網路中介模擬遊戲的增命計畫 – 利用精靈有限度地個人化分配酬賞

Abstract

鉅量多人連線遊戲在遊戲市場上的佔有很大的市場，但卻存在著很嚴重的問題，遊戲者在鉅量多人連線遊戲中的移動率十分快速，遊戲本身一開始都可以吸引大量的玩家，但很快的玩家們會失去新鮮感、失去樂趣而紛紛離開，如此將嚴重影響到遊戲的生存。目前的研究過於偏向於如何改進遊戲的畫面與場景設計來吸引玩家，而忽略了一個精巧的酬賞也能夠有效的增加玩家玩遊戲的意願 故本論文提出並說明了有限度地個人化分配酬賞機制。利用三種異質性的代理人建構出一個多代理人系統來改變原有的酬賞分配機制。這三種代理人分別由玩家的觀點、酬賞的觀點與系統的觀點來看待每一次的酬賞。如此可以在保護鉅量多人連線遊戲原有的特性之下，滿足了玩家的需求並讓每次的酬賞發揮最大的效益。根據實驗結果顯示，本機制能夠有效的提升酬賞的效益，增進玩家的樂趣，增加玩家玩遊戲的意願，以延長遊戲的生命期。

Massively Multiplayer Online Game (MMOG) becomes the most popular game in recent game market. But how to increase the life cycle of game also becomes the core problem for game designer. An MMOG can fascinate a large number of players at the beginning. Soon, players feel that the game is not so newfangled and interesting anymore. Players will leave the game easily. Most of recent researches have put too much emphasis on how to improve the game’s gorgeous tableau and scenery, but haven’t noticed that an elaborate reward assignment mechanism could make the players willing to stay in game. This paper proposes a “Personalize Assignment of Reward Restrictedly” model which is a Multi-Agents System composed of three heterogeneous agents, and uses this model to improve the original reward system in game. The proposed idea is to assign reward “restrictedly” and “personally” by use of Personal Agent, Reward Agent and District Agent in game. The results indicate that proposed reward assignment mechanism can not only help the system spend less reward totally, but satisfy the players’ requirement and desire at the same time.