Results from laboratory PVT measurements, special core analysis studies, and reservoir condition core-flood studies were used to develop and test the accuracy of a compositional simulation model in predicting near-multicontact miscible gas flood performance at the laboratory scale. Excellent agreement was obtained between the core-flood measured recoveries and GOR and the compositional simulation model predictions. No history matching was required to obtain this result. The choice of PVT and special core tests were critical to our ability to model the near-multicontact miscible displacement process without history matching. This result provides a high degree of confidence in using the compositional simulation model in cross-section and 3D element models to predict field-scale performance. Cross-sectional simulation sensitivities were performed in a geostatistical model of a dipping reservoir. Gas-flood recovery was on par with water-flood recovery for this cross-section.