Friday, March 09, 2007

Adding another member to the Class Trinity

If you are familiar with a MMORPG then chances are that you're familiar with the idea of the class trinity. Back in the 70's when the creators of D&D were laying down the framework for pretty much all fantasy based RPGs they originally came up with only three roles. These were the infamous tank, damage, and healing roles that are still used today as the basis of any class oriented MMO. Over time more classes that fulfill these rolls have been created but very few improvements or differences in the classic roles have been made.

Current MMOs have been using the class trinity system for so long that things have become a bit stagnant. Numerous attempts have been made to abandon it and go with something a bit more free form like a skill based system. However, games based on free form skills usually come across as unnecessarily complex to average gamers and have limited appeal. Most MMO companies have instead adopted several ideas over the years to modernize the class system and try to add some freshness to the fantasy combat genre.


Ideas that have been added to the Class Trinity

Hybrid Classes:
Hybrids were created to try to blur the roles and allow players to be able to fulfill more then one role in a group. However, developers quickly found out that idea of hybrids didn't really work out as intended because of the imbalance in class popularity. Hybrids that could deal damage would often stick with that role over their other options simply because most people found it more fun. Hybrids that could heal were often pigeonholed into that role since the pure healer role was so unpopular.

Crowd Control Classes:
MMO developers have always had a problem designing complex encounters and usually fell back on just including more mobs to make a fight more interesting. A couple games introduced crowd control classes like bards or psionists in an attempt to give groups a chance at these harder encounters. This slowly caused crowd control classes to be required by groups at the higher levels and caused group formation times to increase. One of the best design decisions World of Warcraft made was to give every class a limited amount of crowd control.

Buffing/Debuffing Classes:
The idea of the buff class was to provide an additional role to the class trinity which would augment the abilities of fellow players while limiting the abilities of their foes. Classes meant for this role were usually given a wide assortment of spells but had limited survivability (armor and hp). Often the class was given a pet to allow some solo utility when not in a group. Most modern MMO's have since turned this idea into a pure damage class with some limited abilities to buff or crowd control.

You can see how these three ideas have changed the class system since the early days of table top gaming and early text based computer games. However, within the last five years not much has been added to the class system. World of Warcraft has introduced perhaps the biggest change by allowing almost all of its classes some hybrid and crowd control functionality. This was done more as an attempt to achieve a fair pvp system then to increase popularity of the class system but has seemed to pay off for the company.

My own ideas for introducing something fresh into the class system revolves around a class which would handle threat or aggro. Basically they could produce some threat themselves but their main area of expertise would be taking it from others and giving it to the tank. This would allow a unskilled group a chance at higher end encounters as long as their Threatmaster was skilled at partitioning off the aggro produced by heals and damage. Plus if the game allowed the class to view how much threat he was moving in numeric format it would give much the same sense of accomplishment as causing giant damaging crits.

While the class would have some greifing potential it would be a great boon to a casual oriented MMO where players often find themselves grouping with people of differing skill levels. How many times has a group you've been in had its damage classes pull mob aggro by either accident or through bad luck? In such a situation the mob usually ping pongs back and forth and settles on the healer if not taunted off. A threat based class would be able to prevent such situations and thus could guarantee large groups of mobs would stick on the tank. Tanks might not losing the threat management aspect of their class but at least this is not a class that would compete directly with them.

Anyways I'm sure there are many other ideas out there for making a completely new class or even introducing a new role into the class trinity. Either way leave your suggestions in the comments and try not to include any ideas that might have been used in EQ, AC, DAOC, DnD, or VG.

13 comments:

Bosh said...

The problem with that kind of class would be that it would be almost completely useless for both soloing and PvP and it has the potential to be useful enough in group PvE to be a required class and it would be VERY annoying to end up having to need one of those, and a healer, and a tank, and a crowd controller to make a group be effective.

I think it would be a better idea to scrap in-combat healing and taunting altogether since neither are used much in PvP (taunting is completely useless and focused fire makes healing mostly useless). What I would rather see is AI that isn't more stupid than a random number generator when it comes to target selection ("I know I'll attack the guy wearing metal and ignore the guy wearing a robe, that makes sense!")and more reliance on damage mitigation and defense such as:
-Being able to take hits for your party member.
-Knockback effects to keep people away from your squishy groupmates.
-Temporary invulnerably spells that can be cast on groupmaters.
-Damage shield spells that damge peple attacking you.
-Spells that deflect ranged attacks back on their originator.
etc. etc.

Relmstein said...

You raise a good point that a threat handling class would be useless for pvp and soloing. Even if given a pet for soloing the class would still bring very little to a PvP team.

I believe MMOs should either be balanced for PvP or PvE. I think the current way WoW is bouncing back and forth between the two is hurting the game.

