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

Determining best ISO settings

  • Please log in to reply
13 replies to this topic

#1 8472

8472

    Sputnik

  • -----
  • topic starter
  • Posts: 29
  • Joined: 20 Jan 2018

Posted 21 March 2019 - 08:31 PM

Hi,

I regularly set my m43 camera (E-PL5) to ISO 800 for my DSO imaging, but was wandering if this is the best ISO to use?

I've scientifically calculated the optimal sub length at various ISOs in the range of 200 to 3200 for my skies (read noise squared divided by LP (e-) x 10).

Now am I missing something, but when I look at the Full Well Capacity at ISO 200 (24777 e-), and divide by the Read Noise at the same ISO (6.7 e-), I am left with a larger number than if I image at ISO 800, or any other ISO over 200.

As i have no guiding issues with the extended sub length which goes with ISO 200 imaging, would I be better off using this ISO? Or are there any other factors I've overlooked that might come into play?

Thanks,

8472

#2 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 15185
  • Joined: 27 Oct 2014

Posted 21 March 2019 - 09:53 PM

That's a very decent method.  Another is to look for a leveling off of the read noise versus ISO graph.  Look at both these pages.

 

http://dslr-astropho...trophotography/

 

http://dslr-astropho...-nikon-cameras/

 

If both methods give you the same result, great.  If not it's a judgment call.



#3 georgian82

georgian82

    Mariner 2

  • *****
  • Posts: 253
  • Joined: 16 Jan 2016

Posted 21 March 2019 - 10:31 PM

Hi there,

I have an Olympus EPM2 (same sensor as EPL5) which I use for astrophotography and I determined that ISO 800 is the best for this camera based on data from Sensorgen.de (no longer available).

Cheers

Edited by georgian82, 21 March 2019 - 10:31 PM.


#4 whwang

whwang

    Mercury-Atlas

  • *****
  • Posts: 2564
  • Joined: 20 Mar 2013

Posted 21 March 2019 - 10:47 PM

Hi,

I regularly set my m43 camera (E-PL5) to ISO 800 for my DSO imaging, but was wandering if this is the best ISO to use?

I've scientifically calculated the optimal sub length at various ISOs in the range of 200 to 3200 for my skies (read noise squared divided by LP (e-) x 10).

Now am I missing something, but when I look at the Full Well Capacity at ISO 200 (24777 e-), and divide by the Read Noise at the same ISO (6.7 e-), I am left with a larger number than if I image at ISO 800, or any other ISO over 200.

As i have no guiding issues with the extended sub length which goes with ISO 200 imaging, would I be better off using this ISO? Or are there any other factors I've overlooked that might come into play?

 

Full well changes with ISO, of course.  And full well divided by read noise is called dynamical range.  So it should not be surprising at all that you find higher dynamical range at lower ISO.

 

It's good that your have stable guiding so you can afford the very long subs to overcome read noise at low ISO.  However, guiding is only one of the factors.  Another factor is the number of subs.  Suppose you need a total of one hour of integration, you can either do two 30-minute subs at ISO 200, or do 10 6-minute subs at ISO 1600.  Which one to choose?  I will probably go with the ISO 1600 combination and dither the 10 exposures.  This will give much better hot/cold pixel rejection during stacking and will give you cleaner images.

 

The importance of dynamical range is over-emphasized in astrophotography.  The advantage of dynamical range at low ISO usually is too small to justify the sacrifice in number of exposures and/or the effort to maintain very good guiding.  If you do encounter saturation problem when you image objects with very bright cores (M31, M42, M8, etc), the easiest way is just to take additional sets of very short exposures, rather than going to low ISO.  In many cases, even if you go to a lower ISO, you will still get a saturated core after you increase the length of the subs to overcome read noise.  You will need those additional short exposures any way, so why bother with low ISO?

 

With all the above, my recommendation is to go to higher ISO for lower read noise (and a larger number of shorter subs).  Most of those who have troubles at high ISO are those who adopt subs that are too short.  However, you seem to have the capability of calculating the right lengths of your subs, so this should not be a problem for you.

 

For exactly what ISO to use, one can go to http://www.photonsto...Charts/RN_e.htm to check how the read noise changes with ISO, and find the lowest ISO where the read noise is low enough.  This is typically between ISO 800 and 3200.  Unfortunately I cannot find your camera model in that page, so I can only give you a rough range.

 

Cheers,

Wei-Hao


  • mxpwr likes this

#5 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 15185
  • Joined: 27 Oct 2014

Posted 22 March 2019 - 10:23 AM


 

For exactly what ISO to use, one can go to http://www.photonsto...Charts/RN_e.htm to check how the read noise changes with ISO, and find the lowest ISO where the read noise is low enough.  This is typically between ISO 800 and 3200.  Unfortunately I cannot find your camera model in that page, so I can only give you a rough range.

 

Note that, for modern "ISOless" cameras it's typically a lot lower.  With my D5500, on the good mount I use 200, on the tracker 400.


  • 8472 likes this

#6 8472

8472

    Sputnik

  • -----
  • topic starter
  • Posts: 29
  • Joined: 20 Jan 2018

Posted 22 March 2019 - 01:39 PM

That's a very decent method.  Another is to look for a leveling off of the read noise versus ISO graph.  Look at both these pages.

 

http://dslr-astropho...trophotography/

 

http://dslr-astropho...-nikon-cameras/

 

If both methods give you the same result, great.  If not it's a judgment call.

Hi,

 

I've read that article a good while back to find the ideal ISO for my 1100D. Very useful, but no numbers for my Oly.



