when using a simulated cmd, if there are stars outside where there were artificial star tests the simulated stars got tossed. Fine when we're talking about super faint F814W=30 but not when the model produced a MS when the data show only ancient SF.
The fix was to assign NO ast correction if the star is found in a magnitude above the 50% completeness limit.
Better? Randomly apply mag_err to that to give a more realistic cmd distribution.
when using a simulated cmd, if there are stars outside where there were artificial star tests the simulated stars got tossed. Fine when we're talking about super faint F814W=30 but not when the model produced a MS when the data show only ancient SF.
The fix was to assign NO ast correction if the star is found in a magnitude above the 50% completeness limit.
Better? Randomly apply mag_err to that to give a more realistic cmd distribution.