Results of 3 nights combined

[TiffsAndAstro]

I can just about live with a green gradient but with a red one too, its tricky to process out.. hopefully ill improve. i've attached the post processed sum of all 3 nights, total of 5h22m integration time. Also a screen shot of the auto stretch view of the stacked file from DSS after a photometric colour calibration.  I've overly post processed it again, but i think its the best i can do so far. I think i have the start of the dust lanes, but it could be my over processing. 

Would welcome any tips/advice or highlighting any problems. Focus and back focus seems ok. No obvious egg shaped stars, etc. 

 

 

[alacant]

10 minutes ago, TiffsAndAstro said:

its tricky to process out

Hi

If you can't eliminate the issue with the colour bands at source...

In Siril: Send the pp sequence (not the single  final image) through both Background Extraction and Banding Reduction (vertical) before registering and stacking as normal -> really clean background.

HTH

[Elp]

Exactly what I was going to suggest.

[alacant]

Still thinking: Part of the problem is that there are stacking artefacts and irregularities. The latter can be removed by pre processing with flat frames. Maybe you can still do that? If so, it would vastly simplify  processing. The former? Just crop, but don't attempt to process with them remaining. Please see the luminance below.
600d? So no in-camera bias or dark frames of any type. Instead simply subtract the offset (2048) during pre processing.

Cheers

Edited April 30 by alacant

[TiffsAndAstro]

30 minutes ago, alacant said:

Hi

If you can't eliminate the issue with the colour bands at source...

In Siril: Send the pp sequence (not the single  final image) through both Background Extraction and Banding Reduction (vertical) before registering and stacking as normal -> really clean background.

HTH

I had to stack in dss, as i couldn't get sirilic to fully finish due to it not being able to find .seq files.

I can't stack manually in siril yet, i still haven't read through the documents explaining it.

[TiffsAndAstro]

8 minutes ago, alacant said:

Still thinking: Part of the problem is that there are stacking artefacts and irregularities. The latter can be removed by pre processing with flat frames. Maybe you can still do that? If so, it would vastly simplify  processing. The former? Just crop, but don't attempt to process with them remaining. Please see the luminance below.
600d? So no in-camera bias or dark frames of any type. Instead simply subtract the offset (2048) during pre processing.

Cheers

This is really odd, as first step i do on the stacked outut file from dss is load it into graxpert and crop it. I guess ill have to have another go and make sure i appy the crop

I had a look at how to subtract offsets but as i can't stack manually in siril, i don't think i can do that either. 

Before (if?) i learn to stack manually in siril AND apply 2048 offset, is it worth me continuing to add to this one target?

[TiffsAndAstro]

And thanks everyone for the comments and help, much appreciated. I have rewired my wires a little better now. Not much more i can do until i get a nano windows pc to put on it and a better camera is a higher priority

[Elp]

All you do is:

1. Rhs pane, plus button to open your lights, select all lights, open, under the file list type in a name for the sequence, drop down menu next to it choose image sequence, process button.

2. Now you have a sequence containing all your lights (it'll take up a lot of HDD space if you keep making sequences which Siril does throughout the process). See the second tab, your active sequence will be there, the active sequence tends to be the most recent one created.

3. Go to background extraction menu item from the top menu, a BE popup menu will appear, as you're going to run it on all lights, select algorithm polynomial and set order to 1 as it's a basic linear gradient, generate your sample points or do it automatically, the points should not be on top of stars or nebulosity, press compute extraction and it'll preview the result, then there's a button underneath in the BE popup menu to run it on a whole image sequence. Simple.

Sirils RHS panes are self explanatory, you go through them left to right. You'd have to check the Siril documentation for specific items, or just mouse over them a popup explanation which is comprehensive pops up.

It really isn't difficult to use, and the results are far better than with DSS.

[Elp]

If you want to stack master stacks (which have been pre calibrated) you simply load the masters (still in linear unstretched form) as if they were lights, run the sequence, then go to registration and run that, then stack tab and run that. Done.

Edited April 30 by Elp

[TiffsAndAstro]

4 minutes ago, Elp said:

If you want to stack master stacks (which have been pre calibrated) you simply load the masters (still in linear unstretched form) as if they were lights, run the sequence, then go to registration and run that, then stack tab and run that. Done.

Ok ty for this. will have a go doing as you said this evening, will try and watch a video and read those docs first. I really haven't looked at any of those tabs or attempted anything with them except look at console output. 

[Elp]

If you want to go through calibration you load your calibration frames as if they were lights (one type at a time), sequence them, then go straight to stacking to create a master say, master dark. The written files will appear in your designated Siril working folder.

You can then use the master as part of your calibration tab when calibrating the lights. You need to enable the debayer option before saving and CFA check box option as you're using a colour camera.

[alacant]

1 hour ago, TiffsAndAstro said:

is it worth me continuing

If you have working flat frames for each session, yes. If not, probably best to make a new start.
Flat frames are not really optional.

Edited April 30 by alacant

[TiffsAndAstro]

6 minutes ago, alacant said:

If you have working flat frames for each session, yes. If not, probably best to make a new start.

I took flats for each session and they seem to be working ok.... Its possible some lint dropped from the cloth i stretched over the dew shield. Gonna try just using my tablet with no white cloth and hope its diffuse enough. I also had some difference in rotation and didn't do a 'get rotation from camera' before framing.

I think my result, above, is ok, especially for a first attempt at combining three sessions, but the "scores" in dss vary alot: (roughly) sess1 = 900-1400, sess2 = 600-900 and sess3 = 800-1900. 

Im reluctant to continue with it if my initial capture data is too ropey.

[TiffsAndAstro]

23 hours ago, Elp said:

All you do is:

1. Rhs pane, plus button to open your lights, select all lights, open, under the file list type in a name for the sequence, drop down menu next to it choose image sequence, process button.

2. Now you have a sequence containing all your lights (it'll take up a lot of HDD space if you keep making sequences which Siril does throughout the process). See the second tab, your active sequence will be there, the active sequence tends to be the most recent one created.

3. Go to background extraction menu item from the top menu, a BE popup menu will appear, as you're going to run it on all lights, select algorithm polynomial and set order to 1 as it's a basic linear gradient, generate your sample points or do it automatically, the points should not be on top of stars or nebulosity, press compute extraction and it'll preview the result, then there's a button underneath in the BE popup menu to run it on a whole image sequence. Simple.

Sirils RHS panes are self explanatory, you go through them left to right. You'd have to check the Siril documentation for specific items, or just mouse over them a popup explanation which is comprehensive pops up.

It really isn't difficult to use, and the results are far better than with DSS.

So im ploughing through this and wondered how it applies to multiple night data ? Do i process each night individually then stack them, or?

I just tried loading all my lights are doing as you say above but not sure its right way to do this? Im still very unsure about doing this manually

[Elp]

The example you've quoted is for one sessions data.

If you want to stack each sessions master stack you simply load each master stack into the list, create the sequence, go through registration tab, stacking tab.

Siril is capable of stacking multiple sessions data I believe but you need to use a script which I don't do as I like to see each sessions data, calibrated etc step by step, it's easier to fault find if you run into an issue.

[TiffsAndAstro]

23 minutes ago, Elp said:

The example you've quoted is for one sessions data.

If you want to stack each sessions master stack you simply load each master stack into the list, create the sequence, go through registration tab, stacking tab.

Siril is capable of stacking multiple sessions data I believe but you need to use a script which I don't do as I like to see each sessions data, calibrated etc step by step, it's easier to fault find if you run into an issue.

Bollox, sorry. Multiple threads/post, sorry

Manually stacking one night looks ok to do via siril, but im also trying to combine two sessions (and if that works, adding a third) via sirilic. It *almost* seems to work in sirilic but i have an error, and trying to work out what the problem is is tricky

[Elp]

Think I've asked this before, any reason why you're not using the GUI version of Siril? You just do as I said:

If you want to stack each sessions master stack you simply load each master stack into the list, create the sequence, go through registration tab, stacking tab.

It'll be very quick as each master stack will have already been calibrated, in your case you're only loading 3 images into the list, registering and stacking. Takes less than a minute.

I'd also edit your post...

Edited May 1 by Elp

[TiffsAndAstro]

21 minutes ago, Elp said:

Think I've asked this before, any reason why you're not using the GUI version of Siril? You just do as I said:

If you want to stack each sessions master stack you simply load each master stack into the list, create the sequence, go through registration tab, stacking tab.

It'll be very quick as each master stack will have already been calibrated, in your case you're only loading 3 images into the list, registering and stacking. Takes less than a minute.

I'd also edit your post...

Im still struggling reading through siril manual stacking guides, so i had a quick blast following your bullet points and it worked ok.

I am using the GUI version of siril, its just that for multiple sessions, sirilic seems to be the way. can save sirilic file list like in dss so it should save time in future.

[alacant]

2 hours ago, TiffsAndAstro said:

multiple sessions

Rather than having to make four independent stacks, perhaps easier/quicker...

Siril:

Pre process each night to obtain a pp_  file for each frame.
Then it's...
Conversion tab -> +
Add each pp_ file from all nights -> give it a name -> convert [1]
Register -Plot (remove the failed frames) -> Stack

That's it.

I'd recommend learning Siril manually to begin with. Perhaps move to Sirilic when you understand what the Siril modules are doing.

Cheers and HTH

[1] on Linux file systems we use symlinks to the files. Don't know if you can do that with ntfs. 

Edited May 1 by alacant
Spell

[TiffsAndAstro]

11 minutes ago, alacant said:

Siril:

Pre process each night to obtain a pp_  file for each frame.
Then it's...
Conversion tab -> +
Add each pp_ file from all nights -> give it a name -> convert [1]
Register
Stack

That's it.

I'd recommend learning Siril manually to begin with. Perhaps move to Sirilic when you understand what the Siril modules are doing.

Cheers and HTH

[1] on Linux file systems we use symlinks to the files. Don't know if you can do that with ntfs. 

I think dynamic links are the windows equivalent.

I will try as you suggest first chance I get.

Also I agree I need to learn manual stacking in siril but I'm finding it slow going.

I was going to try  manual stacking each session then try and manually stack the three results as part of my learning process

Appreciate all the help, ty

Edited May 1 by TiffsAndAstro

[Swillis]

If you were to stack each session individually, then as elp says you can check the quality of each session and identify and issues specific to that night. You could then load in all the preprocessed (pp_) calibrated files as a new sequence then register and stack as one batch.

I originally used the scripts for Siril, but I started using it manually and found it to be quicker. I took me a little time to figure it out and I made a few mistakes so had to go back and forth a bit, but I would go back to the scripts now. Overall it was a lot easier than I thought it would be.

I like that you can view the graphs showing how good the data is, which can help in deciding what cutoffs to choose for which frames to stack.

[TiffsAndAstro]

15 minutes ago, Swillis said:

If you were to stack each session individually, then as elp says you can check the quality of each session and identify and issues specific to that night. You could then load in all the preprocessed (pp_) calibrated files as a new sequence then register and stack as one batch.

I originally used the scripts for Siril, but I started using it manually and found it to be quicker. I took me a little time to figure it out and I made a few mistakes so had to go back and forth a bit, but I would go back to the scripts now. Overall it was a lot easier than I thought it would be.

I like that you can view the graphs showing how good the data is, which can help in deciding what cutoffs to choose for which frames to stack.

I think you might want to check out deep sky astro video on sirilic. Whilst I can quite get it to fully work, it's likely noob error.

It's (lol) easy to use and a nice way of storing file lists in multi sessions.

It even shows a colour flow chart of how you set the processes

of course, once I get it to actually fully work.

Will do as you and elp says above as soon as a get a chance. Otherwise I can only do multi session in dss and those two gradients are annoying. Might be there in siril though too, I guess. 

TY for the help. I'm slowly getting places. Above image is the my first I'm ok enough with to use as my desktop:)

Edited May 1 by TiffsAndAstro

[Swillis]

I did look at Sirilic a while back but it looked a little complicated, and I just ended up figuring out to use Sirilic manually. Maybe I'll take another look again

You've certainly made some very quick progress. Keep posting, I have enjoyed following your threads.

[TiffsAndAstro]

14 minutes ago, Swillis said:

I did look at Sirilic a while back but it looked a little complicated, and I just ended up figuring out to use Sirilic manually. Maybe I'll take another look again

You've certainly made some very quick progress. Keep posting, I have enjoyed following your threads.

Ty kind of you to say.

Definitely check out deep sky astro video on sirilic.

[TiffsAndAstro]

3 hours ago, alacant said:

Rather than having to make four independent stacks, perhaps easier/quicker...

Siril:

Pre process each night to obtain a pp_  file for each frame.
Then it's...
Conversion tab -> +
Add each pp_ file from all nights -> give it a name -> convert [1]
Register -Plot (remove the failed frames) -> Stack

That's it.

I'd recommend learning Siril manually to begin with. Perhaps move to Sirilic when you understand what the Siril modules are doing.

Cheers and HTH

[1] on Linux file systems we use symlinks to the files. Don't know if you can do that with ntfs. 

Getting closer, maybe. How important are the 3 stars i manually chose? Im guessing one or more are bad choices? will have a couple more goes on different combinations of stars. 

17:09:54: Reading FITS: file Test all nights PP_00061.fit, 3 layer(s), 5202x3464 pixels, 16 bits
17:09:54: Sequence processing partially succeeded, with 241 images that failed.
17:09:54: Execution time: 39.43 s
17:09:54: Registration finished.
17:09:54: 242 images processed.
17:09:54: Total: 241 failed, 1 registered.
17:09:54: Checking sequences in the directory: G:\astro\data\projects\m101\2024-04-27.
17:09:54: Closing sequence Test all nights PP_
17:09:54: Reading FITS: file r_Test all nights PP_00024.fit, 3 layer(s), 5202x3464 pixels, 16 bits
17:09:54: Sequence loaded: r_Test all nights PP_ (24->24)
17:10:10: The AutoStretch display mode will use a 16 bit LUT

seem to be getting further using a non default registration option - not the 123 star one  Global Star Alignment. Or is there a better option? Registration seems to be going ok now.

doesn't seem to be any option to remove the failed frames within plot. i assume they are the ones under the green curve? Goinna run stacking anyway and see what happens

ignore this, rejection options on stacking tab seem to do this. apologies, i was literally follwoing line by line, but this process has definately helped me understand siril better.

Edited May 1 by TiffsAndAstro