Flat Wizard Multi Mode Camera Gain override
Issue #891
resolved
In the flat wizard it is currently possible to override the camera gain at a global level. When running the flat wizard in multi mode, exposing flats for both broad and narrow band it uses the single camera gain specified. It is typical to use different gain settings for narrowband and broadband filters especially in heavy light pollution.
My request is for an enhancement to the multi mode to allow the override of the gain on a per filter basis.
It is of course possible to run the flat wizard multiple times, one run for each gain needed, but this involves covering and uncovering the aperture for the dark flats which this enhancement would eliminate.
Thanks
Graham
Comments (10)
-
-
reporter
I don't use bias frames. Instead I use a dark flat which incorporates the camera bias. The bias is a reading of the inherent camera noise it is affected by the gain setting. For that reason I believe the dark flat should use the same gain as the lights. The only way to do this and still match the exposure time of the flat, would be be shoot the flat and dark flat at the same gain.
-
flat calibration with flat darks is decoupled from any aspect of the lights. therefore it does not matter if you shoot your flats and lights at different gains. Consistent gain would matter if you did use bias frames and wanted to use the same master bias for calibrating both flats and lights. There is no problem to fix here.
-
I am sorry, but it does matter, since e.g. PI needs a flat and a matching dark flat to calculate the proper parameters for „Mure Denoise“ to denoise the stacked lights. If only one is using another gain or offset it completely fails. It is mandatory that they do match exactly.
-
I’m honestly not familiar with this Mure Denoise and have never used it. Classic flat subtraction does not require matching gains, however. The original author’s explanation never mentioned this Mure Denoise.
-
I just wanted to add this crucial information, since it is really absolutely critical to have the matching flats and dark flats. It was one one of my failing points when starting to use Demure.
I also have noticed different sensor behavior with changing gain. E.g. flats for 3,5 nm UNB filter take up to 180s each. Gain has significant impact on exposure time and I noticed different fixed pattern noise, which is then differing from the lights. So you over- or undercorrect some pattern, which you will see in the end.
I have got used to strictly be consistent with gain and offset for flats and darks, since this was identified as the root cause for an issue for which I had requested help in the PI forum when calibration failed in WBPP. The clear culprits were not identical exposure / gain flats and flat darks in respect to the lights. Since I carefully take identical exposures the problem has been gone.
-
repo owner
- removed version
Removing version: 1.11 Nightly (automated comment)
-
The question of gain matching flats to lights was studied and discussed in detail recently. In short, gain matching flats is needed for cameras with dual gain modes to get proper results. On such cameras, many people recommend shooting broadband at gain 0 in low conversion mode and 100 (or whatever the trigger value is for the camera) for narrow band in high conversion mode. Given the recent studies and discussions by expert imagers confirming the need for gain matching flats on some cameras in particular, please reconsider adding a gain setting per filter in the flat wizard.
https://www.astroworldcreations.com/blog/understanding-flats-part-1-gain-matching
-
repo owner
- changed status to resolved
gain, offset, binning will all be adjustable on filter level for version 3.x
-
repo owner
Issue
#1114was marked as a duplicate of this issue. - Log in to comment
Gain doesn’t matter much with flats, hence the single knob. You’re presumably shooting into a flat panel or the lit sky anyhow, so light pollution isn’t a factor. Your flat darks just need to match, which the Flat Wizard arranges for already.