In previous works, the general, one-parameter friction theory (f-theory), models have been applied to the accurate viscosity modeling of reservoir fluids. As a base, the f-theory approach requires a compositional characterization procedure for the application of an equation of state (EOS), in most cases a cubic EOS, to the description of the oil phase properties. Normally, such a procedure provides a method for characterizing an oil into a number of compound groups along with the critical constants and parameters that are required by the EOS. Thus, in this work, a compositional characterization method based on an accurate description of the fluid mass distribution is presented. The characterization procedure accurately matches the fluid saturation pressure. Additionally, a Peneloux volume translation scheme, capable of accurately reproducing the fluid density above and below the saturation pressure, is also presented. The combination of the mass characterization scheme presented in this work and the f-theory, can also deliver accurate viscosity modeling results. Additionally, depending on how extensive the compositional characterization is, the approach,presented in this work may also deliver accurate viscosity predictions. The modeling approach presented in this work can deliver accurate viscosity and density modeling and prediction results over wide ranges of reservoir conditions, including the compositional changes induced by recovery processes such as gas injection.