We compare the near-equilibrium solution of a global ocean/sea ice model at a horizontal grid resolution of 1¿ ¿ 1¿ to near-equilibrium solutions obtained in two configurations at 4¿ (longitude) ¿ 2¿ (latitude) resolution. All simulations use realistic, smoothed topography and monthly averaged climatological forcings. Our comparison of the results emphasizes large-scale features relevant to global climate change. Since neither the 1¿ ¿ 1¿ simulation nor the 4¿ ¿ 2¿ simulations resolves ocean eddies, our results do not address the possible importance of resolving eddies in ocean-climate simulations. There are significant differences between the 1¿ ¿ 1¿ and 4¿ ¿ 2¿ simulations, most notably in the Arctic Ocean. However, the large-scale features of the model solutions are very similar at the two resolutions and in many cases are more sensitive to a large difference in horizontal viscosity than to the difference in resolution. This suggests that other approaches to improving the solution of ocean-climate models will be more effective than increases in horizontal resolution outside the eddy-resolving regime.