If you tried to scrap in combat healing I think most dps class fans would scream bloody murder. A lot of dps are very used to being able to just go all out on a mob and have the healers and tank take care of the little things like aggro and hitpoints.

Ticker said...

The class limitation to those types of interaction exists because the combat systems only facilitate healing, taking damage, and dealing damage, essentially. To break beyond the trinity, more types of mob interaction will have to be invented. The problem isn't the lack of classes, it's the lack of interaction between your team and the opponent.

Relmstein said...

An interesting idea: Balance for a set of tactics that works on both players and mobs. If the players are smart enough to focus fire on the healers then so should the AI.

The only problem with that is it would require a complete redesign of combat abilities. Basically you would have to give tanks and dps classes some other way to keep healers alive besides taunt. Your suggestions of stuns, knockbacks, and limited vulnerabilities is good and reminds a bit of CoH's playstyle.

The only problem CoH had was that it was so hard to balance such a system that including items would have made it impossible. I think one of the big draws for MMOs is still statistical improvement through equipment.

Anonymous said...

I was a debuffer class in AC2 - Enchanter and solo'd just fine. It was harder in the beginning because the pet was all but useless but as I worked my way up the talent tree, it got much better.

All of my skills translated as is to the KvK, except Mezmerise - don't know why, I should have been allowed to "sleep" players like you can in EQ2. Which brings me to another game where I was a debuffing/CC class - Illusionist. Again, the PVE and PVP mechanics were the same and I didn't have any problems with soloing any PVE content.

For the most part, WOW has no debuffing class. There are a few classes with a single spell that can debuff such as the Priest's Dispell Magic and the Shaman's Purge. However, with all the damaging DOTs and beneficial buffs in the game, it seems like a debuffing class would work well.

And even though the Mage can Sheep and Frost Nova and the Rogue can sap, it still seems to me that there could be a primary debuff with more CC added to the staple of classes.

Anonymous said...

Don't get me started on healing spec bitterness. My WOW priest was the first healer ever and likely to be my last. I doubt I'll ever take the risk again of rolling a healing class as a replacement for a main character.

Funny thing is that I LOVED leveling her solo - had no problems what so ever. It's the group experience that totally sucks.

Anonymous said...

No game ever did a healer right. WoW's continuous nerfing of healers is beyond reprehensible. They also have no idea about balance or class design.

Even if healers were overpowered, no one really wants to heal.

Oh well. I've already quit my 70 priest and the game itself.

Relmstein said...

Yes so far the healing class is the least fun and only ever seems to get played because it allows people to get group and raid spots.
Me and my friends try to get around it making sure everyone has a healer alt then taking turns healing.

The Psionicist class in vanguard looks like an interesting take on the CC class. The basically make illusions that have hitpoints and taunt but don't do damage. They can also stun and sleep it looks like.

Taquelli said...

The Commander – Character Class that has a function without being an absolutely necessary member of the group. Basically provides a briefcase of abilities that try to maintain control of the situation. Not a pure crowd control class, not a buffer/debuffer, but a class that supports in many ways.

- Buffs and Debuffs: Naturally, they will have some buff and debuff capabilities, more than any other class, as well as abilities that hurt the enemy’s attack capabilities, lowering armor, attack power, mana supply,
- Crowd Control: Probably will have the most powerful form of crowd control, but not to the degree that they are necessary, just helpful. Probably mage sheep + sap capabilities.
- Heads-up: The class has the capability of access enemy’s hit points, statistics, abilities, hints, loot tables, pull ranges, and reputation; as well as pertinent information about the dungeon and zone. Can also shout out when a huge ability is going to hit. Basically all the functionality of CT Raid in an avatar.
- Command: The class can issue commands to various members of the party, basically forcing them to perform certain tasks. Also can short-range teleport any member of his party, to move them out of the way. Possibly even speed up cool-down timers, and quickly bounce aggro back to the tank.
- A tiny army: Has the ability to fill a party with his pets, creating level-appropriate but stupid drones that perform the functions of a tank, a DPS, and a healer. They aren’t nearly as powerful as a PC-equivalent, and need to be ordered by the commander. They cannot be summoned if the party is full.

Issues: Still not terribly fun. The drones add to the solo experience, nothing like always having a party to help with your quests, but that alone causes issues with dungeons: why would a commander ever join a party if he always has one? Also seems like a required class for a raid leader, and may cause tension if abused or unexpected in a party. “But I didn’t want to heal him!” Also, heads-up may destroy the mystery of the game, but it beats alt-tabbing, I suppose.

banzaimonkey said...

The solution to the "trinity" issue is not to add more classes, but simply to dissolve the trinity outright. Hybridize every "class" or character. This is what games that have an entirely classless system attempt to do.

