Original problem:
Running OSC preprocessing script.
Gives an error with the master_dark.fit file saying the darks contain too many negative pixels.
Based on a similar report on Reddit, I wanted to try and calibrate/stack manually instead of with the script and then checking the box ‘optimize’ next to the dark fit file field.
This is when I encounter a second problem:
11:05:43: Sequence has no registration info, cannot use wFWHM weights, aborting
11:05:43: Stacking failed.
11:05:43: Stacking failed, read the log to fix the problem.
The screenshot in the tutorial shows to chose average stacking with rejection. Even when I change this to something without rejection (eg means stacking) I get the same error. Why does it throw an error for weighted FWHM when I’m not trying to use it for rejection?
It refers to a log to read to find the problem, I don’t find a log anywhere
What am I missing here?
EDIT/
Think I found the reason (bug?) for my second problem.
When I chose average stacking with rejection, there’s an additional field visible in my version that shows a menu what to choose for weight. This is not visible in the screenshot on the tutorial. This seems to be set to weighted FWHM by default.
But when changing the method, the field disappears, but the script seems to still hang on it.
When I set it to none, then change the method. It does work.
the second problem is a known issue, fixed for the next release I believe.
the first issue is that your calibration files do not match the level of your light images. Most modern cameras will not support dark optimization, so that’s probably not the solution. Look at the image statistics of your darks and lights, if the dark is brighter than the lights, it’s not right.
For issue 1. When I look at my stacked darks, they indeed look a lot lighter than my stacked lights. That’s normal for modern camera’s then if I understand correctly? So I shouldn’t bother with darks & biases but only focus on flats then?
I’m not familiar with modern cameras, I wouldn’t say it’s normal but it may be due to some settings that should not be activated. It’s likely that if you don’t have bias you will not be able to use flats because they will overcorrect if not calibrated.
I have the denoise function DEactivated (this is an in camera option that actually takes a dark frame after every regular frame and processes it internally). So then I don’t understand why my darks are lighter than my lights.
I took biases as wel, just forgot the flats.
Somehow, when I stack my lights only, I get better results then when I stack my darks, my biases and then calibrate the lights with those 2.
Sorry, Siril ( & astro photography in general) noob here. What do you mean with ‘use same offset level’?
Light leaks might be the cause, a lens cap not sufficient?
That depends. If you have lights around your camera you could have infrared leak.
If you use DSLR, make sure darks has same iso level. If you use astro camera, then OFFSET is one of the settings. OFFSET and GAIN must be the same between darks and lights.
Aha, I see. Thanks for explaining. Maybe there were indeed lights around me from the neighbours garden to the right of me.
The difference between stacked without calibration vs with calibration is very apparent here Siril, 750 lights, stacked without any calibration vs stacked with 80 biases and 50 darks, autostretch to get a quick view:
