Although 3D-printing is common in dentistry, the technique does not produce the required quality for all target applications. Resin type, printing resolution, positioning, alignment, target structure, and the type and number of support structures may influence the surface roughness of printed objects, and this study investigates the effects of these variables. A stereolithographic data record was generated from a master model. Twelve printing processes were executed with a stereolithography Desktop 3D Printer, including models aligned across and parallel to the printer front as well as solid and hollow models. Three layer thicknesses were used, and in half of all processes, the models were inclined at 15°. For comparison, eight gypsum models and milled polyurethane models were manufactured. The mean roughness index of each model was determined with a perthometer. Surface roughness values were approximately 0.65 µm (master), 0.87-4.44 µm (printed), 2.32-2.57 µm (milled), 1.72-1.86 µm (cast plaster/alginate casting), and 0.98-1.03 µm (cast plaster/polyether casting). The layer height and type and number of support structures influenced the surface roughness of printed models ( ≤ 0.05), but positioning, structure, and alignment did not.