Go Back   IceInSpace > Beginners Start Here > Beginners Astrophotography

Reply
 
Thread Tools Rate Thread
  #1  
Old 15-10-2019, 09:56 PM
HeavyT
Registered User

HeavyT is offline
 
Join Date: May 2019
Location: South Australia
Posts: 48
Does stacking different shutter speeds impair noise reduction?

Hi folks

I have been working on a nebula image. Having stacked about 5hrs worth of 150 and 300 second subs (ISO 640 and 320 respectively) with my unmodded dslr, I thought I'd achieved a masterstroke last night by acquiring a further 2hrs of 600 second subs at ISO 160. Certainly more faint nebulosity looked to be poking through the gradient in the subs, I thought it was going to get close to finishing my image.

Unfortunately after the initial process of the stack (previous stack Plus twelve at 600s), there's so much noise it might has well have been a stack of 30 minutes!

Does Deep Sky Stacker need me to add say 40 more at 600s in order to remove the noise?

Only other thing I can think I did wrong is setting one of the 600s frames as the reference frame in DSS... restocking at the moment using one of the tried and true 150s subs as a reference.

Apologies if this is a daft question, it's just that I've stacked varying exposure times before without this happening. Maybe 10 minutes just heats my sensor up too much, even though it was Low ISO and a cool night?

Any advice will be well received.

Thanks and clear skies all.
Reply With Quote
  #2  
Old 15-10-2019, 11:34 PM
raymo
Registered User

raymo is offline
 
Join Date: Sep 2011
Location: margaret river, western australia
Posts: 6,070
Not really enough info Todd. I assume that you stacked darks and maybe flats
with each different group, and enough of them. DSLRs have a sweet spot for
astrophotography, which is the point on an ISO graph where the two different types of noise intersect. With popular Canon DSLRs it is usually 800, 1000,
or 1600. For example the EOS 1100D is 1600 and the EOS 600D is 800.
The further you deviate in either direction from the sweet spot, the more noise you will get, so your choices will not get a really good result, especially
160.
If you are using in camera noise reduction, and hence no darks, it is fine to stack different sub lengths and ISOs simultaneously.
If using darks, flats, etc, you need to stack them separately.
raymo
Reply With Quote
  #3  
Old 16-10-2019, 06:49 AM
HeavyT
Registered User

HeavyT is offline
 
Join Date: May 2019
Location: South Australia
Posts: 48
Thanks Raymo

Quote:
Originally Posted by raymo View Post
Not really enough info Todd. I assume that you stacked darks and maybe flats
with each different group, and enough of them. DSLRs have a sweet spot for
astrophotography, which is the point on an ISO graph where the two different types of noise intersect. With popular Canon DSLRs it is usually 800, 1000,
or 1600. For example the EOS 1100D is 1600 and the EOS 600D is 800.
The further you deviate in either direction from the sweet spot, the more noise you will get, so your choices will not get a really good result, especially
160.
If you are using in camera noise reduction, and hence no darks, it is fine to stack different sub lengths and ISOs simultaneously.
If using darks, flats, etc, you need to stack them separately.
raymo
Thanks Raymo

I will go hunting for an ISO graph for my Nikon to see where that wmsweet spot lies. Hopefully with shorter exposures at higher ISO I can build up the same signal strength I was seeing at 600s, because it looks phenomenal.

The second stack seemed a little better but still too noisy for 8hrs. I do need to catch up on dark and flats as I didn't shoot any of these for 600sec. Also will try to stack in groups - have never used that function before, always got away with stacking everything in the one list, and have rarely used any darks, flats etc. But I've never gone to 600 seconds before.

Clear skies mate
Reply With Quote
  #4  
Old 16-10-2019, 07:01 AM
RyanJones
Registered User

RyanJones is offline
 
Join Date: Jun 2018
Location: Melbourne,Australia
Posts: 1,439
Hi Todd,

I have experienced the same issue my self. I’m going to assume you have separate darks for each different length of sub ? I found on my Canon that anything more than 5min subs introduces issues no matter what the outside temperature is. I also found that I lose dynamic range. I’ve tried so many different combinations of different ISO and sub lengths with calibration frames for each and no matter what I do, I can never seem to get as good an image as i can with just one length and one ISO and just lots more of them. As Raymo suggested and lots of other people do, there is a “ sweet spot “ but it’s not just about thermal vs read noise. It’s also about dynamic range. My canon’s “ sweet spot “ for noise ISO 800 but I lose a lot of colour at that so I’ve settled on ISO 400 and 3min subs. I get very few noise, colour and other issues at this setting but it does mean 10hrs of subs as a good amount rather than 4 or 5. My advice for what it’s worth is to take time to find what works for your setup at a single length , single ISO and quantity. Only use the much talked about sweet spots as a guide of where to start from. Good luck and I look forward to seeing your images

