Jonathan Steinhauer's MMO ColumnSteinhauer's Opinion: Breaking the Level Barrier, Part 2Jonathan Steinhauer's MMO Column - RSS 2.0
Editor's Note: This is the second part of this article. For part one, click here.
Before we dig any further into debunking levels, I have a confession to make. I do, paradoxically, like the "cosmetic" side of levels and in the first half of my previous article was playing the devil's advocate. They serve a useful purpose as a measuring stick. Games are by their very nature competitive, so it's important to have bragging rights. "I hit level 20!" or "I defeated that level 40 ogre with my level 35 archer" and so on. Also, it's helpful (though not entirely realistic) to use them to assess odds against a potential foe. "Can my level 15 warrior survive against that level 18 goblin?" Admittedly, this is a departure from real-life imitation, but a large piece of game design is the balance between playability, realism, and complexity. This aspect of levels, however, is where their usefulness ends.
The key problem is that character growth currently is defined by levels. Instead, levels should be defined by character growth.
The first step in finding the solution is to look closer at learning. As I said last time, real-life growth is more often on a learning curve or slope, rather than in the stair-step method common in games. Take running, for example. Anyone (without a physical disability) can run, the only question is how fast. Limitations are based on fitness, training, genetics, physics, and so on. A person who exercises will gradually get faster, but they won't suddenly jump from running a six minute mile to running it in four. Another example would be vocabulary. A person gradually learns words over time either through studying them or encountering them in life. A man's vocabulary won't jump from 5,000 to 10,000 words in an instant.
That's not to say that step learning doesn't exist. But it is generally found in activities were you either can or can't do a thing. For example, in the quest to fly, the Wright Brothers didn't gradually fly. The first time occurred in a single instant and once that was achieved, they could repeat the feat (although I would argue that the gaining of knowledge that enabled flight was still slope-learning). Some growth in life tends to be a blend, such as studying mathematics. The gaining of knowledge and skill in math occurs on a learning curve, but there are "steps" along the way, such as basic addition and subtraction, Trigonometry, and Calculus.
In games, the same sloping growth should be used. This can be achieved without the complete removal of levels by simply removing their defining influence. In other words, the gaining of a new level shouldn't unlock new skills, cause an instantaneous leap in health, or any other great jump in competency. In most games, progression is measured in a similar fashion: through the gaining of experience. At certain points, say for example 1,000, 2,000, 4,000 and 8,000 experience, a level transition occurs. And that is fine. This is the "cosmetic" side of levels: "I've reached 4,000 experience, therefore I am Level 3." The problem is, when levels are used to define the character: "I've reached Level 3, therefore I get a 100 point boost in health." The solution is to stagger that health growth across the period of time it takes the character to move from 2,000 to 4,000 total experience. Whether that means a flat slope where a health point is earned every 20 experience or whether it should be a curve with faster growth near the beginning of the level and slower growth at the end (or vice versa) is less significant and should be determined by the nature of the game and the nature of the stat in question.