OBJECTIVE During laboratory fabrication procedures and/or clinical adjustments, pressable materials: IPS Empress and Empress 2, may be ground, polished or sandblasted. These treatments may affect their strength by introducing microscopic flaws and defects. This study investigates the effect of these procedures on the mean flexural strength of these materials. METHODS One hundred and forty disc specimens (14mmx1 mm) of IPS Empress and Empress 2 were prepared, and divided into seven groups of 20 specimens for each material. Groups were untreated, polished, polished and glazed, ground, ground and glazed, sandblasted, sandblasted and glazed. Surface roughness, mean biaxial flexural strength and Weibull modulus were appraised, and a scanning electron microscope was used to describe surface features. Statistical significance among groups of population was analysed using one-way Anova and Tukey's multiple comparison tests. RESULTS Untreated and sandblasted groups showed significantly the highest roughness values, and polished the lowest for each material (p<0.05). Ground groups showed significantly lower roughness values than the sandblasted groups, and significantly higher roughness than the polished groups for each material (p<0.05). Polished groups for each material demonstrated significantly the highest mean flexural strength values (p<0.05). No significant difference in the mean strength values was found between untreated, sandblasted and ground groups for each material (p>0.05). Heat treatment had no effect on roughness or strength values of all treated groups of both materials. The Weibull modulus values for both materials varied with different treatments. They showed higher values for polished and untreated groups, and lower values for ground and sandblasted groups. CONCLUSIONS Surface roughness may not be the only feature that determines strength. Other issues such as porosity, microstructural residual stresses, surface and bulk defects may also be pertinent.