In view of future high-precision large-scale structure surveys, it is important to quantify the percent and subpercent level effects in cosmological N-body simulations from which theoretical predictions are drawn. One such effect involves deciding whether to zero all modes above the one-dimensional Nyquist frequency, the so-called “corner” modes, in the initial conditions. We investigate this effect by comparing power spectra, density distribution functions, halo mass functions, and halo profiles in simulations with and without these modes. For a simulation with a mass resolution of ${m}_{p}\sim {10}^{11}$ ${h}^{-1}\,{M}_{\odot }$, we find that at $z\gt 6$, the difference in the matter power spectrum is large at wavenumbers above ∼80% of ${k}_{\mathrm{Ny}}$, reducing to below 2% at all scales by $z\sim 3$. Including corner modes results in a better match between low- and high-resolution simulations at wavenumbers around the Nyquist frequency of the low-resolution simulation, but the effect of the corner modes is smaller than the effect of particle discreteness. The differences in mass functions are 3% for the smallest halos at z = 6 for the ${m}_{p}\sim {10}^{11}$ ${h}^{-1}\,{M}_{\odot }$ simulation, but we find no significant difference in the stacked profiles of well-resolved halos at $z\leqslant 6$. Thus removing power at $| {\boldsymbol{k}}| \gt {k}_{\mathrm{Ny}}$ in the initial conditions of cosmological simulations has a small effect on small scales and high redshifts, typically below a few percent.