Fix Version/s: None
DM-12058 revealed some clear problems in how PSFEx handles masking. We're hoping some of these will just go away when we rewrite the code to fix the bugs in PSFEx upsampling, but that is not guaranteed. We should follow up after the upsampling work is complete to see if we still have problems in masking, and create new tickets to fix them if we do.
I think we should keep these ones alive, and try to schedule someone to look into it - this was not an issue we fixed, and I'd completely forgotten about it. It's even possible this is in play in some of the recent warping/coaddition PSF problems we've seen in prepping for the next HSC release.
Thanks Jim Bosch — and paging Yusra AlSayyad so she can bear this in mind for the next round of DRP sprint planning.
The upsampling has been done on
DM-12411. Arun Kannawadi is going to follow up on this when he does DM-12058 this month. Marking it invalid as a duplicate
Robert Lupton, Jim Bosch — on
DM-12411, we established was no more work planned. Does that mean we're now happy with PSFEx performance, and we can regard this as a “won't fix” and DM-12058 as “done”?