The Problem with Multiplayer? Those Damned Other Players!

I've been playing and studying massively multiplayer online role playing games for a long time. I find them fascinating, especially from the perspective of a game designed to be played by so many thousands of players simultaneously. I find them fascinating from a business perspective, especially in the microtransaction sense. But I also find them fascinating from a social perspective: all those people playing together, interacting, competing, grouping... and potentially building alternate personalities and histories.

Part of me still clings to the romance of the MMORPG, the kind of tale told in .hack or 1/2 prince or Sword Art Online. No, not the whole "die in game, die in real life" nonsense, but the sense of such a deep immersion in the game fantasy that characters can establish reputations beyond statistical game mechanics. The online worlds in these tales, where powerful high level characters are few and far between couldn't be more different from real MMOs like World of Warcraft, inhabited almost exclusively by max-level characters decked out in various tiers of epic gear.

I sometimes wonder what it would take to make that romance a reality. Would the game have to be designed to be slower, with more emphasis on the "game" rather than the "end-game?" That romantic feeling was a lot more alive in Asheron's Call, where players could engage in "end-game" style activities long before they reached the technical level cap, and when players did reach the technical level cap, it was raised. Unlike World of Warcraft, where "leveling" is just some annoying tutorial before the real game starts, Asheron's Call allowed and even embraced level variance.

Part of this came down to the core system of how the game treats levels. In WoW, Rift, and EQ, and even good old pencil and paper Dungeons and Dragons, the difference in level between character and monster plays a mathematical role in determining the success of an action between the two. In D&D 4th edition, for example, a player's Armor Class and To Hit bonuses include, among other factors, 1/2 their level. All other things being equal, a level 1 character has a -10 penalty to hit a level 20 character. Likewise, in WoW and Rift there are signifnicant penalties to hit monsters more than 2 levels above a character's level. It's not just that a level 1 character doesn't have enough damage or ability to kill a level 20 monster, they can't even overcome the game's level-based mathematical imbalance and hit it. No matter how good your skills are or gear is, that 3 level difference equates to some kind of 40% higher miss rate and a 5 level difference is probably insurmountable.

In contrast, level in Asheron's Call was just a measure of accumulated XP. That's it. A level 100 character had more xp than a level 1 character, but that 100 or 1 never themselves enter into equations. Thus, in some sense, it was possible (no matter how implausible) for a level 1 character to defeat a level 100 character, provided the level 100 had a low enough defence skill (which, again, is not strictly influenced by level). Thus, while level was a measure of relative power, it was not part of the mathematics. So while level 1 characters killing level 100 characters wasn't plausible, level 40 characters fighting level 60 characters was. It was tough, but there wasn't some arbitrary "only fight things x levels higher than you" clause putting a hard cap on things.

Another thing games like WoW and Rift did was introduce hard level requirements on content. You HAVE to be level 50 to participate in end-game content in Rift. Period. The game will not let you enter the dungeon if you are not 50. In Asheron's Call, there weren't hard level limits. So if you were 45 and your friend was 55, you could go into the level 50 dungeon together.

When you look at the romantic treatment of MMORPGs, statistics are "soft" rather than "hard". A dungeon is "too challenging for a character of your level, but maybe with a powerful friend you can overcome it" rather than "it won't even let you zone in to the instance, so go grind more quests." Clashes between characters of very different power levels were on the scope of "ugh, he's just too strong! But if I give it my all...!" rather than "well there's no point really, 10 levels of difference equates to a -300% chance to hit." It's hard to convey epic excitement through the cold, hard math that dominates modern MMO design.

So I think "soft" statistic MMORPG design would go a long way to restoring that romantic treatment of online gaming, a return to the sense of "virtual worlds" which give rise to their own set of reputations, histories, tales, and destinies. You know, rather than the "grind - raid - quit" formula WoW made so popular.

...

There's another mechanistic issue in MMORPGs, and that's grouping up. First of all there's finding similarly-minded individuals to group up with in the first place. Then there's finding a mutually agreeable activity. In games with "hard" level restrictions, that means that you can only group up with people one or two levels (or sometimes no levels) different than you. And in games with heavy dependence on quest structure, that means grouping up with someone on the same step of the same quest as you. These sort of requirements can make grouping a chore that most players would rather avoid.

In an effort to overcome "hard" requirements, a lot of games are introducing more and more automatic grouping systems. Systems that auto-group players of the same level together, send them to the same content together, etc. But the system works to the point where players no longer have to interact together to complete the content, and in many cases the players are taken from separate servers in order to reduce wait times so they can't actually see each other again after the content is done.

