such, your enemy stats will need to increase by that same percentage if you wish to maintain the same level of damage (or even increase that damage). If your percentage increases become too high, as well as your HP becomes too high... all challenge goes out the window as your enemies need exponentially more stats to maintain any semblance of challenge.

I think the only way in which a flat percentage works is if you get very little, if none at all, HP increases.

You also run into this issue where the percentage reduction may not matter at all. If you take 500 damage from an enemy and equip a 10% damage reduction item, you may go from dying in 4 hits to 5 hits. It gives you one extra hit before death. (for example, you have 2000 HP). Now, if you suddenly get a 15% damage reduction armor... it still takes those 5 hits to kill you. The percentage reduction is meaningless. Okay, so how high does that percentage need to be to go from dying in 5 hits... to 6?

21%

To get a single hit more before death, you have to increase the armor by 11% more. Or...

251 more HP (2251 total HP).

What about 7 hits to die? Damage reduction of about 33% for one more hit or about 2766 total HP (an increase of 515 HP over the last upgrade).

The numbers skyrocket from there. This assumes that all enemies will deal 500 damage to you. Smaller numbers than 500 will have even more negligible effects on percentage of damage reduction. Larger numbers mean you need even large jumps in those percentages just to increase the amount of hits you can take by ONE.

The numbers get out of hand very quickly.

It gets even worse if your enemies attack power goes up alongside this. Look at what happens with a static amount of damage. Can you imagine the insanity that would result from not only increasing defense, or increasing HP, but then also increasing enemy attack power?

This says nothing of what a player may end up spending on Consumables or HP to even get rid of the damage being dealt to them (a single attack taking 1 fourth of your HP? A player healing at only two hits requiring items or spells equal in value to healing 1000 HP?).

I'd reserve the damage reduction for special pieces of equipment or for games in which HP doesn't go up at all, and all enemies deal about the same damage to you. That way, it makes the armor actually valuable to a player.

What is your approach to the notion? Perhaps I'm just being stupid, but I can't help but wonder about the effectiveness of such plan.

My approach was to include certain special pieces of equipment which would reduce incoming damage like you describe. They are special because the damage reduction is valuable alongside the normal stats they'd also be employing.

For example:

Monster A has 500 Attack Power.
Character B has 300 Defense Power and 2000 HP.
My damage formula is a.atk - b.def.
500 - 300 = 200. 2000 / 200 = 10 Hits to die.

Now, as far as I understand it, the percentage rates are dealt with "on the back end" after the damage formula. That means, it simply reduces whatever damage amount would've been there.

So, a 50% reduction turns that 200 damage into 100 damage. 20 hits to die.

Furthermore, I use it on special types of equipment because I use 3 stats to deal damage and provide defense (Attack and Defense offset each other, Magic Attack and Magic Defense offset each other, Agility and Reflex offset each other). Some skills are physical and use Attack or Agility to deal damage. Some skills are magical and use Attack or Magic to deal damage (there are instances of these being used by all three stats, but the ones not mentioned are far more rare). So, a Physical damage reduction can cover a stat weakness in a character. Let's say I've got a knight without much Reflex. He's going to get wrecked by any skill that uses "Agility" as its primary stat. But, if he's got a 20% damage reduction to physical attacks... he'll take less damage provided the skill is physical.

Not many of my items have this reduction on them. It is meant to be used on top of what is already there, or used to shore up weaknesses in characters/classes.

 

[GLM]

I'm a fan of keeping numbers low to make the math easier for the player and for development.

That being said, I'm making a game where there is no traditional leveling which uses a pseudo-tabletop dice-based combat. The armour is a straight subtraction from damage, and dodging and blocking attempts are limited per round.. so I might be just thinking of my needs.

 

[Aesica]

My approach has been to have the armor values scale up in a linear fashion at first, then gradually transition to flatter bonuses, but with other benefits. To compare some things from my current project:

• Dragoon Helmet: +68 Def, +30% Fire/Ice/Lightning resist
• Crystal Helmet: +68 Def, +50% resistance to attacks from Undead or Demon foes

Yes, it's a mini tribute to FF4, but you can see how each of the two helmets, while equal in raw defense, offer different benefits

I also think it's fun to go outside the box a bit and get creative with some of the equipment--beyond just basic stats. To use more examples:

• The Vortex Barrier is a shield with a mere +1 Def, but when guarding, it reduces wind damage vs the entire party for that round.
• The Force Armor's defense and magic defense scales with the user's current MP.
• The Bunny Girl Costume scales with level and applies a Taunt to the user while guarding.
• The Shield of Strength heals the user for a small amount when they guard. (yes, a tribute to Dragon Quest)

And so on. So overall, you can say my approach to gear is to make equipment a strategic choice rather than "oh boy, a longsword +2 to replace my longsword +1." It's also more fun when the gear adds something to the player besides just extra stats, such as the Vortex Barrier's party-wide wind resistance.

 

[GLM]

I also like the "offer the same defense, but have different special effects" approach.

I will take that into consideration for my own project, especially since with really low numbers it's hard throw much equipment at the player before they're at the realistic maximum value.