We investigate the influence of scene illuminant on perceived image quality. Given two multispectral images, an original and a reproduction (e.g. compressed, gamut mapped,..), we seek redundancies of perceived difference through changes of illuminant, and with regard to 5 so-called image difference features (IDF). In order to do this, we employ an information-theoretic perspective to measure variations of entropies in each IDF, w.r.t. various scene illuminants, and in the case of two particular kinds of distortions: spectral gamut mapping and a spectral reconstruction from a six-channel camera model. Our results indicate that changing the scene illuminant has a lesser influence on achromatic image difference features.