Logo image
Performance Measurements on a Cloud VR Gaming Platform
Conference paper

Performance Measurements on a Cloud VR Gaming Platform

Yen-Chun Li, Chia-Hsin Hsu, Yu-Chun Lin and Cheng-Hsin Hsu
QoEVMA 2020 - Proceedings of the 1st Workshop on Quality of Experience (QoE) in Visual Multimedia Applications, pp.37-45
10/2020

Abstract

cloud computing computer games measurement prototype virtual reality Computer Science Applications Computer Networks and Communications Software Computer Graphics and Computer-Aided Design Human-Computer Interaction
As cloud gaming and Virtual Reality (VR) games become popular in the game industry, game developers engage in these fields to boost their sales. Because cloud gaming possesses the merit of lifting computation loads from client devices to servers, it solves the high resource consumption issue of VR games on regular clients. However, it is important to know where is the bottleneck of the cloud VR gaming platform and how can it be improved in the future. In this paper, we conduct extensive experiments on the state-of-the-art cloud VR gaming platform - Air Light VR (ALVR). In particular, we analyze the performance of ALVR using both Quality-of-Service and Quality-of-Experience metrics. Our experiments reveal that latency (up to 90 ms RTT) has less influence on user experience compared to bandwidth limitation (as small as 35 Mbps) and packet loss rate (as high as 8%) . Moreover, we find that VR gamers can hardly notice the difference between the gaming experience with different latency values (between 0 and 90 ms RTT). Such findings shed some lights on how to further improve the cloud VR gaming platform, e.g., a budget of up to 90 ms RTT may be used to absorb network dynamics when bandwidth is insufficient.

Metrics

1 Record Views

Details

Logo image