Content Comparison

5min

project status

15min

Carlos feedback

15min

Next steps

• Current implementation of the the dism2.0 is mostly a wrapper to Luca’s jupyter notebook.

• When the input catalog is an IRAF-based dsimulator output, the dsim2.0 output matches the output of IRAF-based dsimulator.

• dsim2.0 generates no slits when the input catalog has no pre-selected objects.

• dsim2.0 does not allow the user to input the mask center.

• The dsim2.0 code works on mm on the focal plane from the beginning. The IRAF-based dsimulator works in arcsec and does the transformation to mm only at the very end of the mask generation process.

30min

Discussion

• We need to decide which approach to take, whether to obtain the mask solution using the jupyter notebook adaptation from IRAF to Python (option 1), or to obtain the solution empirically based on observations (option2).

• There is value in obtaining the solution empirically, because this is the approach used by LRIS. It can be beneficial if/when dsim2.0 is extended to LRIS.

• The advantage of using the adaptation to Python of the IRAF-based code is that we know the IRAF dsimulator works. It has been producing scientific papers for nearly two decades.

15min

Next steps

• The minimum requirement for dsim2.0 to be accepted should be to reproduce the same results as the IRAF disimulation. Acceptable exceptions are targets close to the detector gaps, vignetting areas and bad columns, because these regions of avoidance are better represented in dsim2.0 than in the old dsimulator.

• We will continue the approach of using Luca’s jupyter notebook as a baseline for dsim2.0.

• We need to better understand the part of the code that does the automatic selection of targets.