This conference paper describes the analysis of the potential of GaInP/GaAs/1-eV/Ge four-junction solar cell to improve on the efficiency of the state-of-the-art GaInP/GaAs/Ge benchmark. We emphasize three factors: (1) The newly proposed terrestrial concentrator spectrum has a lower ratio of red to blue light than does the old AM1.5 direct standard spectrum. (2) Standard two-layer antireflection coatings do not provide near-zero reflectance over the full spectral range of interest for these devices. (3) GaInNAs junctions used to date for the 1-eV junction have quantum efficiencies less than(approx)75%. These factors all limit the device current, adversely affecting the four-junction efficiency. We discuss strategies for ameliorating this problem, including going to alternate structures such as a GaInP/GaAs/0.9-eV three-junction device.