Objective: To evaluate and compare the error in cephalometric measurements by manual and computerized methods using the softwares Dolphin Imaging version 11.0 and Dentofacial Planner version 7.02. Methods:The sample comprised 45 lateral cephalograms. A calibrated operator conducted 90 manual and 180 digital cephalometric tracings, using 8 angular measurements (SNA, SNB, ANB, FMA, 1.NA, 1.NB, SN-Occlusal, SN-GoGn) and 5 linear measurements (1-NA, 1-NB, Co-Gn, Co-A e LAFH). For analysis of results, the dependent t test was used to compare the error of each method (intragroup error); comparison of measurements between the three methods was performed by the one-way ANOVA and Tukey test in the presence of significant result. Results:The results showed statistically significant differences, especially between measurements evaluated on the Dolphin software and by manual tracing (p>0.05). Conclusions:According to the present results, the three cephalometric methods have proven to be accurate, since they presented few systematic errors. The computerized method using the Dentofacial Planner software showed the highest reliability, followed by the manual method, while the Dolphin Imaging software was the least effective and more likely to produce systematic errors in the identification of points.