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

How do you know when to stop processing? M1 example

  • Please log in to reply
13 replies to this topic

#1 steinw

steinw

    Explorer 1

  • -----
  • topic starter
  • Posts: 65
  • Joined: 07 Aug 2014

Posted 22 February 2020 - 08:04 AM

Pixinsights learning curve is not only steep, it's very long as well. And it's not only about mastering a few tricks, there is an endless amount of them as well. I always seem to get to a point where I'm not completely happy, but after xx hours I think I need to move on. I know this is just a part of the learning curve, and for most of my stuff I will probably go back to redo at some time. Like the example below: I really would like to do something better with the stars.

 

So what is your strategy? When do you know that enough is enough?

 

get.jpg?insecure


  • Stelios, elmiko, zakry3323 and 1 other like this

#2 Tapio

Tapio

    Vanguard

  • -----
  • Posts: 2,415
  • Joined: 24 Sep 2006
  • Loc: Tampere, Finland

Posted 22 February 2020 - 08:30 AM

Your target looks good and stars are not too 'distracting'.
However, I would try bring some colour in stars.

When to stop, very subjective decision.
When I'm satisfied, and still satisfied next day.
  • elmiko likes this

#3 steinw

steinw

    Explorer 1

  • -----
  • topic starter
  • Posts: 65
  • Joined: 07 Aug 2014

Posted 22 February 2020 - 08:35 AM

Your target looks good and stars are not too 'distracting'.
However, I would try bring some colour in stars.

When to stop, very subjective decision.
When I'm satisfied, and still satisfied next day.

Thanks, that's an excellent "rule".

I don't have any RGB data for the stars. For the strict SHO version at Astrobin, the stars looked almost natural in color, but for this odd blend narrowband version I just used Luminosity stars. I should probably make a habit of capturing a few RGB frames as well.
 



#4 elmiko

elmiko

    Skylab

  • *****
  • Posts: 4,164
  • Joined: 27 Jul 2009
  • Loc: Arizona

Posted 22 February 2020 - 08:37 AM

Looks really great to me! I usually take a break from processing. Come back the next day and see if I still like what I see.


  • wrnchhead likes this

#5 wrnchhead

wrnchhead

    Surveyor 1

  • *****
  • Posts: 1,590
  • Joined: 28 Aug 2017
  • Loc: NE Kansas

Posted 22 February 2020 - 09:29 AM

Looks really great to me! I usually take a break from processing. Come back the next day and see if I still like what I see.

I usually wind up, after a day or a few passes, that I liked my original processing more than I did at the time. 


  • TOMDEY likes this

#6 Gipht

Gipht

    Vanguard

  • *****
  • Posts: 2,204
  • Joined: 12 Nov 2016
  • Loc: Prescott Valley, AZ.

Posted 22 February 2020 - 09:50 AM

Very nice picture.   With the stars you might try a version aimed at star color.  Sometimes the central object is so faint that many of the stars are clipped with the exposure time needed.  If you can find more color, then making a star version, adding more saturation and blending the two versions can sometimes help bring some color to the  star field.



#7 zakry3323

zakry3323

    Apollo

  • *****
  • Posts: 1,408
  • Joined: 11 Apr 2016
  • Loc: Pittsburgh

Posted 22 February 2020 - 10:07 AM

I feel this pain with every image.

My strategy is to identify 2 or 3 processes to practice on that I've recently learned, and learn (or at least dabble with) one new process per image. That way I'm always progressing and challenging myself. I stop when I start to agonize. If I'm agonizing, I'm not having fun, and I know that I'm not going to learn as well or want to come back to processing so much the next time. So I get every image to where I say "This is acceptable for my current abilities", keep all of my files very well organized and stored, and take a break. After several months keeping things interesting by working on new images, I can come back to where I "left off" on a prior image with fresh eyes, a new perspective, and hopefully some better honed skills. Sometimes this means going way back in the workflow and doing things "more better", which can feel like a waste of time, but that's the cost of learning. 



#8 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 18,956
  • Joined: 27 Oct 2014

Posted 22 February 2020 - 10:13 AM

When a new version is the same or even clearly worse than the last, and I have little idea about how to improve it. 

 

I usually do 10-50 hours, depending on the complexity of what I'm trying to do.  Mixing narrowband and broadband data generally has me toward the high end.

 

But it's like total imaging time.  Simply a personal decision.


Edited by bobzeq25, 22 February 2020 - 10:13 AM.


#9 TOMDEY

TOMDEY

    Fly Me to the Moon

  • *****
  • Posts: 5,939
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 22 February 2020 - 10:51 AM

Whatever you do, keep your original original data! Revisiting one's "perfectly-processed" trophy-image... is kinda like waking up from a hangover, looking in the mirror, and and reaching for the makeup removal cream.    Tom

Attached Thumbnails

  • 22.1 clown makeup.jpg

  • Stelios, steinw, zakry3323 and 1 other like this

#10 DrGomer

DrGomer

    Mariner 2

  • *****
  • Posts: 220
  • Joined: 07 Aug 2019

Posted 22 February 2020 - 11:05 AM

Thanks, that's an excellent "rule".

I don't have any RGB data for the stars. For the strict SHO version at Astrobin, the stars looked almost natural in color, but for this odd blend narrowband version I just used Luminosity stars. I should probably make a habit of capturing a few RGB frames as well.
 

Could one use an inverted star mask and then apply a photometric calibration to get stars  close to the right RGB color?



#11 DrGomer

DrGomer

    Mariner 2

  • *****
  • Posts: 220
  • Joined: 07 Aug 2019

Posted 22 February 2020 - 11:10 AM

 

But it's like total imaging time.  Simply a personal decision.

what helps me make that decision is knowing what a S/N=2 improvement looks like and then see how many more hours it'll take to reach 4x the current integration time.  In the southwest where I'm at, I tend to have a lot of clear skies (not right now :p) so I can afford a 4 day image session for a single target. Others where clear skies are a luxury, 1 day per target is about it since there aren't a ton of good imaging days in the year. 



#12 pfile

pfile

    Fly Me to the Moon

  • -----
  • Posts: 5,467
  • Joined: 14 Jun 2009

Posted 22 February 2020 - 11:25 AM

Could one use an inverted star mask and then apply a photometric calibration to get stars  close to the right RGB color?

i don't think the right wavelengths are there in the SHO data to get the star color correct - H and S are both deep red and O is teal of course. and then they are mapped funny with the Ha's red color going to the green channel...

 

rob


  • bobzeq25 likes this

#13 steinw

steinw

    Explorer 1

  • -----
  • topic starter
  • Posts: 65
  • Joined: 07 Aug 2014

Posted 22 February 2020 - 12:08 PM

I feel this pain with every image.

My strategy is to identify 2 or 3 processes to practice on that I've recently learned, and learn (or at least dabble with) one new process per image. That way I'm always progressing and challenging myself. I stop when I start to agonize. If I'm agonizing, I'm not having fun, and I know that I'm not going to learn as well or want to come back to processing so much the next time. So I get every image to where I say "This is acceptable for my current abilities", keep all of my files very well organized and stored, and take a break. After several months keeping things interesting by working on new images, I can come back to where I "left off" on a prior image with fresh eyes, a new perspective, and hopefully some better honed skills. Sometimes this means going way back in the workflow and doing things "more better", which can feel like a waste of time, but that's the cost of learning. 

Pain - I can relate to that. It's good to see that a lot of you struggle with the same issues. Thanks for all your replies. But I sure wish I could adopt the systematic approach above.

 

For the image in this thread my decision to end the processing was based on the following:

1) I had already spent probably 25-30 hrs

2) I had problem remembering the last step I did

3) I had too many alternatives and ideas, and the whole thing just seemed to branch out in complete chaos with different versions ++.

4) I have som other data I really want to process, but I also really wanted to finalise one image. (Too many unfinished works)

 

So I decided to post versions with three different blends as I couldn't make up my mind.



#14 TOMDEY

TOMDEY

    Fly Me to the Moon

  • *****
  • Posts: 5,939
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 22 February 2020 - 03:49 PM

Processing can be obsessive-complsive, addictive and sense-numbing. When enough seems never quite enough, there are three psychological affectives at play:

 

>desensitization of one's sensory-stimulation/aesthetic sensibilities

>inadequate, poor quality data becoming overtly manifest

>wanting the target objet to be somehow more spectacular than it actually is!

 

I usually wind up, after a day or a few passes, that I liked my original processing more than I did at the time. 

Me too! When I process, I try light, medium and heavy processing, and save all three versions. Next day, it is the subtle version that wins the show. The others look ham-handed. I have come to conclude that it's my first bullet up there. Too much time in the processing seat had dulled the senses. It's like going to a Rock Concert... and realizing that your a bit more deaf next day... and that the music really wasn't all that good as it seemed the night before.    Tom

Attached Thumbnails

  • 23.1 overprocessing.jpg

Edited by TOMDEY, 22 February 2020 - 03:50 PM.

  • Stelios, steinw and wrnchhead like 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