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

Best ISO short exposures DSLR

astrophotography beginner
  • Please log in to reply
12 replies to this topic

#1 SiriusZ

SiriusZ

    Lift Off

  • -----
  • topic starter
  • Posts: 12
  • Joined: 18 Jan 2020

Posted 19 January 2020 - 08:59 AM

Hi guys, I´m starting astrophotography, I have been reading about ISO issues, but can´t find a solid info, sometimes I found contradictory answers related to this matter in different forums.

 

As far as I know, higher ISO doesn´t mean more details, and each camera has it´s own proper ISO in astrophotography (noise-signal), so, questions: 

-why then shooting higher ISO than 100/200 if it´s doesn´t add details and apparently it´s better for noise except for certain older cameras?

-The "optimal" ISO for each camera is applied in the short time exposure like 1 min or only in long exposures? that´s my case, 8 bortle sky and no guiding so I´m stuck in 30/60 sec maximum.

 

 

I have a Nikon D3200, 200 ISO seems to be best for my camera according to http://dslr-astropho...nikon-cameras/ 

 

So guys, Do I have to shoot on ISO 200 even with 1 min max exposure and light pollution?  what would you suggest?

 

- Skywatcher Telescope N 150/750 Explorer BD EQ3-2 GOTO.

 

Thanks and sorry for my english! 



#2 kathyastro

kathyastro

    Vanguard

  • *****
  • Posts: 2,177
  • Joined: 23 Dec 2016
  • Loc: Nova Scotia

Posted 19 January 2020 - 09:16 AM

Unlike in the days of film, where a higher ISO (they called it ASA back then) meant greater sensitivity of the film, the ISO control on a DSLR does not alter the sensor in any way.  Increasing the ISO does not make the sensor more sensitive, so it can't increase details.  All that the ISO control does is apply a mathematical gain control to the sensed data.

 

Feel free to use whatever ISO setting gives you good results.  The most common recommendation is to use the setting closest to your camera's "unity gain" value.  Unity gain is where the gain control is neither increasing nor decreasing the data values from the sensor.  Most camera sensors have their unity gain at about ISO 800-1600.

 

If your sky conditions and tracking dictate 1-minute exposures, then use that setting and experiment with ISO settings.  You don't "have to" use any particular setting.  Try unity gain.  Try higher and lower.  See what gives you the best results.


  • Knucklehead, Michael Covington, Simcal and 1 other like this

#3 klaussius

klaussius

    Viking 1

  • *****
  • Posts: 709
  • Joined: 02 Apr 2019
  • Loc: Buenos Aires

Posted 19 January 2020 - 09:21 AM

ISO settings in digital cameras control signal amplification at the sensor. Read noise is somewhat constant regardless of amplification, which means higher ISO settings get you more signal with less noise.

 

Photon noise will be amplified as well, so at some point you don't get any improvement. The optimal ISO for DSLRs usually sits between 800 and 1600 ISO. Sometimes 400. Each camera is different.

 

There's an ISO where you get 1 ADU per electron (sensed photon), and that's "unity gain". This is the best ISO for low read-noise cameras. DSLRs tend to have high read noise, though, so you probably want to go a bit higher for DSLRs. ISOs lower than unity gain do waste some signal, so it's usually not worth going lower than that.

 

On the other side, higher ISOs will cause a brighter image, and with light pollution in the mix, that will play against you, brightening the background and decreasing dynamic range. If you lose too much dynamic range, you'll start to lose the color of bright stars, so it's a tradeoff.

 

Longer exposures will decrease photon noise, but you'll have to lower the ISO to retain dynamic range, increasing the relative effect of read noise. Again, it's all a tradeoff - more read noise, less photon noise.

 

With nikons, I've seen good results with ISO 400 and 800, so you probably should try those. Then play a bit with the settings and pick the one that works best for you, the differences aren't huge anyway, the most important factor is, as always, integration time.


