Jim Sterling over at Destructoid wrote a post the other day claiming that linear games provide pacing and structure that nonlinear “sandbox” games do not:

Indeed, if every game was a huge open world, you would soon find yourself growing bored, or at least overwhelmed as you struggle to find time in the day to explore sandbox after sandbox. After hours spent in the hustle and bustle of Liberty City or Tamriel, a game with clearer focus and a set beginning, middle and end can be just what the doctor ordered, providing some experiences that total freedom just can’t manage.

Sterling’s got a point: open games that allow for actual player agency over the path of the plot do tend to have inferior pacing and emotional impact when compared to games with a linear plot. However, Sterling falls into a common trap when it comes to game design: just because they tend to be inferior doesn’t mean they can’t manage to provide that pacing. Game designers just don’t pay enough attention to it. More after the jump.

When I write a short story, or come up with the plot for a linear game, pacing is involved in the process as a matter of course. I choose, usually consciously, exactly when a given event will occur. Pacing is always in the front of my mind, because I can see how far along the scrollbar is in the window where I’m writing up my plot outline. When I split my game up into levels, I know that Ravenholm, for example, is Level 6, so I can foreshadow it in Level 5.

However, a potential plot of nonlinear game is at best a partial ordering of events, and so pacing seems to be impossible to implement. After all, if I don’t know what order in which events will occur, how can I make sure they have emotional impact? It’s not as easy to do as with a linear plot, but it’s not at all impossible.

When I construct the outline for a nonlinear work, I am placing implicit constraints on the plot. Let’s use Morrowind as an example. By placing the god Vivec within a solidly locked palace, I can reasonably expect that the player will not encounter him until she is given the key. This builds suspense throughout the game, as the player slowly understands the story and nature of the gods of the Tribunal.

“But,” one might argue, “isn’t that just a linear plot in disguise? If the player must have the key, and the key can only be attained near the end of the main quest, then doesn’t the emotional impact stem from the linearity of that part of the plot?” Good point. But in Morrowind, the Palace of Vivec can be opened without the key… with enough effort. At any time during the game, with sufficient skills and/or spells, the player can unlock Vivec’s abode and visit him. This requires a significant amount of effort, though… which means that the player has both recognized the palace as an important location to visit, and has worked hard to gain access. Thus, the emotional impact is maintained despite the challenge being essentially nonlinear.

The plot in this case is constrained by the rules of the game world while not being forced into a linear path. By keeping in mind the effects of my design decisions on the player’s potential plot paths, I can encourage experiences that are just as good as those in a linear game. Myst‘s plot consists of six segments, the middle four of which can be traversed in any order. However, the player’s experience is still strong, with the gradual realization that Sirrus and Achenar are not hapless victims, but instigators, and the final reveal where the player finds out what actually happened to Atrus.

It’s not the case that non-linear games lack focus and have fuzzy pacing. It’s just poorly focused and paced non-linear games that do that. Linear games can be quite good, but they typically don’t take advantage of the unique quality of interactive entertainment: its interactivity. I can give the player agency over the game’s plot without removing my authorial control over her experiences. It just requires me to keep in mind that pacing is just as important for non-linear games as it is for linear ones.