Cephalometric analyses using computer software have more advantages than manual analyses. However, the software should be evaluated for the accuracy and reproducibility before it can be used. The aims of the present study were: 1) to compare the differences in accuracy and precision between utilizing scanned images and soft copy images with the software 2) to assess the reproducibility of software and manual cephalometric analyses. Fifteen cephalograms were selected randomly from the Record Unit, Hospital Universiti Sains Malaysia. All films had 10 fiducial points marked and were scanned at 75 dots per inch (dpi) and 300 dpi. Manual and digital measurements were compared to obtain magnification factors. Seven digital x-ray images of a caliper were taken. The mean differences were measured between the fixed caliper measurement and digital measurements. Subsequently, 37 cephalograms were traced manually and digitally with Computer-Assisted Simulation System for Orthognathic Surgery (CASSOS) software to assess its accuracy and reproducibility after applying the obtained magnification factor. Steiner analysis was utilized and T tests were used to evaluate the mean difference. P<0.01 was considered significant. The magnification factors of 300dpi and 75dpi for both horizontal and vertical measurements were 0.50 and 0.95 respectively, while magnification factor for softcopy image measurements was 0.89. Cephalometric comparisons between original and digital images showed statistically significant differences for several variables but the mean differences were clinically insignificant. Although some distortion was noted, it was clinically acceptable after correction of the enlarged images with magnification factors. The reproducibility of CASSOS is excellent and as good as other commercially available cephalometric software.