Cameras, Perspective, and Empathy

I am currently undergoing the process of designing my first game, and I have a huge toubt in my mind. Long story short, I’m working on a 3D graphic adventure that will be based in one single house that undergoes major transformation during the gameplay, and I’m having a hard time deciding how to set up the main camera.

I believe that watching the protagonist’s face is a key role for empathy, people are way more likely to feel connected to a character they see and perceive, seeing facial expressions and reactions in real time create a sense of attachment that is crucial, especially in a game where the story is everything. On the other side, a total indoor game is not the best option for a Mass-Effect like camera as it would need a lot of objects culling to run smooth on the entire building. There are many options that came to my mind, but I really can’t decide which one could produce better results.

1) 3D person fixed-side: like This War of mine, I could create a lateral camera that renders the scene from one side. While it would probably be the better way not to hide stuff from the view, it is the most used camera of history, and being the project a one-house show, it’d take away not one but five view sides out of six, giving me way less space for the puzzles.

2) 3D person fixed camera: very much like Siberia, each space will have its own camera set up, and when a player enter that area it gets activated. This solution would give me three sides to put the puzzles, but it would also make the game look old, static and probably give the player too many cues about how to solve the challenges.

3) Orbital camera: an orbital camera would be a camera that rotates around the room in a 360 space, always focusing on the room center. When the camera is above, the roof is culled, when it is under the scene, the floor is. With this type of settings, I can use all the sides of a room, but it would be long and hard to code, and it would make the entire scene very artificial. Looking at a room from below really sucks, especially if you’re on the second floor and there’s another room you’re hiding below.

4)First Person Controller: this solution seems to be the best one in terms of gameplay, but it goes straight against what I’ve said in the post introduction. Unless I make the player live inside a house full of mirrors, it will take away a lot of empathy, and it will be harder for the player to link with the character.

All those solutions have pros and cons, there are also a lot of hybrid possibilities that could be used, but still, I can’t decide what to do. I’ll probably be testing most of these solutions over time, but I still wanted to start a conversation about it. What do you people do? Does FP == less empathy? What camera would you use for such a game?

What if you go with the first-person camera, but then also have a head shot of the player character in a panel somewhere? This is commonly done in FPS games to show health, but you could instead use it to show emotional state (assuming you have some way to simulate that). That might give you the best of both worlds.

To play devil’s advocate to Joe’s suggestion, I’d prefer a holistic arrangement with a single view instead of an overlaid head shot.

Traditional thinking is that first person is better for games that are focused on interacting with the world, while third person is better on games that are focused on the player’s character (PC), since that’s what the player’s literal focus will be on.

What about a third person camera that follows over the shoulder when the PC is moving? When the PC is stationary and turns, the camera could stay in place until the PC moves. That’s a fairly modern camera scheme. Over the shoulder works well for movement, and a stationary camera when turning in place allows the player to see the PC’s front.

And if the PC does something interesting, such as interacting with an object in the room, the camera could zoom into an appropriate close angle, which could either be programmatic or specified manually for each interactable object.

This way the player can see the PC’s body language, even if the camera is just over the shoulder. And more interesting events could take advantage of what should feel like a fairly natural adjustment in camera angle.

5 Likes

A thing to keep in mind with first person views is that there ostensibly is no player character. The PC is the player and vice versa. Add to that that FP is only seriously beneficial to shooting gameplay, and it starts looking a lot less rose tinted. You can argue that it’s more immersive, but relying on FP for immersion isn’t a good sign.

If games handled nuance better (at all), I would be more inclined to agree, but it’s far less important than it should be. In the middle of action, the only way for the player to reliably pick up on the character’s emotional state is if it’s expressed mechanically, because it’s almost assured that the player has a thousand other things to worry about.

Even in a cutscene, there are still the usual issues of cinematography to deal with on top of the costs involved. Character animation ain’t cheap, much less when it’s expressive, and that cost is in performance as well as labor.

2 Likes

I do agree with this. Besides, I wasn’t planning to have any UI, and having just a big face in the corner won’t help.

You know what? I like this hybrid approach. Camera could even be moved independently around the character when it’s not moving, and get closer to its shoulders when it moves. This way, all the room (with the exception of the roof, maybe) is usable, but I still won’t have any problem moving in closer spaces. Thanks, that’s a great hint!

Generally speaking, I tend to agree, but there are some exceptions, for example Portal, while being a TP has managed to make a memorable character, and The Vanishing of Ethan Carter has been a huge success while being a FP puzzle game.

