Is it correct to say that N-body simmulations of structure formation in LCDM model use as initial conditions the density perturbations at last scattering suggested by CMB measurements but do NOT use the last scattering velocity perturbations (peculiar velocities)? I've read description of N-body simmulations and it seems that they take the last scattering density perturbations as initial condition but compute the corresponding initial velocity perturbations uzing Zeldovich approximation. Is that the same or different from the velocity perturbations suggested by CMB measurements? Would the results of the N-body simmulations be any different if one sets the initial peculiar velocities to zero exactly? The reasoning is that the velocities at last scattering do not matter anyways so why bother with Zeldovich approximation???? Such reasioning seems to be corroborated by section 6.3.2 of The Mukhanov texbook which claims that the initial velocity perturbations "do not play any role in the formation of the large-scale structure of the universe" since they decay with time. He says, such perturbations are generated later after nonlinear structure has formed.