"Mixing" Exposures
 
Share:
Notifications
Clear all

15th Feb 2024: Astro Pixel Processor 2.0.0-beta29 released - macOS native File Chooser, macOS CMD-Q fixed, read-only Fits on network fixed and other bug fixes

7th December 2023:  added payment option Alipay to purchase Astro Pixel Processor from China, Hong Kong, Macau, Taiwan, Korea, Japan and other countries where Alipay is used.

 

"Mixing" Exposures

8 Posts
4 Users
1 Likes
1,475 Views
(@oopfan)
Neutron Star
Joined: 5 years ago
Posts: 109
Topic starter  

In the past, I've read that it is ill-advised to mix, for example, 60s luminance with 120s luminance in the same stack. Is that still true? Perhaps I misunderstood the context. What if I assigned 60s luminance to Session 1, and 120s luminance to Session 2. Followed by assigning both stacks to the luminance channel using Color Combine. Is that a recommended solution? I'd hate to have to choose between 60s or 120s. I'd like to use both, if possible.

Thanks!


   
ReplyQuote
(@col)
Neutron Star
Joined: 3 years ago
Posts: 127
 

Mine are OSC images, but I do the same with exposure lengths and with gains, for dynamic range. I used to really, with a 12 bit camera, but stick to single settings for my 16bit one. You should be OK for dynamic range or tonal range integrating as sessions (singly or as separate integrations from one process in the integration tab). So long as you assign your calibration frames (if you use them) to the right sub exposures, it should be OK.


   
ReplyQuote
(@rickwayne)
Neutron Star
Joined: 5 years ago
Posts: 70
 

My most-recent image mixed 15", 30", 200", and 600" exposures. Whether it was inadvisable I couldn't say, but it came out pretty ok. APP integrates the various exposures pretty well; I have had less success with mosaicking images that differ wildly in exposure, but APP will handle that too if you have enough patience to sit through the LNC iterations.


   
ReplyQuote
(@vincent-mod)
Universe Admin
Joined: 7 years ago
Posts: 5707
 

It's fine to mix, but it isn't super beneficial yet. What you're probably after then is a HDR integration and APP doesn't support that yet (it's on the list of things to add though).


   
ReplyQuote
(@oopfan)
Neutron Star
Joined: 5 years ago
Posts: 109
Topic starter  

Thank you, everyone.

I want to take every opportunity to increase the total integration time of a faint object. I've been given several hours of additional luminance frames by a generous person, but the exposure is different and the optics are slightly different as well. I have watched the video tutorials on this topic.

So, I have a good follow-up question: "If adding different luminance frames captured by the same equipment is not very beneficial, then what is the benefit of a multi-person project?" The question is silly, but why is it beneficial in one context but not another? Or am I reading too much into the reply? Clearly, there are various degrees of "beneficial".

Thanks again.


   
ReplyQuote
(@vincent-mod)
Universe Admin
Joined: 7 years ago
Posts: 5707
 

Ok, so to be clear... more data = good. 🙂 The more data you have, the lower the overall noise and the better the signal. Of course you wouldn't want to combine very bad data with good data preferably. But if you combine a lot of nice data, even from different scopes, that will be really nice as an end result for sure. The thing with different exposures is that you usually want to use that to get more detail in over-exposed stars and galaxy centers etc. That specifically is not yet a feature really.


   
ReplyQuote
(@col)
Neutron Star
Joined: 3 years ago
Posts: 127
 

@oopfan

More data is generally better, especially if it is good data. APP doesn't have a dedicated HDR exposure mixing, but its not so important since lower exposures (maybe your case) might have better star color balance (sometimes is the case for OSC cameras). Anyway, it will help SNR assuming there are no major sky background differences and improve rejection algorithms for the background levels.

You could try one curiosity test to see for yourself. Integrated 50 exposures at the longest length, then 50 with the shortest length (or whatever number you have, but ideally more than 20 subs). Then add half and half of each up to the same number of subs and compare the outputs in a similar region at high magnification non the viewer to satisfy your self you are not worsening anything. Then the integration with everything together should in principle be the best scenario for the data assuming no massive deviations in the data sets aside from exposure.


   
Rick Wayne reacted
ReplyQuote
(@oopfan)
Neutron Star
Joined: 5 years ago
Posts: 109
Topic starter  

Thank you, Colm. That is great advice!


   
ReplyQuote
Share: