Please use this identifier to cite or link to this item:
Title: Improving Mobile Game Performance with Basic Optimization Techniques in Unity
Authors: Koulaxidis, Georgios
Xinogalos, Stelios
Type: Article
Subjects: FRASCATI::Natural sciences::Computer and information sciences
Keywords: mobile game performance
optimization techniques
low poly
object pooling
Issue Date: 2022
Source: Modelling
Volume: 3
Issue: 2
First Page: 201
Last Page: 223
Abstract: Creating video games can be a very complex process, which requires taking into account various hardware and software limitations. This process is even more complex for mobile games, which are limited to the resources that their platforms (mobile devices) offer in comparison to game consoles and personal computers. This restriction makes performance one of the top critical requirements, meaning that a videogame should be designed and developed more carefully. In order to reduce the resources that a game uses, there are optimization techniques that can be applied in different stages of the development. For the purposes of this article, we designed and developed a simple shooter videogame, intended for Android mobile devices. The game was developed with the Unity game engine and most of the models were designed with the 3D computer graphics software Blender. Two versions of the game were developed in order to study the differences in performance: one version that applies basic optimization techniques, such as low poly count for the models and the object pooling algorithm for the enemy’s spawn; and one where the aforementioned optimizations were not used. Even though the game is not large in scale, the optimized version achieves a better user experience and needs less resources in order to run smoothly. This means that in larger and more complex video games these optimizations could have a bigger impact on the performance of the final product. To measure how the techniques affected the two versions of the game, the values of frames per second, batches and triangles/polygons were taken under consideration and used as metrics for game performance in terms of CPU usage, rendering (GPU usage) and memory usage.
ISSN: 2673-3951
Other Identifiers: 10.3390/modelling3020014
Appears in Collections:Department of Applied Informatics

Files in This Item:
File Description SizeFormat 
modelling-03-00014.pdf6,32 MBAdobe PDFView/Open

This item is licensed under a Creative Commons License Creative Commons