Thursday, September 13, 2007
End game options
No part of a MMORPG is discussed as much as the end game. In most games the end game is significantly different than the leveling game leading up to it. And this change of direction doesn't go down well with everybody. Furthermore there are different approaches in different games to how the end game should look. So I thought I'd compile a list of the different options, discussing their advantages and disadvantages.
The first option for an end game is not to have one. Or as Mind Bending Puzzles proposes: a game over screen with a cut scene showing you as having vanquished the game. While there aren't any MMORPGs I know that do that (only ATITD has an end at all), a game over screen is what you'd expect in any single player RPG. Games like Legend of Zelda or Final Fantasy don't have an end game. The obvious disadvantage, and reason why this doesn't exist for MMORPGs, is that people tend to stop playing after reaching the game over screen. Game companies simply don't want to lose the monthly fees. The key to overcome that disadvantage is replayability. Games like World of Warcraft or the upcoming Warhammer Online have great replayability, because starting as a different race, preferably with a different class, gives you a completely new experience, with new quests and zones to explore. A MMORPG with a game over screen could also use a feature many console games have, unlocking new character classes every time you reach the game over screen. Blizzard's decision to have hero classes start already at high level is a missed opportunity; added low level content for new Deathknights would have been better.
Another way to avoid an end game alltogether would be to make a game with infinite levels. That isn't as crazy as it sounds, as many parameters in the game are already scalable. The only problem is that you obviously can't have an infinite number of developers creating new content, thus the content would have to be randomly generated. But a game which plays in randomly generated instanced dungeons, recycling the same dungeon tiles and monsters over and over could have an infinite number of levels. The reason why this isn't done brings us to the one big advantage of having an end game: When everybody is at the same level, it is easier for people to play together. Sure, there are still differences in equipment and player skill. But it is much easier to play together for a group of level 70 characters than for a group with a big level difference between the highest and the lowest player. So now we are at a point where we want leveling to stop, and while we still want to hand out rewards, we don't want these rewards to cause huge differences in power between the players, because that would be as bad as them having different levels. So, what can we do?
The model proposed by games like Everquest or World of Warcraft is to fill the end game with raids and other activities which take an enourmous effort for a small chance to get a small improvement to your character. If you drew a curve of power increase versus time the raid end game is where the curve goes asymptotic, approaching a ceiling slower and slower, never really arriving there. But speed isn't the only difference between the raiding game and the leveling game. The other very important difference is the number of people you play with. Leveling is done very often solo, or in small groups. Raid groups are much larger. This is both an advantage and a disadvantage. Advantage because you get to play with a large group, which can be a lot of fun. Disadvantage because organizing a large group to all turn up at the same place at the same time and stay together for several hours is difficult. If for Real Life® reasons you can't play at the same time as everybody else, or you can't play for several hours in one session, you are effectively excluded from the raiding end game. While there is a bit of soloable PvE end game, in the form for example of reputation grinds, they are not as much fun as raiding.
The alternative to a PvE end game is a PvP end game. Experience tells us that this is bloody hard to get right. One of the earliest lessons in MMORPG history is that free-for-all PvP isn't working very well. PvP by definition has a winner and a loser. If the fight was "fair", both winner and loser might come out of it with some sort of satisfaction. But if the fight was very unfair, like 3 guys ambushing one guy, and then camping his corpse, the loser is going to be very, very unhappy. And the satisfaction the winners get out of ganking somebody is also limited. Thus the overall sum of happiness that the fight produced is negative. Nobody wants to pay $15 a month to get beaten up virtually, so the more ganking goes on, the faster the game bleeds subscribers. More successful PvP systems have various restrictions, for example consentual PvP, in which you either need to flag yourself or enter a special PvP zone to indicate your consent to participate in PvP. Even more restrictive are PvP features like battlegrounds, where the numbers and levels of the two sides are more or less evened out. But as battlegrounds constantly reset to the initial state, people often get bored and demand a PvP with more consequences. The current best idea around is realm versus realm PvP, of Dark Age of Camelot fame, and about to be taken to the next level in Warhammer Online. But making a good PvP end game is still hard, and we don't know yet whether WAR will succeed in that. One problem with realm vs. realm PvP is that just like raiding it isn't a soloable activity. Even worse than PvE raiding, the best time for a PvP raid is often in the small hours of the morning, where you'd expect little resistance. Which is great for the small number of people who can afford to raid from 3 am to 6 am, but not really a viable way to play a game for people with a normal life.
I haven't really seen games with good soloable PvP. I would imagine that it would be possible to make lets say a Gladiator PvP game (better than Gladiatus) in which you duel other players in an arena in a large variety of different combat types. The Romans invented a lot of those, with different weapons, with and without savage animals, and with some other rules variations. Should be possible to reproduce that in an online game. But even duels are hard to get right in a MMORPG, where classes are often balanced to perform different roles in a PvE group. Thus the better you make your PvP, the more you need to diminish your MMORPG part, until you finally arrive at some variation of Counterstrike and no MMORPG at all.
The biggest problem with PvP games is that very often the developers think that if they have end game PvP, they don't need anything else. Unfortunately that doesn't work. While human opponents are more intelligent than computer opponents, they are limited by the restrictions of their characters and the PvP system. If you fight two characters of the same class one after the other, the combats won't be all that different from each other. The idea that PvP never grows old because it would be a form of player-created content is wrong. Usually players try the PvP system as intended for a while, get bored quickly, and then either cancel their accounts, or start harassing other players out of sheer boredom, regardless of the consequences.
A totally different form of end game comes from the surprising observation what the players of Ultima Online did once they were offered a chance to escape PvP. It turned out that those players who stayed very long in the game often spent large amounts of time with fluff, with features that didn't increase the power of their characters at all. The biggest fluff thing was decorating your house with ultra-rare items, which were purely decorative. The end game becomes the point where you stop playing a game, and you start living in a virtual world. Not easy to pull of either, but features like player housing obviously help a lot here.
Neither end game option is able to retain players in a MMORPG for long. So ideally an end game has many different elements, PvP and PvE, group and solo, adventuring and economy and fluff. World of Warcraft is nearly there, although their PvP still is far from perfect, and WoW is weak on fluff. A dream game would have you live in a virtual world, having a house and possessions there, being active in defending those from PvE and PvP threats, while working to increase your possessions in various ways from crafting to treasure hunting. How to realize all that in an actual game nobody knows. But I think the genre will be moving in that direction, because games offering a lot of different content will over time fare better than games concentrating on single features.