Cheers
Reply With Quote
  #5  
Old 17-10-2019, 07:16 AM
Startrek (Martin)
Registered User

Startrek is offline
 
Join Date: Dec 2017
Location: Sydney and South Coast NSW
Posts: 5,990
I’ve also experimented with stacking different sub lengths and ISO settings on my Canon 600D and found that it just doesn’t get a satisfactory result
My sweet spot is ISO 800 and my sub exposure times vary with each object I image at my 2 locations , Sydney Bortle 8 and Narrawallee south coast NSW Bortle 3
As Ryan said take plenty of subs, take plenty of darks ( ratio of at least 3:1 ) at same settings preferably straight after your subs and make sure your histogram curves are in the 1/3 zone towards the right side where possible ( obviously City and Suburban sky glow blows your histogram way out )
For exposures beyond 1 minute or so using a DSLR , dithering is a must do for fixed pattern noise reduction and dark current noise reduction
All the best
Good Luck !
Reply With Quote
  #6  
Old 17-10-2019, 07:35 AM
ChrisV's Avatar
ChrisV (Chris)
Registered User

ChrisV is offline
 
Join Date: Aug 2015
Location: Sydney
Posts: 1,737
Unless you are going for an object with high dynamic range I don't see the point of different sub lengths. I know it's a long time since I studied maths but I can't imagine how it will improve SNR.

Just pick your camera's sweet spot and go for it. Some combination of exposures length and iso that will swamp read noise and not saturate the bigger stars too much. As Ryan said.

Maybe try dithering. That might help with fixed pattern noise etc?
Reply With Quote
  #7  
Old 17-10-2019, 07:51 AM
sil's Avatar
sil (Steve)
Not even a speck of dust

sil is offline
 
Join Date: Jun 2012
Location: Canberra
Posts: 1,474
i suggest giving Astropixelprocessor a try (free 30day trial), far superior to dss and works well with whatever subs you have to throw at it. doubling the time is introducing noise, halving iso as well doesnt mean zero noise. there are different type of noise at play which are not noticable for regular photography (since you're going off the exposure triangle) so the more subs you throw at the stackthe more Signal is gathered, true, but also the more Noise is gathered too. So if you decreasing SNR in the stack the less you can stretch the data. Make sure to take darks, flats and bias shots too with each set of lights so each set of lights can be preprocessed best to increase SNR in that set and therefore in the overall stack. If you didn't you can try doing those tonight under similar conditions and same camera settings. its not ideal but it should help. I dont think you've wasted time, maybe just your preprocessing needs improvement. if you give APP a try I find its default settings produce a stretched result not far from the best i can achieve in Pixinsight, so if APP gives you pretty much identical to your dss result then youve probably done the best possible with those subs and maybe look to your subs to solve the problem. All subs.
Reply With Quote
  #8  
Old 20-10-2019, 12:53 PM
HeavyT
Registered User

HeavyT is offline
 
Join Date: May 2019
Location: South Australia
Posts: 48
Too low ISO

Thanks all this is good advice.

I think I've figured out my acquisition problem. I'm imaging the Tarantula and went for longer subs to get more of the outer faint nebulosity. I kept the ISO low to try not to blow out the stars and more importantly the core. I can stretch the core without noise and it looks good. When I bring up the mid tones they come out and so does the noise.

So I guess I need to do some exposure bracketing - take a new stack at higher ISO to get the faint stuff up out of the noise, then mask in the core over the top?

I was worried about the core, and forgot that noise lives in the shadows. Until recently I had only been apply linear exposure stretches then raising the black point until the light pollution and noise went away. This works ok for very bright nebulae like M16, M17 etc and I've been able to reprocess those and extract more mids without noise, but the fainter objects like Helix and Tarantula are very noisy, so I conclude I've not managed my histogram well.

I'm very amateur and until doing astro never really post processed any photos, except in Instagram haha!

Thanks again for your assistance, you've helped me work out where I've gone wrong 🙃 argh learning curves!
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 01:24 AM.

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