On Saturday, I attended Gamecamp, an unconference bringing together 225 people who set their own agenda on the day for burning topics they’d like to discuss. One of the topics that came up was how to improve your game, with five tips from Graham McAllister who comes from games usability testing company Vertical Slice. I’ve had an interest in website usability for some years, so it was interesting to see some familiar concepts applied to the arena of game design, especially at a time when the range of devices and control methods available is booming.
Here are his five tips for improving your game:
- Get the control method right. Start by looking at what movements the game requires the player to be able to make, and then see how they match up with different control options available. In a 2D space shoot-em-up game, for example, you might need to make small movements to dodge bullets, quick movements to dash to collect weapons, and also be able to stand still. Tilt control isn’t good for small movements or standing still; whereas direct manipulation on the touch screen satisfies all those conditions. A virtual D-pad might be best where it’s important not to block the screen. McAllister said they had about 10 game states they check the controls against.
- Get the tutorials right. Tutorials suck, said McAllister. They’re the first thing players see, and yet they’re the last thing to be programmed, thrown together in a weekend. He suggests that you make a list of everything you want the ideal player to be able to do, such as using all the powers and features in the environment. Then find a way to teach players how to do those things in a way that is timely and feels real in the game (so it is memorable). If you put a meaty tutorial at the start of the game, people will have forgotten about some of the powers introduced in it when they need them half-way through the real game, so introduce things in context. People abandon games that confuse them, so nurture your players’ skills.
- Involve real users, not your friends. McAllister said that some games pass quality control without ever being played by real people, which is worrying. Don’t just ask people what they thought after playing, because they’ll only remember the start, the end and something big in the middle perhaps. Instead, record what they do and use biometrics to identify when they’re most emotionally engaged. Play them back the video of their own game and ask them what they were thinking when they were at their emotional peaks.
- Recruit players using psychographics. It’s not enough to say that you’re testing a game for 24 year-old men who have played Gears of War, because people play games for different reasons. Instead, use psychographic profiling to make sure that you’re testing people with the right motivation for your game. For example, is your game one for people who enjoy completing games, or one for people who enjoy the social aspect of gaming more? Combine demographics and psychographics to recruit players who will give you the best feedback. McAllister recommended Jesse Schell’s book The Art of Game Design for more on this topic.
- Do usability acceptance testing. Set usability goals at the start for what a successful game will look like. McAllister gave the example of a game where players were saying it was too hard and the developer didn’t agree. Then they quantified it: how many times should be player die in the first section? The answer was twice, but the usability testing was showing people were dying ten times as often. Articulating what the player’s experience should be like at the outset makes it possible to do acceptance testing for the game’s usability.
What techniques have you found useful in testing the gameplay of your games?
Filed under: Game development Tagged: | game development, GameCamp, gameplay, playtesting, testing
