Click "Sleep" for a dark background.
Click "sleep" again if text isn't dark.



The Coefficient of Clean pt.7

I can't stop pointing out additional cluttered design features and explaining more clean games. I think it's important to see a range of different clean games that vary in depth, complexity, skill, genre, style, theme, etc. Plus, doing more examples helps me calibrate my scale. 

Input buffering is when the input signal is sustained or carried over for longer than it is inputted into the "controller" by the player. This feature is the intersection of input, player/character states, and mechanics design. Traditionally, games are designed with no buffering. When you hit a button to execute an action, if the game state isn't ready for that action to happen, you get nothing. I call this the dead state for that input or action. Typically, games do not provide any feedback for dead state inputs. You would have to release the button and press it again for another attempt at the action. This kind of design is great for creating a very clean relationship between intent (inputting into the controller) and action. It also works well for digital inputs that are either on or off like buttons or the individual directions of a D-pad. 

In the same way that hitboxes and cancels can be tuned to make a game play and feel more like its fiction, buffer frames can actually be used to make a game feel more precise. In a simple platformer with no buffer frames on the JUMP mechanic, you can only JUMP when your character is standing on the ground. So if you're trying to time a JUMP just as you land from the air, if you hit the button in the frame just before the character hits the ground, you'll get nothing (dead state). A buffered JUMP can make this precise timing challenge a little easier so that players that are slightly off still get the result they want.  

Precise timing challenges and tight windows can benefit greatly from a buffering system. Take the classic Pac-Man game. When you move through the hallways, the motion is smooth and continuous. When you want to dart down an adjacent path, you'll find that as you approach the turn if you start holding the new direction early, Pac-Man will make a very crisp precise turn. 

Super Smash Brothers Brawl has an input frame buffer of 8 frames. That's 8/60th of a second (.133 s), which is faster that your fastest reaction times. Don't believe me? Take the test for yourself. We, B.E.S, designed the game REFLEX so you can measure your own skills. As long as the buffer for inputs is faster than your fastest reflex speeds, it becomes difficult for a player to notice the slight disconnect between inputting and seeing a buffered action.

Buffer frames can take the edge off of playing by widening precise timing windows. One downside is with greater input buffering the player is more likely to feel a disconnect between his/her inputs and actions. Guitar Hero is a very popular music-rhythm game. One notable difference between it and Japanese music games is the timing window for hitting notes. Japanese games love to be very precise and reward players for being exactly on beat within a few frames. Guitar Hero has much more leeway and counts a hit as a hit no matter how well you fall on the beat. The result? Losing players spam more in Guitar Hero hoping to catch a few notes. Also in long sections of repeated notes, it's easy to lose the beat and slowly trail off so that your inputs are steady, but they're just steadily off of the real beat. During these momements it feels like I'm not even playing the game/music. 

Another downside to input buffering is sometimes the game will carry on with actions or convert your inputs into moves you never intended simply based on the way the game/character states are programmed. For this reason, there can be a glaring disconnect between the intended action and resulting action for buffered mechanics. For these reasons even well designed bufferable mechanics can easily clutter a game. The more complex the game with buffered moves, the more likely there will be buffering issues. Noticable buffered mechanics detract from the cleanness of mechanics under the "direct" category. 


Take BOXLIFE for the Nintendo DSi. (watch the trailer here


 .9*(1*.959) - 0 = .8631

  • M = .959 4 mechanics; CUT. FOLD. MEND. SEND. Each button (toch screen or physical) has only one mechanic mapped to it (individual). Most mechanics use the touch screen to manipuate the game elements (direct). There's only one thing to do in the game; manipulate the paper to fold boxes. This makes all the mechanics as dynamic as possible. Interacting with the images on the touch screen is intuitive. There are no cancels. There are no simultaneous actions. Unfortunately, the CUT mechanic is highly buffered if you draw lines faster than the scissors animation can cut. The game stores all of the your inputs too. So if you're very speedy, or manic, you might have to sit and wait for many seconds as the game catches up. 
  • F = 1. The boxes fold with appropriate isometric animations. If you try to fold a shape that isn't completely cut free, the animation will respond accurately showing you that you missed a spot. The shapes you cut are automatically color coded. Each mechanic has a distinct sound effect. The HUD is minimal and the stop watch timer is very informative. All the hitboxes are equal sized squares. Even the bombs in Factory mode are preceded by a widening shadow that functions as an organic timer
  • C = .9 The only camera issue is a very slight one. When you only partially fold a box (or any shape) the 3D folding animation actually obscures some of the isometric flat paper behind it. If you fail to fold a complete box, you have to wait about 2 seconds for the sides to fall back down flat. 
  • S = 0.  There is really not much more to the game. In Factory mode, the shapes you create when you cut and remove boxes changes the remaining boxes you can cut out. This helps keep the static-space down. But the random bombs that drop do more to keep the experience fresh. Depending on where they fall, you must encase bombs in boxes to remove them safely. If you take too long, they explode burning nearby squares thus altering the field even more. You have to keep your eye open across the entire field to play well. This mode stresses all the DKART skills making for an engaging and challenging experience without being a complex or cluttered one.  
  • For non-gameplay features, the player high score is neatly displayed in a graph before you enter any mode. The player progression in terms of an upgradable avatar is displayed in all the menu pages. The overall progress in the Factory mode is reflected in the wealth of items surrounding your home on the title page. All the features of the game are efficiently displayed in 3 menu pages. 


Multitask is a perfect example of a game with only a cleanness issue of simultaneity. 

1*(1*.833) - 0 = .833


  • M = .833. This game is pretty abstract and simple so I'm giving its mechanics the points of intuitive design by default. Each action has its own button. We can think of the game as a hybrid game of 4 very simple games. Because each game is supposed to be separate in this case, the dynamics of the mechanics are isolated as well. Holds and taps work for all the mechanics (direct). However, being able to do all 4 minigame actions at once clutters the design. Yes, the game (which I like) is designed around this simultaneity. But doing 4 things at once inherently splits your attention in 4 directions. Instead of being aware of all 4 screens at once, I found myself rapidly switching back and forth. When I watch other player's videos, I'm not even sure what they're focusing on at any moment to get the job done. 
  • F = 1. Everything is communicated clearly and simply. The tilt of the ball and platform. The incoming spikes. The timers on the boxes. And the sliding obstacles. When you fail, the game pauses (hit pause) and a red "X" appears where you made the mistake. 
  • If Multitask featured sound effects for feedback, it would probably only clutter the experience. Think of this idea like watching four TV screens at once. Your eyes can focus on one at a time, but your ears would be more easily overwhelmed due to the inability to filter out sounds and the low badwidth of adio information. And in a one-hit-kills game, all the sounds would have to be warning sounds as opposed to hit-sounds to be useful. 
  • C = 1. Each of the 4 games is a single screen game. The game screen is divided into four. There are no camera issues whatsoever. 
  • S = 0. There's really nothing else to this game. 


If you think my cleaness expression arbitarily reflects my on preferences and biases as a gamer, I encourage you to play all the games for yourself and compare your experiences. With the math, I can more easily see that BOXLIFE is cleaner than Multitask, which is not an issue of style, graphics, or genre. 

« The Coefficient of Clean pt.8 | Main | The Coefficient of Clean pt.6 »

Reader Comments

There are no comments for this journal entry. To create a new comment, use the form below.

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>