MMORPG Healing: Evolution?

[Sky]

And there's also the kill-to-get-xp bullshit.

If you lose that, you open up the possibility of actually running away when you're getting beaten down too badly, but still getting character advancement gains (assuming you even want that in your game, natch).

Maybe the need for healing is from the fact that the characters in these mmorpgs are total pussies, excepting CoH. And the more you gain in levels in a game like EQ, the more you become a pussy compared to an equal con mob. Silly imo.

[Kail]

MY SUPER-SECRET, SUPER-ADVANCED AI ROUTINE THAT TOOK YEARS TO FINE-TUNE:
First, attack the guy that heals the most.
Second, attack the guy that damages you the most.
Third, attack the guy who serves no purpose other than to be attacked once all his healing and damaging friends are dead.

I think you'll find that making enemies behave somewhat logically would totally change the dynamic. Maybe the guy who is just tough would actually have to, you know, protect his buddies somehow, rather than spam taunt over and over - or else find heartier buddies.

The implicit assumption in the "tank, damage, healer" paradigm is that the good guys can dictate what the bad guys do. (Namely, attack the tank) If they can't and the enemy has a brain suddenly the black mage who dies in two hits doesn't look so hot anymore.

I agree with most of what you're saying in concept, but if I may be allowed to channel my inner catass, it looks like you're just getting rid of the "holy three" by making two of them basically useless. Tanks work because they're able to suck up damage that would otherwise go to the mage or cleric; take away that ability (by coding in AI that always goes for healers first) and all your tanks are basically useless.  Healers work because they're able to heal damage and thus help the other members of the team fulfil their roles.  Put healers on the top of the aggro list, though, and they need to spend all their time frantically healing themselves, making them useless, as well (though maybe they could function as a kind of crappy tank or something).  Suddenly, the only viable class is the damage dealer.  Now, admittedly, that's the way fighting in real life works, but I don't see how that, in and of itself, is going to improve things.  Now, instead of having three really boring classes, you've only got one...  That's an improvement?

To toss out a better example: let's say I'm playing a game and my character gets hit by an arrow.  So, the next time I level up, I put a point into a skill that protects me from any arrow attacks.  Suppose someone's coded this game's AI to react to players intelligently: the result is that no enemies will fire arrows at me (they'll all use swords or whatever other option they have).  So, as a player, while the skill may indeed be extremely powerful, there's no time I can ever say "Wow, it's a good thing I spent points on that skill, I made a good choice," because the AI (acting intelligently) is never going to put me in a situation where it's useful to me.

I agree with most of the rest of what you said; the idea of diversifying your damage dealers sounds like a good one to me.  I just don't know that cranking the AI would help very much.

[HaemishM]

There's a whole lot wrong with MMOG combat from the AI side.

First, AI monsters have almost NO group tactics. There is none of the cooperation that exists in player groups, other than the healer heals (sometimes) and the casters cast and debuff, and all of them generally have the same target for these abilities, the tank. Thus, we have aggro control, which is really borken. AI groups can only have one target at a time. They have no overriding mission, they have no reason to be there, other than as MBI's, to use Lum's terminology. They don't guard jack shit, other than the little area that triggers their aggro. They don't go out and try to raid anything. Occasionally if there are quests involved, they do something related to the quest, but for the most part, they are completely passive. They exist TO BE KILLED FOR IMPROVEMENT UNITS. There's the real problem.

Make the AI exist for something other than to be achievement barriers for the players. Give them a purpose other than trigger point guards. Once you start thinking about mobs having purpose, then you start thinking of them as more than MBI's. Once they have a purpose, you can start thinking of how they should achieve that purpose. DAoC actually had one interesting part of AI, with certain mobs being considered scouts who would run from the players and go grab more friends to attack the players. That's a start. Take it further. Let mobs go after towns and cities. Don't let the town guards be safe zones for players; essentially, the town guards in MMOG's are "Get out of fucking up" cards. Stop that. That doesn't mean remove safe zones, but it does mean removing some of them where it makes sense in the environment.

