So we’ve discussed in depth why cinematics can be a problem, and seen some specific examples of these issues. Now it’s time to wrap things up with some ideas for doing things right. That’s not to say that many games haven’t already done a fine job, or haven’t made significant progress. Still I have a few ideas I think will help move things along. So let’s dive in.
Clearly the most glaring contrast between cinematics, and any other part of the game, is the fact that they are most often non-interactive. This of course varies depending on the genre. In RPG’s conversations tend to be a custscene diced up; with intermissions allowing you choose how to splice it together into a cutscene unique to you. That’s definitely a good way of handling it and expanding upon, Bioware has introduced interrupts in the Mass Effect games. These allow you to further splice in certain actions into cutscenes, giving you a feeling of spontaneity. These are a good start for genre’s that require dialogue. Even so, they could be enhanced to provide a much more engaging level of interactive storytelling. How so? Well let’s look at those two pieces.
The way dialogue is designed in games like these breaks the fourth wall by necessity. The reason is that they require input from the player in order to proceed. This usually involves having the user select from a list of onscreen reactions, be they text or symbols. There are two resultant issues there. First, the storytelling is halted; the universe frozen until such a time as the player makes a choice. Second, the interface is an overlay, a menu screen existing over the world with text or icons that are a clear break from the story and it’s universe, to a reminder that this is a game, a piece of software requesting input from its user.
Don’t Break The Flow
So how do we get past this? Well there’s a couple of ways, several of which are already in use. For instance, let’s start with the dialogue break. There are some effective techniques here, one of which is to not lock the player into the cutscene. This is something thankfully more developers seem to be figuring out. You can see a gradual progression in this regards in the Elder Scrolls titles. Note the change from Oblivion to Skyrim. In Oblivion, whenever you entered a conversation, your camera zoomed into the face of the person you were talking to and locked there with their gaze intent on you. This is clearly not conducive to natural conversation.
However, in Skyrim, the system changed, the NPC’s were no longer locked in place. They were free to act naturally, and carry out manual tasks or even lounge about when they conversed with you. Beyond that, your view as also free to some degree, allowing your gaze to wander while you conversed. This was a definite improvement over the previous games. Yet, this is only one way to handle it and while it’s an improvement, it’s still not ideal.
Some games, take the other extreme, and which also has its benefits. In these cases, the games simply take the dialogue out of the cutscenes wherever possible. When it is done well this can be a great achievement, but it doesn’t always work out. The Fable games for instance, often have cutscenes replaced with pre-scripted events where dialogue is delivered by an NPC who is standing near you, while you have complete freedom of movement. This gives you a greater sense of having a natural conversation, but it also comes with its share of problems.
Just as you are allowed freedom of movement, so is the world around you. This can be a huge problem in storytelling, and is one of the reasons many games freeze the world around you during cutscenes. This is because if the world around you is free to move, it’s also free to interfere with the storytelling as well. This can be quite obvious in both Skyrim and Fable, where NPC’s and other components of the world interfere and sometimes ruin crucial plot events.
So where do you find the balance? Absolute freedom can be chaotic, uncontrolled. However, too much control makes your cinematics stiff and unyielding. What about games that don’t even need interactive dialogue from the player?
Know When You’re Not Wanted
A balance can be struck, and a significant portion of it is knowing when to do what. For instance, there are many games that have no interactive dialogue. These games still manage to tell engaging stories, and they have some interesting tricks up their sleeves. For instance, by having non-crucial plot information relayed to the player outside of cutscenes, you can minimize the need for constant interruption from these cinematics.
In cases of pure exposition, there’s often no need for it to be revealed in the form of a lengthy dialogue-heavy cutscene. Often times, the dialogue can be carried out fluidly during something as simple as small breaks in action during gameplay, or sometimes even in the midst of combat or other gameplay. People are capable of multitasking and non-essential information can be provided in many forms. Such as through dialogue from NPC’s or through the use of visuals such as in-world signs or radio chatter. The key is to be imaginative, concise and unobtrusive.
One common method that has become popular recently is the use of interrupt events. These are when your attention is called to some key even taking place in the world around you. This is often accompanied by some prompt from an NPC and usually also by an in game prompt to press a button. This then re-focuses the camera in such a way as to allow you to see some pre-scripted story event. While this is again a step in the right direction, something as simple as the onscreen prompt in the middle of your screen is another fourth wall breaking element, that brings you back to the knowledge that you’re just inputting commands into a piece of software.
But you really do need and want player input, this is a game after all. How is it possible to do that without stopping the game? How do you execute dialogue? Interactions with the world? What about Quick-time events? Loading and death screens? Interrupt events? It’s not a matter of if. It’s a matter of how.
Subtlety Is An Art
So if you’ve paid close attention, I’ve gone through a lot of examples of improvements and solid methods, but with the caveat that none of them is quite right. They all lack three key ingredients; discretion, foresight and subtlety. So what does that mean in practical terms?
Discretion – This means you need to know when a cutscene is needed and when it isn’t. Not every bit of explanation or exposition needs to interrupt the game. Know when to use alternative methods of conveying story or information by using tools such as interrupt events, NPC dialogue, or visual cues. However, when you do need a cutscene, try to make them as concise as possible without compromising the storytelling. It can be hard, but with extra effort, a balance can be struck. Finally, give the player, NPC and world a degree of constrained freedom. Not enough that he/it can ruin the experience, but not enough that it feels unnatural or surreal either. How do you accomplish this?
