The Newmark displacement (ND) method, which reproduces the interactions between waves, solids, and fluids during an earthquake, has experienced numerous modifications. We compare the performances of a traditional and a modified version of the ND method through the analysis of co-seismic landslides triggered by the 2022 Ms 6.8 Luding earthquake (Sichuan, China). We implemented 23 ND scenarios with each equation, assuming different landslide depths, as well as various soil-rock geomechanical properties derived from previous studies in regions of similar lithology. These scenarios allowed verifying the presence or absence of such landslides and predict the likely occurrence locations. We evaluated the topographic and slope aspect amplification effects on both equations. The oldest equation has a better landslide predictive ability, as it considers both slope stability and earthquake intensity. Contrarily, the newer version of the ND method has a greater emphasis on slope stability compared to the earthquake intensity and hence tends to give high ND values only when the critical acceleration is weak. The topographic amplification does not improve the predictive capacity of these equations, most likely because few or no massive landslides were triggered from mountain peaks. This approach allows structural, focal mechanism, and site effects to be considered when designing ND models, which could help to explain and predict new landslide distribution patterns such as the abundance of landslides on the NE, E, S, and SE-facing slopes observed in the Luding case.