|
[In this thought-provoking design piece, writer Sande Chen (The Witcher) takes a look at how to ratchet up emotional intensity - through narrative design, visuals, and music - to create more meaningful games.]
While games have come a long
way, developers have yet to take full advantage of the medium by incorporating
meaning within the very structure of a game. Time-tested techniques from film
can help, but first they must be adapted to fit the unique properties of video
games.
One of the strongest tools to convey meaning is thematic expression,
which can be integrated not only in narrative, but also in other game elements,
such as art direction, game systems, sound design, and music. Developers who
yearn for mass-market appeal can use this multidisciplinary approach to create
meaningful and more emotionally charged games.
Narrative Design and You
"If we're going to build
really powerful games, we need interdisciplinary teams," says Sheri Graner
Ray. As a freelance game designer and production consultant, Ray knows
firsthand the level of collaboration that can occur when artists, writers,
designers, programmers, and composers work together.
In particular, Ray advocates
bringing in a writer or narrative designer early enough in the design process
so that the game can have more depth. A programmer wouldn't be expected to know
which colors evoke which feelings, Ray points out, so why would anyone without
training or experience presume to take on the role of the narrative designer?
As champions of the story,
narrative designers are more than just dialog writers. By using themes and
other tools of storytelling, they oversee the process of crafting a meaningful
experience throughout the game. Themes, like an emotional heartbeat, enable
stories to deliver a universal message, yet also resonate with players at a
personal level. Thematic expression can surface in the narrative, art
direction, game systems, sound design, and music. As such, the role of the
narrative designer is inherently multidisciplinary.
Themes also are likened to
the game's or author's point of view. "Even in games," says narrative
designer Stephen E. Dinehart, "we seek to communicate a message."
The
theme gives the reason why this particular story should be told. Often, to find
these meaningful themes, writers soul-searchingly delve into a personal credo
to discover where their passions lie - because to convey passion, one should be
passionate about the underlying message. Because themes are best felt at the
subconscious level, a writer must be careful not to be preachy and instead, let
the theme flow from the situations presented in the story.
The integration of themes is
what elevates a film to cinematic art. Developers of video games can adapt the
very same techniques used in film to create more meaningful games. Just as
films have stirred passions, evoked emotions, and even encouraged real-world
action, so too can games.
The Visual Structure of Games
Much as the Munsell Color
System provides a way to specify a particular color of value 8 and chroma 2
without confusion, Bruce Block's visual theory of cinematography gives a common
language for developers to discuss visual components.
As described more fully
in his book, The Visual Story, Block discusses how film directors use the
narrative structure as the basis for cinematography decisions. Color, tone,
shape, line, angles, lighting, and camera movement are all choices that reflect
theme and affect how audiences feel about a character or setting.
For example, ambiguous space
can create anxiety, confusion, or tension in a horror flick.
Curved lines have
a more whimsical quality than straight lines. A car zooming diagonally across a
frame looks sleeker and speedier than if it were travelling horizontally. Likewise,
a composition built around triangles will be more visually intense than one
centered around round shapes.
Furthermore, a director can use affinity and
contrast to control visual intensity.
|
Printer-Friendly Version
It's a concern that there is even a need for an article like this. I'd have thought this mentality was so obviously sensible as to be accepted wisdom.
Though when I consider some recent games I've played I can see that it clearly isn't.
Still, what really intrigues me is the thought of "cheating" the player with cutscenes that do not look like ones. I actually never thought of such tiny scripted events as cutscenes. Though I believe the possibilities of using them are quite limited, since I need a player action that triggers a mechanic that is normally not controllable by the player until it is finished. This requires good timing to not unmask the cheat. Taking the mentioned example of jumping over the chasm: if I as a player do not have the ability to cling to a ledge in the game, why can I do so now? But when it is integrated as a gameplay mechanic it loses its dramatic effect since it is repeatable at various locations. I do not want to say this approach will not work in reality but in my opinion the use of scripted events and director agents needs special caution to not break the credibility of the game world for the sake of dramatizing effects. This would lead to just another case of razzmatazz like the particle effects.
