2005A Known Features

Old colour equations from Nov 2004 : see SV report for changes but main effect is to reduce H-band mags by ~0.1m.
Non-linearity caused by stacking bug : subtle effect particularly at bright end and therefore also affects 2MASS calibration of faint objects.
Interleave and cheesegrater pattern : if the components of an interleave are not in the correct positions then interesting but not very informative patterns result.
Incomplete MBSs & checkerboard pattern : same problem as above. If the MSB is not completed this can result in equally interesting patterns.
Trailed images - telescope not settled before exposure started; unguided, guided but drifted off e.g.: edge of AG?, too faint ? residual streaks after filter changes- if this a problem -> flush. (see example)
Incomplete sky crud removal - problem with (mainly) varying scattered light on residues on inside of field lens. Problem should be mitigated by cleaning prior to 2005B.
3x3 interleave and "designer" tiling : caused by offset bug for observing the interleave components, fixed for 2005B.
Cross-talk : fascinating unpredicted differential effect, solution in hand. (see examples)
Persistance : residual images can persist for minutes, tricky one to deal with since seems to depend on several factors. (see example)
Banding at channel level : could remove but only at expense of removing real large structure.
Out-of-focus images : see DQC information in summary plots and summary.list files.
Inter-pixel capacitance : see SV report.
Incomplete catalog parameter set : get version 2 which has the full standard set including the missing Petrosian fluxes and error estimates for positions and fluxes.