Foresight -Think things through before you plan these events. Can random NPC’s pass by and make noises to drown out essential dialogue? Can the Player Character walk away from or accidentally terminate the dialogue, causing him to miss crucial plot points? Could multiple scripted events overlap causing confusion and disarray in what were intended to be dramatic or key events? All these can be avoided with a bit of planning. Why not have the ambient noise reduced automatically when important scenes are triggered? Could you also make triggered events be mutually exclusive? So that two major events can’t overlap. Perhaps limit the player character in simple ways that leave him feeling free to interact as necessary, but not free to act in ways that are disruptive to the storytelling.
This could be accomplished by pre-scripting context specific actions for all planned cutscenes. Limiting the areas a character can move during a cutscene by making the events take place in locations with clear spatial limits. You could also make less obvious limits by guiding the player through slowing his movement speed and just slightly distorting his input and characters movement to ensure he doesn’t stray too far. Also, through the use of visual cues that lead the character to environmental elements that can be interacted with.
The player could pick up and peruse through objects in the environment. Or perform basic actions one might see in a conversation. These might even be integrated into the dialogue, slightly changing the way a conversation plays out. They could even be used as a method for expanding dialogue options for the player. And yet, sometimes you just need to get a clear and decisive input from the player, how do you go about it?
Subtlety – This is where things get tricky. So you want to have an interactive cutscene, where do you go? Why to the quick time event! You want two-way dialogue with input from the user? Text options on the screen, or icons! You want to call attention to some special event taking place? Press indicated on screen button to view! Did the player just die for the hundredth time? Press indicated commands to re-load your game!
I may come off critical of these methods, but the fact of the matter is, most of those items are perfectly valid and often necessary components of a game. The problem isn’t their presence, but their presentation. Notice that they all share one key element; the superimposition of real world elements over the game world. It’s by these means that they elicit your input, but it is also by these means that they break your immersion. So how do you get past that?
Visual Cues and Player Conditioning. An excellent example of a game that almost pulled this off was the 2008 Prince of Persia game. You see, instead of always relying on telling you to mash a certain button, the game conditioned you over time to know when you needed to push what. This then could be extended to use in what would usually be non-interactive sequences, as well as to engage in dialogue and even reload the game. Each of these essential actions were tied to visual cues, which made it easy to know when certain actions were required of you to complete or trigger the scenes or events. However, as before, Prince of Persia didn’t quite pull it off. The issue there lay in the lack of variety in gameplay and oversimplification of its implementation. However, let’s look at how it could be improved.
The biggest element is conditioning the player. This must be done in the early stages of the game, with control schemes that are intuitive and easy to remember so that the player can understand what is needed from them in-game. Once the player understands how to take actions in-game, subtle visual cues should be used to make the player aware that their input is needed and what their possible actions are.
Doing it Right
The Shenmue games had a great example of this, in the manner in which you were taught fighting moves. The NPC who taught you would explain the move and perform it. You would then try to replicate it. The key was that, in fighting, the buttons had clearly understood uses. So when he told you to use your arms, you knew to push a certain button. When he said, you kicked too early; you knew you had pushed a specific button too early. All of this was handled in-game, in a manner no different than would be done in reality, thus preserving immersion.
You then begin to layer complexity. Who’s to say a cutscene must always turn out one way? Or that you can’t have subtle input in its direction? I mentioned before the idea of player interaction directing the course of dialogue, but Shenmue again presents us with an interesting way of handling this. In Shenmue’s case it was through the use of QTE’s. Multiple QTE’s could be distributed throughout a cutscene, allowing multiple branching paths through the story or just the cutscene. Fail or pass, doesn’t mean you restart, it means the story alters slightly, or even diverges greatly depending on how linear the game needs to be. However, let’s take it one level higher. Why not allow for multiple diverging paths in each QTE? Why not chain them?
It might not sound clear, but say, in many combat QTE’s you’ll often be forced to have things play out a certain way. However, with a little foresight, couldn’t multiple paths through the QTE be present, allowing a completely pre-rendered scene to play out differently for each person, not only based on which events they made or missed, but which of multiple actions/attacks they choose to take in a scene? If slashing with my sword is X in a game, why not allow for that in a cutscene? Instead of forcing me to press Y to shoot, let me choose how the scene will play out much the way we do with dialogue.
Then where you seal it all is with subtlety in the visual cues. We don’t need giant X, B, Y, A, RT, LT buttons to appear on screen. If the controls are intuitive, and the situation self-explanatory, introduce small visual clues early on and condition the player to their use. Train them to recognize and use them early on in non-essential situations, so that when you need them to act, they’ll know that they have to. You can also introduce more visual cues to be used to represent multiple options. Simple things like changes in lighting or a small glow at the edge of the screen corresponding to a button color – Perhaps an action or stance of a character on screen. It’s only limited by the imagination, yours and the players.
Keep On Pushing
So we finally come to the end of this series. These are just a few ideas, and believe I could go on expounding on these for a few more weeks. In fact, I likely will, but we’ll be moving the topic away from the use of cinematics and into other topics. Hopefully I’ve given you some food for thought on this matter and helped you to see that cinematics aren’t bad in and of themselves. However it is time their implementation was handled better so as to supplement immersion and gameplay instead of distracting from them. I’m sure you all have your own ideas, critiques, and musings. So I’ll leave you to them in the comments! I am eager to hear what you’re all thinking.