[jpark]

The underlying design goal was actually a business goal - the assumption that subscribership retention is most served by enforced grouping so that people would make the social bonds that keep them paying.  (Implicit was the assumption only addictive behaviors/game mechanics made for profitable games, since well-refuted by CoH and WoW)  There might have been something to that when the early market were techie geeks with no social IQ.  It does seem that developers are again acknowledging that there are more well adjusted people using their services than formerly realized.

Player retention?  Yup, I buy your point.  Ironically, especially in EQ, grouping provides another (mal)function - content.  Half the night in EQ was spent just trying to get a group together.  In EQ just havinga  tank and healer was not enough - you needed a specific group composition (enchanter, monk for pulling etc.) depending upon the encounter in question.  This was "content" since putting together a group took so much work.

Bad "content" to be sure.  CoH has a nice twist to this - if you can't find a good healer - get a good controller.  In CoH, one support class can often substitute for the other if your group fights differently.  EQ never had that flexibility - if you needed a healer or enchanter - you could never swap.

I think this has contributed to the ill feelings towards support classes like the healer.

Either way, there needs to be 1) something you can do to stop your opponent from healing his way out of every situation, and 2) some kind of risk associated with healing, in order to balance out the risk/benefit analysis (which, right now, usually has no risks at all, meaning it's always a good idea to heal anyone who needs it).

Point #2 is a great insight.  I remember sometime ago suggesting that to do what you just mentioned, that some of the best heals might be ones where the cleric has to walk up to the tank and "touch him" thus bringing him into risk of aggro from the target that tank is fighting.  That is a risk for the Healer - if he wants to heal better - he has to get closer to the tank in combat risking more aggro himself.  The closest mechanic I see to this currently is the CoH Empath defender - absorb pain - a high level heal - causes damage to the healer.

[Shockeye]

I may bitch and moan about crappy AI in monsters in MMOGs, but if it was there I might be one of those people avoiding it because it's "too hard".

[Polysorbate80]

MY SUPER-SECRET, SUPER-ADVANCED AI ROUTINE THAT TOOK YEARS TO FINE-TUNE:
First, attack the guy that heals the most.
Second, attack the guy that damages you the most.
Third, attack the guy who serves no purpose other than to be attacked once all his healing and damaging friends are dead.

I think you'll find that making enemies behave somewhat logically would totally change the dynamic. Maybe the guy who is just tough would actually have to, you know, protect his buddies somehow, rather than spam taunt over and over - or else find heartier buddies.

The implicit assumption in the "tank, damage, healer" paradigm is that the good guys can dictate what the bad guys do. (Namely, attack the tank) If they can't and the enemy has a brain suddenly the black mage who dies in two hits doesn't look so hot anymore.

I agree with most of what you're saying in concept, but if I may be allowed to channel my inner catass, it looks like you're just getting rid of the "holy three" by making two of them basically useless. Tanks work because they're able to suck up damage that would otherwise go to the mage or cleric; take away that ability (by coding in AI that always goes for healers first) and all your tanks are basically useless.  Healers work because they're able to heal damage and thus help the other members of the team fulfil their roles.  Put healers on the top of the aggro list, though, and they need to spend all their time frantically healing themselves, making them useless, as well (though maybe they could function as a kind of crappy tank or something).  Suddenly, the only viable class is the damage dealer.  Now, admittedly, that's the way fighting in real life works, but I don't see how that, in and of itself, is going to improve things.  Now, instead of having three really boring classes, you've only got one...  That's an improvement?

You'd have to give the meatshields some tactical way of actually defending some person or spot.  Blocking the doorways, using your body or shield to intercept enemies (a la DAoC), grappling/tripping enemies, whatever.

To toss out a better example: let's say I'm playing a game and my character gets hit by an arrow.  So, the next time I level up, I put a point into a skill that protects me from any arrow attacks.  Suppose someone's coded this game's AI to react to players intelligently: the result is that no enemies will fire arrows at me (they'll all use swords or whatever other option they have).  So, as a player, while the skill may indeed be extremely powerful, there's no time I can ever say "Wow, it's a good thing I spent points on that skill, I made a good choice," because the AI (acting intelligently) is never going to put me in a situation where it's useful to me.

