Stat complexity

[wallacethepig]

I think the stats should only be as complex as your game needs them to be. If you have one character, you can toss out like three stats from the get-go. MAT and MDF don't need to be there if nobody uses magic, and agiltiy can likely be eliminated as well because you're either going to go first or you aren't.

Personally, I think of MAT and MDF as "extra" stats, and honestly, you don't even need one of them if you're willing to change the formula (since most characters with high MAT also have high MDF, and the stats serve the same purpose, there's no reason to have both).

It's fun to play around with numbers, though. Try out stuff and see if you like it, but I would say trim all the fat first, and then add in stats that you need.

-Wallace

 

[Milennin]

I find consistency and clarity to be very important - the player should never have to wonder "why did this low-HP enemy take me so many hits to bring down... was it high defense, elemental resistance, or maybe I did something wrong?"  While I like low numbers too, I personally put a higher priority on providing this kind of direct clarity than on keeping numbers low.  Maybe we just have different priorities here?

I don't see how a player could be confused about an enemy taking low damage if the enemy graphics show something that looks tough or sturdy (wearing armour or made of rocks).

Elemental resistance, once again, is easily shown through visuals. If I see an enemy made of fire, I can expect fire skills to be less effective against them. If there is an enemy that's strong against both physical attacks and also has fire resistance, then maybe their graphic would be of a fire elemental wearing plate armour or is covered by rocks.

Sure, once you see the enemy party is composed of Knights and Dire Wolves, you know you're facing Physical damage.  Is there anything you can do to react at the point where you're able to make this prediction, though?  You don't usually get to see the enemy party until you're actually in the battle, and at this point your DEF and MDF are locked in and you can't really do anything to counterplay around their magic-heavy Faerie party.  (Not to mention even faeries usually have an "Attack" command that they're using ~40% of the time!)

You actually can, by looking at the area you're in. In a forest area you can expect to mainly face physical enemies, because most of them will be wild animals or bandits. In a crystal cave you could expect enemies to be magical, because crystals = magic in RPG logic. The boss of a water area would be magical, because water is one of the four elements used in magic. The boss in a castle would be physical, because it would be a huge knight. Unless the castle is a mage guild, in which case you would expect a magic boss to show up.

(Faeries using attack command is fine, but they would be hitting low because most of their points are put into magic. But really, this is all something a developer can work around and give that Faerie a magical auto-attack instead).

In competitive Pokemon you can work around the meta and also swap out one Pokemon for another if the Physical/Special breakdown is unfavorable.  In League of Legends you buy your equipment throughout the game, once you already know the opponent's composition.  These are good examples of counterplay and they highlight why having multiple types of damage and defense can be excellent game mechanics.  But I'm struggling to think of places in Dragon Warrior or Star Ocean or Skies of Arcadia that offered similar counterplay opportunities, except where you've already lost to a boss (or read the strategy guide) and you know exactly what they can throw at you in advance.

Not sure why you're comparing PvP with PvE here. They're completely different.

Not seeing how counterplay in a PvE game is necessary either. I find it much more interesting to have to change tactics on the fly during combat itself instead of playing around enemy strategies outside of combat by organising gear and stuff in menus. If you have a boss that consistently beats first-time players that don't find out how to beat it in their first encounter, then there's something wrong with the boss, not with not being able to counterplay before the fight starts. I'd rather engage players with the encounters themselves than having them constantly sit in menus to prepare to counterplay upcoming encounters. This also applies to the previous paragraph; not everything needs to be known beforehand.

(I only just noticed that this entire thing has moved away from talk about stats...)

 

[Kes]

(I only just noticed that this entire thing has moved away from talk about stats...)

It might be nice to keep on topic so as to discuss the OP's question.

 

[jwideman]

I think the stats should only be as complex as your game needs them to be. If you have one character, you can toss out like three stats from the get-go. MAT and MDF don't need to be there if nobody uses magic, and agiltiy can likely be eliminated as well because you're either going to go first or you aren't.