  • Gucky and Simcal like this

#4 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 21,150
  • Joined: 27 Oct 2014

Posted 19 January 2020 - 11:12 AM

Hi guys, I´m starting astrophotography, I have been reading about ISO issues, but can´t find a solid info, sometimes I found contradictory answers related to this matter in different forums.

 

As far as I know, higher ISO doesn´t mean more details, and each camera has it´s own proper ISO in astrophotography (noise-signal), so, questions: 

-why then shooting higher ISO than 100/200 if it´s doesn´t add details and apparently it´s better for noise except for certain older cameras?

-The "optimal" ISO for each camera is applied in the short time exposure like 1 min or only in long exposures? that´s my case, 8 bortle sky and no guiding so I´m stuck in 30/60 sec maximum.

 

 

I have a Nikon D3200, 200 ISO seems to be best for my camera according to http://dslr-astropho...nikon-cameras/ 

 

So guys, Do I have to shoot on ISO 200 even with 1 min max exposure and light pollution?  what would you suggest?

 

- Skywatcher Telescope N 150/750 Explorer BD EQ3-2 GOTO.

 

Thanks and sorry for my english! 

This is pretty simple, though often misunderstood.  High ISOs amplify the signal some, but it's almost always unimportant.  In astrophotography you accumulate signal with long total imaging time (not long subexposures, a different topic).  And stretch the data in processing.  So high ISO does not help much, if at all.  What it does is reduce dynamic range, which is important for good images.

 

In general, you use the lowest ISO that doesn't hurt in other ways.  Early Canons have a lot of read noise at low ISO, so you use 800 or 1600.  Late model Nikons are "ISOless', don't have the problem.  But they do funny things at ISO 100, you use 200 or 400.

 

That's "good enough" for most situations.  More information and detail at the website you cited.

 

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

 

That doesn't change with light pollution and subexposure time.  I use 200 with my Nikon D5500 in Bortle 7 with my good mount which tracks well, 400 with my camera tracker to shorten subexposure time.


Edited by bobzeq25, 19 January 2020 - 11:15 AM.

  • Michael Covington likes this

#5 TelescopeGreg

TelescopeGreg

    Vanguard

  • -----
  • Posts: 2,096
  • Joined: 16 Jul 2018
  • Loc: Auburn, California, USA

Posted 20 January 2020 - 01:18 AM

All the comments, above, are good, but there is one issue not reflected.  You said you are limited to 30-ish seconds exposure. 

 

The "rule of thumb" is to expose your images such that the peak of the histogram, as viewed in the back of the camera, is about 1/4 or so of the way off the left-hand edge.  You want enough light to get the bulk of the image just enough off the edge that you aren't losing stuff in the noise, but not so much that you saturate the brighter stuff (stars).  Invariably you will do a little of each; the trick is to find a good balance between them, for the particular target you are imaging.

 

So here's the problem.  I have the same camera, a similar telescope (203mm f/5.6), and somewhat the same limitation on maximum exposure length (though for different reasons).  With 30 seconds and an 150mm f/5 telescope, you're not going to move that histogram hump off the edge at ISO 200.  Or even 800.  I have to bump the ISO up to maximum (or nearly so) in order to do that. 

 

That will absolutely increase your noise, and decrease your dynamic range; neither is good.  But I find that unless I increase the ISO, I cannot pull the details out of the black depths in the image.  Your situation might be different, as it also depends a bit on the types of light pollution you have vs the sorts of targets you are imaging.  I'd recommend experimenting a bit to see what works best for you.



#6 HowardSD

HowardSD

    Vostok 1

  • *****
  • Posts: 160
  • Joined: 25 Nov 2018
  • Loc: Palm Springs, CA

Posted 20 January 2020 - 05:41 AM

I'm in the same boat atm, i have a Nikon D5500 with an Eon 72ED F/6 with 0.8 focal reducer/ FF and struggling to find the best combination in a bortle 6 zone.

