Saturday, January 15, 2011

The problem with stealth

Stealth games have always been a little bit of an anomaly when put up against larger, more successful genres.  Even since games like the original Metal Gear on the MSX, it's always been a bit of a risky proposition to build a game where the player's greatest asset isn't a gun, but hiding and moving slowly.  Especially when put next to fast-moving arcade titles of the day, Metal Gear's emphasis on sneaking seemed almost counter-intuitive.  Common thinking states that players like to feel empowered when they play games, that they enjoy being able to do things that they'd have no hope of in the real world; most developers interpret this as the distinctly masculine act of performing excessive acts of violence against others, or in placing first in a competition.  Stealth, by nature, is somewhat contrary to what most developers think players want.

The suggestion made by stealth games over the years is that basic sneaking just isn't enough to keep a game interesting.  The Thief series was able to gain a niche interest by providing tools the player could use to stay safe and escape from danger (water arrows to turn off lights, climbing gloves to scale walls, rope arrows to grapple and swing), but when it came to directly dealing with threats, the player was often at a severe handicap.  It was only as of the third Thief game, Deadly Shadows, that players were reasonably equipped to deal with their enemies head-on.

The same trend followed in the rise of so-called "stealth action" titles, including Metal Gear Solid and Splinter Cell, games which attempted to buck the conception that stealth was all about moving slowly, hiding, and generally keeping out of harm's way.  Their protagonists embodied masculine power: Sam Fisher, a whiskey-drinking war veteran worth a thousand men, and Solid Snake, a cloned super-soldier who was a master of using his environment to outwit enemies.  While both still relied upon stealth, they were also both capable in direct confrontations, so much so that their respective games could be nearly played as straight-up shooters, if it wasn't for the occasional mission which forced non-lethal measures.  Interestingly, both of these games departed from the steampunk-medieval theme of the Thief games, using the advanced technology of the modern era to justify the extreme abilities of their protagonists.

This isn't the

This isn't the stealth I thought I knew...

Now, with the recent release of Splinter Cell: Conviction, stealth action has taken a very bold step in the direction of action.  While Ubisoft's latest game does allow for the use of stealth, but amidst the chaotic gunfights and action movie venues, the game certainly tries its best to encourage players to play less like a ninja and more like a commando, a one-person fighting force who only relies upon stealth and deception until his or her enemies are riddled with bullets.  The question that remains in the wake of Conviction is, "where can we go from here?"

The cynical approach...

It's very easy to take a look at Conviction and echo those tired old words: Ubisoft sold out.  After all, the latest Splinter Cell seems to have more in common with Epic's Gears of War than any other game, and the supposedly "too old for this" Sam Fisher is now more agile and capable than he was ten years ago.  In fact, the stealth in Conviction nearly mirrors similar gameplay mechanics in the Gears of War series - sneaking is only a tool that the player uses to approach enemies from unexpected angles before attacking them.  Splinter Cell, of course, takes greater advantage of stealth, but the focus of the game has shifted radically: no longer is the goal of the game to sneak through and complete objectives, with weapons only as a last resort, but rather, it's to take out all opposition in the way.

Given the relatively slow evolution of the stealth genre, the transition may seem a bit more gradual and the differences a little superficial, but it's clear precisely which mode of thinking informed the design decisions surrounding Conviction.  The player's new ability, Mark and Execute, for instance, is geared entirely towards killing enemies quickly, and many portions of the game are designed in such a way that sneaking around is extremely difficult or even impossible.  Where it used to be perfectly possible in previous Splinter Cell games to finish entire missions or even the whole game without alerting, killing, or even laying a finger on enemies, in Conviction, that seems like a near impossible task.

Of course, Splinter Cell isn't the only game series to move in this direction.  Metal Gear Solid 4 also shifted towards action in a big way by offering control options and scenarios which mirrored successful Western third-person shooters.  While stealth is still a component of the gameplay, and much of the game can be completed in such a fashion, there are also vehicle chase sequences and hectic gunfights where subtlety is clearly thrown to the wind.  Once again, the cynical eye would examine Metal Gear Solid 4 and argue that this was done in order to appeal to a greater market segment, and this may be true to some degree given the recent popularity of the third-person cover-based shooter.  However, I'd like to propose an alternate analysis of the situation...

How do we make stealth exciting?

Stealth games, as I mentioned above, have always faced the challenge of providing excitement to the player, while at the same time attempting to maintain that they are thoroughly about sneaking and hiding from danger.  To say that these are contrary goals isn't quite accurate, but given the settings that recent stealth games have employed (i.e. modern military), there are only a handful of ways to spice up hiding from danger.  The easiest way to do this?  By adopting shooter elements.  Solid Snake and Sam Fisher both have guns, after all, so why shouldn't they use them more often if their missions are so critical?

The game flow seen most commonly in the stealth genre goes something like this:

  1. The player enters a level/environment/room/etc.
  2. The player performs reconaissance and assesses potential goals, threats and opportunities
  3. The player plans a method of approach to that goal
  4. The player executes this plan, attempting to avoid obstacles along the way
  5. The player, if impeded, deals with the obstacle using the appropriate mechanic
  6. The player reaches the goal, and continues on to step 1 again

While good environment design and story can help to drive the player along, the optional step in this sequence, step 5, is where the majority of tension in a stealth game comes from.  While planning and executing a sequence flawlessly is enjoyable for the player, there's no real threat, and thus no tension, if the player does not risk or encounter some sort of obstacle.  In most stealth games, these will be enemies, though the obstacles can also be of an environmental nature (locked door, electrified fence, snowstorm, etc.).  This is the proverbial "wrinkle" in the plan where 90% of game's fun really comes from.

As said, the modern military setting of current stealth games has been rather limiting.  A quick brainstorming session will likely reveal that, if a developer stays within what are conidered to be "realistic" boundaries, the potential problems the player can face are actually quite limited in scope; avoiding repetition within the existing setting is already difficult enough as players continue to tire of the same old "brown and grey" military themes, but when the number of potential obstacles is also highly limited, coming up with unique, plausible and fun challenges is quite the task.

Barrels? Check. Crates? Check. Sneaking suit? Check. Originality? Hm...

Barrels? Check. Crates? Check. Originality? Hm...

I believe that the gradual shift towards more and more action isn't simply a result of market demands and executives dictating that stealth games have "more action", although I'm sure there is some truth to that, whether the efforts are explicit or more emergent trends after the same developers have grown tired of making the same types of games over the last decade.  Rather, the shift that has occurred is a result of increasingly overcompensating for the lack of problems designers can reasonably present to the player.  Put simply, if you've already done the bank robbery, the "no alarms" mission, the "no fatalities" level, the "outdoor" mission, et al., where can you go from there?  The set pieces have to get bigger, the stakes have to get higher, and the action has to get more intense.  Players aren't willing to tolerate the same old ideas; those ideas need to be supercharged, electrified and intensified.

The solution is setting

In light of this assessment, there's really only one option I can see to truly invigorate the stealth genre, and that is to ditch the modern military theme, or at the very least, to let loose and stop being so concerned with maintaining illusions of reality.  Videogames are all about creativity and excitement, and when something ceases to be both creative and exiciting within its existing framework, it takes a shake-up to bring in new ideas.  Mind, I'm not suggesting that the next major stealth franchise take place in a sci-fi environment, or that it takes a page from Tolkien, but a change in setting is exactly what the stealth genre needs in order to become relevant and exciting again, a genre that feels more like its own and less like a subset of the shooter.  Looking to the future, I'm hoping that the new Thief and Deus Ex games on the horizon will be the kick in the pants that stealth needs to get on its feet again.


  1. What do you think about "Alpha Protocol"? It still had the military setting, but it seemed to be ditching the gritty realism feel of stealth games.

  2. Sorry about my late reply! After the initial wave of no comments here (I got plenty on Gamasutra, strangely), I assumed there weren't any on this article as well.

    Alpha Protocol is a pretty strange game because it straddles the lines of so many different types of genres, at different times in the game. Sometimes it's a straight-up action shooter, other times it's a stealth game, sometimes it's got silly boss fights, other times it's going for realism, other times still it's got RPG worked in pretty prominently... it's clearly a case of a game that can't quite decide what it wants to be, and the history behind its development suggests that's exactly what happened.

    As far as stealth goes, I think Alpha Protocol is both a success and a failure... success because it does have stealth as a viable option most of the time, but failure because it relies a lot on weird game mechanics like showing arrows over characters' heads and time-limited invisibility for the player. Some scenarios in the game lead you to believe you can get away with no combat at all, but there are also certain portions which pretty much require you invest in some weapon skills. My first play-through was ruined because I tried a pure pacifist route, and then got forced into a boss fight I couldn't finish, for instance.

    Ultimately I think it would have been stronger if it just wasn't quite so schizophrenic... as an RPG it's passable, as an action game it's passable, and as a stealth game it's passable, but it doesn't excel at any, and often punishes the player for trying to specialise too much in something by switching up the rules of scenarios suddenly. The quality of the writing and voice-acting saves it for me, despite its issues.

  3. people find ninjas awesome so thats why proper stealth games can still be rather big hits if you do it properly and market it as such.

  4. "Players aren't willing to tolerate the same old ideas"

    They obviously are.