Background: Three-dimensional (3D) visualization systems, also known as heads-up systems, are now available for eye surgery and as with every new device there is need for a specific evaluation. Objectives: The aim of this study was to compare the efficiency, surgical comfort, and safety of a 3D visualization system to a standard binocular microscope (BM) in routine ophthalmologic procedures. Method: After a 4-week training period, a 3D visualization system (Ngenuity, Alcon®) available in one of the Robert Debré Hospital Ophthalmology Departments’ operating rooms was compared to a standard BM (OPMI LUMIRA 700, Zeiss®), in the process of a call for new device evaluation. From December 2017 to March 2018, 5 surgeons and their respective residents were asked to fill in a questionnaire for all procedures. Before the surgery, the surgeon recorded: (i) the type of surgery (cataract [PK], retinal detachment [RD], epiretinal membrane peeling [ERM], macular hole, vitreous haemorrhage [VH]), (ii) the type of visualization system chosen (3D or BM), and (iii) the estimated surgical risk (low, intermediate, or high grade). At the end of the procedure, the primary surgeon recorded the remaining parameters, including: (i) surgery duration, (ii) intraoperative complications, (iii) percentage of endoillumination for posterior segment surgeries, (iv) status of the operator (senior or resident) and operator switch if necessary (senior only, resident only, or resident with help of the senior), and rated: (i) the visual comfort (low, normal, excellent), (ii) the operative fluency (low, normal, excellent), (iii) backaches (none, low, moderate, important), and (iv) headaches (range from 0 to 10). Age and sex were collected retrospectively. The procedures performed with 3D and BM were subsequently compared using univariate (χ<sup>2</sup>, Fisher, Wilcoxon) and multivariate analysis (generalized linear model), allowing us to identify parameters independently associated with PK surgery duration. Results: A total of 102 valid questionnaires, relative to 73 PK and 29 vitreoretinal procedures, respectively, were analysed. As regards PK (3D, n = 25 vs. BM, n = 48), the mean age, sex ratio, surgical risk, intraoperative complications (1/25 vs. 4/48), visual comfort, backaches, and headaches were similar between the two systems. The use of 3D allowed faster PK surgeries (16.44 ± 4.36 vs. 21.44 ± 7.50 min; p = 0.007) and slightly enhanced the operative fluency. In vitreoretinal surgeries (3D, n = 14 vs. BM, n = 15), no obvious differences between the two visualization systems were observed, although the use of the 3D system was found to slightly decrease the operative fluency. Parameters independently associated with PK surgery duration were 3D visualization (β = –4.4 ± 1.4; p = 0.002), high preoperative surgical risk (β = 6.2 ± 2.4; p = 0.012), intraoperative complications (β = 8.7 ± 2.6; p = 0.001), and surgeon status (β = –4.4 ± 1.3; p = 0.001) in univariate and multivariate analysis. Conclusions: 3D visualization can be safely used in routine practice. It slightly improves the operative fluency, allowing faster PK surgery.