6.6. Failing Faster¶
In general, the more times you iterate, the better your final game will be. You will have learned from your failures with time to correct them and avoided sinking as much time into things you never got to finish or decided were not fun. As the extra credits guys say, your goal should be to Fail Faster:
Warning
This is literally the most useful piece of advice you will get in this course. Not just about game design and development, but about any activity that involves creative problem solving.
Iteration can often seem hard. Every stage of creating a video game takes work:
You can’t really reduce the time it takes to design the rules of the game, without compromising your goals. You can’t rush creativity.
But... You can avoid the distraction of trying to come up with the perfect idea instead of starting to work on a good idea. And you can focus your design work on what is immediately needed in your game. Don’t start by designing 100 different weapons for your action RPG - design one weapon and get it working. Then you can loop back and design more using the lessons you learned from implementing the first.
Before you can even build game play you need months of work to build your new game engine.
But... Who says you need to start with your own engine? Even if you are planning on implementing your own engine (or heavily modifying one), it is common practice to build a prototype using existing tech: Gamemaker, Unity, or an older version of your own engine. Test your game play ideas in an existing engine if possible before committing to developing an engine.
Designing or finding art will take time.
But... You can test mechanics with placeholder art that takes 30 seconds to make. It is easier to make a fun game prettier than a pretty game more fun.
Play testing a game takes lots of time - there is no way to safely remove that.
But... You can be efficient about testing. Make cheat codes, rearrange rooms, do whatever it takes so that you can test the current feature fast. If you are working on a boss, you better not be playing the game for 5 minutes to get to him every time you make an adjustment.
Evaluation doesn’t take very long; you’re making a simple yes/no decision of whether the game is “done” or “good enough” based on playtest results. There is little to be gained by rushing through this further.
In short, build it as fast as possible. Make it as ugly and cheap as you can get away with. Cut corners where they don’t matter - if random enemy movement is almost as interesting as the AI system you have in mind go with the random movement until you are convinced there is nothing more important to spend time on than the AI.