I don't think Sande was saying scripted events or animations are cut scenes.
Repetition doesn't automatically make something less dramatic. Repetition can heighten drama as it raises expectations which can be twisted slight for effect. You've jumped across a chasm repeatedly during the course of the game, and, although you've hung by the edge of the ledge a few times, you should be able to jump across this one with no problem. But suddenly, you do end up hanging on the edge. Why is this time different? Is my character exhausted? Is the world shaking? Will I be able to save the princess in time? Curiosity and drama ensue.
Though not strictly cut-scenes or even animation, the context sensitive taunts the game uses are personalised to each charater and each specific event or combination of events. Their use is outside the direct control of the player but they are used to reinforce the differents of the individual "characters".
Even though the player had experienced a mechanic such as clinging to a ledge before, maybe now, in this scenario the player barely avoids the persuer as they miss the jump and fall to their demise. Or maybe the player falls into the chasm after the ledge crumbles. This type of direction can be used in any number of events without tearing the player from the games imposed reality.
These problems have been accounted for and eliminated at the end of Call of Duty 4's Wetwork level. It is the first mission (after the training), on a boat at night while raining. After defeating the final group of enemies in the level, a cut-scene informs the player that the ship has been attacked and is sinking. The player is then forced to follow the squad as they escape the inside of the now slanting, rocking ship. Once outside, the ship starts to slant significantly as if the ship is about to capsize. Their helicopter is hovering on the other side of the ship. As the ship is tilting back, an increasingly steeper angle for each of the squad members to run up and jump off of is created. The player, being the fourth and final person to make the jump, knows the jump WAS possible since the player has seen three others make the jump, but the now steeper angle has both slowed momentum and increased the distance to the helicopter. If the player jumps at the right time and grabs the helicopter, hanging on the edge, and they are required to press the jump/climb button to pull themselves onto the helicopter. Also note that the jumping target is only wide enough to accept a well aimed jump; a diagonal jump will result in failure and a nearby respawn.
Yes, this approach needs to be articulated again and again. Film language has been mis-used so often in games (e.g. cut scenes) it's important to show how it can be translated to gameplay.
Another example: In addition to scripted events, we can mess with entirely systemic gameplay as well to serve narrative. I'm reminded of the first time I messed with a combat system back in 1999 to force a player defeat no matter what the player did in order to further a story.
However in that game players were already educated to the fact that the narrative was king, and expected things like that. It's very important to ease players (as much as developers!) toward such techniques to avoid anyone calling foul.
But as you say yourself, you try to encourage using tools from every department to convey story. My guess was and still is that there are still few tools that are special to games, so it is actually neccessary to fall back on tools from other forms of art, not just mandatory. So I think it still needs some time and experience to develop these tools - and getting narrative designers on game projects early will help, I believe.
Once on a project a narrative designer should be able to experiment though and these experiments are what I refer to as (financial) risk since they may fail. Hopefully we will see these experiments happen more often in future games that even if they fail we still get to learn somethihng from it.
Quite quickly a player will develop a precise sense of exactly how far he can jump. Scripted sequences are spotted immediately by players and knowing that things are going to play out in a predetermined fashion does the exact opposite of raise tension in a dramatic moment - it saps all life and energy from the game while the player waits for the sequence to complete.
In order for there to be dramatic tension, the player must believe that there is a chance for failure.
Such situations occur all the time in multiplayer games due to the human element involved. A 1v4 comback in Counter-strike. The time you backstabbed the entire team. Getting killed by a flashbang. Simultaneous kills. In some sense the beauty of these moments relies on their rarity, however, so I don't know whether the player would believe it if every moment of the game was a thrill ride no matter the skill of the player. We game designers need to learn to cheat as well as our players have.