 

I struggle to get the histogram to that optimum area, last night i shot an hours worth of 2 min subs at ISO 400 but my histogram was only halfway, probably should have dropped it to ISO 200 but on my test shots i didn't really see a noticeable difference on the histogram. I was shooting at ISO 800 and 20-30 sec subs prior to using guiding and the histogram was further to the left. So if i getting this right from these comments ISO 200 should be my starting point for 90 sec + subs then either 100 or 400 depending on where i am on the histogram? Just shooting at 200 makes me think i'm not going to capture all i could but guess i'm wrong in that thought?



#7 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 21,150
  • Joined: 27 Oct 2014

Posted 20 January 2020 - 11:39 AM

I'm in the same boat atm, i have a Nikon D5500 with an Eon 72ED F/6 with 0.8 focal reducer/ FF and struggling to find the best combination in a bortle 6 zone.

 

I struggle to get the histogram to that optimum area, last night i shot an hours worth of 2 min subs at ISO 400 but my histogram was only halfway, probably should have dropped it to ISO 200 but on my test shots i didn't really see a noticeable difference on the histogram. I was shooting at ISO 800 and 20-30 sec subs prior to using guiding and the histogram was further to the left. So if i getting this right from these comments ISO 200 should be my starting point for 90 sec + subs then either 100 or 400 depending on where i am on the histogram? Just shooting at 200 makes me think i'm not going to capture all i could but guess i'm wrong in that thought?

Histogram "halfway"?  Halfway to 1/3, or halfway, period?  Halfway to 1/3 is actually not that bad.  Halfway, period, is overexposed.

 

You are wrong in that thought.  <smile>  The important thing is total imaging time.  _That's_ where the dim stuff comes from, not more subexposure.  Shoot more subs.  My rule of thumb for total imaging time.  One hour is a bare minimum, OK for bright targets like globular clusters.  Two is better, four good.

 

Don't use 100, Nikon does something strange there.  With my good mount, I use 200.  D5500.

 

Minor point, but eventually you'll get there.  The reason the histogram doesn't seem to move much is that it's "stretched" or "non-linear".  The linear histogram moves proportionally.  The non-linear version is fine for now, eventually you'll use the linear one, and digital measurements.

 

But for now, shoot more subs.


Edited by bobzeq25, 20 January 2020 - 11:45 AM.

  • HowardSD likes this

#8 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 6,572
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 20 January 2020 - 12:17 PM

The histogram only needs to be definitely off the left edge.  With newer sensors, it doesn't need to be as far from the left edge as with older sensors.


  • Kendahl likes this

#9 HowardSD

HowardSD

    Vostok 1

  • *****
  • Posts: 160
  • Joined: 25 Nov 2018
  • Loc: Palm Springs, CA

Posted 20 January 2020 - 05:13 PM

Histogram "halfway"?  Halfway to 1/3, or halfway, period?  Halfway to 1/3 is actually not that bad.  Halfway, period, is overexposed.

 

You are wrong in that thought.  <smile>  The important thing is total imaging time.  _That's_ where the dim stuff comes from, not more subexposure.  Shoot more subs.  My rule of thumb for total imaging time.  One hour is a bare minimum, OK for bright targets like globular clusters.  Two is better, four good.

 

Don't use 100, Nikon does something strange there.  With my good mount, I use 200.  D5500.

 

Minor point, but eventually you'll get there.  The reason the histogram doesn't seem to move much is that it's "stretched" or "non-linear".  The linear histogram moves proportionally.  The non-linear version is fine for now, eventually you'll use the linear one, and digital measurements.

 

But for now, shoot more subs.

To clear a point the Histogram was halfway across the little graph instead of being more over towards the left hand side.

 

Thank you for your input, especially seeing as you the same camera! Next time out when the clouds dissipate (unusual for here... Palm Springs) i'll go back to my attempt at capturing Andromeda and try for 1-2 hours of subs (2mins @ISO 200). I've just started using a guide scope with my AVX and PHD 2 and working out the kinks, feel comfortable with 2 mins exposure wish i could do more but i get some bad spikes every now & then on my guide graph. I am however getting better and seems my mount is also (not sure if it's me doing something or my mount learning to correct better).

