Objective: To carry out a comparative analysis of the accuracy of the adherence of tooth-supported cast and milled cobalt chrome alloy frameworks in the shoulder area depending on the length of the prostheses. Methods: For the clinical trial 314 patients who needed metal-ceramic fixed dental prostheses (FDPs) were recruited. Metal frameworks of prostheses were made with conventional lost-wax technique (CLW) and computer aided design and manufacturing (CAD/CAM). To assess the accuracy of the adhesion of the frameworks to the ledges of the prepared teeth stumps, the silicon film between the abutment teeth and inner surface of crowns was obtained, cut into 8 pieces and measured in the shoulder area. The marginal gaps were measured using ImageJ2 application from digital photos obtained with calibrated digital microscope (250×). The data obtained were analyzed using the non-parametric Kruskal-Wallis H-test and Mann- Whitney U-test. Results: The average values of the marginal gap in the CLW group ranged from 71.9±7.6 μm in single and double crowns to 117±3.7 μm in prostheses with a length of 7-8 units. With a corresponding increase in the length of the prostheses, the average values of the marginal gap in the CAD/CAM group were from 67.1±6.8 μm to 90.8±8.3 μm. Differences in marginal clearance parameters between all groups were statistically significant (p<0.05), except for the group of single and paired crowns with milled caps. Conclusion: The values of the edge gap in all groups were within the acceptable range. However, CAD/CAM frameworks were significantly more accurate in long-length prostheses. Keywords: Teeth prep, CAD/CAM, accuracy of the fit, ledge, precision of the fit, marginal gap.