Personally, I think of MAT and MDF as "extra" stats, and honestly, you don't even need one of them if you're willing to change the formula (since most characters with high MAT also have high MDF, and the stats serve the same purpose, there's no reason to have both).

It's fun to play around with numbers, though. Try out stuff and see if you like it, but I would say trim all the fat first, and then add in stats that you need.

-Wallace

The reason to have separate stats for magic and physical is that they are, well, separate. It would be different if you simply removed magic. Though, as I said earlier, that's not enough. We need stats for psionics too.

I've been considering renaming the stats and changing all the formulas to be more D&D-like, with Attack being Strength instead and Magic Attack would be Intelligence. Then defense would be a function of armor.

 

[M.I.A.]

I'll chime in on this one.. for fun.

As someone who opts to use very low/simple battle algorithms.. I always try to find a creative use for "non-essential" Stats like Luck.

For me, to keep strategy live and foremost, all foes (sans bosses) and actors have the same HP/MP. [30hp, 10mp]

When I calculate DMG, it's simply (A.ATK - B.DEF ) + 1 [so any attack will minimally deal 1HP Dmg], so even if a defending target has really high DEF, they still take at least 1DMG.

Skills do a steady DMG amount (+/- a certain %), and don't use M.ATK. Example, Physical Skills might deal 20DMG (+/-) 50% - B.DEF, where Magical Skills will deal 20DMG (+/-) 50% - M.DEF. .. So M.ATK in my games are pretty useless by themselves. So I make a special TP Skill.. say.. a multi-strike Limit Break. The cost is 50TP, and the formula will simply be (A.MATK) and it will strike 15 random times to random targets. Excellent against bosses, terrible against large troops.. but I don't even consider DEF or M.DEF in the equation.

I also almost always have a skill that uses LUCK explicitly for its DMG calculations, free to cast, but with a heavy Cool Down.

With smaller algorithms, its easier to balance formulas (in my opinion) for battle and skills.

For me:

HP = Health (also, Brawling Skills Cost)

MP = Magic Skills Cost

TP = Physical Skills Cost

ATK = Usually total DMG dealt in an attack

DEF = If used, to buffer Physical attacks

SPI = Sometimes used to formulate MAG attacks, usually a base integer to deal MAG DMG in a Special Attack

RES = M.DEF to buffer Magical Attacks

AGI = Determines attack order (aka Speed)

LUK = Nonsensical stat used for a Special Skill formula.

I also use crits, evasion, etc. very consistently. All actors and foes either have a 10%, 25%, or 33% chance to Crit or Evade. (So, 1 in 10, 1 in 4, or 1 in 3).

I have a skill that uses LUK & Crit to deal massive DMG, but only if the Actor (who is an Archer, the skill is Blind Shot) has the Blind state on them.

I also realize that this post does not read as clearly as I think I would have liked it to, but my Rx has kicked in and I'm quite loopy.. lol.

 

[jonthefox]

@Miacuro:  I actually really love a lot of those ideas.  Have you released any games? Would be interested in playing!

 

[M.I.A.]

@Miacuro:  I actually really love a lot of those ideas.  Have you released any games? Would be interested in playing!

Thanks!! I try to keep things simple when I work on projects since I have so little time to work on them.

I don't have anything released yet. I have some projects that I have made great progress on, but mostly I look back on them and view them as practice for the one or two big ones I really want to work on. And I hate to sound snobby, but RMMV is looking like it's going to be my favorite platform!! I have been working pretty vigilantly on the two main projects with MV and I've never been as excited to continue working on them. With RM, RM2K3, RMXP, RMVX, Ace.. it all seemed.. like it was so much more work. I looooove MV!! I think it's the platform I've been waiting for!!

Friend me if you want. Or follow me. Or whatever it's called here. I will get around to posted in Early Projects, etc. as I make headway.