"As an example, picture the player chasing after a criminal across the rooftops of closely spaced buildings. At one point, the player must jump across a gap separating two building, and the designer wishes to increase the drama of this action. The animator creates a sequence where the player character falls just short and must grab the ledge. A brick breaks loose under one of the character’s hands, but the character still manages to hang on. The character grabs the ledge again and waits for the player to direct him to climb up or shimmy over to a nearby fire escape platform. To make this look good, the character must come down at just the right place to miss the ledge by only a little and to grab a brick at just the right place."
This is the tricky part. I also believe video games need to work on failure. Right now most games have simple failure systems: bad performance whittles away your health until you are dead at which point you must retry. I'd like to see a game where failure was part of the gameplay, where failure impacted the story in response. I'm not sure if failure without gameplay penalty would be effective, but its the extreme end of the scale and there are many points in between that would make gamers happy. Auto-adjusting difficulty (when it works well) is merely a subset of a huge number of dynamic responses to player failure that should occur in games.
The key here is that if the player doesn't perform an action, she falls. That's why there's drama - versus the example "The character grabs the ledge again and ***waits*** for the player to direct him to climb up". i.e. the character will hang there while the player goes to grab a coke. Not much drama.
The trick to all this in story telling is the result. Does the player 'die' and start again from a check point? Or do they fall and lose whoever they were chasing/get caught by whoever was chasing them, thus opening a different story branch. The more branches, the more complexity, the more time, and thus, the more budget. And I think budget may truly be the limiting factor to adding great drama/story to games.
The two recent Tomb Raider games from Crystal Dynamics involve situations where Lara jumps for a ledge and grabs it with one hand instead of two. I've played both games for several hours and I still don't know precisely why or when I might fail to grab a ledge completely after a jump. I know it's possible but it doesn't happen that regularly so when it does I feel lucky that I made it.
If such a mechanic was tied to the narrative of the game, with failed jumps occuring more often during dramatic sequences it would serve to increase the tension and the feeling of "only just making it".
The player already knows such a mechanic (That they might fail to grab the ledge fully) is in place but without being able to exactly define the conditions required they are unlikely to feel cheated if it happens more frequently in dramatic moments. Of course with any such system overuse would greatly decrease the impact.
I suppose it's possible after hours of playing that you still don't know your limits in a game, but my experience suggests otherwise. I played Assassin's Creed and got a pretty good handle on jumps I could make and couldn't make.
I suspect in the TR games that at some point during your jump you realize 'uh oh, it's too far'. If at the moment you would otherwise fall, you suddenly find yourself hanging on a ledge, you might consider yourself lucky once. Maybe twice. But if you keep making jumps that you otherwise wouldn't make specifically in dramatic situations, you'll realize you have no chance of falling and that control has been taken away from you. And actually, the more often it happens, the LESS tension it will create.
But the topic of discussion is whether *scripting* adds drama. GTAIV has plenty of scripted events - barrels rolling out of the back of a truck, police cars ramming buildings all around you - but none of it is dramatic because the player quickly realizes he has nothing to do with the script - it's happening around him rather than to him and all that's required is to wait the script out.
This response is about to get long...
Think about this jump script in a larger context. You're chasing someone and you have to catch them or you fail the mission. You come up to a place where you have to jump and it's a huge jump - possibly the longest you've ever had to jump. You think 'there's no way I can make that', but you have no choice, so you jump. Low and behold, you barely make it. Is that dramatic? I'd say no because a) as a player I recognize that this impossibly long jump is just that - impossible - and b) shortly after realizing there's no other option but to attempt the jump, I realize that it's been placed here purposefully so that the game can make the impossible possible for me. I know that I will make the jump because there is no other option for me.
