Thursday, February 01, 2007

Game design

I'm a casual gamer. I tend to get frustrated by certain types of games, and lately I've been trying to identify which aspects of games I like or dislike. Here's what I've come up with so far.

  • Easy to learn, hard to master

    If I need to go through 10 minutes of tutorials before I can understand the basic gameplay, chances are I'll turn off the game and move on. Like any entertainment, games should hook the audience. Story-wise, jump straight into the action and leave the exposition for later. Game-wise, start off with a ridiculously obvious control scheme (from the developers' point of view), hint at the unique aspects and depth of the game mechanics, and let the player delve into them later while actually playing the game.

    Good example: Prince of Persia: Sands of time does a good job of extending standard platform gaming in an interesting way.

    Bad example: EVE. I stopped playing after a good 40 minutes of just creating a character and sitting through what seemed like endless tutorials. I guess that game isn't for me. Another bad example is Prince of Persia: The two thrones, which demonstrates the other extreme in how it throws the dark prince at you and leaves you to try to figure out the control scheme on your own, on a time limit. Don't know which button to press, through some mysterious telepathic connection to the game designer? Time's up, start over.

  • Soft obstacles

    This is a term I made up for something that I want to see more of in games. Most games are basically made up of obstacles; different problems and situations that you have to get past in order to reach the goal. A hard obstacle makes that goal seem unattainable and frustrates the player. A soft obstacle doesn't stand in the way of progress. Every time you try to overcome it you either partially succeed, gain some new insight into how to overcome it, or noticeably increase your skill. You may even receive an explicit reward for trying. In other words, soft obstacles maintain a level of hope and bait you to keep playing by keeping the goal in sight. A good way to achieve soft obstacles is to have several mutually independent obstacles (i.e. parallel missions) and open-ended gameplay. Let the player take a break from the game without actually shutting it off.

    Good example: Even if I lose a race in Need for Speed: Carbon, I feel that I've honed my skills, learned the layout of the track, and come closer to unlocking another cool upgrade for my car. Another good example is Grand Theft Auto, where the open-ended gameplay allows you to blow off steam between missions.

    Bad example: A lot of adventure/quest games, where you customarily get stuck and can point-and-click your way around for several hours without even understanding the overarching goal, much less which pixel the designer wants you to find, or which inventory objects you're supposed to combine for no obvious reason in order to advance the unrelentingly linear story. Getting stuck represents absolute downtime; you're just watching the character walk around. That's not entertainment. I might as well watch my screensaver (currently set to "Blank").

  • Failure as a game mechanic

    If the player regularly sees a "game over" or "mission failed" screen and has to start over or load a saved game to complete the game, then failure is an integral part of the game mechanics. I've gone back and forth on this issue, since some games actually do this well. My conclusion is that you should be very careful with this. If you absolutely need failure, make it as non-frustrating as possible. Don't punish the player for playing within the rules of your game.

    Good example: Somehow, Splinter Cell: Chaos Theory doesn't annoy me with its constant save/load cycle, and instead makes me think that I can sneak past another obstacle after the next load. Another good example are the LucasArts point-and-click adventure games, in which failure simply doesn't exist.

    Bad example: Any old Sierra quest game. In fact, I habitually refer to this kind of poor game design by simply calling the game a "Sierra game", usually with added expletives for flavour.

  • Replayability/moddability vs extensibility

    The business model of games usually aren't a deciding factor in whether I play them or not (although I really can't justify spending money on subscriptions for games that are practically void of content). As a player, I welcome open-ended gameplay because it generally softens the obstacles. However, I'm also interested in developing my own games. From a game developer's point of view, income becomes an issue if players are content with replaying old open-ended games indefinitely. If a game is infinitely replayable and moddable, it might be hard to create an incentive to buy new games. One possible solution is to sell extensions for popular franchises.

    Good example: A lot of multi-player games never get old, especially console games in 2-player cooperative mode. I recently played International Karate + with my brother on his Atari 1040 STFM for hours on end. As for extensibility, I like the Guild Wars business model, in which extensions are the only source of income. Half Life is a good example of a game that has sold extremely well thanks to a mod, in this case Counter-Strike.

    Bad example: Fable doesn't have a satisfying end, and the replayability is largely cosmetic. World of Warcraft already has a subscription fee, so selling extensions seems rather unfair.

  • Hate the player, not the game

    Win or lose, the skill of the player should be the only deciding factor. Before the days of multi-gigahertz processors, simulating real-time physics for more than one car was nearly impossible on retail hardware. Therefore, most racing games cheated and let AI cars drive around unaffected by physics. Of course, any kind of cheating frustrates the player to no end if it's discovered, and can be conjectured to be the sole reason for failure. Why would you even try to win against a game that cheats? Avoid bad interfaces or input schemes, and avoid even the impression that the game has an unfair advantage over the player. The player should never be able to blame the game for losing.

    Good example: Again, Need for Speed: Carbon.

    Bad example: Mario Party, apparently. I haven't played it.

Additionally, games should be fun, like Space Channel 5. Seriously, please make some effort and don't include "boring" in your design goals. I realize that I may have just eliminated every other word in the World of Warcraft design document. So be it.

No comments: