Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Adding new subs to a previous imaging session

  • Please log in to reply
28 replies to this topic

#26 Desertanimal

Desertanimal

    Vostok 1

  • -----
  • topic starter
  • Posts: 116
  • Joined: 20 Feb 2021

Posted 10 April 2021 - 05:44 PM

So I ran two stacks in DSS.

 

It was two night's of data. Two different exposures (:30 and :45), same ISO. 

The first stack I just tossed all of the data, lights, darks, flats and bias into the main group. The second stack I put only bias into main group and lights, flats and darks into groups 2 and 3 from each night. 

 

Looking only at the resulting .tif images, what I see is the grouped version has a slightly wider histogram. They each have two sets of peaks which baffles me on how to stretch it but that' another point. 

 

I get your point of using calibration subs vs masters equals the same output. So this test only checked to see if they were handled differently loaded as groups or simply tossed in and labeled. 


  • Mike in Rancho likes this

#27 Mike in Rancho

Mike in Rancho

    Viking 1

  • -----
  • Posts: 659
  • Joined: 15 Oct 2020
  • Loc: Alta Loma, CA

Posted 10 April 2021 - 07:33 PM

Are there any other concerns I should have by having a calibration Library?
Clear Skies.

Hard to say.  You're venturing into territory I usually don't go.  That means you get to let us know how it works. grin.gif

 

Generally speaking, it seems to me that the most reusable frame is the master bias per your ISO.  The only thing I can think that could go wrong there is the master is too out-of-date, or, whatever you are trying to calibrate now has different RAW settings.  Sometimes that happens:  For example if you underexposed and need to bump the gain so DSS can even see any stars, your master bias created with normal RAW settings probably doesn't match now.  Recently, I changed my default DSS setting to have the no WB box checked off.  So, I had to create a new master bias, which I can do as the subframes were also still stored away.

 

Master darks and dark flats probably come next.  Same concerns as above, but also add in whether or not a bias frame was used in their creation, or not.  Also temperature - not always possible with a lot of DSLR's.  In fact at the moment I am not using them as I can find little improvement.  Will see if that holds into the summer temps, or maybe I'll start trying LENR.  For now though, I'm just stacking lights, flats, and bias.

 

Master flats I would put last as far as library capability, for all the previous concerns plus dust motes and possible vignetting orientation.  Here's the part I don't know yet - at some point, DSS scales the flats (or master flat) to the lights.  Now, is that done during creation of the master flat, or during stacking after the master has already been made?

 

Being unsure, what I have done is actually just duplicate the flat subs from a different session (same ISO and exposure) into a new folder, and then let DSS register and stack as if they were new flats taken for that purpose.

 

Best thing for reuse and library though is to have all the details and settings noted, and maybe also archive DSS' info file with it, in case you need to open it up and look.


  • galacticinsomnia likes this

#28 galacticinsomnia

galacticinsomnia

    Viking 1

  • *****
  • Posts: 768
  • Joined: 14 Aug 2020
  • Loc: Pacific Northwest - Oregon

Posted 10 April 2021 - 07:35 PM

So I ran two stacks in DSS.

 

It was two night's of data. Two different exposures (:30 and :45), same ISO. 

The first stack I just tossed all of the data, lights, darks, flats and bias into the main group. The second stack I put only bias into main group and lights, flats and darks into groups 2 and 3 from each night. 

 

Looking only at the resulting .tif images, what I see is the grouped version has a slightly wider histogram. They each have two sets of peaks which baffles me on how to stretch it but that' another point. 

 

I get your point of using calibration subs vs masters equals the same output. So this test only checked to see if they were handled differently loaded as groups or simply tossed in and labeled. 

IMO, taking an average of all your data, including calibration files, in one bulk run will give you less noise and overall a better final data set.  I don't know the mechanics, and admittedly, is a total guess, and maybe someone can school me on it, but it seems taking sigma clipping or median averages out of a greater set of data would provide better results since all the data is evaluated against each other for the best possible outcome.

Clear Skies !


 


  • Mike in Rancho and Desertanimal like this

#29 galacticinsomnia

galacticinsomnia

    Viking 1

  • *****
  • Posts: 768
  • Joined: 14 Aug 2020
  • Loc: Pacific Northwest - Oregon

Posted 10 April 2021 - 07:42 PM

Hard to say.  You're venturing into territory I usually don't go.  That means you get to let us know how it works. grin.gif

 

Generally speaking, it seems to me that the most reusable frame is the master bias per your ISO.  The only thing I can think that could go wrong there is the master is too out-of-date, or, whatever you are trying to calibrate now has different RAW settings.  Sometimes that happens:  For example if you underexposed and need to bump the gain so DSS can even see any stars, your master bias created with normal RAW settings probably doesn't match now.  Recently, I changed my default DSS setting to have the no WB box checked off.  So, I had to create a new master bias, which I can do as the subframes were also still stored away.

 

Master darks and dark flats probably come next.  Same concerns as above, but also add in whether or not a bias frame was used in their creation, or not.  Also temperature - not always possible with a lot of DSLR's.  In fact at the moment I am not using them as I can find little improvement.  Will see if that holds into the summer temps, or maybe I'll start trying LENR.  For now though, I'm just stacking lights, flats, and bias.

 

Master flats I would put last as far as library capability, for all the previous concerns plus dust motes and possible vignetting orientation.  Here's the part I don't know yet - at some point, DSS scales the flats (or master flat) to the lights.  Now, is that done during creation of the master flat, or during stacking after the master has already been made?

 

Being unsure, what I have done is actually just duplicate the flat subs from a different session (same ISO and exposure) into a new folder, and then let DSS register and stack as if they were new flats taken for that purpose.

 

Best thing for reuse and library though is to have all the details and settings noted, and maybe also archive DSS' info file with it, in case you need to open it up and look.

When I say a master set, I don't mean a Master.tif files.
I actually keep a library of calibration raws, and I stack them with my lights, depending on the camera, iso, scope used, for my target. 

I don't like the idea of using the "Master" precalibrated stacks because when running a file compare utility on them, have revealed when not making any changes to the settings in DSS, and not changing any options, there were definite differences in both file size and data, even if it wasn't visually noticable which is why I do create fresh stacks, but my raw images are used in the organization as outlined. 

Clear Skies !! 


  • Mike in Rancho likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics