Watch Me Play

When considering how a game is perceived by its player(s), it is also interesting to consider how a game is perceived by pure observers. How engaging is it for those merely watching? There is probably little financial incentive to explore this aspect of a game, but it's a curious design experiment if nothing else.

Thinking about this, it seems there are a few qualities that significantly affect how watchable a game is.

1) Nature of Challenge

Platformers have obvious challenges that are usually approached at the player's pace. Shooters are largely reflex challenges with a high degree of surprise. Puzzle-based games rarely require manual dexterity of any sort. How readable the goals are to an observer affects how enjoyable it is to watch. Having little understanding of what's going on, or being unable to follow the player's rapidly changing focus, makes for poor watching. But being able to gasp at whether the player will make that jump, or better yet being able to actively assist the player ("What if you put the red block there instead?") hold the attention far better.

2) Camera Perspective

Something of a corollary to the above, first-person games are probably less watchable than third-person ones. First-person allows, and often necessitates, changing the orientation of the viewport rapidly and dramatically. If you're not the one doing that adjustment, it's easier to become disoriented. Third-person tends to change camera alignment less often and more slowly, with all movement being respective to a fixed object (the player's avatar). Without the ability to mentally map controller input with the camera's output, having a representation that can be understood purely visually is much easier to watch.

3) Limited Systemic Information

Games that take place largely in the player's head, e.g. strategy games, are likely difficult to enjoy watching. When a great deal of gameplay takes place in the player's mind, there is little to watch. For the player, the game is really about internalizing and reacting to the game's state. Learning to parse that state from the game's visual representation takes time. If an observer doesn't have this systemic knowledge, there is little to observe.

Now there are some games that could be very engaging to watch for those familiar with it. Arcade fighting games, e.g. the Street Fighter series, represent this distinction well. As the crowds at Evo will attest, watching two masterful players going head-to-head can hold attention rapt ... if you're familiar with the game. If you aren't, well, it's a lot of confusing sound and fury that ends in 90 seconds or less. Starcraft is nationally televised in South Korea, but it's also so ubiquitous that just about any viewer is at least passingly familiar with the gameplay. I'm not sure I'd ever find watching someone play Team Fortress 2 or Civilization to be especially engaging, but seeing someone pull off a near-impossible stunt in Burnout is only enhanced by knowing just how hard that stunt was to accomplish.

There are other obvious things too, e.g. single player or local multiplayer is more watchable than online multiplayer, just due to physicality, console is better than PC or handheld, etc. Ultimately, it's about being able to understand the game's drama. Action tends to be better at this, since character-based drama obviously requires familiarity with said characters. Without that familiarity, character-based drama has to hang on the quality of writing and acting ... yeah. That one probably isn't to be relied up.

If there's any functional design goal here, it would be encouraging observers to become players. As you may have been thinking already, Rock Band does this better than almost any other game. It has a simple, understandable goal, creates a context where play feeds directly into the experience and it has an intuitive interface. Plus, given that you'll likely be playing with others, all someone has to do is push that plastic guitar in your hand and say, "Take a turn." No single player is put on the spot either. Many of Rock Band's lessons are design specific, but it's still interesting to see how they make their game so damn inviting.

As an aside, the co-op portion of DeathSpank was designed with these principles in mind as well. It's too long to get into here, but I believe I'll be writing something for DeathSpank.com about it next week. Look for it and I'll try to remember to put up a link too.

Are there games you found fascinating just to watch? Have others watched you play games? I'm curious as to what qualities other folks find engaging when someone else has the reins.

The Maker's Lens

This post is a little more personal and less pointed than many previous. But it's an observation that's been vaguely bouncing around my head for a while and my wife (and talented editor!) helped crystallize it last night, so I figure now is as good a time as any.

It feels like my perspective on games changed after I started making them professionally. It's not for the better or for the worse, merely different. I've seen how the sausage is made. The process is a little unnerving, but damn, it's still delicious.

What I gained was an understanding of how all the pieces fit together. Even the simplest of games has a tremendous amount of interlocking pieces, nearly all of them inter-dependent. At times I'm amazed games get made at all.

I am able to better recognize the components now. I see how the components can be built into systems that challenge, engage and emote ... or at least attempt to. It's a little disheartening at times, like touring the set of a favourite movie or TV show and seeing there is no ceiling and only two walls. The vast, vibrant world you imagine really looks something like this.

It's not a negative perception though. In some ways, it's even more fascinating, both on the part of the creator and the player. The creators provide just enough substance for the player's imagination to take over. The player doesn't need to try to sneak a peek past that unclimbable wall, because they believe the city continues outward.

I am now more aware of what's beyond the wall though (nothing). It's very similar to when I read The Design of Everyday Things and started noticing how nearly every door in the world is poorly designed. But it also helped me recognize and appreciate instances of really good design. Games are not much different.

I will notice clever tricks with lighting to guide the player's eye. Seeing an NPC behave strangely, I start guessing at the logic bugs in his state management that might have caused this. I'll note the way certain abilities drive particular behaviour. The different ways enemy AI behaves has become especially noticeable.

But when a game is so engrossing that the developer lenses come off for a while, I cannot help but admire it. All the various gears and levers recede and it's just about the experience. Rather than appreciation of various components, it's the whole artifact that's engaging. Weirdly, it makes me take notice because of all the things I'm not noticing. Eventually, I'll probably start trying to peer under the hood again. Although this time, it's likely to try and understand how all the bits fit together so well and how I can replicate the same.

One interesting side effect of this is I feel more empathy for other developers. I have a lot of difficulty completely lambasting a game, because I realize there were dozens of people not too unlike me who worked quite hard to bring this thing into existence. Mistakes are not absolved, but I find myself asking, "How can I avoid doing this?" rather than frothing about the developers' intelligence and parentage, as the howling hordes of the internet so often do.

I've always been pretty dogmatic about finishing the games I start. But now, I finish games because I want people to finish mine. By finishing games, I feel as if I'm fully appreciating the efforts that went into them. I've probably sunk some unnecessary hours into games that peaked early and I understand why others would bail early. If there's just some smaller interesting facet I want to examine, I'll grab a demo. But as soon as I've laid down cash, I feel compelled to finish.

There likely aren't any stunning revelations here. It's just interesting for me to reflect on how the way I see things has changed over the past couple of years. I'm curious to hear if your perspective on playing games has shifted as you started making games, playing different kinds of games or even just playing more/less.

More than anything else, I have found it very valuable to recognize that I don't see all the exact same things others do. This makes their perspective invaluable. Find these people, be they team mates, playtesters or just other games, and let them help you see things with a different set of lenses.

The Best Disguise

Author and journalist Tom Bissell recently had his book Extra Lives: Why Video Games Matter published. In concert, Michael Abbott recorded a fantastic interview with Bissell. It's one of Michael's best and I highly recommend giving it a listen. I have Bissell's book on my desk, but haven't had a chance to dive into it yet (the fantastic Darwin's Bastards occupies my limited reading time at the moment). Hopefully the topics addressed in their conversation are borne from sections of the book though, as Tom has an interesting perspective on a number of subjects.

One such subject, and this ties into this series on perception, is how game systems are presented to the player. As games have modernized, certain games have attempted to disguise or justify familiar gameplay tropes within the game's fiction. Bioshock explained the game's checkpoint system via the Vita-Changes that rebuild the protagonist from his stored DNA should he be killed. Most likely, you completed Bioshock not with the original protagonist, but with his umpteenth clone.

Bissell praises the Disguise, referring to Jon Blow's claims that the needs of a game's narrative (progress with good pacing) and the needs of a game's challenge (hold the player back until some goal is achieved) are inherently opposed. By cleverly utilizing this concealment of game systems, it's possible to ameliorate or even exploit this conflict. Portal and Dead Space are cited for their superb execution in this regard.

