Game Off 2018

18 Dec 2018

It’s been a long time since the last post. I’ve been translating Keith Burgun’s “Clockwork Game Design” after my submission for NYU’s game design MFA 2019; first chapter is almost finished, and there are 4 chapters in total. Actually the little game I created for Game Off 2018 learned from this little book.

It is about defining “elegance” in strategy game design.

Elegance here is efficiency. Do as much as possible with keeping as simple as possible. For strategy games we may use efficient mechanics to achieve efficiency. That is, make tough decisions.

I have considered 3 things in the game playthrough:

  1. There are abilities in this game. The abilities you gain also enable enemies extra behaviours.
  2. You must clear out the room to reveal more levels. Awards should be granted doing this. Therefore tech points are collected eliminating an enemy, in return it unlocks new abilities.
  3. You are upgraded at a certain point - every enemy you kill gives a tech point and you upgrade after collecting 10. But you don’t die, instead, you lose an ability (so are the enemies) and revive, unless you’ve lost all.

The first rule gives a tradeoff. The complete mechanic I wanted to implement is creating multiple abilities, and randomly show up for an upgrade. So there is the actual tradeoff - players now have to make decisions. Yes, you may shuffle the abilities as well, hoping something better showing up, but must with a price.

The combination of abilities must be carefully designed. If a player is able to use better strategy, just like keeping a handful of good cards, a more efficient combination of abilities must be granted. And so, what drives a player to pursue the best combination of abilities, if the abilities themselves are useless to beat the game?

Spore successfully completed all these goals in its design.

The “don’t die” mechanic adjusts game difficulty. When a player is not upgraded, the game will not feel intensive. A master of action games will encounter more dangerous enemies while rookies will not. This idea comes from creating flow.

As to implement these core mechanics, I also had to consider how to make enemy AIs not stupid. I first tried pathfinding, but later realised it’s not a necessity. The actual result I want is to make the enemies more powerful when they “cooperate”. Think about pacman. Think about cellular automatons. They don’t need a central coodinator to control each’s behaviour but they still produce more complex coordinated behaviours when they act together. Most time I spent on this game was testing different enemy behaviours.

I finalized with the following:

  1. Vision. When out of range, enemies will not chase the player. They breed, build walls, wanders and stay together instead.
  2. Break and build walls. Walls should be interactable. Everything in this game should be included in the main feedback cycle. The main cycle is: eliminate enemies - reveal new message - reach the end of story. So I added these new enemy abilities.
  3. Breed when there are one or more enemies staying together. And the player may distract them from breeding. Breed rate is linear, instead of logarithm.

Moreover before the end. If you are interested in Keith Burgun’s book, also make sure to have a look at his “3 minutes game design” on youtube. That includes all essential ideas for his book.

Published on 18 Dec 2018