Choosing a Game Engine for a Project

So, you have a new project, lots of shiny new game engines abound, but which one to choose?

Forget the marketing, forget the bells and whistles, forget the hype. Instead get down to the basics.

1. What do your people know? Learning a new engine is time consuming, and without experts there will be numerous "gotchas" that can crop up in even the most tried and tested engines (someone will say "you need to do it the ____ way"). If you have to hire new people, make sure there is a market for those and the budget to get some expert skill level. This can be the biggest hurdle, and often the reason that you may not be able to go with the newest engine. Having a good community that supports the engine is also important.

2. Does the engine do everything you need? Does it really? Really, really? Doing your homework on what the engine can and can't do comes back to the first point. If you don't know what the engine is capable of, you should reach out to as many people as possible to get real experience with that engine.

3. Look at the history of the engine. What kind of games have been created with the engine? What kind of timeline, investment and skill set did it require to accomplish this? A game engine is not going to do everything for you, and knowing what developers, level designers, pipeline and server developers need to do to interact with the engine will let you know approximately what the engine can do.

4. How "unique" is your game? This is often a deal breaker for which engine to choose. If you are trying something that has never been done before, or has a high level of customization of different parts, you will likely hit a wall (in some engines sooner than others) where completing that task the "right" way can not be accomplished because of restrictions on the APIs the game engine provides. If you need to do things like access internal buffers, memory or anything else to accomplish your task, make sure your engine supports this flexibility. Unfortunately these kinds of requirements happen too late, but being aware of these kinds of issues will help you choose a flexible engine.

5. Where (and on what) is your game going to run? If the engine does not support multiple platforms this can be a significant amount of effort to port over to other operating systems. Some things can port easier than others, but each platform that needs to be supported will complicate the code base.

6. Do you have the source code? It is great to have the source in order to control your own destiny (although be sure to read the license agreement), but some game engine source code is not meant for public consumption. Often times the internal workings of the game engine are very difficult to modify or extend, and touching one line of source code will make merging future updates to the engine from the distributor one more complex task to manage. Open source game engines are pretty good, but they are still a bit behind the leading engines.

There is no "perfect" engine, all have strengths and weaknesses, all can make great or bad games. Be careful with this decision though, as it can be one of the biggest decisions for the path and people that work on your game.

Good luck and do your homework,
Michael Hubbard
http://michaelhubbard.ca