You’re not actually asking how to make the player feel empathy with their character. You’re asking how to make the player feel empathy with you. The problem you’re trying to solve is that you want to show the players how they are supposed to feel.

If you have written a good story, they will feel that way automatically. You won’t have to show them the character’s face, and even if you did, if they don’t already feel that way they will think you got it wrong. Nothing good lies down this road.

2 Likes

That’s a really good point. When I first played Mass Effect, I felt a certain dissonance when the protagonist took initiative to do things that I didn’t directly tell him to do. It was a departure from the traditional 1:1 mapping of player to protagonist in RPGs. It does distance the player from the protagonist, but many people felt like it still worked. I think it’s worth exploring how far one can take this distance. After all, when you watch a movie or read a book, you’re completely separate from the protagonist and yet you can still empathize. But it’s certainly something a designer should approach while keeping your advice in mind, too.

Something I was thinking about after that post, incidentally: sometimes the protagonist isn’t supposed to be you. It’s supposed to be someone you control, or can direct. Like imagine a sniper team, where you’re the spotter. You tell the sniper what to shoot, and he shoots it, but how he feels about shooting it is another form of feedback that tells you whether you’re doing a good job.

So there are (or at least could be) games where that’s a valid part of the gameplay. The question is whether your protagonist is the player, or not.

2 Likes

Pardon me for responding out of order, but…

I would say that the ambiguity in the original post (and their specific use of “protagonist,” which implies an actual character) means that no, the player is not the protagonist (nor is the protagonist the player–I’m struggling to find an appreciable difference between the two phrases).

No, they’re asking how to make the player feel empathy with a character in the game–the protagonist. They didn’t specify whether that would be a character separate from the player (such as Shepard, or Booker from Bioshock) or a self-insert. But, again, the use of the word protagonist implies the former. Additionally, note the use of the word “empathy.” An assumption can be made, though clarification is probably needed.

Worth pointing out–I agree with you about making a player feel a certain emotion in response to the story, but this isn’t necessarily the case for empathy. Empathy requires an empathetic character. Sam Fisher evoked empathy in Double Agent when he lost his daughter, because of the effect it had on him as a character. If he hadn’t changed at all, if a bunch of horrible things happened to him but his character in the game didn’t change at all–he would not evoke empathy.

Miriam Webster:

We’re (the OP is) not talking about evoking just any emotion from an audience, but specifically the emotion governed by the emotions of another person. Showing the emotions of that other person is critical.

The character in the game does not actually have emotions. What it has is a visual approximation of what the developer thinks emotions look like, with which that developer hopes to communicate what emotions the player should be feeling right now.

But no matter how you slice it, players have emotions of their own. And if that character is supposed to be the player, when it doesn’t look like the character feels like the player feels, the character is wrong. The player has real emotions that really exist. The character has fake emotions invented by a developer.

It can be productive, however, to show someone else’s emotions so you know how they are feeling. That will impact and affect the player’s emotions, but because it is not supposed to be the player, it’s not universally wrong if the player feels differently about the events.

I’m glad I could actually generate such a positive and intriguing discussion about the topic. :slight_smile:

I think your comparison perfectly hit the spot. Mass effect is a game where you’re supposed to chose between A and B many times, and modify the world accordingly. The game i am planning instead, is closer to a book or a movie, the story is there and you just have to bring it forward by solving enigmas and puzzles. The dissonance in ME is also created by ambiguous choices which usually don’t produce the desired action by the character, but this is nother story…

I’ve already replied to this, but it’s worth expanding the answer probably. I believe that there is a huge difference between crafting an environment where the player builds his own story and letting him play the story you’ve designed. There are many games where player can create their own custom avatar, pick up names and nicknames, and forge their experience. In that case, you are totally right, and emotions are just mechanical responses to what the simulation believes the player is feeling right now. In the other cases however, players are reading a novel, they are experiencing the story of their character and seeing his emotions. Empathy comes a step later probably, you can feel it or not, but surely the aim is to tell what the character is feeling, not what the player should feel. You can have a dramatic scene and make the player lough at the same time.

Trying to be more specific, my game is going to be focused on just one character, with very few interactions with other agents around the world. It’s going to be HIS story, and I’ll try to let the player understand what he’s feeling and what he’s going through. With that I don’t mean that the game will be sad and break everyone in tears, but there won’t be much space for custom experiences. I’ll tell a story, and the player will “read” it through the game. For this reason the camera design is important, it’s really just like creating a movie script.

That’s true. The question is whether anyone wants to play the story you’ve designed. I’m not saying they don’t - the “visual novel” is a thing that exists - but it’s worth asking whether the people who want what you’re making are the people you want as players.

I had a VN-style introduction planned for my current game, but no matter how much I love the idea and think it’s awesome, I am pretty sure the average player will be frustrated and annoyed by it - because the kind of person who likes a visual novel is not the kind of person who wants to play the game that comes after the introduction.

The person who wants to play the actual game, by contrast, wants me to shut up and get out of his damn way so he can play. And that means by the time he gets to the part where he might need a tutorial to understand what he’s doing, he’s not reading anymore; I’ve burned up all the goodwill I had, and now he’s just closing every window I open without reading it. So he doesn’t know how to play, and he isn’t learning how to play, and he probably ends up ragequitting the game and leaving a bad review.

I’m probably going to turn the VN introduction into a YouTube video. That way, people who like that sort of thing can go watch it, and people who don’t… never even need to know it exists. I’ll get to go stroke my cane on the internet, and some people will watch anything that looks even sort of like anime, so I’ll get my validation from the audience. But I won’t get it at the expense of the player experience.

A lot of what game developers (including me) like to do in their games is just masturbatory, and doesn’t make anything better for the player. We’re all fundamentally going “look at me, look at me” with everything we make, and the inherent promise is that whatever we’re doing is worth looking at. Every time it’s not, we have a “crying wolf” situation, and eventually people stop looking.

1 Like

Depending on your input system it might be worth using cinematic cameras. Cinemachine is free since Unity hired the developer. There’s lots of camera techniques that can guide and enhance the emotional impact.

Big assumption to make. I’ve read everything from G-Senjou no Maou to Saya no Uta to Katawa Shoujo, and played everything from Watch Dogs to Mass Effect to Sim City 4.

Not really.

Ah, nvm. I thought you were talking about a different game.

Will the delivery of emotion/state of mind be during gameplay, or during cut-scene/monologue/dialogue sections?

Honestly I can only remember 1-2 games that have delivered emotion through ‘animation’ enough to cause empathy, and of those none were never the result of facial animation.
First game that comes to mind as I sit here typing is Ori. Ori delivered emotional tone through animation, full body action, movement and poses.
Another one is either Uncharted 2 or 3 where your wounded and hanging off the cliff in the train car. Nathan is visibly injured and can’t move well, the animations of his movement are labored and when he gets slammed into objects the cinematography of the character and camera combine to deliver a pretty believable scene.

I can’t think of any game (certainly not Mass Effect) that delivered believable empathetic facial animation - during gameplay. Do you have a game reference you are using?
Even the highest quality productions have to over exaggerate key facial poses to portray the simplest of emotions during cinematics.

1 Like

Good point. Don’t underestimate body language. That’s one of the reasons why I think a third person camera is good for this game concept; you can always see the PC’s body, even if you can’t see his face.

1 Like

What a great thread, such good ideas around really difficult goals such as empathy.

Isn’t it interesting how game genre is a description of camera? First Person, 3rd person, top down, isometric, etc.

The games which have connected with me the most at an empathic level:

Ico - 3rd person, sculpted user-controlled camera. Maybe the best game ever

Heavy Rain - Cinematic / quicktime events. One of the most emotional opening scenes

Red Dead Redemption - 3rd person orbit camera

Basically anything Naughy Dog does ! - 3rd person sculpted orbit cameras.

Is the camera a factor in the empathy? Sure but not as much as good storytelling and performances.

You’ve got all the tools you need in Unity with Timeline and Cinemachine: FreeLook orbit cameras, cinematic cameras, collision detection, procedural cutscenes of variable scenarios.

Unity 5.6, Timeline, Cinemachine

Cinemachine v1.5 + FreeLook

(scroll down to see posts about how to use if you’re not using Unity v5.6 / Timeline / PostFX)

Here’s a quick video of Timeline and CM working together

We know the documentation is lacking, so until that’s done if anyone has any questions, please don’t hesitate to ask.

I generally prefer not to see a character’s face in third person, whereas I really like the third person perspective in general. Even when I see body language there’s still a sense of continuity between me and the character, it kind of feels like I’ve just got a really long neck hehe, and I can pretty much ignore the back of their head. But the moment I see the face it crosses the line and I really feel disconnected, I start reacting to and not with the face.

So if I were going to try to convey emotion I would definitely try to do it through body language, since there’s still some sense that it’s just my body, my gut, reacting to something that my mind hasn’t consciously processed, and in that way it’s very suggestive to me as a player in terms of feeling an emotion.