Death By Interface

Death by Interface is a long known pandemic of video games. The most common subset is death by ladder, a problem to this day. In the old days, ladders were notoriously slippery, and players plummeted to their deaths by an inadvertent control-slip. These days, ladders can be too sticky, so the player is stuck on the ladder, sometimes helpless, in the middle of a firefight or other hazardous situation.

More generally, death by interface is what happens when a player should be able to respond to hazards intuitively, but has been disabled because the interface is non-intuitive. The player gets stuck, exposed to the hazard, and expires.

This is not to be confused with actual environmental hazards. In Subnautica when I’m penetration diving and suddenly realize I’m low on air and in the middle of a labyrinth of twisting corridors and air vents (all alike), and then I panic and can’t find my way back, that’s not an interface problem, but an environment problem. I perished because I went into a dangerous environment and didn’t mind my air. Had I known the way back, I’d have been able to maneuver myself to safety. Lots of real people really die doing penetration diving, and I’m glad to only be doing it in a video game.

To the contrary, death by interface happens when a hazard is navigable, but non-intuitive elements of the interface interfere with a player working the controls to reach safety, resulting in death.

In The Division, it happens a lot.

Duck and Cover

Tom Clancy’s The Division uses a lot of contextual movement. Some of its interface designs are closer to those of tactical games than a shooter. I actually like The Div’s cover system, when it’s not killing me.

I tap the cover button (I’m using Numpad-Enter, which I commonly use to crouch), and my agent slams herself into the nearest wall. I can then point at other walls, and hold the cover button to sprint over to that wall. In the meantime, aiming pops me out enough to aim (sometimes not quite enough to aim at my target of choice). But generally The Div’s cover system keeps me out of harm’s way long enough to decide how to approach the situation tactically.

But then…

Sometimes it’s difficult to detach from the wall when a grenade is heading my way, but other times it’s too easy to detach, say when I want to crawl around under cover. I’ll detach, pop up and eat bullets. Still, when it works, my agent keeps her head down well.

But then comes the problem of when there’s no cover to be had. The Division’s Manhattan is chock full of spots that don’t look like cover but are (e.g. sidewalk storefront shingles of plywood, or a telephone pole).

Manhattan is also full of places that look like cover and totally should be cover, but the cover interface doesn’t want me to go there. Amusingly, the level designers tried to pile garbage bags or other sloped detritus against walls that aren’t really cover, yet any GI worth his MREs would happily embrace that pile, stench and refuse be damned. Indeed, our extensive library of WWII photos shows soldiers of all walks deeply enamored of sloping piles (usually of rubble) so long as that pile is willing to absorb bullets.

And then there are some parts of levels where it seems intentionally stripped of cover to devise difficulty where there shouldn’t be, especially when they appear in the higher-level areas of the city.

So there are three special kinds of Division cover-related death-by-interface (not necessarily a comprehensive list):

Sudden bad guys, but there’s no viable cover. It’s especially bad if in close quarters, but all the cozy walls are unviable as cover. Not Cover-Viable!

Sudden bad guys, and this time the cover faces the wrong way. Most of the time the agent can circle around or jump over first, but sometimes the interface doesn’t want to do that.

The agent loves his cover, and doesn’t want to leave, even when a grenade lands nearby. Also rushers (that is enemies that charge, brandishing a club or fire axe). Rushers will flank the agent and he finds himself getting beaten with a baseball bat while stuck to this cover wall. Variant three is when it’s time for the tactical retreat but parting from that cover is such sweet sorrow…

Jump and Climb

Jumping is also contextual. This is actually noticeable in previous Tom Clancy branded tactical shooters, which don’t allow jumping generally, but do have context points where a context key can be used to jump. In The Division the jump key is used to mantle up ledges and jump over low obstacles. (IRL jumping and climbing through windows makes one super-vulnerable to getting shot. Also sliding down ropes from helicopters. Avoid doing that.) There’s also a convention that fences with spikes and barbed wire designate nontraversable fences

But then there are places where it looks obvious you could jump up (or down) but can’t. Because the game says no.

To be fair, in old fashioned run-and-gun games in which you can jump to your hearts content and climb on anything low enough, level designers would just try to put some kind of obstruction to prevent players from going where they shouldn’t go.

