I created a figure showing how the fibers are distributed, using the information from https://github.com/Subaru-PFS/pfs_utils/blob/master/python/pfs/utils/dummyCableB.py and nominal detector map. Lines are colored according to the bundle color (yellow, blue, green and orange, + red are individual fibers.) In the bottom, I have added the median value of the x-coordinate of each fiber to give you the idea where the fibers are.

Could you please add your code to pfs_utils?

Added. The code is also available as a part of the larger notebook in https://github.com/Subaru-PFS/dev_2ddrp/blob/master/2d_PSF_code/Zernike_Work_Mar_Apr_2020.ipynb

I don't think this is useful as a notebook, especially if it's part of another notebook that's already archived. I suggest just adding the notebook to this ticket.

Done

Can you add faint lines for the amps?

“The better we serve, the greater the fulfilment.”

I am copying below the email that I sent to Jim, regarding the DCB and our need for possible shutters.
########
Without a shutter, we can get 18 isolated fibers/ 22 total fibers (orange, yellow, blue, and red3) or 13 isolated / 29 total fibers (orange, yellow, blue, green, and red3). As you can see green bundle overlaps 6 times and gives only 1 fiber that does not overlap. So including the green bundle gives us 1 new isolated fiber and ``destroys’’ 6 isolated fibers, for a total loss of 5 isolated fibers. In case we use 3 bundles, we could potentially include one more isolated fiber at the absolute edge of the detector (red1 or red7), but those are just a bit above 60 pixels from new fibers from fibers bundles and I do not think that the risk outweighs the benefits.

With one shutter we can get 19 isolated fibers / 29 total fibers. (We can get all of the fibers and with turning on/off the offending green fiber we can get more isolated lines when we need them)

With two shutters we can get 23 isolated fibers / 29 total fibers (with three runs needed to get all of the isolated fibers).

The only concern is that the green bundle is the bundle that we have now! So if the green bundle is not available we would not be able to compare the results across time or with LAM data at a high level (on the other hand if we assume that instrument properties do not change and we get some green fiber data now, we would basically cover the whole thing, although with green fibers having different data quality).

#######

Updated the notebook attached to this ticket with the code.