GDC: Last session of the day

I'm sat in Cory Ondrejka's lovely house eating lovely chili and being unsociable while I just put this last bit of stuff up. Then I'm off to play card games.

The Games Studies Download, Presented by Mia Consalvo, Jane McGonigal and Ian Bogost.

Just my notes.

10.
“How does games music impact a player’s effectiveness?”
(Gianna Cassidy et al, Glasgow Caledonian University)

This study looked at a player’s ability to do missions and do well and complete the game. They picked missions in PGR2 and added different soundtracks; aggressive music, relaxing music, any music (player chosen), silence and soundtracks. They measured speed of player, accuracy, emotional arousal and attention to the game.

The most interesting thing was that the emotional impact of music had no correlation with player effectiveness in the game. The music could really pump them up but it wouldn’t help them play better. Marked improvement came from when the player picked the song themselves.

Scores went up dramatically when players choose their own music.

So game music is not just about world building or impact and player success hangs on the music in the game.

How are you using game music to support or challenge your players?
How are you going to use the fact that players selecting their own soundtrack makes them better gamers?

9.
Moto GP and Unreal Championship. Usability of Xbox Live voice channel.
“What do players really think about voice chat and its usefulness in gameplay?”
(Kevin Hew, Martin Gibbs, Greg Wadley, University of Melbourne.)

They found that players thought voice would be an advance. Good for tactics, freeing up hands from typing in text, etc. However, after the gameplay experience with voice chat, they found problems with usability. There’s a lot of noise in voice chat. Non conversational sounds, conversations not relevant to them, people making annoying motorbike sounds, trash talking and spam. Also players had difficulty figuring out who on screen was the person talking. No connection [between voice and avatar].

Players also disliked the lack of control – no private messaging, no message-reception confirmation. No confirmation if someone had heard what they said or not.

So – voice isn’t always an advance. Players slowly adapted, but most people just took the headset off when it became troublesome.

What specific elements of your gameplay does voice chat […enhance?]

8.
Controllers and interfaces. “Gestural and embodied controllers are fun, but are they good for gameplay?”
(Stephen Griffin, Georgia Tech).

This work was done as an industrial design study; this guy just won an award for a self cleaning kitty litter box. Might be an interesting controller! So what they found out:
There’s a trend toward transparent interfaces. The more simple and transparent interfaces, the better they are.
These weird controllers, they create interaction that accounts for the body. We have to wrap our hands around controllers in an awkward way;

[got moved, missed a bit]

…. buttons are best for complex symbolic action. Designs for new gestural systems should take this into account: don’t throw away the buttons! Think of DDR.. you’re involved, stepping, but the gesture of stepping/pressing down on the pad is very symbolic.
If you saw the guitar hero talk, they said it’s not a guitar simulation, it’s a rockstar performance simulation. You can grip the guitar, and get the performance.. the tilt sensor just adds to that.

Are you choosing the right gestural vs. symbolic control system?

7.
Halo 2 for Xbox Live. Networked PC games. MMOs.
“Does the presence of other players make an online game more or less immersive?”
(Cheryl Campanella Bracken, Cleveland State University.)

We know that introducing players means more variety over time: are there ways to use other players to make a more immersive experience? The process was very intense ethnographic research, including game journals and interviews.

There are three different kinds of strong presence that are very pleasurable to feel. Spatial-physical is one, when you’re feeling in the space. Then social presence: to what extent you feel the other people are present with you. TV vs. film, when you watch TV live you’re mildly aware that there are millions of people watching the show, but it’s not very present to you. But go to a movie theatre, and you have a stronger social presence. Thirdly, co-presence, which is the first two presences combined. Do you feel the social presence of those other people inhabiting the environment with you?

The most interesting thing was that adversaries … other players they needed to beat .. they were the same as bots. Depersonalized. They didn’t feel any social presence with antagonists, but the strongest form of presence were in collaborative online environments. Any situation where everyone is working together in a digital environment.

Takeaway: Collaboration is an extremely powerful driver of immersion and stickiness.

Where could you add moments of multiplayer collaboration in your game?

6. “Are players cheating as much as we (and other gamers) think they are?”
(Dale Miller et al, Stanford University).

This was a psyche test. They had people watch a 10 yr old boy take a test. They told these people that he was likely a cheater: and so lots of people saw him as cheating, even though he wasn’t.

People tend to be highly vigilant for signs of dishonesty. One dishonest act will label you as a cheater. This is especially true in hi-stakes situations.

Perceptions are really important for players and developers. If players THINK there is a lot of cheating in a game, they’ll see it - even if it’s not there. The other thing that can happen is, if you put people in positions of surveillance over others, they’ll become hyperaware and begin to see cheating where there’s really none. The cynical GM is a good example here.

What concrete steps can you take to assure players that a competition is fair?

Do you really need all of the ranking and stats systems? Are there elements of cooperative play that you could add?

5.
Player controlled cameras.
“What innovative game design users are there for player-controlled cameras?”
(Michael Nitsche, georgiatech.)

Interactive equivalent of cinematic montage is rare. Think of a sniper rifle as an example or if you have multiple cockpit views in a racing game. The player is performing some action to make this shift. Siren had this weird concept of ‘sight-jacking’ where you could see the environment from the enemy’s perspective. We’re not so confused in cinema when there are rapid cuts in montage…

The game here was Fatal Frame 2 (Project Zero: Crimson Butterfly in Europe) which uses a fixed perspective. This idea of manipulating the interface to show us another perspective should be meaningful. That cut should be meaningful. This attempt to get into the eyes in the enemy has meaning. So player-controlled camera movement can be thought of as an adaptation of cinematic montage.

[I didn’t understand a bloody word of this :o) ]

How can your game make more creative use of player-controlled camera cuts?