I cannot disgree with that either. The narrative setup in Portal allows the challenges to make complete sense while still leaving room for GLaDOS' excellent characterization. Dead Space makes the entire game's HUD endogenous, greatly closing the gap between what the player is meant to experience and what the protagonist is. Elsewhere he praises Mirror's Edge for other reasons, but I imagine its minimal use of unjustified gameplay tropes contributes to its acknowledgement.

He decries the coffee thermos collectables in Alan Wake, a condemnation I wholly agree with. Rather than attempting to integrate that collectible into the game's fiction (which is doubly absurd considering there already was a different collectible that did fit into the game's fiction), the thermoses exist exclusively to provide supposed challenge and give the player "more to do." Poor use of the Disguise like this reeks of design-by-committee: "(Game X) had collectibles and it did well, so we need to find a way to add collectibles to our game."

The conversation on this topic is great, but it does fail to acknowledge one thing I feel it should. Bissell speaks at length about "games," but most of the time, he's really referring just to single player, story-based games. There's no problem with this, of course, but I hope his book either calls this out or discusses other types of games.

That omission is notable because the Disguise can apply equally to multiplayer or games not based on story. There's no question the aesthetic and characters of Team Fortress 2 give the game much more appeal than the generic soldiers of the original Team Fortress. The promotional comic for the free Solider vs. Demoman update is more compelling than the entire narrative of a good chunk of games. But TF2 doesn't use the Disguise just for charm and personality.

As each class in TF2 has dramatically different abilities, being able to recognize them quickly based on silhouette alone is very important. By creating unique characters for each TF2 class, their highly readable silhouettes are reinforced by the fiction. As each character literally has a unique voice, their barks are immediately recognizable. In most modern-day military shooters, every "Tango down" sounds exactly the same. But there's no difficulty telling the difference between the Demoman belting out "Medic!" in Scottish brogue and the Heavy's thick Russian cries. These are just a few examples. Valve elegantly utilizes the Disguise to present information about the TF2 classes without needing to explicitly explain anything.

It's also the case that a great many games don't need the Disguise. Games where story isn't important shouldn't care about justifying their gameplay. Mario needs to find stars because ... I guess it helps save Princess Peach, who has inevitably been kidnapped again. But really, the stars exist because it's a serviceable objective in the challenge to navigate fantastical and dizzying environments. No more is needed and even that thin veneer could probably be lost without doing much harm to the experience.

It is really about being clear on what you're trying to make. As critics and consumers of games, we ought not to tolerate brain-dead artificialities that exist only to provide "challenge" while doing great harm to a game's story. Conversely though, we shouldn't condemn a game for lacking these things if that game is built to be about the challenge and the narrative is, at best, a little colour. The Disguise can be a powerful thing and those developers that know how to utilize it correctly have produced some of the engaging, immersive games seen to date. But perhaps the important aspect of the Disguise is knowing when to use it. Rather than timidly flitting between both sides, having the clarity to know when such things are necessary would demonstrate the focus necessary for a project to have a shot at being great.

Don't forget, read the book and listen to the interview. Both are definitely worth your time.

Make a Perception Check

I think the readability series I wrote over a year ago is still one of my favourite pieces on the site. I think the connectedness has an advantage over the more reactive, "here's what I'm thinking this week" writing I tend to do. So it's time to roll out another, this time about "perception." The connections may be a little more tenuous than the readability series, but I'm hoping to keep the focus on how different methods of constructing elements of a game (or even wholly different forms of games) can significantly affect how players act.

Out of the gate we're going to talk about one of the fundamental actions in tabletop RPGs: "Make a Perception check."

For those unfamiliar with tabletop RPGs, a Perception check is the mechanical representation of your character's senses. Different systems have different parlance, but the basic action is nearly universal in some form or another. It determines whether or not you sense something intentionally concealed or merely notice small details. Most systems allow eagle-eyed characters to improve their sensory skills. And therein lies the complication.

