\begin{table}%t3
\caption {\label{Tab:psferr}Overview of consequences of errors in the estimated PSF.}
%\centerline
 { \begin{tabular}{llllll}
 \hline\hline
 \noalign{\smallskip}
  &  $\Delta_{\rm mag}^{{e}}$ & $\sigma_{\rm mag}^{{f}}$ & $\Delta_{\rm pos}^{{g}}$ &  $\sigma_{\rm pos}^{{h}}$ & Diff$_{\rm mean, \sigma, max}^{{i}}$\\
  & [mag] & [mag] & [pix] & [pix] & [counts]\\
\hline
a$^{{a}}$ & 0.02 & 0.02--0.11 & 0.05 & 0.03--0.1 & 2.45, 0.50, 0.4\\
b$^{{b}}$ & 0.03 & 0.01--0.13 & 0.02 & 0.02--0.10 & 2.70, 0.26, 1.9\\
c$^{{c}}$ & 0.04 & 0.02--0.13 & 0.05 & 0.025--0.10 & 2.34, 0.12, 1.0\\
d$^{{d}}$ & 0.075 & 0.01--0.12 & 0.03 & 0.016--0.10 & 2.12, 0.27, 1.68 \\
\hline
\end{tabular}}
\par
\smallskip
$^{{a}}$ PSF convolved with a 2 pixel FWHM Gaussian, i.e., too broad (mainly in the core).
$^{{b}}$ PSF with all pixels raised to the power of~$1.1$, i.e., too narrow.
$^{{c}}$ PSF derived from several stars within $5\arcsec$ of the guide star, but ${\sim}4$~mag fainter, i.e. PSF  wings underestimated.
$^{{d}}$ Moffat function (fitted to guide star PSF) used as PSF.
$^{{e}}$ Approximate mean offset between input and measured magnitudes of the stars.
$^{{f}}$ Range (for magnitudes 12 to~19) of the standard deviation of the difference between input and measured magnitudes.
$^{{g}}$ Approximate mean offset between input and measured pixel positions of the stars (given for~$X$, similar for~$Y$).
$^{{h}}$ Range  (for magnitudes~12 to~19) of the standard deviation of the difference between input and positions (given for~$X$, similar for~$Y$).
$^{{i}}$~Mean, sigma and maximum deviation of estimated diffuse background.
\end{table}