In later games, invisible walls were added to make sure visible walls didn’t fail. This was frustrating, but only because players want to explore everywhere, and unravel cleanly designed levels with their precision acrobatics and logical paradoxes. I’ll talk about invisible walls more, momentarily.

Jump / Climb related death-by-interface in The Division:

Rather than bumping up against a shelter for cover, the agent climbs up on top, instead. This draws gunfire from every enemy in range looking for a target. Extra points if the agent is, for some interface reason, reticent about climbing back down again.

It’s time to make a tactical retreat, only I can’t go that way. Behind the trucks I’m standing on is a barbed-wire fence below the roof level. It doesn’t matter that I didn’t see it before and should, by all plausible interpretations of general mechanics, be able to hop right over it without trouble.

I’ve been sniping from scaffolding and it’s time to make that tactical retreat again. Only there’s one way down, the stairs (or the ladder!) where all the bad people are. And it’s never a good idea to climb a ladder in combat. (Stairs suck too.)

In and Out

Context control design is nothing new. Thief was updated to include a jump/block control so that the player didn’t jump but instead blocked when his sword was drawn (an interface liberal, Thief was, it also had separate jump and block controls, in case a player wanted to keep them separate). In The Division, I’ve done, as I have in many games, bound my context action button (open box, turn key, push button, etc.) with my reload button, since both functions don’t interfere with each other (much). There is a place for context.

But context-based design is opt-in design. You can only jump where we let you jump. The player is allowed to do only what devs specifically allow. Contrast opt-out design in which player can do anything but the developer specifically disallows.

Most games use a combination of both. General movement in The Division (or almost all games) is opt-out. You can go anywhere unless it’s specifically blocked off. Most games that allow for jumping and crouching simply extend movement options to cubbies and low platforms and short ravines. Thief extended this further with mantling and the ability to place ladders (aka ropes), which were still easy to fall off to one’s demise.

Efforts to solve the death-by-ladder problem are what got opt-in movement interfacing started in the first place. Some games figured that if you had to specifically engage a ladder through a context control (say by using it with an action button), then they’d be safer. It was an easy step from there to turn jumping contextual when there were only few situations where leaping was necessary or appropriate.

We’ve also seen the rise of invisible walls, which started as a way to reinforce walls and cliffs to keep players within the confines of the level. In later games, perfectly good (appearing) corridors would be blocked off on the mere excuse that the story isn’t in that direction. (It’s not too offensive when used sparingly.) In a step in the opposite direction the Serious engine could surround the level with miles of traversable, explorable and remarkably empty desert (which, while liberating to players, presented its own problems).

The good thing about erring on the side of opt-out play is that there are happy accidents. Players can utilize all of the terrain’s features, even those that the developer didn’t account for. Of course, this can also lend to unhappy accidents as well, in which critical logical sequences can be short circuited, or the player finds himself off the map, sometimes unable to get back, or falling into infinity (a transcendental experience by which many players have found enlightenment, albeit briefly).

Opt-in design gets neither benefit. No accidents, happy or unhappy, but if the dev isn’t super thorough, it can become as if like invisible walls are everywhere. Opt-in also requires the necessity of perceptible interface queues: cover has to look like cover, and non-cover look like non-cover, or the player gets frustrated. The emotional impact of vagaries and arbitrary exceptions in opt-in games becomes exaggerated, driving players to paranoia. They also can cost (virtual) lives. When an opt-in escape becomes closed off (say if Batman has to fight a boss monster), if it’s not acknowledged or justified by story, it reads as developer disinterest. Slipshod storytelling. Like too many invisible walls, it feels contrived and confining. (More so when, in Batman canon, he depends more on tactics and second-guessing than he does on sheer rage to drive his dispensation of justice in fist-sized servings)

To be fair, Manhattan in The Division is huge. A few level-design flukes and awkward interface moments are inevitable and forgivable. But ambition of magnitude requires commitment of magnitude. A few danger spots where the cover is all wrong can be counted as anomalies, omissions in quality control. But when they appear all over town, and more consolidated in higher-level districts, then it feels wittingly contrived, like someone’s using interface difficulty rather than environmental difficulty to make the game unnecessarily harder.

Then it feels like the developers are just intentionally trolling the players.


One thought on “Death By Interface

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s