I am thinking about which coordinate system we are going to use for the "hole mask". Perhaps MCS coordinate is good enough just for engineering purpose, providing that the position for the "hole mask" and MCS are fixed every time.

I think that it has to be in PFI x,y, just like the cobras. We are trying to confirm, or establish, the coordinate transform from MCS to PFI, which is available from the updated transform function which will soon be in pfs_utils. That function would take the MCS xpix, ypix coordinates and generate PFI x,y.

Thanks for clarifying this for me. After we install the mask, actually we can measure or use some calculation to know the PFI x,y for those holes. Then we get some images and MCS return the MCS xpix, ypix for those holes. I suppose there is a function in pfs_utils to establish or update the coordinate transformation from those data. Right?

We could use a similar design for cobra but this one only contains serial number and x, y positions. Is there any other concerns like how to number those holes or any additional parameters?

I do not think we will use pin-hole anymore. Let's close this ticket.