You wouldn't want to have most enemies *never* fire arrows at you; that would be an unrealistic "cheat" on the part of the AI (never mind that games cheat that way all the time).  It shouldn't have any way of knowing that you're invulnerable to missile attacks without trying at least one (unless your SuperAntiMissileShield includes a big sign floating over your head that says "Neener neener neener!")  Any additional combatants coming into a fight could be assumed to know your missile-immune status, since presumably any intelligent enemy would be able to communicate that amongst themselves. 

Stupid monsters, well, who knows.  Maybe they *are* dumb enough to do it indefinitely, since they're running on instinct instead of problem-solving skills.

A better method is giving the power trade-offs.  In some games like CoH this already happens in some instances--your uber-power is timed, or drains your stamina like mad, or gives you some other weakness.  You can't run it all the time, but only when the benefits outweigh the drawbacks.

(Whups, complicating the game, there go the players over to something else....)

*edit for shit-poor grammar*

[Xilren's Twin]

I may bitch and moan about crappy AI in monsters in MMOGs, but if it was there I might be one of those people avoiding it because it's "too hard".

While I take the point you and Lum touched on, it's important to remember "better AI" doesn't have to mean simply "harder".  I think what most people would be happy with is simply some order of variance in combat behaviors by mobs.  Every combat should not play out the same as every other combat across all mobs types. 

Even simple stuff; i.e. have an AI for aggressive animals with no/low awareness.  Say a rabid tiger may choose their targets purely randomly in a combat and be immune to "taunting" b/c um, it's a tiger.  But, some tigers may fight to the death, some may run when barely wounded.  Do you pursue a wounded tiger into a dangerous wilderness?  Passive animals like deer may default to run from any agressive action at all, so if wanted to actually hunt one of them you may have to use teamwork to flush a mob towards a person with a melee weapon or set trap, or use ranged attacks and pursue, etc etc  That sort of stuff. 

Intellegent, weapon wielding humanoid should be able to have varied behaviors.  This doesnt have to be extreme, but hell even in NWN when designing monster spawns you could set it up so an orc encounter randomly choose from 5 different orc types like two handed axers, bowmen, sword and shield, tough orcs, priests and the like.  A group of 4 bowmen and 1 shaman shouldn't be the same fight as 2 axes and 3 swordsmen. 

Someone else mentioned it, but the sheer fact that in most mmorpg's today there is really only 2 possible options, victory or death.  That is highly limiting.  But as long as exp points are only awards on death of mobs, anything that changes that stucture wont jibe with the rest of the game system.

Redoing combat is something thats a ground up redesign. Healing is just one aspect of the typical overly ordered and predictible game system in most mmorpgs today.

Xilren

[Margalis]

So the problem with this is that when monsters are coded to use "smarter AI" (sometimes tearing after the casters, healing smarter, etc) the players consistently hate it, and vote with their feet to the areas that have monsters with the least amount of AI. Stupid monsters mean lower risk, which help skew the risk/reward ratio in the players' favor.

Damion Schubert has a good discussion on why we don't really want good AI here: http://www.zenofdesign.com/?p=337  Basically, players are already very well trained to break combat down into a problem that CAN be easily solved. If combat is changed so that it cannot be easily solved, players will either solve it easily anyway (ie raiding in EQ1) or decide that your combat sucks and refuse to play.

Again, I think you are defining away the problem. Why not just say "players don't want a good game" and leave it at that? Before WoW you could make the argument that players were trained to sit in one spot and endlessly pull the same MOB over and over.

If you change the creature AI, you will also have to change the classes in the game. And yes, that would break up the trinity but it would also introduce new things to take it's place. The pure damage dealer and pure healer would both fall by the wayside, but new character classes (with hopefully more variety) would take their place. In real life you do have battlefield medics,  combined arms, etc.

