Friday, May 28, 2010
Perfect MMORPG: Challenge
What we think that we are doing and what we actually do when playing MMORPGs are two very different things. Imagine I had hacked into your computer, turned on your webcam, and filmed you during your last long play session. If I asked you what you have done during that session, you would tell me about how you have slain the princess, and rescued the dragon, or the other way around. But the video would show you more or less motionless sitting in a chair in front of your computer, with one hand on the keyboard and the other on the mouse performing constant but tiny movements. All the dragon-slaying is only happening in your head, with at best some facial expression on video to show it.
But because what we think we do is so different from what we actually do, we also tend to discuss subjects like challenge in the terms of the imaginary world. The discussion of challenge in games like World of Warcraft has been endless, but always revolving about things like boss abilities or the famous “moving out of the fire”. If you formulate it like that, you end up with simplistic but wrong conclusions: Why would anyone be so stupid to not move out of the fire he is standing in? But in reality the player is *not* standing in any fire, he is still sitting in his chair, hands on keyboard and mouse, eyes on the screen. If he doesn’t move out of the fire, it is most probably because he didn’t notice he was standing in the fire, because his eyes were glued to some other part of the user interface, e.g. the raid’s health bars for a healer. Which is why we have addons that play a warning sound if we stand in the fire.
So to discuss challenge for my perfect MMORPG, I will zoom out of the virtual world and into the real world, and discuss the subject in real world terms. What challenge *can* a MMORPG possibly have, given that you don’t even leave your chair while playing? Obviously the physical challenge is minor, beyond not getting carpal tunnel syndrome, and timing your bio breaks right.
Challenge in a MMORPG is limited to pressing the right key (or making the right mouse movement) in time in reaction to some visual or audio input from the game. Success or failure depends on both speed and you ability to choose the right reaction to the given input.
So let’s first talk about reaction times: Studies have found mean reaction times to simple inputs in people of college-age to be 190 ms for visual input and 160 ms for audio input. Women react slower than men (is it sexist to quote a scientific result?). Another study with people around the age of 56 found average reaction times of 360 ms, reaction time goes up with age after the teenage years. The same study showed that reaction time doubles if people have to make a decision what button to press, and that there is a correlation of that reaction time with IQ, but only a small one. Reaction time when measured on the same sort of machine goes down with practice, to a degree that in the context of video games means that practice has a bigger influence on reaction times than intelligence.
From the science follows some conclusions about game design for our perfect MMORPG: Relying too much on reaction times is fraught with danger. One problem is that a given level of reaction time challenge is easier to overcome for a male teenager than for a middle-aged house wife. The other big problem is that the order of magnitude of human reaction times is similar to the order of magnitude of typical pings of online connections. The best possible connection you could have is about 30 ms, but a large percentage of players have to live with pings of around 200 ms, and some players are plagued with 500 ms of lag due to location (the famous “oceanic servers”) or a bad ISP. Thus if you design a MMORPG in which the challenge increases by requiring shorter and shorter reaction times, you will find that the people who killed the final boss are all male teenagers living close to the server and having a perfect internet connection. And they had to practice that fight a lot. Does that sound like a game you know? I would argue that while there is a market for fast reaction games for teenagers, there is a good argument for doing such games in single-player and LAN-multiplayer mode, where the population is more homogeneous, and lag doesn’t favor anyone.
Thus in my perfect MMORPG, the reaction time requirements would be generous enough throughout the game so that gender, age, or the quality of your internet connection don’t have a major influence on your chance of success. Which leaves us with the challenge of having to press the *right* button, or making the right mouse movement. “Aiming” like in a first-person shooter is out, because lag again gets in the way: One interesting observation of multi-boxing is that you’ll find that the relative positions of your characters on the two screens are not the same, due to predictive algorithms MMORPGs use to make lag less obvious.
So what is the “right” button to press, and how do players know which one it is? The classic method is to give the players a bunch of slightly different abilities, and let them figure out which ones to use. The problem with that is that classic combat systems are not very interactive, and the same ability button has the same result on many different monsters in many different situations. Thus which buttons is the right one to press is independent of the combat you are currently in, and can be calculated with some math. Thus you end up with a so-called “spell rotation”, which players get from some theorycrafting website, and which tells them which buttons to press in which order. The same problem prevents you from putting fixed puzzles in your game, some people will figure everything out and put the solution on the internet, where other players just look it up instead of thinking themselves.
The obvious solution for the perfect MMORPG is to design combat and other challenges in a way that the best solution isn’t known in advance. It is somewhat curious that some people are very much opposed to that idea, and think that “you can’t have randomness in a MMORPG, because then people would randomly win or lose”. That is nonsense. Just look at simple games like Tetris, where which block falls down next is completely random, but it is nevertheless your skill in reacting correctly and quickly to each block which determines your high score. Imagine how boring and bad a game Tetris would be if it was possible to beat the game by using a “122333 – 122333 – 122333 – etc.” keypress rotation. The fundamental reason why some people oppose having to react to unpredictable random events is the reaction time science quoted above: People are used to faster being better, and the science says that if you have to see what is happening, think about the right response, and then press a button you are much slower than if you don’t have to think, and can improve your reaction time with practicing the same button press sequence over and over.
Now a lot of people point to games like Farmville and proclaim that “players do not want to think”, which is based on some completely faulty logic. Other casual games also have millions of players, games like Solitaire or Bejeweled, and in all these games the challenge is to press the right button in a situation which is determined by random factors. People do like to think as long as the challenge is something that looks doable, and not solving differential equations in your head. Non-thinking Farmville “works” in the context of being a free game on the social network of Facebook, but it is extremely unlikely that anyone would actually want to pay $15 a month for a MMORPG based on the same principles.
Thus in my perfect MMORPG the main challenge would be having to see what is happening in the virtual world around you, e.g. what the monster you are fighting is doing, and having to react to that in a reasonable amount of time by pressing the button for the optimal response. The outcome should not be a simple yes/no one, where any fault means you don’t kill the enemy mob, but there should be a noticeable difference in efficiency between choosing the right response and random button mashing, even in solo play at low levels. The degree of efficiency needed to overcome a challenge can then slowly go up with level. There should *not* be a challenge-free leveling game up to the level cap, followed by an endgame with completely different and much harder requirements. Instead the leveling game should actually train players in the skills they would need for the endgame. Killing a giant at high levels *should* be harder than killing those boars at level 1. And if you can play your character well enough to do solo quests at the level cap, going from there to raiding should only be a small step up in challenge.