 |
|
|
| |
Hitman Designer: Metrics Plus Personas Equal Fun
by N. Evan Van Zelfden, Staff [PC, Console/PC, Exclusive, Design]
|
|
| |
|
July 3, 2008
|
| |
During the recent, Gamasutra-attended Dutch Festival of Games, Alessandro Canossa, a designer at Hitman developer IO Interactive and a doctoral candidate, gave a lecture on how developers can create games by keeping a specific type of player throughout development.
“When you measure things, you can understand them. You can properly talk about them,” opened Canossa. "From the game point of view, it makes sense, because you can finally start making science out of things that were almost only conjectures until now.”
The Persona
During his speech about game metrics and how to use them, Canossa described metrics as “a set of techniques to gather data from players." He continued, "It’s a way for us designers to know exactly where you are, when you are, what you’re doing, how many doors you’re opening, and all these data that can be extremely useful to inform designing things.
But, says Canossa, they’re only useful “if you find a way or a lens through which to read these data, and to make sense of it. Personas are this lens that allows us to make sense of this great mass of data.”
He elaborated on the "persona" concept: “The personas are a mental construct that allow designers to imply players. So, as a designer, I can imagine what kind of player will play my game.”
Canossa gave a literary example of a persona being used by a creator. “Joyce wrote Finnegan’s Wake with an implied ideal of an insomniac reader," he explained. "He was writing for an insomniac that would read at night, because he can’t sleep very much. And the whole book became the masterpiece that it is, because of his idea of implying a reader before.”
Relevance To Games
But how about the Finnegan's Wake model applied to games? Canossa noted: “The value proposition of game isn’t the product itself, but is actually the way it makes people feel. They don’t buy games, they buy experiences.”
On the other hand, “the problem with this approach is... potentially really expensive adaptive technologies, like ad hoc solutions, or AI systems that can adapt and create and shape," he pointed out.
“At the same time, designers desire to maintain control, meaning that they have an agenda, they want to tell stories, and this is our psychological limitation of what can actually be opened up.”
He continued by pointing to a specific example. “I know you’ve all played Grand Theft Auto IV," said Canossa. "But have you tried to take your girlfriend Michele to the strip joint? She doesn’t react! She doesn’t say, ‘Oh, I like it,’ or ‘Why did you take me here?’ There’s absolutely no emotion. It’s like we’re in a burger joint.
“It’s really puzzling, because that’s exactly the clash of telling a story - providing you with a girlfriend to take places - and giving you the freedom to go to a strip joint without actually backing it up by having lines and responses.”
Further variables were raised: “Another irritation to [the clash of] freedom versus storytelling is ultimately the individual nature of emotion, meaning that we develop shortcuts between states of the world, through our experience.”
For example, “if you were bitten by a dog, you’re afraid of dogs," Canossa pointed out. "If you were not, you’re not. Emotion and emotional response is extremely connected to memory. This makes it extremely individual, and difficult to target and engender.”
Measuring And Predicting Fun
Looking at metrics, and tying into the ambiguous concept of "fun," Canossa floated the idea of using biometric feedback to measure fun.
It's an interesting concept, he said, but technologically difficult. For now, designers should try to fill in those gaps by making use of existing metrics as well as the persona concept, trying to implicate a player state.
“From the player’s point of view, the persona is actually an informative, sense-making metaphor provided beforehand by the game’s designer," he said. "As an example, Hitman: Silent Assassin - it’s a title of a game, and the silent assassin is also the preferential persona that designers offer to players as a mode to go through the game.”
Designers can use personas as tool sets by which to compare designer expectations with actual player experiences seen in playtesting. “Once I used personas in combination with metrics," he said, "as a designer I can start evaluating the goals that I set beforehand.”
Successfully implicated personas will guide a game experience, even if the player is unaware. Concluded Canossa, “From the player's point of view, personas are constructs that layer and build, consciously or not, to unify their own actions and to make sense of the game world.”
|
| |
|
|
However, the article implies that Canossa believes that biometric feedback is the only way to get feedback on whether players are having fun in the game--and that personas are a surrogate because that that technology becomes feasible.
While I agree that biometrics aren't ready yet, I don't think they are actually necessary. People know when they are having fun, and more importantly, we can infer whether someone is having fun just by watching them play and asking questions like 'how fun was this game?' This process is called 'user-research' by those who have training in the HCI and psychology.
Microsoft has been doing using these formal user-research (note: not focus groups) since 1997. Bungie, Valve, and InfinityWard have all publically cited user-research as a major reason for the success of Halo 3, Portal, and Call of Duty 4. In the past few years, THQ, Activision, and WB Interactive have begun doing formal user-research (namely, by hiring former Microsoft employees). There are several independent vendors who do user-research work for games (myself included). No need to wait for biometrics. :)
To see how user-research was done on Halo 3 see the article in Wired (Aug 2007): http://www.wired.com/gaming/virtualworlds/magazine/15-09/ff_halo. They tested over 600 users, integrating 3000 hours of user-feedback into the game prior to Halo 3 hitting the shelves.
I love this quote. Quite true. I don't think many developers understand that.
Thank you, Bill, for the compliments; however I never claimed that biometrics are the only way to get feedback regarding player enjoyment.
Regarding the quote "The value proposition of a game isn’t the product itself, but is actually the way it makes people feel. They don’t buy games, they buy experiences." it is actually from Nicole Lazzaro.
If anybody is interested, here is the link to the presentation I held at NLGD - Dutch festival of games:
http://www.itu.dk/people/alec/NLGDAlessandroCanossa.ppt