search

LSSTDESC / imSim

GalSim based Rubin Observatory image simulation package
https://lsstdesc.org/imSim
BSD 3-Clause "New" or "Revised" License
36 stars 15 forks source link

Add surface effects like Annealing/Bamboo Patterns as GalSim surface operations #108

Open cwwalter opened 6 years ago

cwwalter commented 6 years ago

We should implement fringing. Some formalism and a standalone simulation are here.

The place to put this into the sensor model now exists in the GalSim API as part of the sensor work. This might be implemented as a surface operation (like DCR is now). A list of effects to add below (please add to it):

Note other effects (like bleeding) are best handled elsewhere.

craiglagegit commented 6 years ago

Chris,

The annealing/bamboo patterns are fixed pixel distortions, and as such the logical thing would be to incorporate these into the tree ring code, which just moves the pixel boundaries slightly. However, I would need data or some sort of proposed model on how much to move each vertex, and I don't currently have this.

Craig

From: Chris Walter notifications@github.com Sent: Wednesday, April 11, 2018 7:14:43 PM To: LSSTDESC/imSim Cc: Subscribed Subject: [LSSTDESC/imSim] Add fringing and other surface effects as GalSim surface operations (#108)

We should implement fringing. Some formalism and a standalone simulation are herehttps://confluence.slac.stanford.edu/pages/viewpage.action?pageId=215845587.

The place to put this into the sensor model now exists in the GalSim API as part of the sensor work. This might be implemented as a surface operation (like DCR is now). A list of effects to add below (please add to it):

Note other effects (like bleeding) are best handled elsewhere.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/LSSTDESC/imSim/issues/108, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AI-5-hB2Ol8T4n7clT_bjnmLjdgurcXuks5tnriTgaJpZM4TRBsu.

cwwalter commented 6 years ago

The annealing/bamboo patterns are fixed pixel distortions, and as such the logical thing would be to incorporate these into the tree ring code, which just moves the pixel boundaries slightly.

Interesting. I thought it was also kind of an overall QE effect since the surface was modified. Does it really just change the pixel size?

craiglagegit commented 6 years ago

The annealing/bamboo patterns are fixed pixel distortions, and as such the logical >>thing would be to incorporate these into the tree ring code, which just moves the >>pixel boundaries slightly. Interesting. I thought it was also kind of an overall QE effect since the surface was modified. >Does it really just change the pixel size? I guess I don't really know. I thought the bamboo was due to tiny variations in the location of the gates and channel stops, which would manifest as pixel size variations. I'm less sure about the annealing features. Craig

From: Chris Walter notifications@github.com Sent: Thursday, April 12, 2018 10:25:18 AM To: LSSTDESC/imSim Cc: Craig S Lage; Comment Subject: Re: [LSSTDESC/imSim] Add fringing and other surface effects as GalSim surface operations (#108)

The annealing/bamboo patterns are fixed pixel distortions, and as such the logical thing would be to incorporate these into the tree ring code, which just moves the pixel boundaries slightly.

Interesting. I thought it was also kind of an overall QE effect since the surface was modified. Does it really just change the pixel size?

— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/LSSTDESC/imSim/issues/108#issuecomment-380882765, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AI-5-hfgDzht6emK9qVda2HmK7itk_BIks5tn439gaJpZM4TRBsu.

aaronroodman commented 6 years ago

Craig,

I’ve always assumed that these annealing patterns are surface-based QE variations, since there are only visible at Blue wavelengths.

if these were variations in gate/channel stops wouldn’t this be seen at longer wavelengths too?

Aaron

On Apr 12, 2018, at 2:46 PM, Craig Lage notifications@github.com<mailto:notifications@github.com> wrote:

The annealing/bamboo patterns are fixed pixel distortions, and as such the logical >>thing would be to incorporate these into the tree ring code, which just moves the >>pixel boundaries slightly. Interesting. I thought it was also kind of an overall QE effect since the surface was modified. >Does it really just change the pixel size? I guess I don't really know. I thought the bamboo was due to tiny variations in the location of the gates and channel stops, which would manifest as pixel size variations. I'm less sure about the annealing features. Craig

From: Chris Walter notifications@github.com<mailto:notifications@github.com> Sent: Thursday, April 12, 2018 10:25:18 AM To: LSSTDESC/imSim Cc: Craig S Lage; Comment Subject: Re: [LSSTDESC/imSim] Add fringing and other surface effects as GalSim surface operations (#108)

The annealing/bamboo patterns are fixed pixel distortions, and as such the logical thing would be to incorporate these into the tree ring code, which just moves the pixel boundaries slightly.

Interesting. I thought it was also kind of an overall QE effect since the surface was modified. Does it really just change the pixel size?

— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/LSSTDESC/imSim/issues/108#issuecomment-380882765, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AI-5-hfgDzht6emK9qVda2HmK7itk_BIks5tn439gaJpZM4TRBsu.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/LSSTDESC/imSim/issues/108#issuecomment-380954158, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AEEa88N_DLrboodjPYmm1dYKC-F-7yGFks5tn8sqgaJpZM4TRBsu.

