Transcript File

Designing Mobile Games
1
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
Speaker Introduction
• Name: Greg Costikyan
• Organization: NRC/Game House
• Location: New York
• How am I involved in (mobile) gaming: 3 decades as a
game designer (30+ titles published), co-founder of one of
US’s first mobile game developers, used to edit game
section of Forum Nokia website, now work as a games
resarcher for NRC.
2
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
What Does the Player Do?
• Game design is not about story or character.
• Game Design is about action.
• Not necessarily fast action, but the player takes actions to affect
the game state.
• What does the player do?
• Media assets are the “nouns” of the game—allowable actions are the
“verbs”
• UI allows you to trigger the verbs.
• Each verb mapped to a UI feature.
• In a mobile game, ideally 1 key = 1 verb
• Possibly to net menus, etc., but preferable to keep actions
mapped to individual key presses
• Write down your list of verbs.
• Possible to build a good game with limited verbs: Doom has only
8 (left, right, ahead, back, jump, shoot, switch weapons, pick up)
• Can you see how your list could produce an interesting game?
3
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
Verbs (con’t)
• Will a single key-press suffice?
• E.g., two used in golf games (direction and power).
• Avoid multiple simultaneous key-presses when possible, as many
J2ME implementations don’t permit this.
• If feasible, avoid mapping multiple actions to a single key, or different
keys to the same action, to avoid player confusion.
• A single key to mean “act” or “select” can work, IF the meaning is
always clear in context.
• Game design has two main components: UI specification and
gameplay algorithm specifications. The two must dovetail neatly, and
it is worth thinking about clean UI design from the start.
4
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
Struggle & Challenge
• A game should be a struggle.
• If too easy, it is boring.
• If too hard, it is frustrating.
• Have to find a happy medium.
• Players enjoy overcoming challenges.
• Difficulty settings help.
• Dynamic difficulty adjustment can be used, but carefully.
• Three basic kinds of challenges:
• Physical
• Mental
• Opponents (AI or players)
5
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
Types of Challenges
• Physical: Depends on timing and mastery of the interface.
• Mental:
• Resource trade-offs
• Tricky placement of game objects
• Interacting systems whose behavior is hard to solve
• Combining game objects
• Even if your game is not puzzle-based, think about how to use
the verbs of the game to produce interesting puzzles
• Opponents
• For multiplayer, this is provided by the other players
• In single-player, this is provided by AI.
• Even simple AI can make opposition more interesting to the player
• Example: Space Invaders
• After defining verbs, defining the sorts of challenges your players will
face comes next.
6
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
Categories of Pleasure
• Marc LeBlanc’s taxonomy of game aesthetics
• Always useful to think about what aesthetic pleasures players will
draw from the game
• Sensation: Graphics, sounds, tactile feeling, etc.
• Fantasy: Consistent and appealing background/world/story.
• Can be simple: “An Italian plumber must rescue his girlfriend from
a giant ape.”
• Narrative: Not necessarily “story,” but narrative arc: Sense of
heightening tension and release.
• Challenge.
• Fellowship: Important particularly for multiplayer, but even with
soloplay games, players enjoy talking about their experiences.
• Discovery: Exploration, new things (with each level?)
• “Masochism:” Submitting yourself to the structure of the game.
• Does your design provide each/some of these?
7
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
Constraints
• Constraints do not limit creativity; they spur it.
• The sonnet.
• Application size.
• If targeting older Series 40 phones <=64k
• Application memory space
• When running, application consumes more than the app size
itself—graphic buffers, objects created at runtime, etc.
• For older Series 40 phones <200k.
• Screen size & format:
• Characters should be ~10-15% of height and width of screen
• If a “HUD” is used, it must be simple—ideally <6 pieces of
information.
• Portrait rather than landscape format.
• Processing power (complicated simulations a problem)
8
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
Constraints (con’t)
• Mobile Device UI
• Can generally rely an an ITU-T keypad, two soft keys, D-pad
• No pointing device
• Variable keypads
• The social space of mobile devices
• Handle interruptions gracefully
• Go easy on the sound (and gameplay MUST NOT depend on it)
• Keep the backlight on
• High color contrast for readability in direct sunlight
9
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
How Multiplayer Games Differ
• Players provide the challenge
• Provide ways to help and hinder each other
• Replayability vital.
• Handle player drops gracefully
• “Civil Disorder”
• AI take-over
• Replacement player
• Or design so that the loss of a player is unimportant
• Player Matching
• “Quick game.”
• Challenges
• “Reserving” a game for friends (buddy lists?)
• Use of rankings to match players of equivalent skill
• Short play sessions
• Ideally <=15 minutes
10
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
Multiplayer Differences (con’t)
• “Balance” no longer = right difficulty level
• Instead = all players have equal chance of winning
• However, asymmetric games can be balanced
• Diplomatic games are self-balancing
• Physical: Depends on timing and mastery of the interface.
11
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
Game and Network Issues
• Server-driven games
• Ongoing cost for game provider
• Secure data storage
• Makes cheating harder
• Bandwidth-to-user not normally a constraint
• Peer-to-peer
• Cheaper for game provider
• Cheating easier
• With large numbers of players, bandwidth becomes a bottleneck
• Particularly for Bluetooth, which is always hub-spoke
configuration
• Not feasible with legacy phones (requires IP address, SIP, or
Bluetooth)
• Player matching/discovery becomes a problem
12
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
Dealing with Latency
• Always a problem with networks
• On wired Internet, 100-200ms latency rules out street fighters
• On 2G networks, ~1 second latency
• If HTTP must be used, ~5 second latency
• NRC tests show that UTMS can produce >100ms latency
• ---But in lab, actual network deployments may be slower..
• And generically, “3G” doesn’t solve all problems—EV-DO in
deployment ~500ms latency
• In general, unless targetting UTMS, must always work around latency
issues
• Approaches:
• Turn-Based games (round robin or simultaneous movement)
• Act-whenever
• Slow update games
• Shared solitaire games
• Mask latency with game fantasy
• Untie game outcomes from specific play configuration
13
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
Designing for Community
• Shared high-scores, tournaments, etc.
• But many pitfalls:
• Avoid incentives for player drops
• Don’t encourage newbie-bashing
• No “perfect” scores
• Permanent high scores can be a deterrent
• Chat
• Keypad text entry a problem—taunts?
• SMS for persistent/long term games
• Pathway to Glory use of VoIP
• Friend Finding
• Buddy lists
• Phone number/User ID query
• SMS challenges
• Diplomacy
• Web presence
14
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
The Metagame
• Richard Garfield & Magic: The Gathering
• Anything surrounding the game that increases player interest
• Tournaments/seasons
• Trading
• Offline activities
• Stable strategies
15
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential
No Single Methodology
• Tried to provide a coherent framework here
• This is an art, not an engineering discipline
• Kipling: “There are four and twenty ways/of writing tribal lays/And
every single one of them is right.”
• But in general, if you think about “what the player does,” what
pleasures players draw from the game, and what techical and
business constraints apply, you’ll start from a solid base.
16
© NOKIA
Presentation_Name.PPT / DD-MM-YYYY / Initials
Company Confidential