The impact of spectral power distribution of daylight simulators (ie, D65 simulators) on surface whiteness specification was investigated by focusing on how CIE whiteness and tint values of 8 whiteness samples with fluorescent whitening agents (FWAs) vary under different D65 simulators. Large variations in both whiteness (approximate to 16 points) and tint (approximate to 1.6 points) are observed under the D65 simulators above BB grade, as characterized using the CIE metamerism index. However, it is found the variations of the whiteness and tint values are smaller under the D65 simulators whose radiant power in the UVA band (ie, 300-340 nm) was within +/- 30% in comparison to CIE standard D65 illuminant, as defined in BS 950, which may be a better alternative for evaluating the quality of a D65 simulator for surface white specification using CIE whiteness and tint formulas. The findings also suggest the necessity to fine-tune or revise the CIE whiteness and tint formulas to characterize the surface whiteness under nonperfect D65 simulators or arbitrary light sources.