4.
If you had to read only one paper, this would be the one you should read. It’s an incredibly applicable read. “What strategies to gamers invent to communicate to other players in online games, and can our games be designed to better support these strategies?”
(Tony Manninent and Tonmi Kujanpaa, University of Oulu)

Lots of ethnographic research. Players wanted to communicate three different things in the game space: intentions, actions and effects.
•    How can a player communicate to another player what they intend to do?
•    How can I show you what I’m doing when I’m doing it?
•    What are the effects of what I’ve done?

So if I’ve communicated to you that I intend to kill this person, and I’ve done it in a really amusing way, can I leave the body somewhere for other people to see after? [How gory!]

There are three strategies really undersupported in Battlefield 1942.
The first was gesture. Silent communication between characters. There’s no pointing! So players were aiming their weps in a certain direction, so they hacked gesturing.

They wanted more nonverbal audio. It’s one thing to have dialogue but you won’t always be able to communicate within the fiction of the game world. Can I bang on something that you’ll hear from across the environment? Can I make sounds that other people can hear?

Also they wanted non-violent physical contact, the only physical contact was violent. No high-fives. I love the tickling in Sims, which is probably not appropriate for Battlefield 1942 but still, they want something other than just hacking each others arms off?

So, have you explored non-standard possibilities for interaction forms?

3.
This one was on 2 games: Sacrifice and Half-Life. “Can alternative controllers like eye tracking devices offer a PC gaming experience that is more fun and involving than mouse control?”
(Erika Jonsson, royal institute of tech, Sweden.)

She created modded versions of the 2d Sacrifice and 3d Half Life game. players aimed by looking at the monsters, and firing with the mouse. With HL she tried two methods. Looking in a certain direction, field of view changed with eye direction.

With the 2d, everyone rated the game as more fun when played with the eyes (rather than the mouse). Their scores for using eye tracking vs. mouse were double.
HL was more mixed: the people who had the system with field of view combined with aiming control, they didn’t like it. But with the moving-with-mouse-aiming-with-eyes version, they responded well.

Eye tracking could be an interesting addition to your game. It might make games easier, may make them score better.

What novel input devices are you considering for your PC game?

2. “How can we generate facial animation that combines speech and variable emotion?”
(Yong Cao, UCLA.)

Speech driven faces are common.

But if we take the same speech and change the tone and context, we look different. I perform different if I’m angry or happy. The sounds and tenor will change. This isn’t being captured at the moment and would be interesting to capture.

So you can either manually script this stuff or use something like a support vector machine..

Could your characters’ facial expression be more emotionally specific during speech?

1.
The number one research finding from the past year – and it has to do with monkeys and bananas! This is a study of Super Monkey Ball 2. “How do game events marking success vs. failure affect a player’s level of engagement?”

So you know it’s a bowling game, really. How are players affected emotionally and their attention levels, immersion and emotional arousal, when they’re doing well or badly?

So they brought subjects in, and asked them how they felt when they were doing well or do badly. Then they measured sweat, heart rate, facial EMG activity (facial reactions), the eyes.. they were really getting a sense of how happy they were, engaged they were, etc.

There was more pleasure and excitement in their active failure than in their success. Knocking the monkey off right at the worst moment was more pleasurable for them than getting a straight. This was a cool experience, the visual expression of the monkey falling off into space was so exciting it didn’t matter that they’d done badly.

Passive experience of failure is the worst thing you can do: so if you replay it, they really disengaged. They hated it.

Being actively responsible for something not working can be pleasurable, whereas watching yourself being rubbish is really bad. What does this mean for replay!

Players would be happy when they did well but their arousal and interest would go down. You’ve done the thing you want to do.. that’s less exciting that trying to do it itself. So failure is an unexpected hotspot for pleasure.

How much fun is failure in your game?

[out of battery]

Comments

[Cormac Russell]

I see no sign of you actually putting the latop down slowly and backing away.

[Alice]

I can't believe you're sat behind me and typing that. :D

[Cormac Russell]

Heh, what good is technology if we don't overuse it. Grab me some chips? :)

[Thomas Malaby]

Thanks for the great report, Alice.

There was more pleasure and excitement in their active failure than in their success.

Actually, studies of gamblers have shown this repeatedly. It's the experience of engagement with contingency itself, rather than simply success or failure, that matters.

[Dominik]

very interesting, thanks! Is there anywhere where the papers / presentations can be found? Maybe even a video? That would be great!

[Raph]

My battery did not die, and hence #0 plus the location of the slides are available on my blog now:

http://www.raphkoster.com/2006/03/25/gdc-day-5-top-ten-lessons-from-game-studies-number-zero/

[Jane McG]

The slides and handouts (with links to the full text of all of the studies online) are at www.avantgame.com/top10.htm . Woot!

[Brian 'Psychochild' Green]

A heroic effort in transcribing these sessions, Alice. Very well done.

I found #6 interesting, because I had observed that often: people accused of cheating tend to get a lot of reports against them because the perception becomes that they are cheaters. It becomes hard from a CS point of view because people feel you are lax on cheating if you don't ban all the "cheaters" reported. It's good to see some research backing this up.

My thoughts,

[dominik]

Woot indeed! Thanks Raph and Jane!

[pantomimeHorse]

I wonder if this'll work: "But wait, don't kick me! Studies show that you're far more likely to falsley identify cheaters due to this games reputation for having cheaters!"

I've had exactly this problem with COD2, getting kicked for doing too well.

I find #9 and #4 particularly relevant to BF2 and COD2. It's always frustrated me how no one is trying to tackle the problem of player-cooperation beyond tacking on voice chat.

One big problem with cooperation in these games is how death comes too quickly: even if two players try to stick together, the game is against them as soon as one of them dies.