The problem, however, is that the rest of the design elements in these games are not designed to work in this fashion. Changing the way classes work (or getting rid of classes all together) requires a departure from the standard "kill X, get Y, level up, repeat" 'gameplay' dynamic.

I really don't think it's possible to eliminate the "trinity" because the entire system upon which most of these MMOs are built is reliant on such as system. There isn't really much to do in an MMO other than kill enemies or heal something. The entire gameplay dynamic is focused around hitpoint bars. In some cases, mana bars are involved also. In order for this "tank, healer, DPS" thing to change, that core dynamic has to change first.

Now don't get me wrong, I don't mean that hitpoints or mana should be eliminated entirely, but they need not to be the primary focus of every aspect of the game.

If you want to see more interesting things than the "trinity", you'll need to change the way combat works. You'll need to add things like morale, diverse devices and skills, smarter enemies, and smarter players.

AI and/or scripting is probably the linchpin here. The less effort made to make combat more intelligent and dynamic, the more it relies on predetermined figures and statistics to determine the outcome. That relies on predictability, which in turn relies on simplicity, which inevitably rests on a simple system that noone wants to change.

Anonymous said...

Your synopsis of the state of the trinity in the first half is excellent, but the 'threatmaster' archtype is a bit flawed.

I have to agree with what some of the other commenters have stated, we should either dissolve the trinity or at least work to lessen its presence. I think this is done through the removal of the black and white "agro system" and the development of less predictable encounter AI, that forces players to think on their feet.

I don't believe that getting rid of classes altogether could ever be a good thing. Skill based games have never flourished among the casual to average player. Its too easy to wander down the character development path that you think is good, or is good but won't be good in 10 levels--or 10 days from now after a patch, and be screwed later on.

Inevitably within 3 months the very best skill templates will be tested and determined, listed on guide sites--all labeled with names like "Warrior," "Cleric," "Paladin," "Mezzer,"... sound familiar?

People like classes. People like to know whether they're a good player, of if they need to tighten their skills. They know this by fulfilling their role which is framed by their class.

It all comes down to efficiency in combat. Sure, I'd like to say that we should think outside the box and come up with new ways to advance besides killing monsters... but how many people are going to be lining up to "parley" the dragon to death?

This is why combat is the key. Get rid of the simplistic threat mechanic and make enemies behave more like a real player would. Introduce more mechanics like the collision system in AoC, where tanks can physically block enemies from getting to the squishies. The real "tanks" in ye olde battlefield combat, the front line footmen, put on armor and shields and stopped the enemy from advancing. They didn't do it by shouting insults at them.

Relmstein said...

WoW very old thread that is getting some google love so I thought I would leave another comment with some updates:

Basically VG was horribly executed but had some interesting classes designed around PvE.

WoW continues to have problems with classes being good at either PvP or PvE.

Age of Conan is trying some weird things with combat but still using the threat system so the trinity is probably alive in that game.

Warhammer redesigning the holy trinity for PvP combat. Tanks will actually be useful with taunt reducing damage done by an enemy unless they are fighting the tank. This mechanic should make tanks very handy at keeping casters and healers alive.

Sean said...

I've been toying with how to make the merchant class a viable option for mmorpg's. I'm a merchant at heart when I play these games spending more time in the AH then in instances. I would like to believe there are others like me who would be happy with a traveling merch class if it had the following abilities:
1. Questgiver: Can give quests to other players in the game. These would grow in number/coolness as the merch leveled. If I need to cross a field why not have some parties clear the way. This would be the primary ability to attract people like me.
2. Blacksmith: Can repair players items anywhere, maybe not in instances/raids. This puts a social spin on the class as you could make repairs much slower than normal hopefully leading to conversation while the items repair.
3. Trader: Can trade with other merchs (real or computer) and maybe special auction house abilities. The trick is to make this beneficial in other ways besides gold. ie not a farmer class see monetary system.
4. Weak/High Aggro Radius: Keep merchs close to towns unless they get brave.
5. Cart Puller: Bigger carts hold more stuff, but cost more to get em.
6. Price Control: To some degree I should be able to control my prices. Also some leeway for bartering would be cool.
7. Monetary System: I think instead of dealing in plain gold/silver the merchants would have their own money similar to the battleground fame system. This money can be used to bring items into the game for sale or to pay other merchants with. This could prevent farming as well.
8. Merchant Quests: Gotta have merchant dailies :)

I haven't thought about talents, but I have seen systems where as your merchant talent increases so does your profit on each deal. Talent Tree's might be "Black Market", "Traveling Salesman", and "Antiques".

I would like to know if anyone would actually play a class like this or what you think it would take for people to really get into it like they get into their dps's, tanks, and healers.

Sean