The simulation routines have been run through a preliminary set of exercises. They give results that are consistent with our understanding of the data. There are several additional enhancements to the software that need to be made to more closely approximate ``real'' data and allow us to test the functioning of the survey pipeline. First is the bad pixel mask; variation of this will help us understand how good the detector need to be. Other observational refinements will include: systematic variation of the background (ramp), systematic variation of the seeing, image jitter (non--uniform offsets), and the inclusion of extended sources. Improvement of the offset determination will help the processing part of the simulator.
There are many different tests that can be made with the refined simulator. We can map the effects of variable numbers and locations of bad pixels, understand the severity of background and seeing variations, probe completeness and reliability, and evaluate the band merging procedure. More specific to photometry, we can revise the algorithm for identifying and splitting double stars, test ways of merging aperture and PSF-fit photometry, and refine how the PSF is characterized.
Figure 3: Photometry/Position errors -- Standard Run / b=90 / FWHM
Figure 4: Photometry/Position errors -- Standard Run / b=25 / FWHM
Figure 5: Photometry/Position errors -- Standard Run / b=10 / FWHM
Figure 6: Photometry/Position errors -- Standard Run / b=5 / FWHM
Figure 7: Photometry/Position errors -- Standard Run / b=0 / FWHM
Figure 8: Photometry/Position errors -- b=10 / FWHM 
Figure 9: Photometry/Position errors -- b=10 / FWHM 
Figure 10: Photometry/Position errors -- b=10 / FWHM 
Figure 11: Photometry/Position errors -- b=10 / FWHM 
Figure 12: Photometry/Position errors -- Subpix Response / b=25 / FWHM 
Figure 13: Photometry/Position errors -- Subpix Response / b=10 / FWHM 
Figure 14: Photometry/Position errors -- Good Dither / b=25 / FWHM 
Figure 15: Photometry/Position errors -- Poor Dither / b=25 / FWHM 