Widget Baron Online A Multiplayer Simulation Game Karl Koscher Firat Kiyak.

Widget Baron Online

A Multiplayer Simulation Game

Karl KoscherFirat Kiyak

Widget Baron Online

Multiplayer “tycoon” game based in virtual Seattle

Players compete against each other to create the most profitable widget empire

Urban environment simulated by UrbanSim “a simulation model for integrated planning and

analysis of urban development“ Urban environment influences game

strategies

Current State of Widget Baron

Single player game developed over the summer as a research project

Players can explore the city, set up widget shops, and vote on urban policies

Game mechanics are a bit weak Multiplayer would dramatically enhance

playability Real competition > simulated competition Real economic effects

Current State of Widget Baron

Widget Baron Online

Target audience: casual gamers Target systems:

Mid-to-high level PCs Broadband Internet connection

Benefits: Fun for the user May get players interested in civic issues May get extra publicity for UrbanSim

Non-goals

Not massively-multiplayer (<100) Not designed for hardcore gamers Not an urban planning tool

Requirements

10-100 simultaneous players Robust economic model Player interactions and communications Playable on mid-level systems Open source Fun!

System Architecture

Game server coordinates, simulates economy

Simulation cluster runs UrbanSim May be possible to

distribute to players’ machines!

File server stores UrbanSim run results

Scenario DB maps votes to UrbanSim run data

Game Server

Scenario DB Server

File Server

Simulation Cluster

Lifecycle Plan – Rough Timeline

Three weeks for basic architecture Network communication Server code

Three weeks for refinement Performance tweaks Game enhancements

Two weeks for beta fixes

Lifecycle Plan

Support: Code maintenance: open source,

UrbanSim will maintain code Servers: UrbanSim will provide Customer support: Documentation, but

that’s it!

Risks

A lot of existing C++ code Some well-documented, others not so

much A lot of work

Six weeks is not much time to develop a multiplayer game!

Somewhat dependant on UrbanSim For data from the runs

However…

We are starting from a “complete” game

Final product would be very polished Opportunity to make a very fun game