Open heather999 opened 6 years ago
This has always been a problem. Not having thought about it very hard, my instinct is to do it on the fork.
Perhaps we can branch this fork for Run1.1p as well, and then update the master of the fork? That way we can hopefully process both Run 1.1p and Run1.2p data. The Run1.1p branch could also be updated as we modify the config files on the master of the fork? Does that make sense?
Sounds good to me.
I think I can try to be helpful here @SimonKrughoff. The instructions are fairly straightforward for making this update. I just had a couple of questions.
Generate the gain/saturation file (with obs_lsstSim setup via eups): Build phosim (see Using Phosim for build instructions). You likely will need to create an SEDs directory (even if its empty): mkdir data/SEDs
Is it ok to use the SEDs that are already in place for Run1_2p (I'd assume so - just checking!)
Run phosim to generate a simple no-background simulation (give this several hours to complete): ./phosim $OBS_LSSTSIM_DIR/description/nostars_allchips -c examples/nobackground
Do we have an estimate of how many hours? I ask since I could try using the interactive queue at NERSC, but I'm limited to 4 hours.. guess it's worth a shot to see what happens.
@SimonKrughoff I set myself up to try this.. but I cannot find $OBS_LSSTSIM_DIR/description/nostars_allchips Any suggestions?
@heather999 I have a question in to John P about this. In essence, you need to generate the headers for the particular version of phosim, but you don't want to simulate any photons.
Should an update to the camera description happen in this fork or in the lsst/obs_lsstSim repo? How do we coordinate this update with others using the lsst/obs_lsstSIm repo? And.. we should determine when to do this update.. We're still working with Run 1.1p data and looking to update the configuration files too. Perhaps the camera description should be added after the configs have been updated to be more in line with the HSC configs.