[PIPE2D-227] Use YamlCamera to describe PFS Created: 23/Jul/17  Updated: 28/Jul/17  Resolved: 28/Jul/17

Status: Done
Project: DRP 2-D Pipeline
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Story Priority: Normal
Reporter: rhl Assignee: rhl
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified


 Description   

Switch from FITS to yaml files to describe the PFS camera and detectors



 Comments   
Comment by rhl [ 23/Jul/17 ]

Not all that urgent, but I'd like to switch all the obs packages to yaml. The yamlCamera support is on master obs_base so it should be in w_2017_29.

Note that I switched the defect naming over too (it made sense to do it together as the chip names are used as keys in the yaml file)

Comment by price [ 26/Jul/17 ]
  • PfsCamera: what about changing the YamlCamera so that it will load from a default (in a class variable), instead of having the subclass doing the defaulting?
  • camera.yaml: I'm not familiar with the higher-order features of YAML, but I think I can guess what's going on — neat!
    • The indentation is weird: 3 spaces for some, 5 for others.
    • Am I right in thinking that you're faking up one large CCD with multiple individual ones? Or do you really have a single CCD with 8 amps?
    • Are you only including a single optical double spectrograph because that's all we've got hardware-wise at the moment (no NIR)?
    • Why are the names 1r and 1b within the YAML, but they're referred to as "r1" and "b1" in comments and descriptions?
  • genDefectRegistry.py looks familiar. Has it been copied from somewhere (like obs_subaru)? If so, can we push it up to perhaps obs_base (and maybe genDefectFits.py too)?
Comment by rhl [ 27/Jul/17 ]

> PfsCamera: what about changing the YamlCamera so that it will load from a default (in a class variable), instead of having the subclass doing the defaulting?

I didn't want to do that until it becomes the accepted camera description. Maybe a mistake, but let's handle the then (I want to rename CCDs to Detectors at the same time); one motivation is that I can get a usable obs_base from w_2017_29 if I don't make this change now. I'll add a note to some suitable ticket.

  • I'll fix the indentation;
  • PFS uses a pair of 2kx4k CCDs and pretends (at the FPGA level) that it's a single 8kx8k;
  • New don't have any IR detectors cold, but a pair of blue chips will be cold sometime within a month;
  • I'll fix the 1r/r1 problem (it's an open issue in DAMD)

The genDefectRegistry.py is lightly modified from obs_subaru (to deal with detector names not ccd numbers). We should open a ticket to do this properly and move to obs_base

Comment by rhl [ 27/Jul/17 ]

See DAMD-12

Comment by rhl [ 28/Jul/17 ]

Merged to master. You need obs_psf, datamodel, drp_stella and drp_stella_data as part of this work involved DAMD-12 (the 1r -> r1 transition)

Generated at Sat Feb 10 15:49:46 JST 2024 using Jira 8.3.4#803005-sha1:1f96e09b3c60279a408a2ae47be3c745f571388b.