In general, new and innovative solar cell technologies have spectral responsivities that differ from the spectral responsivities of standard reference solar cells (e.g. c-Si, filtered c-Si). Hence integral calibration methods (e.g. with a solar simulator) require a spectral mismatch correction (IEC 60904-7). Therefore, the relative spectral responsivity of the solar cell is required. The determination of the spectral responsivity for technologies with spectral nonlinearities (e.g. with respect to the irradiance level, like dye-sensitized solar cells (DSC)) is comparatively difficult. However, even though the calibration of the short-circuit current under Standard Test Conditions ISTC is performed at an irradiance level E of 1000 W m-2, commonly the spectral responsivity measurement is performed at one irradiance between 300 W m-2 and 400 W m-2 in accordance to a simplification mentioned in the IEC 60904-8. This simplification can often be used (e.g. for the most Si solar cells), but is not always correct. If this is the case, then the determined spectral responsivity leads to a wrong spectral mismatch correction factor. In this contribution, an example for this behavior is shown with the consequence that for this type of solar cells in general the simplification may not be applied. The solution is the measurement at different irradiance levels with the DSR Method. However, the DSR method uses the combination of a continuous bias radiation and a modulated monochromatic beam. Hence, a dependence of the spectral responsivity of the modulation frequency has to be considered. The most common solar cells (e.g. c-Si) do not show a frequency dependence with respect to the modulation frequency of the monochromatic beam in the range from 0.1 Hz to 99 Hz. Nevertheless, some solar cells such as DSC exhibit a frequency dependence in this range. These multidimensional dependencies of the spectral responsivity as well as its impact and solutions on the determination of the spectral responsivity and the will be presented.