A time limit usually exists in a real-life visual search task. This study investigated the effect of time pressure on dynamic visual search performance. A 2 (movement velocity: high vs. low) × 2 (time pressure: high vs. low) between-subject factorial design was used in this experiment. The time lengths corresponding to 63.2% cumulative probability of detection (i.e., time constant) and the probability of 50% were selected as low and high time pressure. When velocity was 4 / s , 9.72 s and 6.89 s were chosen as the time limits for low and high time pressure respectively; when velocity was 16 / s , 20.30 s and 14.22 s were chosen for low and high time pressure separately. There were 4 different experimental conditions: high velocity/high pressure, high velocity/low pressure, low velocity/high pressure, low velocity /low pressure, and no time-pressure control condition. 60 university students participated in this experiment. They were divided into 4 groups, i.e., each group included 15 students. Each participant took 20 search tasks in the experiment. The search time and accuracy of each participant were recorded. The difference in search performance between low time pressure condition and no time pressure condition was not significant whereas that between low time pressure condition and high time pressure condition was significant in the course of 4/s movement velocity. In the velocity of 16 / s situation, the differences in performance between low time pressure and high time pressure conditions were not significant. Those results revealed that search performance was sensitive to time pressure under the situation of low movement velocity and the performance was not sensitive in the high velocity condition. The obtained results could be a reference for deciding the time limit (i.e., the permitted longest search time) for a specific visual search task.