Comparison of the scatter in the CMDs of the unrestored and restored images as illustrated in Figs. 1 and 2 indicates that iterative/recursive deconvolution followed by aperture photometry produces photometrically superior results to those of the SPS PSF-fitting software for data from HST in its current configuration. From comparison of the scatter in the diagrams for the brightest stars, it appears that there is little to be gained by deconvolution where the signal-to-noise ratio in the unrestored data is high. However, this conclusion is based on the comparison between aperture and PSF-fitting photometry. The application of least-squares type PSF-fitting algorithms to the restored images should further improve the photometric scatter (analagous to the improvement in ground-based data between aperture and PSF-fitting techniques). We believe that further work is necessary to develop least-squares fitting techniques which are appropriate for the image and noise properties of deconvolved HST images.