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

A question of signal?

  • Please log in to reply
5 replies to this topic

#1 pyrasanth

pyrasanth

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2,881
  • Joined: 08 Jan 2016

Posted 30 May 2020 - 07:01 AM

I wish to go deeper with my images but not at the expense of washing out the background. I'm using the ASI6200 camera & I get good results on the C14 at F11 using 60 seconds at 2x2 bin gain 0 offset 50. My Sky is no better than Bortle 7 and perhaps closer to 8.

 

The question is- do more subs at the same exposure go deeper or does the faint fuzzies captured remain the same but the noise around them decreases?- its hard to tell if more of the same is better than less at a greater exposure.

 

Please let me know what your experience has been- more subs the same or less subs but greater exposure length with the objective of getting deeper data.


Edited by pyrasanth, 30 May 2020 - 07:02 AM.


#2 kathyastro

kathyastro

    Surveyor 1

  • *****
  • Posts: 1,874
  • Joined: 23 Dec 2016
  • Loc: Nova Scotia

Posted 30 May 2020 - 07:09 AM

Basically, more total integration time gives a better signal-to-noise ratio.  Whether you get that integration time with fewer longer exposures or with more shorter exposures doesn't make a whole lot of difference.  If light pollution limits your exposure length, then you should take more short exposures.



#3 ceteris_paribus

ceteris_paribus

    Vostok 1

  • *****
  • Posts: 105
  • Joined: 21 Sep 2014
  • Loc: California

Posted 30 May 2020 - 08:06 AM

 

The question is- do more subs at the same exposure go deeper or does the faint fuzzies captured remain the same but the noise around them decreases?- its hard to tell if more of the same is better than less at a greater exposure.

As I understand it, these two things are pretty much the same thing.  As the noise around them decreases, it allows you more freedom in processing to go deeper and pull out the detail that has always been there, but was just hidden by the overwhelming amount of noise.  So, the more images you have in a stack, the better the ratio of signal to noise will be.  These days, with cameras as good as they are, there is nearly no difference between whether you take longer exposures or lots of shorter ones.  So, if your computer can handle the huge amounts of data, then go with lots and lots of short exposures, and the more you get, the better the images will be.

 

-John


  • elmiko likes this

#4 Jared

Jared

    Fly Me to the Moon

  • *****
  • Posts: 6,354
  • Joined: 11 Oct 2005
  • Loc: Piedmont, California, U.S.

Posted 30 May 2020 - 09:03 AM

I wish to go deeper with my images but not at the expense of washing out the background. I'm using the ASI6200 camera & I get good results on the C14 at F11 using 60 seconds at 2x2 bin gain 0 offset 50. My Sky is no better than Bortle 7 and perhaps closer to 8.

The question is- do more subs at the same exposure go deeper or does the faint fuzzies captured remain the same but the noise around them decreases?- its hard to tell if more of the same is better than less at a greater exposure.

Please let me know what your experience has been- more subs the same or less subs but greater exposure length with the objective of getting deeper data.


First, better SNR and a “deeper” image are generally synonymous since better signal to noise let’s you stretch more aggressively before the image starts to fall apart.

As to optimal sub exposure duration... I can only give general advice since there are too many variables, but general advice may be enough with this camera. You want your exposure length to be high enough that shot noise swamps read noise. With CMOS vs CCD that usually means a shorter sub since CMOS cameras generally have lower read noise. Under Bortle 7 skies it doesn’t take too much before your histogram starts to move to the right, so you may be fine at one minute. Your camera has a large enough full well capacity that you could probably go quite a bit longer at f/11 before saturating much, so you might do a bit better with longer subs, but if read noise is already much lower than shot noise at 1 min it’s not going to make much of a difference either way. The main advantage you’ll get from longer subs will be less storage and processing. Biggest disadvantage would probably be not as good pixel rejection from fewer subs. Try both ways one night and see. Or use something like PixInsight’s sub calculator to check to see if your shot noise is dominating.

One other thing I’ll mention is that, at f/11, you are probably still oversampled quite a bit at 2x2 binning. That’s about 0.4” per pixel. Unless you have truly exceptional skies or it’s an especially good night with an object high in the sky you might do better at 3x3. I know it feels like throwing away a lot of pixels, but right now they are probably capturing “empty” resolution and just increasing noise. Again, try for yourself. Take two subs, process to taste, and at the end up-sample the 3x3 to the same pixel size as the 2x2 and see if you lost any actual resolution. I suspect not, and Th e SNR will be something like 1.5x better—roughly the same as doubling your integration time.

#5 WadeH237

WadeH237

    Fly Me to the Moon

  • *****
  • Posts: 5,784
  • Joined: 24 Feb 2007
  • Loc: Snohomish, WA

Posted 30 May 2020 - 08:53 PM

More total integration time is important in getting the best out of light polluted skies.

 

Regarding the number and length of exposures, there are just a couple of rules of thumb.  First, you want to make sure that the dominant noise term is shot noise and not read noise.  Exposures that are too short will be more affected by read noise - but if you have lots of light pollution, you likely won't need excessively long exposure times.  Under dark skies (or with narrow band filters), you will need to expose longer to bring the shot noise up to a level that is sufficient to dominate the read noise.

 

Once you have the read noise beat, further increases to the exposure time will potentially reduce your dynamic range.  What you want to do is to strike a balance, so that your background is sufficient to beat read noise, but not so long that you start saturating too many pixels (note that bright star cores will saturate really quickly, and that's generally ok).

 

Once you have your exposure times figured out, just keep adding more exposures until you meet your goals.



#6 nimitz69

nimitz69

    Vanguard

  • *****
  • Posts: 2,287
  • Joined: 21 Apr 2017
  • Loc: A barrier island 18 miles south of Cocoa Beach

Posted 31 May 2020 - 11:20 AM

Once you are swamping read noise for your Combination of camera, scope, sky conditions and LP increasing sub length buys you nothing. Increasing total integration time trumps everything, all else being equal.

My skies are 19.5 at my observatory and I collect 8-10 hr/target at the minimum before doing any serious PP. currently working on an HDR M13 image with 12 hrs of data and I’m still not done collecting and ready fo PP

Edited by nimitz69, 31 May 2020 - 11:21 AM.



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