Problems with Dark/Flat correction during livestacking in SC3.3?

Discussions, Bug Reports and Issues related to Beta versions of SharpCap
User avatar
DocHighCo
Posts: 52
Joined: Wed Nov 18, 2020 2:20 pm

Re: Problems with Dark/Flat correction during livestacking in SC3.3?

#11

Post by DocHighCo »

Hello,

I just came in from outside and could take some frames through some gaps in the clouds before they closed. But, not much more.
The result is repeatable with the same settings - it´s bad. I have raw frames. I have a saved 16 and 32 bit stack and I have the "as seen on screen" stack. Older settings with older darks taken with SC3.2 seem to stack fine though in SC3.3 .
The only difference I see is that usually I use gains above 100 with the ASI294. This time I only used a gain of 10. So the frames are pretty dark. But they stacked fine nevertheless. Just the result looks weird.

Regards

Heiko
User avatar
DocHighCo
Posts: 52
Joined: Wed Nov 18, 2020 2:20 pm

Re: Problems with Dark/Flat correction during livestacking in SC3.3?

#12

Post by DocHighCo »

So, I am getting nearer to the problem. I used "Folder camera" to restack the raw-frames.
To make it short. There is a problem with the alignment option and dark frame subtraction enabled together for my case.

The raw frames with dark frame subtraction and alignment option "off" can be stacked (well, ignore the quality and the drift - it was very cloudy):
Image

When I hook the "align frame" option the stack comes out like this:
Image
Changing digital gain and other align parameters doesn´t seem to change the outcome.

Without dark frame subtraction and "align frame" hooked it works in SC3.3 though, as do all the options with the same raw and dark frames in SC3.2.
Very strange!


Best Regards

Heiko
User avatar
admin
Site Admin
Posts: 13350
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Problems with Dark/Flat correction during livestacking in SC3.3?

#13

Post by admin »

Hi,

can you share all the frames needed for me to repeat the test on your data via dropbox or Google Drive or something similar please?

If I can make it happen here then even if I can't spot the problem causing it straightaway I can work back through older versions of the code until I find the change that broke it.

Cheers, Robin
User avatar
DocHighCo
Posts: 52
Joined: Wed Nov 18, 2020 2:20 pm

Re: Problems with Dark/Flat correction during livestacking in SC3.3?

#14

Post by DocHighCo »

Dear Robin,

I created a Dropbox link: https://www.dropbox.com/sh/u38jd4xqenum ... X71-a?dl=0
I hope the download works. This is all a I needed to recreate the effect with the folder camera.

Regards

Heiko
User avatar
admin
Site Admin
Posts: 13350
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Problems with Dark/Flat correction during livestacking in SC3.3?

#15

Post by admin »

Hi Heiko,

Thanks for that – I'm downloading it now and will see what I can work out.

Cheers, Robin
User avatar
admin
Site Admin
Posts: 13350
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Problems with Dark/Flat correction during livestacking in SC3.3?

#16

Post by admin »

Hi,

a new version of SharpCap 3.3 is now uploaded which should fix this issue.

cheers,

Robin
User avatar
DocHighCo
Posts: 52
Joined: Wed Nov 18, 2020 2:20 pm

Re: Problems with Dark/Flat correction during livestacking in SC3.3?

#17

Post by DocHighCo »

Thanks Robin,

I will have a look out for the new version.

Regards

Heiko
User avatar
DocHighCo
Posts: 52
Joined: Wed Nov 18, 2020 2:20 pm

Re: Problems with Dark/Flat correction during livestacking in SC3.3?

#18

Post by DocHighCo »

It works for me as well with the latest SC3.3 version.

Thanks again

Heiko
User avatar
admin
Site Admin
Posts: 13350
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Problems with Dark/Flat correction during livestacking in SC3.3?

#19

Post by admin »

Hurrah!

Along the way I found a bug that I originally thought was the problem, but turned out not to be. I fixed that bug too and it should improve the way that hot pixels are dealt with in dark subtraction, possibly leading to a reduction in problems with raining noise (I hope so, anyway).

Cheers, Robin
Post Reply