Archive for May, 2011


Level Design In 11 Points

May 5, 2011

It’s been a while since I’ve written something about game design. So, I decided to write an article describing how I proceeded when I made the levels of Ninja Senki.

  1. I played many MANY platformers and studied how their levels were built. Doing that kind of research is enlightening and fun ^_^.
  2. I designed a bunch of micro situations to test extreme possibilities and various combinations of situations in small rooms prior to designing the actual levels. Ex: Maximum height and distance that the player can jump. Then, I saved these combinations and I ordered them from the easiest to the hardest. In the end, I discarded the uninteresting ones. This gave me a grasp of what I could do with the game system.
  3. I designed the macro architecture of the levels for the whole game by defining the general shape and orientation of each level (vertical up/down, horizontal, stairs, open area, L shaped, Z shaped, etc.). The shape of a level should also match the visual theme as much as possible.
  4. I planned the sequencing of the ambiances of the game. I gauged dark and bright levels. The idea was to make the player travel. Example: I tried to avoid making a sequence like this: cave, dungeon, and then house interiors… But it could have been a good idea if I wanted the player to feel claustrophobic.
  5. I tried to use level design as a means to tell a story. I believe that thematic progression inside each level is crucial. So, I segmented the architecture of each level in 3-4 parts. For example, for the second rooftop level (Scene 14), I divided it in 3 parts: the first part is some kind of gateway guarded by riflemen, the second part is a tower, where you have to climb up, and the last part is balcony where you have gaps to cross and jumping ninjas to defeat. If this progression is planned in the visuals of the backgrounds as well, this can make a level very memorable and engaging.
  6. I chose a limited amount of ingredients for each level (my test rooms mentioned at Point 2 were useful to help me choose what ingredients I wanted). I didn’t want to spoil all of the ingredients in the first level. That would have made the first level overwhelming and the rest of the game irrelevant. I chose only 3-4 enemies and obstacles for each level and then I created a progression by introducing them alone first, and then by combining them in more and more complex manners, thus creating a nice difficulty/learning curve.
  7. I used level design situations to teach the systems of the game. For example, I created some situations where the player had low chance of dying, but had to use a specific ability to progress. This can save a lot of explaining in texts and tutorials.
  8. I used enemies, walls or pickup items placement to hint how the player should solve a situation. For example, I put some pickup items to hint the position where the player should jump to successfully cross a gap. Or I placed a step, a solid object or a groove in the ground to hint where the player should stand to be safe from an attack.
  9. I restrained the height or width of some sub-levels to the height or width of the camera in order to allow the player to see further. Because, when the camera is centred on the player, the maximum distance that he can see is only half the size of the screen. Locking the camera to the level can give more possibilities for platform and enemy placement.
  10. I tried to let the player breathes sometimes. The difficulty shouldn’t be constantly climbing up. From time to time, I intentionally put some easier situations during a level to relieve the tension a bit.
  11. I kept levels short or checkpoints not too far from one another, especially if the difficulty was high. I like to keep playtime below 5 minutes between each checkpoint. It helps keep away the frustration.