Tabletop RPGs almost always involve a group of players. The GM will ask for a Perception check and then every player rolls. But in the vast majority of situations, only a single player needs to succeed. The result is conveyed to the group as a whole, no matter how much effort is made to rein in metagaming. This dampens that limelight moment for the keen-eyed player. And it feels very game-y when the group always sees with the best set of eyes. Possible alternatives, like passing notes to those that succeed their check, can work but are logistically infeasible to be frequent.

This also ties into the other major issue with Perception mechanics in tabletop games, namely that Perception checks without interesting outcomes for both success and failure are probably not worth having. Asking for a Perception check and then responding with "You don't see anything" immediately sets the players to sniffing out what they missed. In the worst case, they'll begin scouring every square foot of an area, probably ignoring whatever situation was previously demanding their attention. Making checks for the players in secret is a possibility, but it somewhat robs them of their ability to make interesting decisions. Or at least know the consequences of prior decisions made, like whether or not they chose to improve their Perception skill.

I'm becoming more and more convinced that Perception mechanics are strongest when they have immediate, readable outcomes. If everyone in the group is rolling, the consequences should apply for everyone. E.g. failing a check means being surprised by the ambush and losing a turn. Calling for individual Perception checks is almost far more interesting (does the character walking ahead of everyone else notice the old rotting floorboards before he steps on them?) when there are immediate outcomes. I haven't really pushed on this methodology enough to say if it has any unfortunate consequences (like always making the most Perceptive character take point), but I'm going to be more rigorous about it in the next game I run and see what happens.

Digital games often reverse the presentation of player information, with very little perceptual information being shared between players. Naturally this only applies to networked multiplayer games, as local multiplayer games naturally share their information, since everyone is looking at the same screen.

Any avatar-based game will naturally facilitate each player having a different perception of the game world. Usually the degree of sharing is driven by function, e.g. can I see my teammates' position on a map, know their statistics, etc. The perceptive skills of each player are symmetrical and their ability to share information is limited to meta-game tools (i.e. text and voice chat).

A few games have really experimented with the idea of offering different perceptive abilities. Chromehounds, a mech combat game released in 2006, featured a fascinating commander role. The commander player had a very lightly armoured mech with limited combat abilities. However, their perception abilities outstripped every other player. They could see further, have better radio control, affect the minimap, etc.

Other games have attempted the same, notably the commander role in the Battlefield series and more recently, in Sony's MAG. Savage, released in 2003, took a somewhat different tack and attempted to blend RTS and FPS gameplay. One player was given RTS-like control of the base, build tree, etc. while the rest of the players served as soliders on the ground. In actuality, I'd say none of these were as successful as Chromehounds in creating a role where differences in perception abilities are meaningful and rewarding. I think they largely fall down because the games are simply too big.

It seems the thinking behind commander roles in games like Battlefield is to simulate real-world coordinated military action. However, they lack what makes a commander on a real battlefield work- a chain of command. With games like MAG supporting 128 players per team, you're guaranteed to be playing primarily with strangers. Strangers who have no interest or incentive to listen to the commander. Without actual threat to life, or even just court martial, coordinating 127 strangers is basically impossible.

Chromehounds supported only 6 players per team, meaning there was a good chance you'd be playing with five friends. Even if you weren't, convincing five people to work together is possible where having over 100 do so is unthinkable. Given that co-op gameplay (along with motion controls, seemingly) are what's vogue these days, I'd love to see further exploration of the path Chromehounds started. A pure co-op experience where significant differences in perceptive abilities were built into gameplay from the ground up could be striking. Left 4 Dead demonstrated expertly that players, even strangers, will work together well if the mechanics ask them to.

And of course, this is only in a cooperative context. The interplay between shared information and hidden information is at the core of a great many competitive games (everything from texas hold 'em on up). That's a whole other post (or series even) that I may not delve into. But it's clear there's a lot of space to experiment with just what players can see without even needing to dive into what the players can do.

Now what happens when what I see informs what you do? I'd hazard it would be something very interesting indeed.