#7 8472

8472

    Sputnik

  • -----
  • topic starter
  • Posts: 29
  • Joined: 20 Jan 2018

Posted 22 March 2019 - 01:42 PM

Hi there,

I have an Olympus EPM2 (same sensor as EPL5) which I use for astrophotography and I determined that ISO 800 is the best for this camera based on data from Sensorgen.de (no longer available).

Cheers

Hi,

 

I also got my noise, DR and FWC info from Sensorgen, but didn't find anything mentioning ideal ISO?



#8 8472

8472

    Sputnik

  • -----
  • topic starter
  • Posts: 29
  • Joined: 20 Jan 2018

Posted 22 March 2019 - 02:17 PM

Full well changes with ISO, of course.  And full well divided by read noise is called dynamical range.  So it should not be surprising at all that you find higher dynamical range at lower ISO.

 

It's good that your have stable guiding so you can afford the very long subs to overcome read noise at low ISO.  However, guiding is only one of the factors.  Another factor is the number of subs.  Suppose you need a total of one hour of integration, you can either do two 30-minute subs at ISO 200, or do 10 6-minute subs at ISO 1600.  Which one to choose?  I will probably go with the ISO 1600 combination and dither the 10 exposures.  This will give much better hot/cold pixel rejection during stacking and will give you cleaner images.

 

The importance of dynamical range is over-emphasized in astrophotography.  The advantage of dynamical range at low ISO usually is too small to justify the sacrifice in number of exposures and/or the effort to maintain very good guiding.  If you do encounter saturation problem when you image objects with very bright cores (M31, M42, M8, etc), the easiest way is just to take additional sets of very short exposures, rather than going to low ISO.  In many cases, even if you go to a lower ISO, you will still get a saturated core after you increase the length of the subs to overcome read noise.  You will need those additional short exposures any way, so why bother with low ISO?

 

With all the above, my recommendation is to go to higher ISO for lower read noise (and a larger number of shorter subs).  Most of those who have troubles at high ISO are those who adopt subs that are too short.  However, you seem to have the capability of calculating the right lengths of your subs, so this should not be a problem for you.

 

For exactly what ISO to use, one can go to http://www.photonsto...Charts/RN_e.htm to check how the read noise changes with ISO, and find the lowest ISO where the read noise is low enough.  This is typically between ISO 800 and 3200.  Unfortunately I cannot find your camera model in that page, so I can only give you a rough range.

 

Cheers,

Wei-Hao

 

Some interesting points there.

 

What I question is, I understand the concept/importance of total imaging time, but in reality, we are not catching photons between our multiple short subs while we are dithering and or allowing the sensor to cool or guiding to settle.

 

The shorter the subs, the more this is the case. In your 1 hour example, if we say we (albeit generously) allot 1 minute dither/settle duration between subs, the long exposures will lose 1 minute of imaging time, but the shorter subs lose 9 minutes. Will dithering offset this deficit enough to negate the lost imaging time?



#9 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 15185
  • Joined: 27 Oct 2014

Posted 22 March 2019 - 05:45 PM

Hi,

 

I also got my noise, DR and FWC info from Sensorgen, but didn't find anything mentioning ideal ISO?

You plot the RN versus ISO, look for where things level off.



#10 whwang

whwang

    Mercury-Atlas

  • *****
  • Posts: 2564
  • Joined: 20 Mar 2013

Posted 22 March 2019 - 09:22 PM

Hi,

My dither overhead is typically 25 seconds, and this is already very long comparing to what I heard here. If your mount has good response to guide commands, it should be very easy to make the overhead below 15 seconds or even shorter. For subs of 6 minutes, this is just a 4.2% loss of time, which is nothing comparing to other overheads like pointing, focusing, and loss to bad guiding/wind, etc. Psesonally I am very willing to pay up to 10% of overhead to dithering, given the enormous advantages it brings.

Cheers,
Wei-Hao

#11 georgian82

georgian82

    Mariner 2

  • *****
  • Posts: 253
  • Joined: 16 Jan 2016

Posted 23 March 2019 - 05:33 PM

Hi,

I also got my noise, DR and FWC info from Sensorgen, but didn't find anything mentioning ideal ISO?


Correct, but you can figure out the best ISO by looking at the graph of ISO vs read noise. At ISO 800 is where the read noise flattens...

#12 8472

8472

    Sputnik

  • -----
  • topic starter
  • Posts: 29
  • Joined: 20 Jan 2018

Posted 23 March 2019 - 06:20 PM

Correct, but you can figure out the best ISO by looking at the graph of ISO vs read noise. At ISO 800 is where the read noise flattens...

Thanks - that is the ISO i have been using, so thanks for clearing that up.

 

I've seen some folks using crazy high ISO numbers, but just wanted to be sure.



#13 georgian82

georgian82

    Mariner 2

  • *****
  • Posts: 253
  • Joined: 16 Jan 2016

Posted 23 March 2019 - 06:24 PM

Thanks - that is the ISO i have been using, so thanks for clearing that up.

I've seen some folks using crazy high ISO numbers, but just wanted to be sure.


No problem. Read this article when you get a chance which explains it all really well:

https://dslr-astroph...trophotography/

Cheers

#14 8472

8472

    Sputnik

  • -----
  • topic starter
  • Posts: 29
  • Joined: 20 Jan 2018

Posted 06 April 2019 - 11:32 AM

Just a thought.

In the past, i took to using a CLS clip filter, more often than not.

Is there a reliable exposure multiplication factor I can apply for using such a filter in my imaging train?

I've heard 2x to 3x bandied about, but is this accurate? I know referring to the histogram on the back of the camera isn't the most scientific method.

Thanks

Edited by 8472, 06 April 2019 - 11:33 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