MITx Introduction to Game Design

Playtest the core ideas/concepts of a game early and in basic form. If possible, playtest on paper (rather than computer) as it is much faster to create and get feedback.

Don't become  too invested in ideas - be prepared to switch or even abandon them. Early play testing helps here as not much time will have been spent developing the idea.

"I wouldn't recommend doing more than maybe like two iterations of a paper prototype. The paper prototype is just to test out the flavour of an idea." Move on pretty quickly, don't spend more than a week or two. Discard everything if necessary.

Try to playtest with other game developers. Don't ask "is it fun" to playtesters, instead ask "how did you feel when playing?"

Don't always follow a tester's suggestions. Instead try to determine (by watching them play) why they suggest things. Address the underlying problem. For example, if they say the game needs to be more blue, is it perhaps because the contrast is problematic. If they do an obviously incorrect action, is it because the game is hard to understand?

Is there always something interesting for the player to do?

If a game is designed for yourself then it will be too hard for other players.

Playtest, playtest, playtest!

Start with the core idea, test for fun, remove superfluous ideas. Iterate on the concept, refine until everything (concept, pitch, graphics, audio) is beautiful.

Try to test every week or two, so the game is always in a playable state.  

Pay attention to controls and people having problems in testing - real players will be less forgiving!

"The point, to my mind, of a prototype, is you really want to spend a small amount of time now to learn the things you're going to need to learn, to avoid spending a large amount of time and someone else's money later. You don't want to be including anything in your prototype that is not necessary to answering your current round of questions."

Know what success and failure looks like at every stage. What does success/failure in a playtest session look like? What does success/failure for the game's release look like?

The UI is the game explaining itself. Good UI makes it clear what can and can't be done. If the game is played badly, the UI should ensure players understand their mistakes.

Usability has to be developed iteratively, start at the beginning and polish all the way through development. Usable UIs are learnable, efficient, memorable, discoverable, pleasurable and handle errors gracefully.

Optimise after the fact - don't over engineer the game. Bounce around the game focus on what will make the biggest impact at the time.

People are most influenced by the last and first moments of an experience. Carefully design the beginning - make it entertaining. Games have little time to impress players. In the game's first moments sell: the narrative (create mystery, selling characters is hard in a short time period, selling the world is easier); the mechanics - show why the game will deliver a good player experience; and/or, spectacle.

Try to find unstated/unrecognized assumptions in the game. Developers are so close to game that there can be large differences between them and players. Test to find the differences. Sometimes there is too much frustration instead of challenge, or more confusion than mystery.

Super success stories in game development are the exceptions. There are not many and are they are unpredictable. Most indie gamedevs do it tough and do it because they love it, not to make money.

The barrier to game development is low with good tools and digital distribution. So there are many competitors. Need to outwit and outlast to rise above the mass of competitors. Use business/marketing/press abilities. Create novel and unique games.

Advice to someone who wants to be in game industry - start making games (not play games, everyone does that) and meet people who are making games (network!). Start with little games to learn what you don't know you don't know. Expect failure, so finish quickly and move on.