Purpose: The aim of the present study was to evaluate the reliability and re-producibility of digital shade selection devices, and to correlate the results with conventional human visual shade assessment. Materials and Methods: Tooth color gradation and agreement of two different digital shade selection instruments were determined by employing a spectro-photometer (ShadePilot, Degudent; Hanau, Germany, Software V. 2.41) and a colorimeter (ShadeVision, Ammann Girrbach; Pforzheim, Germany; Software V. 1.20). The devices were compared with three human examiners with a negative history of visual color deficiency, looking at 40 subjects under clinical conditions. In one half of the test persons, the visual and digital shade determination was performed on the right maxillary central incisor that was free of any restoration, whereas the other half was carried out in subjects with single implant-supported PFM crowns in the same region. The computer-based readings across the regions (incisal, middle, cervical) were recorded two consecutive times. Differences between the computer-based readings were evaluated with the χ2 test (Chi-square). Results: The same shade was obtained by all three human examiners in 22.5% (9 of 40 cases), by all colorimetric readings in 35% (14 of 40 cases), and by all spectrophotometric readings in 55% (22 of 40 cases). There was considerable agreement between the first and second reading of the investigated spectrophotometric and colorimetric shade instruments. The spectrophotometer demonstrated color matches of the first and second reading in 81.7% of the in vivo measurements (coefficient of contingency = 0.98; χ2 test p = 0.99), while the colorimeter exhibited matches in 70% (coefficient of contingency = 0.96; χ2 test p = 0.96) of the cases. Hence, the spectrophotometer exhibited the highest agreement between the two consecutive readings. While no significant influ-ence of the measuring point (incisal, middle, cervical) on the reproducibility of color results could be ascertained, both the colorimeter and spectrophotometer displayed a statistically significant difference for the frequency distribution of color categories for teeth vs PFM crowns (both χ2 test p 0.0001). The measuring results of the colorimeter showed distinctly lighter shades than those of the spectrophotometer. Conclusions: Within the limitations of this study, the results suggest that spectrophotometric shade determination is more reproducible compared to conventional visual shade assessment. It can serve as a reliable addition in color matching and enhances the level of shade analysis, communication, interpretation, and fabrication of dental restorations. Keywords: shade determination, spectrophotometric, colorimetric, visual human, measurement agreement, implant crown