What if the jump is extreme but doable? So, now I come to a jump, a jump I'm pretty sure I've made before, but I barely make it. During the time it takes me to scramble up the ledge, my prey has gained ground on me. The more this happens the more annoyed I'm going to be because I will recognize that this is a purposeful delay and nothing I can do will get me to make the jump without having to scramble. Particularly if I have to replay this mission. Again it takes drama out of the mission.
Now, if I come to an extreme jump and there is a random chance on whether I barely make it or not, or if I always barely make it, but I must activate some other game mechanic to save myself, there is the potential for drama. Not knowing what's going to happen means there's an opportunity for surprise - an opportunity that does not exist in a scripted sequence.
The point in Tomb Raider wasn't that I didn't understand Lara's jump distance, but that I have yet to figure out the circumstances which would lead to her having to scramble up onto the ledge (Requiring the use of the action button to grab with both hands). Sometimes I'd make the same jump and not have a problem, but it never felt like the game was cheating rather that I'd somehow not been quite as accurate with my jumping as I had in the past. I could accept that as a minor failing on my part, I'd been close but not close enough. It also never happened enough for me to really be able to calculate the exact criteria for it.
The reason I brought it up was because it provides a situation where you might just manage to grab the ledge but that didn't feel like you were being cheated. It also requires an action on the part of the player to ensure that they don't totally fail the jump.
I think the concept as defined isn't really the best soluton. Instead of always making the player "nearly fail" a particular jump in a particular sequence it should be a case of providing near failure under certain dramatically appropriate circumstances. Sometimes the enemy would be so far away that it would be more appropriate (read less frustrating) to let the player make the jump, and in other the enemy might be so close that failing the jump might simply seem artificial. But if a range of circumstancs, were right it would fail the jump for dramatic purposes.
That's the point I made in my previous response - the result of your jumps weren't scripted AND it required a gameplay action (action buttons) to avoid falling. You didn't know what was going to happen. On the other hand, in a script, **by definition**, the same thing is going to happen every time regardless of what the player does.
"But if a range of circumstancs, were right it would fail the jump for dramatic purposes."
Now you're talking about tracking quite a few variables, which leads me back to my earlier comment about budget, or rubber banding which is just as bad as scripting - i.e. no matter how far behind you get, you know you're still going to catch up.
However, picture this - you're almost about to grab your prey, but this time, a jump you've had no trouble with the last 5 times you tried this mission is suddenly unjumpable. Despite figuring out the best way to close on your prey at this particular point, you're not permitted to catch it in the name of creating tension/suspense. Essentially, you're not allowed to succeed until the mission says 'okay, that's enough drama for one mission'. Is that dramatic or just annoying?
In one of the later missions in GTAIV you have to follow a guy with a trunk full of drugs with the cops after you. You will quickly realize that the cops are scripted - i.e. they won't take you or the guy you're following down until he stops and gets out on foot. There are a lot of sirens, a lot of police cars crashing around you, but there is no threat to you while the script runs its course. You can drive leisurely around until the other guy stops. There is no tension in this mission until you get out of the car and the cops start chasing you for real. Then all of a sudden there's lots of tension - 20 cruisers with cops firing at you, your escape vehicle about to catch fire, and a threat that if you fail to lose these cops you're going to have to sit through that crappy script again creates some real dramatic tension.
So there's where the drama comes from - you have drugs/money, you have to escape the cops. The more destruction and mayhem you cause, the more cops come. Now you can have fun. The player understands the rules and success or failure is in his hands.
How can you add more drama? With gameplay elements - road blocks, special intercept vehicles. With story elements - one of your friends calls and needs help immediately or someone's going to die. These aren't scripted events - they are simply gameplay elements added to the mix.
Now, Hawkins would suggest scripts to increase drama at specific moment, forcing the player into some contrived situation in the name of enhancing their experience. I say, by their very nature contrived moments steal away any tension or drama.
Your article is so obsolete and hardly scratches the surface of the real problems and issues pertaining to good Games Design.