 

[Beamlight]

There is a lot of good discussion going on here, bringing good points and examining various games and discussing their mechanics. Bravo to you all! I want to bring up something from the OP's original post, though, that I think is an interesting take:

Hey everyone, i wondered was what your opinion on stat Complexity in RPGs ( i'm personally making a Tactical RPG but this is more of a global question, even if i do know it depends on the game's aim ). Do you tend to say less is better or do you think there's is something gained from more complex stats ?

For example, with the common RPG maker stats : HP MP ATK DEF MAT MDF AGI LUK

- You could merge ATK and MAT into one ATK as well as DEF and MDF into one DEF, it is indeed very rare to see a character that uses both types of attacks.

-  You could get rid of DEF,  use the weakness system to create a "weak to weapons" character or a "weak to magic" one. After all, defense is just another HP.

- You could merge LUK and AGI. Or even get rid or both, they are not extremely necessary. Removing them would only remove the luck factor.

All of those changes would reduce the stats to only HP MP and ATK ( and maybe AGI ), and honestly wouldn't change the depth that much. Now i do not believe it would be better, but it's a subject i wanted to hear your opinion on.

What do you think is the ideal complexity of stats for an RPG ? Do you think you should get rid of unnecessary stats and try to optimize the stat window or do you think complex stat system actually has another use ?

Specifically physical/magical is an interesting calculation. With the way MV in particular calculates damage, it is 4 * atk - 2 * def, a rather simple equation that can have some profound impact on the way damage is output. A good rule of thumb is that when user's attack is equal to opponent's defense, atk * 2, and any additional points of attack is atk * 4. 

I want to have a little fun with math here, so indulge me in a hypothetical scenario: let's say we just took away defense, and pushed attack to 2* atk, and have, say, and Iron Giant type enemy with a high defense against physical attacks. For my little "calculation", i'll make the attacker's 40 and the Iron Giant's defense 70.

4 x 40 = 160
2 x 70 = 140
               20 damage

Now for elements, say he takes 75% damage from targets with elements, but defense does not exist, and we simply use attack x 2:

2 x 40 = 80
             *.25
              20

Same difference, right? Lets go on ahead and say that our hero has powerleveled and come back to fight the same enemy:

4 x 160 = 640
2 x  70  = 140
                 500

The difference with x2 attack:

2 x 160 = 320
              x .25
                  80

(also note if you scale with elements, the 500 points of damage would be shot down to 125!)

Big difference, right? That's the trick with scaling. I think having attack and defense gives you more control, But I know a lot of people start rpg making purely wanting to tell a story and that feel overwhelmed with stats. I say if it overwhelms you, go ahead and ditch them!  You can still have a lot of fun designing with combat without being overcomplex. 

 

[Wavelength]

Good point, Beamlight - there are definitely consequences of the system you choose outside of simplicity/complexity, some of which will probably be unforeseen unless you really examine it like that.

I do feel like your example shows why multiplicative-based defense (-75% damage) works better than additive for most games, though.  Using a less extreme example, use those same numbers but imagine the character leveled from 40 ATK to 50.  In the original atk * 4 - def * 2 formula, the damage jumps from 20 to 60 - a threefold increase in damage from a 25% increase in attack power!  How hard would such a game be to balance?!  With percentage-based damage reduction, the damage in such a scenario jumps from 20 to 25 - a 25% increase in damage that's representative of the 25% increase in attack power.

 

[Kes]

That is indeed a good point, but I wonder if something gets lost in the simplification?  For example, the enemy with super-nasty spells, but low DEF, or just a melee fighter with high DEF.  In a troop with both enemies, which you go for first is a strategic choice.  Take away the DEF element and the choice gets a bit bland.

 

[Razoir]

Good point, Beamlight - there are definitely consequences of the system you choose outside of simplicity/complexity, some of which will probably be unforeseen unless you really examine it like that.

I do feel like your example shows why multiplicative-based defense (-75% damage) works better than additive for most games, though.  Using a less extreme example, use those same numbers but imagine the character leveled from 40 ATK to 50.  In the original atk * 4 - def * 2 formula, the damage jumps from 20 to 60 - a threefold increase in damage from a 25% increase in attack power!  How hard would such a game be to balance?!  With percentage-based damage reduction, the damage in such a scenario jumps from 20 to 25 - a 25% increase in damage that's representative of the 25% increase in attack power.

Yep i actually felt the same thing ! Though, it really depends on your game, however if your stats raise relatively quickly you might want to use percentages. The issue with minuses is that it scale very quickly and, as such, could be harder to balance. Wanting the same boss to be beatable but still challenging for a lvl30 that plays well as well as a lvl40 would be far, far harder.

That is indeed a good point, but I wonder if something gets lost in the simplification?  For example, the enemy with super-nasty spells, but low DEF, or just a melee fighter with high DEF.  In a troop with both enemies, which you go for first is a strategic choice.  Take away the DEF element and the choice gets a bit bland.

Of course, you do not take away DEF entirely without severely adapting the game at all. If my game had no DEF element, to make a Bulky melee fighter, I'd drastically increase his HP and/or i'd increase his physical damage resistance. A glass cannon would have low HP and/or great weaknesses to physical/elemental damage.

Of course, removing DEF from a game does not make it better, what i meant to say was that we could remove DEF without decreasing depth at all ( if the game is built for it ). It has no link with the fact that you don't like numbers or not, or don't want to be overwhelmed with stats. I was wondering what was the most fun for the player : few stats or the usual, knowing that, with right game design, both can have as much depth.

 

[NichG]

My feeling is that every stat which is in the game needs to do a distinct and qualitatively different thing. Each stat is an opportunity to differentiate characters, abilities, equipment, etc, which makes the game feel like a richer overall experience. So if you can give me 6 equally meaningful, very distinct stats then I'll take that over 3 stats, but if you give me 6 stats which are have lots of redundancy or 'fake diversity' (e.g. 'this stat controls damage for red skills, this stat controls damage for blue skills, ...') then I'd rather just have the 3 condensed ones instead.

For example, in something like a turn based strategy game, movement range, attack range, attack AoE, damage, and defense would all be natural, qualitatively distinct stats. But if you throw in evasion, it risks just being defense by another name. However, lets say now you had it so that skills/abilities take some amount of time before they go off (allowing enemies to move out of range/etc) - a 'speed' stat which lets you fire off abilities with shorter delays would again be a natural stat for that system.

 

[Drackomordain]

I'm gonna cast my lot in for Stat Complexity, but with a condition.  It has to fit the game.  Stat complexity in and of itself doesn't add fun.  

Stat complexity becomes fun when the player has to make a meaningful choice concerning stats.  

"Do I make my paladin a better healer? or a better fighter? or a better tank?"

The more Stat options the more fleshed out you can make your characters as well.  

Ex. A character with 22 Intelligence, 13 Wisdom, 7 Charisma  paints a different picture than a character with 14 Mind.

In  RPG Games with multiple party members, this translates into party roles and how the player decides to customize the character for the role he has chosen for him/her/it.

I have ALWAYS felt, and very strongly, that it is not a designers job to decide for the player what role the characters will serve in battle.

Person A likes the idea of a DPS Paladin, person B likes the idea of a Tank Paladin.  Which is right?  Both and Neither.  It should be up to the person playing to decide what role each character serves.  

 

[Black Noise]

I don't know if this is appropriate or not but I didn't want to create a whole thread on this either.

Edit: It was inappropriate and I am sorry I posted it without thinking.

 

[Kes]

@Black Noise

This isn't really the sub-forum for detailed feedback on a single, specific project.

Neither is it the done thing to hijack someone else's thread for your own purposes.

I am unable to merge your single post into Features Feedback, which is where it belongs.  Please, therefore, re-post your query in that forum.  

And rather than have this thread derailed into something else, I would ask other members to leave any feedback for Black Noise in that forum, not here.

Thanks.