Procedural Content Generation untuk Meningkatkan Replayability pada Novel Visual Menggunakan Algoritma Genetika

Abstract

Replayability adalah konsep yang penting dalam game design. Kemampuan untuk menikmati sebuah game berulang kali dengan tetap mendapatkan pengalaman baru setiap sesinya adalah nilai tambah yang membedakan video game dengan media hiburan linear seperti film dan buku. Novel visual menerapkan konsep replayability dengan membuat cerita bercabang yang dapat pemain pilih pada setiap sesi permainan. Penelitian ini bertujuan membuat dan menguji metode pengembangan novel visual yang diharapkan mampu meningkatkan replayability. Metode yang digunakan dalam penelitian ini adalah Procedural Content Generation (PCG) menggunakan Genetic Algorithm (GA). Penelitian ini terdiri atas 6 tahapan, yaitu scope & requirements, desain game, menyiapkan aset, developing, integrasi, dan testing. Analisis survei menunjukkan nilai rata-rata engagement novel visual dengan nilai per faktor: aesthetics (85.6), endurability (70), felt involvement (85), focused attention (66), novelty (80), dan perceived usability (84). Hasil tersebut menunjukkan engagement novel visual berada pada kategori baik sampai sangat baik. Penyelesaian permainan memiliki nilai rata-rata 2,8 dan menunjukkan keberhasilan novel visual dalam membuat pemain menyelesaikan permainan hingga akhir. Replayability novel visual memiliki nilai rata-rata 3.2 dan menunjukkan keberhasilan PCG dengan GA dalam membuat pemain memainkan kembali permainan.

Replayability is an important concept in game design. The ability to enjoy a game repeatedly while still getting a new experience each session is an added value that distinguishes video games from linear entertainment media such as movies and books. Visual novel games apply the concept of replayability by creating branching stories that can choose players in each game session. This study aims to create and test a novel visual development method that is expected to increase replayability. The method used in this study is Procedural Content Generation (PCG) using Genetic Algorithm (GA). This study consists of 6 stages, namely scope & requirements, game design, preparing assets, development, integration, and testing. Survey analysis shows the average value of novel visual engagement with values ??per factor: aesthetics (85.6), durability (70), perceived involvement (85), focused attention (66), novelty (80), and perceived usefulness (84). These results indicate that novel visual engagement is in the good to very good category. Game completion has an average value of 2.8 and shows the success of the novel visual in getting players to finish the game to the end. The novel visual's replayability had an average score of 3.2 and demonstrated PCG's success with GA in getting players to replay the game.