Skip to main content

Computer Science Department, PhD Proposal Defense , Shengmei Liu " The Impact of Latency on Players in First-person Shooter Games"

Monday, December 13, 2021
12:00 pm to 1:00 pm


Shengmei Liu

PhD Candidate

WPI – Computer Science Department                                                                       

Monday, December 13, 2021

Time; 12:00 p.m. – 1:00 p.m.

Zoom link:


Professor Mark Claypool - Computer Science, Worcester Polytechnic Institute

Professor Carl Gutwin - Computer Science, University of Saskatchewan

Professor Lane Harrison - Computer Science, Worcester Polytechnic Institute

Doctor Jamie Sherman - Senior Product Researcher, Netflix 



Video game play is among the most popular forms of entertainment in the world and eSports is a multi-billion dollar industry. The first-person shooter game is the most popular genre in esports, with 4 of the top 10 games first-person shooters based on prize money.

First-person shooter games are among the most affected by latency. First-person shooter players want lower latency to maximize their chances of winning.  In general, the lower the latency, the sooner a player sees the outcome of their actions. Latency in first-person shooters comes from two sources: local latency due to system configurations (e.g., mouse and monitor) and network latency due to processing and propagation between client and server in the client-server architecture. While most modern first-person shooter games use latency compensation techniques to mitigate the effects of network latency on players, player performance and quality experience still degrade with the increase in both sources of latency.

My research will focus on how reductions in latencies benefit players in first-person shooter games considering both in-game performance and quality of experience (QoE). Since first-person shooter games span a broad range of gameplay configurations, such as types of weapons, sizes of maps and layouts of obstacles, techniques to study individual games, such as through user studies, cannot scale to cover all possible first-person shooter games. We assert that players' actions in first-person shooter games can be composed of multiple atomic actions. Atomic actions represent the indivisible player game interactions which can potentially be combined into more complex gaming activities (e.g., navigation and selection actions can be combined into moving shooting activities). My approach is to study and model the primary atomic actions in first-person shooter games, and use the models as building blocks to simulate first-person shooter games played with different game configurations. I will focus on the two main atomic game actions in first-person shooter games – navigation (get in position to shoot or avoid being shot), and selection (shoot at a moving or stationary target). I will gather data on each action via a user study, and build mathematical models for player performance and quality of experience with latency based on the user study data. By incorporating the models for different actions, I will simulate specific scenarios, validate the simulation results using data from first-person shooter games, and then simulate first-person shooter games with a broad range of games and game system configurations. This should provide a deeper understanding on the impact of latency on players in the most popular game genre – the first-person shooter.