We made dozens of layouts and small variations of certain areas, especially at the beginning of the game; at the same time we felt that we could not over-simplify these situations. We were always trying to keep something of the elements that we believed puzzles should have.

But it didn't work. We keep simplifying and, suddenly, the most "over-simplified" situations turned out to be the most satisfying for the players. I could verify this firsthand when I showed the game at PAX East.

A Practical Example

In this section I'm going to explain how and why we modified the layout of one of the sections in the first part of the game.

If you have played the game, you will probably remember this area:

This diagram represents it:

This part of the game can only be solved in this way:

• The player can only access the exit by jumping over the fence (4) from (3).
• At (3) there is a shock hazard that kills the player.
• The player must interact with the contextual action (1) that deactivates the shock hazard of (3).
• The contextual action is "guarded" by two zombies. At this point of the game the player is defenseless and must not come in contact with the zombies.
• The only option for the player is to taunt the zombies from (2). Zombies are attracted to the player and get killed by the hazard below. The way is then free for the player to interact with (1), get to (3) and jump over (4).
• If the player doesn't taunt the zombies from (2) and jumps over the hazard, the zombies start moving towards him with no possible exit.
• On the top of that, as the player gets to (2), a huge "Tutorial Hint" appears on screen, inviting the player to taunt the zombies.

As you read it, it might sound a bit complex, but as there is only one solution and any other attempt always kills the player, which is the clearest feedback. The players are actually completely guided toward the correct solution.

"Guided," yes, but they feel that they have been smart enough to cheat two zombies, unblock his path and find the exit.

However, this area used to look like this:

And this is the diagram of this layout:

The idea behind is exactly the same as the other version: the player must attract the zombies to (2).

While the zombies are moving to (2), the player must go over the big generator (5), interact with the contextual action (1) to deactivate the hazard at (3), climb to (3) and, finally, jump over the fence (4) to access the exit.

It seems like it would make players feel smarter than the previous one, right?

This is what we thought, as designers:

• The player can navigate the area and it is not restricted to a small safe zone: more freedom.
• There's no need to use the "taunt" action. The player might walk near the zombies and then run away: more choices.
• The player has to perform an action while the zombies are still "active": more challenge.

So more freedom, choices, and challenge -- that sounds great.

However, this is what's going on in the player's head:

• There is not a clear place to start.
• The generator (5) is something big, with a complex shape that seems to have a hidden purpose. In addition, it's not clear that the player and the zombies can walk behind the obstacle.
• There is nothing that tells the player if what he is doing is right, wrong, or pointless.

No clear feedback and multiple solutions: Noise. The worst enemy of the designer.

This quick example shows how a simpler layout, even though it might look less interesting is, in the end, much more satisfying for the players.