 

I just upgraded my scope to an AT72EDII, no chance to use yet (guess that's why we have clouds atm... even in the desert it happens, lol), so looking forward to my next session. At least i have the hour of shots i took the other night of M81 & M82 to process, yet another challenge, for some reason i have some bad vignetting to deal with which is plaguing me on most my shots, another thread probably coming in the near future! LOL



#10 SiriusZ

SiriusZ

    Lift Off

  • -----
  • topic starter
  • Posts: 12
  • Joined: 18 Jan 2020

Posted 01 March 2020 - 04:56 PM

After reading all the answers, still confused.

 

kathyastro: "Increasing the ISO does not make the sensor more sensitive, so it can't increase details."

 

klaussius: "which means higher ISO settings get you more signal with less noise."

 

 

 

 

Am I loosing something or those two sentences are contradictory?  More signal doesn´t imply more details?

 

So after all, It´s seems to me that the configuration for gettting "the rule of thumb" is the best tip for getting better results under my light polluted sky, am I right?

 

Thank´s for your answers guys.



#11 Kendahl

Kendahl

    Surveyor 1

  • *****
  • Posts: 1,611
  • Joined: 02 Feb 2013
  • Loc: Pinedale, Arizona

Posted 01 March 2020 - 07:59 PM

When I started out, a very good photographer in my local club advised that it was sufficient to get the lower edge of the histogram off the bottom of the scale. As long as I could see a gap, I had enough.

 

Three parameters affect position of the histogram. They are subexposure duration, telescope/lens focal ratio and camera gain/ISO. Your equipment will determine how you adjust them. If you guide, duration can be as long as you need. If not, you will be limited by the tracking accuracy of your mount. If you are using a telescope, your focal ratio will be fixed. If you are using a lens, you have options. Often, one stop less than wide open significantly improves the image. Your DSLR will have an optimum ISO but, depending on the flexibility of the other parameters, you may need something higher. I have noticed that the histogram gets narrower as subexposure duration increases.

 

My mentor's DSLR worked best at ISO 800. His telescope's focal ratio was 6.3 (SCT with reducer-corrector). He used an autoguider. He found that five minutes was the optimum subexposure duration for galaxies and nebulae. He would keep the same duration but drop down to ISO 400 for star clusters. My DSLR also works best at ISO 800 but my refractor is f/7.7. Therefore, I extend subexposure duration to eight minutes. I've also used a CLS filter to capture emission nebulae and had to extend subexposure duration to 30 minutes. On occasion, I replace the telescope with camera lenses with focal ratios between 4.0 and 5.6. That enables me to shorten subexposure duration which is fortunate since setup is easier if I don't have to include an autoguider.


  • 17.5Dob likes this

#12 17.5Dob

17.5Dob

    Fly Me to the Moon

  • ****-
  • Posts: 6,770
  • Joined: 21 Mar 2013
  • Loc: Colorado,USA

Posted 01 March 2020 - 09:59 PM

This is a typical histogram/ sub for me.....(from a modded cam). .............(And after stacking and PP)


49606872903_c6a6b004cc_h.jpg


 


Edited by 17.5Dob, 01 March 2020 - 10:03 PM.


#13 17.5Dob

17.5Dob

    Fly Me to the Moon

  • ****-
  • Posts: 6,770
  • Joined: 21 Mar 2013
  • Loc: Colorado,USA

Posted 01 March 2020 - 10:28 PM

This "typical" sub is even darker, and 7 min. at ISO 6.5......( after stacking and PP)


49607708292_2ef8d6ef48_h.jpg


Total time is the most important requirement. I've gotten to the point, where I consider 3 hrs bare minimum any more....


Edited by 17.5Dob, 01 March 2020 - 10:28 PM.



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





Also tagged with one or more of these keywords: astrophotography, beginner



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics