Go Back   IceInSpace > Equipment > Software and Computers

Reply
 
Thread Tools Rate Thread
  #1  
Old 24-06-2019, 12:56 PM
Startrek (Martin)
Registered User

Startrek is offline
 
Join Date: Dec 2017
Location: Sydney and South Coast NSW
Posts: 6,031
Autostakkert 3 - Quality Graph and Percentage Frame Boxes

Autostakkert 3 Stacking Options
Frame Percentage to Stack

I’ve been using Autostakkert 3 for nearly a year now with mixed results and still can’t get my head around looking at the quality graph and using the “Frame Percentage to Stack” boxes to understand their relationship

I know when your frames are Analysed they are placed in the quality graph in order of best to worse from left to right on the graph and you can move the green vertical line along to see the quality of the images.

( refer to attached photos )

Questions
What does the 50% mean at the end of the quality graph ?

Are the 4 sectors of the quality graph related to the 4 boxes in the Frame Percentage to Stack option ?

I usually use 10 , 25, 50 and 75% in the boxes ( without any logical understanding) but when viewing the final stacked images there seems to be very little difference in final image quality both with my planetary and lunar images

I viewed a few YouTube clips on Autostakkert but still don’t fully understand the criteria or logical method to determine how you look at the quality graph and then set your percentage boxes

Appreciate any advice or assistance from IIS Planetary and Lunar imagers with the above

Thanks in advance
Attached Thumbnails
Click for full-size image (E5E9157A-4032-458D-9464-647D1B1DDA1D.jpeg)
95.1 KB97 views
Click for full-size image (8AC25D8D-11D8-49A6-9C8C-8AF81DE68044.jpeg)
85.8 KB67 views
Reply With Quote
  #2  
Old 25-06-2019, 10:34 PM
Startrek (Martin)
Registered User

Startrek is offline
 
Join Date: Dec 2017
Location: Sydney and South Coast NSW
Posts: 6,031
I think I’ve answered my own question today
Watched a YouTube clip by Emanuel Polichronis on Autostakkert ( I think he’s from Australia)
He explained about the quality graph in reference to the Frame Percentage Boxes. You follow the green graph line from the left until it hits the 50% horizontal line , this represents the best frames of your stack , then at this point left click and hit CTRL and a percentage number appears in the first box of your Frame Percentage Boxes. Say if it’s 45% he said to use say 15% in the first box , 25% in the second box , and use the third box as an option at 45% but don’t go over 45%. Generally he would just use 15% in the first box and 25% in the second box on most occasions to keep the results simple
He reasoning and logic seem ok , but not definitive. I guess this planetary stacking is a bit of a gamble in any case
For now I will use his criteria and method until someone else comes up with a better solution or better way of improving results.
Does anyone use Autostakkert 2 or 3 ???
Does anyone have a comment ??
Cheers
Reply With Quote
  #3  
Old 25-06-2019, 10:53 PM
Merlin66's Avatar
Merlin66 (Ken)
Registered User

Merlin66 is offline
 
Join Date: Oct 2005
Location: Junortoun Vic
Posts: 8,904
I use AS3! For solar imaging.
If you look at the graph....
The vertical line as are at 25/50/75% of the quantity of subs
The horizontal lines are the quality rating lines, 75/50 and 25%
I usually use the point where the green line would drop below say 90% to determine, roughly the number of frames to stack....
Using the selection of say “best 10%” allows you to look at the green curve to estimate the number of frames being stacked.
Reply With Quote
  #4  
Old 27-06-2019, 09:42 AM
sil's Avatar
sil (Steve)
Not even a speck of dust

sil is offline
 
Join Date: Jun 2012
Location: Canberra
Posts: 1,474
AS3! here too. I just stick with 25/50/75% outputs and leave it at that. Very little difference between them really and as AS applies wavelets sharpening already I dont do anything extra to overprocess things. I'm usually throwing thousands of frames at it so differences are small between 25/50/75 as they all are using a lot of frames. when using a smaller capture duration with less frames I tend to examine each output for quality differences and possibly redo with values around the best one. sometimes there are also just processing artifacts where redoing gets better results at a similar percentage value.

hint when using AS try adjusting the display panel brightness when placing APs especially if you have faint moons in the shot you want to bring out, but too much brightness and its hard to place surface APs. balance wha you do, the brightness is only for the display panel and NOT applied to the data at all, its an aid to help place alignment points.
Reply With Quote
  #5  
Old 27-06-2019, 01:16 PM
Startrek (Martin)
Registered User

Startrek is offline
 
Join Date: Dec 2017
Location: Sydney and South Coast NSW
Posts: 6,031
Thanks Ken and Sil

It’s a strange quality graph I must say and even stranger to use it , a bit hit and miss
The best frames are definitely from the far left moving to the right with less quality, I moved the green vertical line along from left to right until the end or far right and checked them, they reduce in quality from left to right
So the green quality line drops gradually as it moves from left to right that means that the quality drops as well
So therefore, why can’t we just select the first 15% from far left or start heading to the right until you hit the 15% mark ?? Or even 20% or 30%
It’s damn confusing this graph and it’s function, I’m sure a more intuitive interactive graph could have been designed into the software
I’ve looked at few YouTube clips and all of them have this part as a grey area of explanation. The remainder of the program makes perfect sense and is basically a no brainer
Sorry to complain or pardon my ignorance !
Reply With Quote
  #6  
Old 27-06-2019, 01:47 PM
Stefan Buda
Registered User

Stefan Buda is offline
 
Join Date: Jul 2013
Location: Melbourne, VIC
Posts: 832
Interesting that no one seems to have an explanation for the need for multiple boxes. I never bother using more than one box. Usually I just specify the number of frames in the first box. Probably I'm doing it wrong but I don't understand the need for more than one box.
Reply With Quote
  #7  
Old 27-06-2019, 02:52 PM
Tinderboxsky's Avatar
Tinderboxsky (Steve)
I can see clearly now ...

Tinderboxsky is offline
 
Join Date: Feb 2012
Location: Kingston TAS
Posts: 1,035
I do exactly the same. Seem to work well.

Quote:
Originally Posted by Stefan Buda View Post
Interesting that no one seems to have an explanation for the need for multiple boxes. I never bother using more than one box. Usually I just specify the number of frames in the first box. Probably I'm doing it wrong but I don't understand the need for more than one box.
Reply With Quote
  #8  
Old 27-06-2019, 11:09 PM
Stefan Buda
Registered User

Stefan Buda is offline
 
Join Date: Jul 2013
Location: Melbourne, VIC
Posts: 832
Quote:
Originally Posted by Tinderboxsky View Post
I do exactly the same. Seem to work well.
Yes, but the mystery remains. There must be a good reason for the multiple boxes.
Reply With Quote
  #9  
Old 28-06-2019, 03:10 AM
Merlin66's Avatar
Merlin66 (Ken)
Registered User

Merlin66 is offline
 
Join Date: Oct 2005
Location: Junortoun Vic
Posts: 8,904
The answer:
https://m.youtube.com/watch?v=Ci_aMmO4DBA

About 16:00into the video tutorial he discusses the options.
Reply With Quote
  #10  
Old 28-06-2019, 09:37 AM
Stefan Buda
Registered User

Stefan Buda is offline
 
Join Date: Jul 2013
Location: Melbourne, VIC
Posts: 832
Thanks Ken, mystery solved.
Looks like I won't need to change my processing routine.
Reply With Quote
  #11  
Old 28-06-2019, 04:02 PM
Startrek (Martin)
Registered User

Startrek is offline
 
Join Date: Dec 2017
Location: Sydney and South Coast NSW
Posts: 6,031
Yes I’ve watched this YouTube on Autostakkert plus his others, Registax etc..
He is still not 100% certain how to use the quality graph and the 4 boxes , only his analysis with trial and error ( in particular the part where he mentions percentages etc.. law of averages , he’s going around in circles and you won’t know until you see the outcome )
It’s still guess work based on a model and “hope for the best” outcome !
Every time I use 15, 25, 50 , 75 in the boxes or 10, 25, 50 , the end result looks the same for each, based on up to a 2000 frame stack

I think I’m going around in circles , we’ll leave it there and I’ll keep guessing my way through like before
Reply With Quote
  #12  
Old 28-06-2019, 06:09 PM
Merlin66's Avatar
Merlin66 (Ken)
Registered User

Merlin66 is offline
 
Join Date: Oct 2005
Location: Junortoun Vic
Posts: 8,904
Martin,
I think it comes down to the slope in the quality graph.
The intersection with the required quality would define the settings in each square.
Reply With Quote
  #13  
Old 28-06-2019, 06:50 PM
Startrek (Martin)
Registered User

Startrek is offline
 
Join Date: Dec 2017
Location: Sydney and South Coast NSW
Posts: 6,031
Ken
Appreciate the advice !!
Unfortunately we are all still saying “I think”
A user manual would be an advantage, but I don’t believe one exists ?
It’s extremely good planetary stacking software and very popular around the world with the only grey area in my honest opinion being this quality graph / stack percentage box and it’s use , selection and resulting outcomes
Cheers
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 05:29 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Testar
Advertisement