Revisiting the ghost’s shell: movement maps and the gameplay tell

In a brief rant on the future of interaction design, Bret Victor talks about human capabilities. Specifically, the capabilities of human hands.

We live in a three-dimensional world. Our hands are designed for moving and rotating objects in three dimensions, for picking up objects and placing them over, under, beside, and inside each other. No creature on earth has a dexterity that compares to ours.
— Bret Victor, A Brief Rant on the Future of Interaction Design

If you’ve been following this series of posts, you know where this is headed. Aside from recent developments in VR, thus far almost all game controllers released to this date have relied on hands.

On Bret’s recommendation, I picked up John Napier’s book on hands. It is a comprehensive book comparing capabilities of different hands, particularly differences between those of humans and other primates. It’s guaranteed to change the way you look at your hands.

It also changed the way I look at controllers. Particularly, the way controllers change the way we play games. I’m not going to open up the can of keyboard-and-mouse–vs–controller-debate here, because frankly that’s just not very interesting to talk about. Here’s a more interesting question: how does your game controller change the way you play a game? Here’s another interesting question: what does this change look like?

And let’s get the obvious answers out of the way as well: “you can tell a console player by the autoaim and slow aiming”, etc … that’s not what I’m driving at here, and if you want that debate, it’s been carried out elsewhere and probably still ongoing somewhere on the internet. Get your fix there.

Movement maps #

Mouse Map of Minecraft, 3.9 hours [Geekosystem]
Mouse Map of Minecraft, 3.9 hours [Geekosystem]

Here’s a mouse map of 3.9 hours of Minecraft. And below, a mouse map of World of Warcraft, 3 hours.

Mouse Map of World of Warcraft, 3 hours [Geekosystem]
Mouse Map of World of Warcraft, 3 hours [Geekosystem]

What would these movement maps look like with a controller config? They certainly won’t look the same; thumbsticks control very differently from mice.

What if you could get a bird’s-eye view of the game map? If your avatar left a visible trail everywhere it went, what kind of pattern would its movement trail trace as you keep playing away on that mouse and keyboard? Or on that controller?

Controller affordances #

In the field of interaction design, there is a term to describe what an object lets us do: its affordances. A teapot affords the ability to be held at one hand so as to pour tea out the other. What kinds of capabilities does each game control schema afford us?

A closer example: I once tried playing Burnout 3: Takedown with a driving wheel. I sucked with it, no matter how much practice I had. There’s an argument to be made about how much I suck at using wheel controllers, but let’s give me the benefit of the doubt and assume that I am at least average: what are the affordances of a handheld controller that are not available to a wheel controller?

The wheel controller is great for precision, for controlling the rate of turn exactly as you like it. But steering response is much quicker with the game controller, no doubt. It takes a split second to go from full-left to full-right. It takes at least 3 times as long with a wheel controller. And that makes all the difference in Burnout.

Whatever the affordances of a controller, it does not matter if I am able to put up the same performance with it. From an external observer’s point of view, the difference between playing with an Xbox controller and a third-party game controller would be minimal: one might seem to play better, but the general patterns of movement would be largely similar. But the movement pattern with a wheel controller is easily discerned.

If we could see the avatar movement maps of different games with different kinds of controllers, would we be able to match the maps to their controllers? I think this would be an interesting experiment for some games. I have twin-stick shooters in mind, in particular Assault Android Cactus.

While the game itself does not focus on accurate shooting (its scoring algorithm does not factor it in, to the best of my knowledge), I find myself focusing more on aiming and less on movement when playing with mouse and keyboard. I move generally more slowly, and turn generally more slowly, but with increased accuracy. And the inverse happens with a game controller: I dance around the stage, pirouetting and doing 180° turns, but with decreased accuracy. My scores are better with the controller, but I attribute that to having spent much more time on the controller than on keyboard+mouse; I have yet to measure the results objectively or do regression testing (but you’re welcome to contact me if you’re interested in this).

Control scheme affordances #

On the topic of affordances, a large part of what is possible in a game also depends on what is programmed/set in the game engine and control scheme. A control scheme that maps two actions to two buttons on the ABXY cluster usually means you are not meant to perform both actions at the same time, and hence it is made more difficult or you to do so. Some games might not allow you to jump and shoot at the same time (some platformers), or to move and aim at the same time (Gigantic Army); other control restrictions abound. And more broadly, some play styles are simply discouraged by the control scheme. One example off the top of my head: Gunvalkyrie.

Most third-person shooters have grown to adopt a run-and-gun, strafing playing style—you move sideways to dodge, to move in and out of cover, while generally maintaining your aim on your target. Gunvalkyrie, however, does not quite have the concept of strafing in its control scheme. You can move/dash forward and backward, do quick turns at 90° or 180°, and aim within a relatively small cone of fire. No strafing. This led to a lot of gamers calling the control scheme broken and unintuitive.

Having played and struggled with the game for a bit, I could see some kind of movement map emerging, however. It is a movement map unlike most third-person games, a movement map where one dashes from safe vantage point to safe vantage point, taking out as many enemies as possible at each point before being forced to find another one. The control scheme seems to be designed to minimise the time you spend turning, and maximise the time you spend dashing. (Again, I’m not interested in advocating either playing style, only in exploring their differences.)

Game engine affordances #

On a similar note, complaints about game mechanics (“rules of the game”) now make me think about movement maps, and what kind of movement maps were intended by the game designer. Another example: the battery mechanic in Assault Android Cactus (AAC).

To summarise the argument, there are some who feel that the battery mechanic is too harsh and affords too little time to complete the mission. This game mechanic (in conjunction with others, such as enemy-dropped powerups) was designed to encourage and reward a much more aggressive, short-range playing style. Nonetheless, quite a number of gamers still wish for an option to disable the game mechanic in game options. (I’ll skip the whole “game creators ought to listen to their customers and give them what they want” part of the argument here if you don’t mind.)

Clearly the complainant’s preferred playing style is not afforded by the game’s battery mechanics; similar examples crop up in games everywhere else: games that don’t let you move faster than the AI, games that always let opponent racers catch up with you, games that don’t give you long jumps, and so on.

The gameplay tell #

The whole argument reminded me very much of a post from ribbonfarm early this year, still on the theme of useability. Paraphrasing it, if you find it difficult to do things a certain way in the software, it is likely that there is a conflict between your needs and [an organization’s] needs. In AAC’s case, it is a conflict between the gamer’s preferred way of playing and the game designer’s intended mode of gameplay.

To put it yet another way, the game designers intend to produce a certain kind of movement map. If your preferred movement pattern does not match this, you will probably have a hard/unhappy time with the game. An inability to achieve the game’s goals despite a high level of ability is a gameplay tell, a hint that your intentions and the game designer’s intentions are in conflict.

I have dropped many games simply because I did not like the control scheme. This is not the same as saying that a game has sloppy controls. In the latter case, even if I have the intended movement pattern in mind, I would be hard-pressed to achieve it because of deficiencies in the controller’s response. In the former case, even if the game controls were tuned to my preferences, there would be a fundamental conflict between our intended movement maps.

We classify games in many ways: by genre, by platform, by perspective. Each of these categories reflects something about our psyche, and we seek out our preferred classifications in the hope of finding something that matches our psyche. But even among similar play styles, there can be a wide variation in experience. If there is a way to parametrise our movement maps, I wonder if it could lead to more nuanced game recommendations, based on your preferred movement/control patterns?

I think this digs at a deeper topic, the embodied consciousness of the gamer mind (or, “how does your controller affect your psyche?”). More on this in a future post.