Players will choose to fight the easiest creatures, so don't give them a choice. Maybe between level 40-45 the best place to level is in a forest and in 45-50 it's in some humanoid village. Or in a quest-based system some of the quests are against tigers and some are against smart spellcasters. You might then see groups forming differently based on the mission types, but everyone could have a role at least some of the time. If you just make the game "here are 10 areas appropriate for your level where you can just stand around and pull endlessly, and area C is the best risk/reward" then yes, everyone will go to area C. So don't do that.

---

You need to look at PvP for a clue. In PvP in WoW you have people debating what types of groups are the best, even who wins 1v1 matchups. Some people claim a Shaman is really good, some people claim a Pally can easily kill a Warrior and some people claim the reverse. It isn't easy to min/max that or arrive at the best character or group because tactics and skill vary a lot. Somebody may thing a group of 5 Priests is clearly the best and someone else may think it's a Druid, 2 Rogues and 2 Mages. Whereas it's pretty clear to people how well the classes and groups stack up against the AI.

They key is that in PvP opponents learn and adjust, and different opponents emply very different strategies.

I think it's safe to say that at this point in WoW there isn't one super group or super class that is universally agreed as being the best for PvP. Because there are many variables at work. In AI bashing there are usually NO variables at work other than base stats.

[HaemishM]

Remember, the reason most people don't want to play a tank, or even a healer, is that for the most part, these pure classes are the most boring in the world to play. People wouldn't mind healing being removed if playing a damage-taking melee class was actually fun and/or took some skill. The same goes for playing a healer.

If you can't even be bothered to make melee fun, why should you bother with anything else?

[ajax34i]

I think the whole concept of "healing" and "damaging" is simply tied to the hitpoints mechanic, like someone already said.  Consider a game where no one has HP's; make death a status effect, add a bunch of other status effects (stunned, rooted, blinded, deafened, disabled in some way, reduced ability to pick targets, slowed, etc.) and combat can be made about forcing the enemy through a bunch of status effects, with the purpose of getting to the "dead" status effect.  Make all classes be, basically, crowd control on steroids, no damage whatsoever (because there are no hitpoints to do damage to), and give each class a series of moves they can do to inflict progressively worse status effect disabilities on the target, until they can inflict the "death" status effect.

Voila damageless combat.

[Kail]

Consider a game where no one has HP's; make death a status effect, add a bunch of other status effects (stunned, rooted, blinded, deafened, disabled in some way, reduced ability to pick targets, slowed, etc.) and combat can be made about forcing the enemy through a bunch of status effects, with the purpose of getting to the "dead" status effect.

I don't really see how that would work out better than a hit point system, myself.

Personally, I'm starting here with the assumption that we're trying (in general) to avoid the instant death, kind of "I'm walking along in Stormwind, la-de-deee, oh, look there's-" WHACK -dead.... kind of scenario, yes?  So you've got to have some kind of gradual transition between the "walking along" phase and the "dead" phase.  If you want to call the steps along that transition "status effects" instead of "hit points," I don't see how it would change anything.  If you need the same healing ability to cure every status effect, it's basically identical to the way hit points work.  If you need different abilities to cure every status effect, it's just an extremely complex version of the same system.  If you can't heal the status effects at all, I don't see how it would be any better than just doing a hit point system with no healers.  I think it could work, but again, with just that change by itself, I don't think it would solve a whole lot.

[ajax34i]

You wouldn't need the trinity.  The complexity can be used to eliminate it.  You wouldn't have all combat rely so much on that HP bar, yours and the enemy's.  You wouldn't have the same classes as every other fantasy MMO out there; you could invent new ones with completely new/weird ability combos.

But yeah, I agree, we can't really go very far with this.  Bottom line, the point of combat is to kill the foozle.  You can make everything depend on this one statistic, the HP bar, or implement 15-20 ... 100 different paths-to-death, so to speak, combinations of timed / coordinated attacks (some inexorably, some counterable) that put the enemy in worse and worse conditions, the final one being death.  "I think I'm in trouble.  I'm in trouble.  I'm in deep trouble.  Oh shit, I'm gonna die.  Goddammit I've died."

