Color graininess is one of the most important artifacts affecting image quality. Previous studies have indicated that it is necessary to remove the screen signal on the scanned amplitude-modulated, AM, screen before adopting any granularity metric to achieve satisfactory correlation between the objective measurement and subjective evaluation, indicating the AMhalftone screen has no effect on the perceived graininess. Nonetheless, the inherent characteristics of FMhalftone screens with pseudorandom dot placement transforms the frequency content of the screen signal from spikes to a distribution across the spatial frequency domain. The objective of this experiment is to conduct a visual experiment comparing the perceived graininess on patches with various color generated by various types of AM and FM screens. Correlated with an unified objective granularity metric, we can evaluate the impact of various halftone screens on color graininess, and verify the assertion that the periodic screen signal is independent from the perceived graininess.