Game Developers Conference Notes

GDC Canada 2009 Notes by Scott Inglis

Robert Gutschera

The Amazing Society
Lessons from the world of paper games
Book with Richard Garfield and Skaff Elias
What is luck
Pros & Cons
Luck vs Skill
Hidden information
Luck: uncertainty of outcome = all games have some luck
Not necessarily from dice/cards/rng
Chess contains some luck, Elo rating system
Chance to win (random against best in the world) 1 in 30^50
About as much chance as winning lottery 7 times in a row.
Luck/skill not opposites, orthogonal

      LS       HS
LL   ttt    Chess
HL  slots   poker

Most VGs LL, HS
Luck can increase skill, know when to press your luck
Returns to skill = Skill - Luck
Luck sources: randomizers, RPS, Human Ignorance

Luck increases base of reasonable opponents
Increased variety of gameplay
Gives catchup mechanism
Adds psychological interest
Something to blame losses on
Games now have matchmaking systems, still imperfect

Luck can be confusing, people are bad at probability
Randomness can mask useful feedback
People like to feel like they are masters of their own fate

Experienced players dislike luck if it can trump pure skill
Designers tend to be prone to this thinking
Sometimes you should listen, sometimes you shouldn't
Constantly shrinking ladder removing bottom rungs
Solutions: good matchmaking/luck
Hidden info: private info/complete unkown
Any source of luck contains HI
Some kind of luck is needed for HI
Luck can control calculations i.e. Chess vs minis
Random damage in rts eases player's job
Players accept simultaneous choices and private info over explicit randomizers
Preplan luck (luck then react) v postplan luck (plan then luck)-die chess or chess die
Entrenched audiences are tough, have to fit some expectations
New platforms are an opportunity, popcap would have better chance with luck rts than blizzard
WoW: luck (drops), persistence, and skill phase out luck (hit stats) so skill takes over near the endgame

Don Mattrick

Different business dynamics canada/us
Vancouver in the 80s, banks
DSI almost acquired by konami before eac
Ea sf only 200 people
High school friends formed dsi
Evolution, first game
LOL, remade as spore
Hollywood/tech evolution differences
More global than film industry
Canadian modesty in game industry
1/3 if hit games Canadian
Canucks game :(
Sony/microsoft banter
Rare in a hamlet
Working with UBC and USC
Call from S.S.
Mentioned SFU as good school for games
Bill Gates never graduated before MS
No one entry point for industry
Find a great team
Great meeting industry icons miyamoto
What is this thing you call iPhone
Indie film era of gaming
Big studio employees like to scale down
9 marketing frames of film industry
Great things from MS at E3
Big Park debut at E3

Dan Irish

Threewave Software
Making Great Multiplayer
Not about mmos but fps games
Multiplayer sells more games
Keeps games from being returned
Fewer rentals
Keeps community engadged
Started with quake ctf map
Worked with unreal and source
Objectives and spawners direct action
Team based games should be focused
Need one generic type
Need one specific game mode
The more iconic the better
Distinctive levels/landmarks in levels
Matchmaking tools important
Gotta fix bugs and offer dlc
Know the single player
Establish your pillars
Known your audience

David Freeman

www.freemangames.com
www.beyondstructure.com
6 layers of a great game character
Creating Emition in Games (book)
300 techniques
Remember to check library
Worked in film industry
Worked with blizz employees, jaffe, others
6 Deepening - layers, pain, aesthetics, understated or angled spirituality, wisdom insight, responsibility, self sacrifice, mystery 
5 Empathy techniques - wall-e truck contents/dream/pet as example.  
4 Char arc - how characters grow. Slams are when a character deals with their fear/limitation/blocker/wound
3 Quirks - individualizes character. Lesser traits.
2 Char diamond - rep key facets of char personality. Effects how char see speak think acts. Need 3 - 5 traits to make npcs interesting. Try to be unexpected to avoid cliches. Look for nonverbal traits. Oracle = Gandalf + Ironic/witty and motherly. Skewed differences != masks. 
1 Truthfulness - do your research

Ray Muzyka and Greg Zedchuck

Bioware co-founders
Narritive
Linear/non-linear storytelling
Narritive goes outside game
Delivering genuine emotion
Giving players choice forces them to participate in the narritive
Narritive in games has evolved
Story-in game narritive
Social networking-outside the game
Narritive for each of bartle's character types
Games are traditionally linear, branching out now
User created content as external narritive
First vs third person perspective effecting user interaction
Environments telling a story
They create settings before stories
Activity chains that interrelate
Players having a major influence on the world
Taking a jab at Fox
Non linear and external narritive in television

Lennart Nacke

Biometrics and playtesting
Phd candidate in digital game development
FUGA project fun/player experience research
Numbers vs interviews / quantitative vs qualitative
Looking for objective measures of fun
Playtesting traditionally seen as entry job
Traditional playtesting:
Quality assurance technical/design
Bug reports/balancing/qualitative
Make sure it functions - somewhat automated
Balance - design check/ trial and error
Focus groups
Think-aloud protocol valve microsoft
Traditional playtesting can be generalized and biased/ subjective interpretation of behavior/ problems with accuracy
Next gen testing less biased and scientifically useful
Next gen testing:
Gameplay metrics - event related triggered, continuous logging, spatial
Psychoneyric surveys
Physiological player measurement - biometrics
GM - provide empirical insights to player behaviour, usually event based, player deaths for example, spatial data allow level design, heatmaps, construction of personas/player profiling
Pros-objective, quantifiable, trends and behaviour
Cons-hard analysis, limited viewpoint
EMG GSR EEG eye tracking, accelerometers, position and pressure sensors

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License