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

Mosaics - LRGB vs RGB? and binning?

  • Please log in to reply
5 replies to this topic

#1 jacquesdeacon

jacquesdeacon

    Sputnik

  • -----
  • topic starter
  • Posts: 43
  • Joined: 09 Aug 2011
  • Loc: Kathu, NC, South Africa

Posted 25 October 2021 - 08:42 AM

Hi Everybody,

 

I am about to undertake a big mosaic project - The Small Magellanic Cloud (SMC).  This will be my biggest ever mosaic project; I have only done up to 6 panels previously, this one is 16 panels!  Going to be a good challenge!

 

I am pondering how I should approach this projects in terms of sub-exposures.

 

My equipment is a TAK FSQ106 @ F5 with a SBIG8300M, giving my a resolution of 2.1as/px.  My sky conditions is pretty average backyard - bortle 5-6.

 

My usual approach is shooting Lum unbinned 5min sub-exposures, R,G,B binned 2x2 3min sub-exposures. I usually aim to get at least 3hr total integration time, split 1:1 Lum (+- 18 subs - 1.5hrs) vs RGB (+-10subs - 30min each channel).  This is usually a good starting image.

 

With the big mosaic I am wondering if I should change things up a bit...

 

Options I am considering:

  1. Shooting only RGB, all bin1x1.  Shooting 3min subs each, 20 subs per channel - 3hrs total per frame
  2. LRGB with reduced exposure times.  3min Lum subs (unbinned) 30subs - 1.5hr).  90s RGB (binned) subs (20subs - 30min each channel - 1.5hrs total) - also 3hrs total per frame
  3. My 'regular' LRGB.  5min Lum subs (unbinned) 18subs - 1.5hr.  3min RGB (binned) subs (10subs - 30min each channel - 1.5hrs total) - targeting 3hrs

 

What in your experience is more important with regards to mosaics, going 'deeper' with a few longer exposures, or trying to get better SNR with more shorter subs? 

The SMC is also a unique target really, do you think it is better to shoot at the max resolution possible to resolve the star-cloud the best with better controlled (less-saturated) color in the absence of nebulosity? 

 

Looking forward to your ideas on this.

 

Best Regards,

Jacques

Attached Thumbnails

  • smc_16panel_rota_r.png


#2 the Elf

the Elf

    Soyuz

  • *****
  • Posts: 3,945
  • Joined: 06 Sep 2017
  • Loc: Germany

Posted 25 October 2021 - 03:48 PM

You need a certain minimum of RGB for natural colors. Only you know how much that is for your setup. Once you have enough data for an image that shows little color noise adding L is a good thing. Given the size of the project I'd invest more time on one panel that is somewhat representative for the whole project and shoot some more. The try how many subs you really need and continue accordingly for all the other panels.



#3 imtl

imtl

    Skylab

  • *****
  • Moderators
  • Posts: 4,257
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 25 October 2021 - 06:42 PM

If it helps you than you can see here what I've done with this object. 9 panel mosaic. Details are on Abin. You can also PM me.
get.jpg?insecure



#4 jacquesdeacon

jacquesdeacon

    Sputnik

  • -----
  • topic starter
  • Posts: 43
  • Joined: 09 Aug 2011
  • Loc: Kathu, NC, South Africa

Posted 26 October 2021 - 12:55 AM

Thank @the Elf. I want to start sometime this week, and maybe do panel 7 as a test, and also analyse the data a bit further in PixInsight to measure my sky noise for the target area specific to dial in my sub exposure lengths.

 

Thanks for sharing @imtl. Stunning image, and it is clear that the target likes good sampling, (yours just over 1.1as/px). The resolution of the small details (small stars) is amazing. With110hrs experience on the SMC, is there anything you would do different if you had to do it again? smile.gif

 

Regards,

Jacques

 


  • imtl likes this

#5 imtl

imtl

    Skylab

  • *****
  • Moderators
  • Posts: 4,257
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 26 October 2021 - 02:54 AM

Thank @the Elf. I want to start sometime this week, and maybe do panel 7 as a test, and also analyse the data a bit further in PixInsight to measure my sky noise for the target area specific to dial in my sub exposure lengths.

 

Thanks for sharing @imtl. Stunning image, and it is clear that the target likes good sampling, (yours just over 1.1as/px). The resolution of the small details (small stars) is amazing. With110hrs experience on the SMC, is there anything you would do different if you had to do it again? smile.gif

 

Regards,

Jacques

Hi Jacques,

 

I would start by doing quick few minutes L on each panel so you could build a good startup for this. This should probably take a few hours of the first night with platesolving/slewing etc.

 

What I would do differently? Not sure. Add more panels since there are more DSO in the surroundings that I did not capture. Maybe add Sii as well. But I wanted to go as deep as I could with the time frame I had. It did take me 3.5 months to complete the data acquisition. 12 hours per panel. I dedicated mostly to L so my SNR is pretty good with my (then) bortle 4/5 skies.

 

The processing will be challenging since it has a lot of different DSOs in it that require different kind of processing. The DR is also challenging.



#6 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 10,979
  • Joined: 12 Oct 2007
  • Loc: Melbourne, Australia

Posted 26 October 2021 - 07:47 AM

I would do tests with a fainter part of the region and capture 1 hour rgb binned, 1 hour rgb unbinned, and 1 hour L unbinned.  Then see what combination gives the best results with 1 hour total exposure time.  It may end up just being 1 hour rgb unbinned.

 

With a cmos camera I wouldn't even bother capturing binned - but with 8300 it may be worth binning if you have very dark skies.  But if you have reasonable sky background, read noise may not matter and you can capture unbinned - then bin or smooth later when you combine with L.  If you end up wanting to combine with L.

 

If you try different combinations: 1 hour rgb unbinned, 30m rgb unbinned plus 30 min. L unbinned - or 30m rgb binned plus 30 min. L unbinned - etc. - you should be able to find out what works best in terms of how you want it to look and how you process.

 

The main thing is to compare on an equal total time basis, and to realize there are no firm answers here because it depends on what you prefer and how you do the processing.

 

It would certainly be advisable to find out what works best ahead of time with some experimentation - before committing to a procedure that may not be best for you.  Do you care about high res. and do you want people to zoom in to the pixel level?  Or do you just want a wide but detailed view?  These are important considerations.

 

Frank


  • imtl likes 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