GAME DESIGN:ALGOTRITHM

Download Report

Transcript GAME DESIGN:ALGOTRITHM

GAME DESIGN:ALGORITHM
DMS 316
WHAT IS AN ALGORITHM?
• A step by step procedure for calculations.
• Pattern
• Input -> output
ALGORITHM EXISTS EVERYWHERE
• Military
• Computer/programs
• Medical field
• Life
• GAME
ALGORITHMS IN GAME
• Fun – Casual / Adventure Game (P v E – Player Versus Environment)
• Competitive - Professional / Semi-pro Game (P v P – Player Versus Player)
ALGORITHMS IN (PVE) GAMES
• The Fixed Algorithms made by game designers to challenge players
• The Counter Algorithms created by players to beat the challenges
The Counter Algorithms: Raid Strategies
(its purpose to beat the boss and continue the adventure)
The Fixed Algorithms: Boss Spells
(its purpose to stop player’s adventure)
Fun?
ALGORITHMS IN (PVP) GAMES
• COMPLEX! (I believe this is what creates the Competitive and Intense atmosphere)
•
Balance is a BIG challenge. (The algorithm has to make sure no other-side is over-powered)
•
The division between player’s skill and the algorithm (avoid to create a game that’s to hard to begin”)
•
Character / Hero / Avatar’s advantage (avoid players to think “ you win because you picked that hero”)
•
Smart player’s strategies (After tournaments, Designer has to decide to Nerf or Buff)
• MOBA- multiplayer online battle arena
•
Heroes
•
Usually controls one unit (hero)
•
Defense of the Ancients (Dota-$10m) , League of Legends(LOL-$2m)
• RTS – real time strategy
•
Unique Troops (like our military)
•
Usually controls more than 100 units
•
Start-Craft (SC-$1.6m) , Ages of Empire
Besides the intensity, the awards and fames from wining the world-championships are other reasons why these
two types of games gained popularity.
GENETIC ALGORITHM
• Used in almost all the RTS , MOBA games
• Is a fast problem-solving algorithm that mimics the process of natural selection
• Tries to create the “no strongest” atmosphere by setting up chain relationships between each unique
units. For example 1 counters-> 2 counters -> 3 counters ->1
• Mutation Rates: the out-puts of unique units AND unique upgrades/equipment's
• Feed Forward Neural Network: the neutral force (resource that players need to fight for)
• Activation Function: usually follows a sequence of phases- early game, mid game, late game.
REFERENCE
• http://www.ijcaonline.org/proceedings/icrtitcs2012/number10/10320-1465
• http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6120766&url=http%3A%2F%2Fieeexplore.ieee.
org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D6120766
• Evolutionary Artificial Intelligence for MOBA/ RTS, College of Science of Commerce, Univierty of Mumbai
– Siddhesh V. Kolwankar