Prof. Aaron Roodman Chair Dept. of Particle Physics & Astrophysics Kavli Institute for Particle Astrophysics & Cosmology SLAC National Accelerator Laboratory Stanford University

SLAC National Accelerator Laboratory E-mail: roodman@slac.stanford.edumailto:roodman@slac.stanford.edu 2575 Sand Hill Rd. Phone: 650-926-2705 MS 29 Menlo Park, CA 94025 URL: http://www.slac.stanford.edu/~roodman

craiglagegit commented 6 years ago

Aaron,

Since Chris said the same thing, I'm probably just wrong. I think I conflated the bamboo effects with the annealing patterns. If the annealing effects are only visible in the blue, then your explanation is probably correct. I thought the bamboo effects were due to in accuracies in the placement of the parallel gates and serial channel stops due primarily to errors in mask making. If this is the case, they should be independent of wavelength, as you said. Do you know if this is the case? I don't.

Thanks,

Craig

From: aaronroodman notifications@github.com Sent: Friday, April 20, 2018 12:40:38 PM To: LSSTDESC/imSim Cc: Craig S Lage; Comment Subject: Re: [LSSTDESC/imSim] Add fringing and other surface effects as GalSim surface operations (#108)

Craig,

I’ve always assumed that these annealing patterns are surface-based QE variations, since there are only visible at Blue wavelengths.

if these were variations in gate/channel stops wouldn’t this be seen at longer wavelengths too?

Aaron

On Apr 12, 2018, at 2:46 PM, Craig Lage notifications@github.com<mailto:notifications@github.com> wrote:

The annealing/bamboo patterns are fixed pixel distortions, and as such the logical >>thing would be to incorporate these into the tree ring code, which just moves the >>pixel boundaries slightly. Interesting. I thought it was also kind of an overall QE effect since the surface was modified. >Does it really just change the pixel size? I guess I don't really know. I thought the bamboo was due to tiny variations in the location of the gates and channel stops, which would manifest as pixel size variations. I'm less sure about the annealing features. Craig

From: Chris Walter notifications@github.com<mailto:notifications@github.com> Sent: Thursday, April 12, 2018 10:25:18 AM To: LSSTDESC/imSim Cc: Craig S Lage; Comment Subject: Re: [LSSTDESC/imSim] Add fringing and other surface effects as GalSim surface operations (#108)

The annealing/bamboo patterns are fixed pixel distortions, and as such the logical thing would be to incorporate these into the tree ring code, which just moves the pixel boundaries slightly.

Interesting. I thought it was also kind of an overall QE effect since the surface was modified. Does it really just change the pixel size?

— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/LSSTDESC/imSim/issues/108#issuecomment-380882765, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AI-5-hfgDzht6emK9qVda2HmK7itk_BIks5tn439gaJpZM4TRBsu.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/LSSTDESC/imSim/issues/108#issuecomment-380954158, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AEEa88N_DLrboodjPYmm1dYKC-F-7yGFks5tn8sqgaJpZM4TRBsu.

Prof. Aaron Roodman Chair Dept. of Particle Physics & Astrophysics Kavli Institute for Particle Astrophysics & Cosmology SLAC National Accelerator Laboratory Stanford University

SLAC National Accelerator Laboratory E-mail: roodman@slac.stanford.edumailto:roodman@slac.stanford.edu 2575 Sand Hill Rd. Phone: 650-926-2705 MS 29 Menlo Park, CA 94025 URL: http://www.slac.stanford.edu/~roodman

— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/LSSTDESC/imSim/issues/108#issuecomment-383201006, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AI-5-tSDLb8JdckB4orU_SZFTL7wdTDdks5tqjm2gaJpZM4TRBsu.

cwwalter commented 6 years ago

DCR has now been added in GalSim as a surface Op through https://github.com/GalSim-developers/GalSim/issues/955. When #101 is merged and turned on in the imSim code this option should also be also be added to the list of surface operation in the drawPhot call.

cwwalter commented 6 years ago

Note DCR is now acting as a surface operation (see closed issue #65). Other effects like fringing, bamboo etc are still missing.

cwwalter commented 1 year ago

Zhiyuan in your current work, have you implemented this as a surface operation? Or is it being done another way?

Zhiyuan-G commented 1 year ago

Hi Chris, right now fringing is implemented in parallel with sky gradient and vignetting. When adding fringing, I was trying to follow the same structure as Jim did for the sky gradient. You can see the code here

cwwalter commented 1 year ago

OK that is what I thought. That seems fine.

But, I wanted to check with @rmjarvis that this seems OK and that we are alright not using a surface operation for this as we have sometimes discussed.

rmjarvis commented 1 year ago

I suspect it's fine. It means that fringing won't affect the galaxy or star photons, but probably we don't care about that. It's most relevant for the sky photons.

cwwalter commented 1 year ago

The fringing is being addressed now in PR #356.

cwwalter commented 1 year ago

DCR was previously added (#65). Surface effects like bamboo annealing are not currently addressed.

I am going to change this older issue to just handle the Annealing/Bamboo Patterns (since there is discussion here) and I add a new issue focused on the fringing.