by Andrew Hand, Sylvia Lin, and Mei Xintong Zha
Posted on Feb 28, 2019
IQP Lag can Kill provides the performance data of users playing two games, shoot and dodge, with increasing lag every 30 second. Heart rate and facial emotions data(emotion Detection - a software similar to the online Affectiva) is simultaneously collected throughout the trial. This project aims to discover the relationships between game latency and player’s performance, frustration, and perception of lag.
According to the plot, more lag doesn't imply an increased heartrate for most players. In fact, most all players keep a relitively stable heart rate for the duration of the game.
If we neglect the outliers, we can see that the distribution of lag counts versus ping levels is much more compact for the dodging game. There seems to be a higher positive correlation in the shooting game, which means players actually detect the same level of lag better in a shooting game.
When players are suffering from high levels of lag, there is not a major correlation suggesting that they click more times. In the shooting game, people’s click frequency vary more as the ping level increases and the overall click distribution is more consistent than the distribution of the dodging game. Contrastingly in the dodging game, people’s click frequency vary more at the low ping levels.
In a dodging game, a hit means a failed escape. There is this almost weird situation where people perform the worst when the ping is 400 millisecond, and they perform better at the highest ping level at 500 milliseconds. Whereas in a shooting game, a hit means a successful shot, where people perform slightly worse when there is a high lag.
The scatterplot shows that the players’ performance is not much affected by the lag with the exception of one outlier. They do not miss more shots when the pings are higher.