The issues of grouping made some games, like Dungeons and Dragons Online, basically unplayable for me. Trying to find a group of (A) like-minded (B) same-level players (C) of the right class distribution (D) interested in doing the same dungeon was basically impossible. Likewise, while auto-grouping in Rift for a T2 or Instant Adventure is as easy as clicking "go", finding a group for more challenging raids is harder than actually completing the raid encounters.

The thing is, grouping to do those raid encounters isn't at all embracing that MMORPG romance I described above. It's not engaging in the fantasy, it's just playing through content that you happen to need other players for. I'm actually not sure MMORPG raids would be any less fun with 19 NPCs endowed with really good AI than with the 19 players you have to put up with.

...

And that's the key problem: other players. It doesn't matter how well designed the game is if other players just ruin it. The concept of a Horror MMORPG was discussed in a recent blog post at MMORPG.com and the conclusion was simple: You Jerks Will Ruin It.

Similarly, a reply to my feature article series addresses the need for mechanistic systems in games to help players deal with jerks. League of Legends introduced the Tribunal system to deal with the throngs of jerks, and a similar game, Heroes of Newerth, is so well known for its jerks that "heroes of newerth jerks" turns up over a million hits on Google. And EVE Online is veritably built around jerks screwing each other over.

All of these seem largely based on the "internet anonymity" issue. There may be some important truth to that. Consider that, in most romantic treatments of MMORPGs, players only ever get one character. That character has just as much identity in the game as they do in real life.

But there are good, valid reasons for making multiple characters. I'd hate to have played WoW or Rift or EQ without alts, because I'd hate to have missed out on the joys of playing all the different classes. Maybe that's what makes multi-class games like Final Fantasy XI and XIV so appealing: they get one step closer to that romantic treatment by making the single character identity way more palatable.

It feels like there's a big struggle in MMORPGs these days: a sort of three-way tug of war between helping players to get involved with others, helping players not feel abused by jerks, and helping players play the game and complete content designed for groups. In most big budget games, the latter two take precedence, so much so that I wouldn't be surprised to see MMORPGs add in so much AI companionship that you never need to interact with real people again.

Still, there are some games that are going the other route, like Lucent Heart is trying to do with its Cupid system. The idea is that friends provide something that AIs will never be able to: that sense of companionship that is more valuable than simply a filled PC slot in the raid roster. But does it work? I created a character in Lucent Hearts to test it out. I registered in the Cupid System three times. All three times, I was dumped by my match before I even had a chance to log in and find out who it was. Jerks.

I'm not even sure I wanted to get matched. The idea of partnering up with someone and french kissing them before every fight to get a combat bonus kind of creeps me out. But aren't I, now, just propagating the same problem I complained about a few paragraphs above? Ruining the fantasy of the game by trying to "beat" the game systems? I can't help it! I instinctively read every tooltip that I can get to pop up!

Maybe part of that romantic treatment of MMORPGs requires a certain level of ignorance about game systems. But that fantasy seems so fragile. When I interviewed Turbine's Cardell Kerr, who worked on Asheron's Call, we talked about how AC had so many "secrets" that got "ruined" by players. Like how the secrecy for spell research was distilled to an app that would tell you what your spells were by gaming the spell research formulas, or mapping tools that would calculate routes through the world by cataloguing every secret portal and passage.

The reality is that your player base is going to push that information out to other people and the sad fact is that you only end up with a game with multiple tiers of knowledge. When a person just picks up your game and has no idea what's going on, whereas if they downloaded these one or two other apps they would have a good time.

So maybe even well-meaning people who try to figure out how the game works and share that with everyone else are, in a sense, "jerks" who end up ruining the game for everyone. Or maybe the problem is just that the romance of MMORPGs is inherently flawed: it's based on an idea that core game systems cannot be known or gamed. In fact, every romantic treatment of MMORPGs I can think of includes elements of the games that are unknown to even the game developers themselves.

Still, it seems plausible that the certainty of knowledge of game systems could be suspended as part of the fantasy: a sort of communal agreement not to ruin the game for oneself or others. Some of my best experiences in Asheron's Call lay in that nebulous haze between ignorance and certainty. Take away the tooltips, then get everyone to agree to play nice, right? Still, it only takes one jerk to ruin it for everyone else.

So in a world of jerks, what are multiplayer games to do? Make players interact less? Insulate player interaction so the game compensates for jerk behaviour? Implement systems that punish jerks? Find a way to make players not be jerks?

What are gamers even looking to get out of massively multiplayer experiences, anyways?