In any case, I'm not an expert, so   .   <- grain of salt.

[Viin]

There are some games that do this well, the two I can think of now are Guild Wars and even Battefield 1942/II.

Even though GW has classes, even the "boring three" to some extent, their skills are arranged in a way that no two alike monks would work better than a monk and a warrior. Add in the requirement that you can only take a small subset of all the skills you have, you can have monks with completely different skills and playstyles. Add in the secondary class and it's an even broader range of possibilities. You *want* to play with other archetypes just because it'll help you defeat the foe. Not because you are a gimp in melee and only allowed you heal. Heck, you could be a smite-only monk who only deals holy damage and doesn't have a single heal spell. Next round, you could play the heal-only monk or a mix. Just depends on what you want to do and what you think the group needs. (Coupled with every class having at least a limited ability to heal themselves, pure healers aren't always required).

Healing in and of itself is not a bad thing, though I do think you guys are right: the current formula must be changed. GW does a great job with this by heavily balancing everything against everything else. Want to 'super heal' someone? well you have to touch them. Since you cannot taunt them and _force_ them to attack you (talking PvP here) it certianly is a risky thing to do. Since GW was balanced for PvP and not for PvE it makes sure all the classes are useful in their own right.

Balance the PC classes against each other. Make them useful in a group and solo.
Create the NPC AI and abilities after creating the PC classes.
Figure out how people will play their characters and even collect data on this. You can use this information to create PC-like behavior in similiar mobs. (Do healers always cast a certain spell? Why? Is there a way to block it? Do warriors always do a bash followed by a punt to the head? Then this must be a common tactic, so have the NPCs use it!)
The harder mobs don't need more HP, they just need more smarts.

So.. PvP focused games seem to help negate this problem, maybe they are doing something right.

[Trippy]

I think the whole concept of "healing" and "damaging" is simply tied to the hitpoints mechanic, like someone already said.  Consider a game where no one has HP's; make death a status effect, add a bunch of other status effects (stunned, rooted, blinded, deafened, disabled in some way, reduced ability to pick targets, slowed, etc.) and combat can be made about forcing the enemy through a bunch of status effects, with the purpose of getting to the "dead" status effect.  Make all classes be, basically, crowd control on steroids, no damage whatsoever (because there are no hitpoints to do damage to), and give each class a series of moves they can do to inflict progressively worse status effect disabilities on the target, until they can inflict the "death" status effect.

Voila damageless combat.

Then the "healers" will be the ones who can remove status effects.

[Margalis]

HP is not in itself a problem. There are lots of games where the goal is to reduce HP and there aren't just 3 basic classes or very simple rules.

Hell, in FFXI you can make the argument that there are 6 or 7 distinct character classes. I'll make it:

White mage - healing.
Ranger - DPS
Black mage - Magic DPS (different because of some game mechanics)
Paladin - tank
Bard/Red Mage - downtime reduction, buffs
Theif - hate control
Beastmaster - solo friendly class

The problem is every other class in the game basically falls into the DPS category but is worse than the ranger. (Or the summoner who is a bad white mage, and ninja who is a bad tank) And of course there is the fact that an optimal group HAS to have a white mage, either a red mage or bard, and a black mage and paladin.

Theif is a form of damage prevention but is not a tank. Black Mage and HTH Combat DPS guys are different because of the game mechanics of magic burst, it's better to have a couple of each than all HTH or all magic.

A big problem in games is not so much that one class in the end is a bit better than another, a bigger problem is that some classes are worse than nothing at all. For example in FFXI if a group needs a damage dealer and Monk is handy it's often better to just wait for a better damage dealer to become available than take the monk. So it's not Monk competing with Ranger and Dark Knight, it's monk competing with waiting around and doing nothing. That's because the way the game is structured an optimal XP group can pull in many times the XP of a just OK group.
---

One of the things that really excited me about WoW was when I heard about how encounters would be with groups of enemies. I think those fights are a lot more interesting because they are less scripted. Not everything goes exactly according to plan. The more you can predict the fights the more you can slot people into very specific roles.

Of course, the problem is that in WoW you don't really need groups at low to mid levels outside of some bosses and instances, and you don't really right groups of enemies very often anyway.

[ajax34i]

Then the "healers" will be the ones who can remove status effects.

Not really.  In a sense, yes, the ability to remove one status effect is equivalent to healing a few HP's in the HP-based game.  However, the "healers" set doesn't exist anymore because there's so much grain and differentiation that no one can be branded "healer".   Everyone can cure one effect or another, but that doesn't make them equivalent, in the way that a priest and a druid are.

Bottom line, there's only one "dead" condition, 0 HP (some games have tried 3 bars, more complexity).  That doesn't give much room for innovation.  Having more than one "dead" condition (suffocated, throat slit, backstabbed through the heart, bled to death, all of them you're just as dead), and making combat about the process of getting there, with many paths and choices, may be fun.  Don't know if it is.

This is what I imagine:  You're a rogue, and the only way you can kill someone in platemail is if they're on their back on the ground, disarmed, and not struggling that much, so you can perform your (only) finisher move which consists of inserting your dagger between the platemail joints to stab the heart.  Now, rewind, and imagine you're standing there, facing a pissed off knight with an axe in his hand, and he's certainly pissed off and seeing you.  How do you bring this guy into the kill posture? 

Combat, whatever you do, is not about HP anymore, it's about getting to the "flat on his back" position.  He could very well have "very good balance" (the ability to "cure" your "trip" or "push over" status effect attacks) as part of his knight class skill package, does that make him a Priest?  Druid?  Healer?

[Pococurante]

Damion Schubert has a good discussion on why we don't really want good AI here: http://www.zenofdesign.com/?p=337  Basically, players are already very well trained to break combat down into a problem that CAN be easily solved. If combat is changed so that it cannot be easily solved, players will either solve it easily anyway (ie raiding in EQ1) or decide that your combat sucks and refuse to play.

Which simply riffs off the Monty Haul paradigm.  Which many users like too.

So we're back to the question of how do we make a massive game that pleases the IGE's of the world, the min-maxers, and the casuals.  And the answer hasn't really changed in the last two decades.  Target a niche or be banal.

There's been a zerg rush towards a single business model for too many years.  I like that we're now seeing or hearing promise of games that segregate playstyles.

[HaemishM]

So we're back to the question of how do we make a massive game that pleases the IGE's of the world

Fuck them. Fuck them in their stupid asses.

[Pococurante]

But it's a billion dollar business! It employs starving game players the world over! Old fart gamers regarner lost memories of their pot-hazed timeless youth!

You would crash cross-global economies and the dreams of millions for some quick anal action?  For shame, Sydney Carton, for shame!

[HaemishM]

For that kind of anal action, I'd crash every known MMOG in existence, starting with SOE's lineup and working my way down.

[Margalis]

That link on AI was crap. 100% realistic AI in games might suck, true, but there is a happy medium.

I also hate arguments like "people love EQ, and EQ has EQ AI, so people love and have been trained to love EQ AI!" You can make that same incorrect argument for tons of things. People love very slow treadmills, right? People love forced grouping, right? People love a game with few quests, right?

You can't compare something to nothing and based on that declare that the something is better than every other something. If people prefer water to acid it doesn't mean there isn't a market for coke because water is the very bestest.

What you see in MMORPGs with WoW is what you see in a lot of industries - everyone is basically on even ground but when one guy does something cool and different everyone else suddenly looks behind and has to catch back up. But before that guy went out and did that cool thing nobody even considered the fact that things could improve - often including the users! And they can't vote with their dollars they have no choice because there is no alternative.

If you play a game with crap AI it doesn't mean you like crap AI. What else is there to play?

And the idea that MMORPG combat is a puzzle for players to solve is absurd. It's been the same puzzle for years, and solving it takes 10 minutes.