The performance of two cubic equations of state (EoS) - a translated version of a modified Peng-Robinson and the Jhaveri-Peng-Robinson - in the prediction/correlation of the phase behavior of gas and oil mixtures is presented. Problems associated with the uncertainty involved in the description of the heavy end of the mixtures are avoided by considering synthetic systems. Prediction results are fair for the gas condensate systems and become very poor when approaching near-critical conditions. The results are better for the oils, but deviations are observed with increasing volatility. A systematic evaluation of the improvements obtained by tuning the thermodynamic model to match the saturation pressure alone or together with constant composition expansion data is carried out to determine the number and type of the EoS parameters that need to be tuned. For far-critical gas condensates and for oil mixtures the adjustment of one parameter is adequate to yield satisfactory results. This is not the case for near-critical gas condensates, where adjustment of two parameters is necessary. The overall quality of the results obtained by prediction or tuning justifies the wide use of the aforementioned EoS - and probably other cubic EoS - in the petroleum industry.