Interesting enough, after I read Sande's article I saw a job advertisement for the position of a narrative designer. Reading through it, somehow confirmed the problems that the industry has while it tries to express its new needs through an old vocabulary. When I read a sentence that says something along the lines "We need a writer/narrative designer, BUT s/he should also understand what gameplay is", for me this simply means there is a new and central role here, but we cannot express what that role is and need to use "buts" and "too" and "also"s. In addition, this makes me think that it is yet too difficult (or unimaginable) for company managers to change the typical/traditional organizational structure of their company to give the new emerging creative roles in the development process the position they need to be in.
You need someone who can write, design and direct, but this person isn't at the center of the process, it is in a position that still reports to the Lead designer, cinematics director and lead artist. This looks like the order of things needs a change. Maybe the problem is that noone knows yet how to talk of a hierarchy in which a designer might work under a writer or "narrative designer"? Worse than that: you want something more than a "plain" writer, a "narrative designer", someone who can direct and design too, so this is actually a quite big thing in itself; but then you expect this person also to write the game manual and the press release etc. Suddenly the word "writer" means also "typist", "journalist", "copywriter" etc. The girl/boy for everything regarding writing? And that is exactly where the typical (and dead wrong) perception of the writer in the game industry shines through again: "Well we already have a writer, we are not going to hire someone else to write the press release." It's kinda frustrating to be confronted with this mindset.
-The Ludosphere
IGDA Writing Forums
Good game writers are hard to find, at least currently, yes, but game writing isn't any more wide of a discipline than programming or art or system design is. The center of your process really needs to be a director that understands how they all fit together, not specifically a writer, programmer, artist or system designer. But generalists are *never* as good as specialists in any given field. You shouldn't try to combine those roles, you'll end up with a weaker product in one or more of those areas. The director role needs to be able to focus on directing, not doing the other stuff.
Second, it seems many of those commenting are staring at a tree and ignoring the forest. Debating the merits of one example is hardly the point of this article or my book. I'm obviously not going to go through the whole concept of my book in one comment to an online article, so if you want more depth feel free to read my book.
But I will say a could things. Cheating does work for the mass market. Yes, a few people will discover that the game is cheating and it will ruin the immersion of the moment. But as long as the sequence is done reasonably well, with some consideration for the psychology of the average player, most people will not notice. The benefits of a better experience for many people outweigh the detraction for a few.
One of the commenters mentioned an interesting point. For dramatic tension the player should believe there is a possibility of failure. The key word there is believe. Get them to believe and they will feel it, but it does not mean they actually have to fail. Of course, in my example I do mention that if they go too far outside the parameters of the sequence they can fail just because it can become too obvious, but that is more of a implementation detail than the idea I was attempting to convey. Most people do not want to die. The average person does not want to play to same sequence over 30 times in order to beat it. Hardcore gamers may enjoy it, but they are not the only market out there. When do you hear someone say, "It was awesome, after missing the ledge 30 times I finally made that jump."
I used to game master pen-and-paper RPG's. My policy was not to kill a character unless they did something really stupid. I might penalize them for making a mistake, but no one wants to keep starting over. The people I played with wanted to have fun. If we wanted a mental challenge we would have played chess. And actually we often did that at other times. The point is that there is more than one type of audience, and you should not assume that every audience wants the same type of game. Figure out what your audience will find fun, and do that. Sande is only trying to help provide the tools to do this for a particular audience.
Anyway, I hope this is not too rambling. I did not have time to sit down and craft a better argument, so it may not hold together as well as I would like. Hopefully I have gotten across my main point however.
One final point is that much of this is derived from opinion, anecdote, and personal observation backed up only loosely by data in the case of games. If someone wishes to conduct more thorough experiments to determine what people do and do not find fun and acceptable I would love to hear the results.
My humble thanks to anyone who has bothered to read this far and I hope you find what I wrote useful (as well as the article Sande wrote).