Downscaling of semiconductor fabrication technology nodes brought forth a need to reassess the accuracy of 3D metrology. Accuracy is defined relative to a reference tool measurement. The authors have studied the accuracy of 3D SEM measurement results for various feature geometries and materials, matching the results to Monte Carlo simulations. Analysis of the SEM images based on an analytical model was performed. Careful analysis of the matching curves for 3D algorithm results and reference data (geometric parameters of the feature) reviled an appropriate behavior of algorithm in the vicinity of the nominal process window, and for sufficiently small feature rounding (production node). We performed matching of 3D CD SEM measurement to reference geometry data using Monte Carlo simulation. We analyzed the accuracy of measurement for a wide range of the feature geometry parameters (height, sidewall angle, top and bottom rounding). The simple physical model for corner rounding estimation is considered. We perform the model waveform analysis of the feature rounding influence on the height measurements. Serving as a process-monitoring tool, the algorithm performance was found in agreement with the required tolerance typical of the nominal process window ± 10%. Serving on extreme R&D, where rounding further away from the nominal window ± 10% is counted significant, there lie observable deviations in accuracy of height and sidewall angle measurement. These are explained through extreme corner rounding effects.