Objective: To explore the effect of different types of visual display terminals (VDT) on visual fatigue. Methods: A total of 68 subjects were enrolled and divided into flat screen group, metal screen group and glass bead screen group. The subjects watched a video in the same environment for 60 minutes. Before and after watching, they were asked to fill in a visual fatigue rating scale and underwent visual function examinations. The blink rate of subjects was recorded during the experiment. Results: After watching the video, for the metal screen, glass bead screen, and flat screen group respectively: (1)The median (interquartile range, IQR) of visual fatigue rating increased significantly (before: 1.0 (1.2), 1.3 (1.6), 0.5 (0.9); after: 2.1 (1.9), 2.1 (1.7), 1.2 (1.0); Z=-4.72, -4.83, -3.75; all P<0.05). (2) The median (IQR) of blink frequency increased significantly (before: 11.5 (10.6), 15.3 (11.9), 7.0 (13.0) times/minute; after: 15.0 (13.0), 17.0 (16.0), 15.0 (12.0) times/minute; Z=-2.64, -2.74, -4.12; all P<0.05). (3) The mean (±standard deviation) of critical flicker fusion frequency (CFF) decreased significantly [before: (35.76±2.98), (35.84±2.79), (35.44±2.21) Hz; after: (35.09±2.78), (34.94±2.67), (34.57±2.33) Hz; t=-2.64, -2.38, -4.12; all P<0.05). (4)The median (IQR) of near point of convergence (NPC) became significantly larger [before: 3.9 (2.6), 4.0 (2.5), 4.3 (2.7) cm; after 5.4 (3.8), 4.8 (2.7), 6.1 (3.9) cm; Z=-3.39, -2.27, -3.36; all P<0.05]. (5)The median of negative relative accommodation (NRA) reduced significantly [before: 2.00 (0.50), 2.00 (0.56), 2.00 (0.25) cm; after 1.75 (0.50), 1.75 (0.25), 1.75 (0.50) cm; Z=-1.98, -2.09, -2.53; all P<0.05]. Meanwhile, no significant changes were found for the positive relative accommodation (PRA), accommodative amplitude, accommodative response, negative relative vergence in all groups after watching the video. The comparisons between the three groups showed that, after watching the video, the participants in the two kinds of projection screen group (the metal and glass bead screen group) had more subjective visual fatigue than those in the flat screen group respectively (Z=-2.09, -2.21; all P<0.05), while there was no significant difference in subjective visual fatigue between the two projection screen groups (P>0.05). In addition, positive fusional vergence recovery point and break point decreased after watching the video in the metal screen and glass bead screen group, respectively (t=4.15, 2.07; all P<0.05). However, no such change was found in the flat screen group. Conclusions: Short-term and long-distance VDT operations have significant effects on visual fatigue, which may be due to the decreased positive convergence ability. VDT operations over projection screens may result in more visual fatigue than LED flat panel display, while the effect of different projection screen types have insignificant influences on visual fatigue. (Chin J Ophthalmol, 2019, 55:595-600).