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

First time doing 15 minute subs from the city.

  • Please log in to reply
36 replies to this topic

#1 asanmax

asanmax

    Vendor - DSLR Modifications

  • *****
  • Vendors
  • topic starter
  • Posts: 954
  • Joined: 17 Sep 2018
  • Loc: Vancouver BC

Posted 12 April 2021 - 12:09 AM

So the Owl Nebula has been the center object for imaging for a couple of days.

It's a very faint nebula and requires a lot of exposure time when using small telescopes.

I captured some 480sec subs yesterday but the final result was not satisfying.

Today I am doing 15 minute subs through the L-eNhance filter under the Bortle class 8 skies.

Nikon D5300, ISO 800.

I was really surprised to see the amount of detail in one sub.

Here is a quickly stretched 15 min JPEG image, cropped about 30%.

 

Single__0006_ISO800_900s__NA.jpg


  • tim53, mtc, mikewayne3 and 5 others like this

#2 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 24,180
  • Joined: 27 Oct 2014

Posted 12 April 2021 - 01:56 AM

So the Owl Nebula has been the center object for imaging for a couple of days.

It's a very faint nebula and requires a lot of exposure time when using small telescopes.

I captured some 480sec subs yesterday but the final result was not satisfying.

Today I am doing 15 minute subs through the L-eNhance filter under the Bortle class 8 skies.

Nikon D5300, ISO 800.

I was really surprised to see the amount of detail in one sub.

Here is a quickly stretched 15 min JPEG image, cropped about 30%.

 

attachicon.gifSingle__0006_ISO800_900s__NA.jpg

Nice image.  But seeing that much detail in a sub is not a good thing.

 

There's an optimal subexposure that balances out read noise (too many too short subs) and dynamic range (wrecked by too few too long subs).  It's not 15 minutes at ISO800.  That's too long, too high ISO.

 

I have an almost identical D5500.  The best ISO for these cameras is 200.  And it's a serious mistake to be swayed by what a single sub looks like.  I sometimes can barely see the target in one sub.  What counts is what you get after stacking many of them, and processing the data well. 

 

This is a somewhat extreme example.  I took this with an F2 RASA in Bortle 7 skies, so my optimal subexposure is very short, 10 seconds.  But it's correct.   Below is what a single sub (the data has been stretched) looks like.  Not much.  But stack 650 of them and process, and see what you get.  That's a great deal of dim detail.

 

This is how DSO astrophotography works.  You build an image with subs slowly.  The key to getting dim detail is _not_ subexposure length, it's total imaging time.

 

Exactly what ISO and subexposure are best is a complicated deal, often discussed here.  I just wanted to make the basic point.

 

Pleiadies subv2.jpg

 

Pleadies 2019 V3 smaller.jpg


Edited by bobzeq25, 12 April 2021 - 02:13 AM.

  • coblr, asanmax, Astrolamb and 2 others like this

#3 DubbelDerp

DubbelDerp

    Surveyor 1

  • *****
  • Posts: 1,587
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 12 April 2021 - 07:52 AM

That looks pretty good! Nice round stars for such long exposures. It's amazing how much light is blocked by that filter. I haven't tried nearly that long of an exposure, but 5 minutes at f/2.8 ends up with a pretty bright background. I tend to push the exposure quite a bit longer than conventional wisdom suggests as well...


  • asanmax likes this

#4 zjc26138

zjc26138

    Loved By All

  • *****
  • Posts: 9,548
  • Joined: 24 Feb 2005
  • Loc: Steubenville, Ohio

Posted 12 April 2021 - 09:03 AM

Looks great to me!


  • asanmax likes this

#5 asanmax

asanmax

    Vendor - DSLR Modifications

  • *****
  • Vendors
  • topic starter
  • Posts: 954
  • Joined: 17 Sep 2018
  • Loc: Vancouver BC

Posted 12 April 2021 - 09:48 AM

Nice image.  But seeing that much detail in a sub is not a good thing.

 

There's an optimal subexposure that balances out read noise (too many too short subs) and dynamic range (wrecked by too few too long subs).  It's not 15 minutes at ISO800.  That's too long, too high ISO.

 

I have an almost identical D5500.  The best ISO for these cameras is 200.  And it's a serious mistake to be swayed by what a single sub looks like.  I sometimes can barely see the target in one sub.  What counts is what you get after stacking many of them, and processing the data well. 

 

This is a somewhat extreme example.  I took this with an F2 RASA in Bortle 7 skies, so my optimal subexposure is very short, 10 seconds.  But it's correct.   Below is what a single sub (the data has been stretched) looks like.  Not much.  But stack 650 of them and process, and see what you get.  That's a great deal of dim detail.

 

This is how DSO astrophotography works.  You build an image with subs slowly.  The key to getting dim detail is _not_ subexposure length, it's total imaging time.

 

Exactly what ISO and subexposure are best is a complicated deal, often discussed here.  I just wanted to make the basic point.

 

attachicon.gifPleiadies subv2.jpg

 

attachicon.gifPleadies 2019 V3 smaller.jpg

Thank you.

That's correct. I knew the stars would get blown out in 15 min subs. I'll have to either pull the star colors from the shorter subs that I did the night before of just do some visible light subs.

The purpose of the longer exposures was to get fainter details of the nebula and to try to push the equipment beyond its comfort zone :)


  • Skysmacker likes this

#6 asanmax

asanmax

    Vendor - DSLR Modifications

  • *****
  • Vendors
  • topic starter
  • Posts: 954
  • Joined: 17 Sep 2018
  • Loc: Vancouver BC

Posted 12 April 2021 - 09:51 AM

That looks pretty good! Nice round stars for such long exposures. It's amazing how much light is blocked by that filter. I haven't tried nearly that long of an exposure, but 5 minutes at f/2.8 ends up with a pretty bright background. I tend to push the exposure quite a bit longer than conventional wisdom suggests as well...

Thank you! I think that filter or its brother L-eXtreme, is a must have item for DSLR imagers.

Looks great to me!

Thank you!



#7 jpengstrom

jpengstrom

    Explorer 1

  • -----
  • Posts: 73
  • Joined: 08 Feb 2021
  • Loc: North Texas

Posted 12 April 2021 - 10:01 AM

Nice image.  But seeing that much detail in a sub is not a good thing.

 

There's an optimal subexposure that balances out read noise (too many too short subs) and dynamic range (wrecked by too few too long subs).  It's not 15 minutes at ISO800.  That's too long, too high ISO.

 

I have an almost identical D5500.  The best ISO for these cameras is 200.  And it's a serious mistake to be swayed by what a single sub looks like.  I sometimes can barely see the target in one sub.  What counts is what you get after stacking many of them, and processing the data well. 

 

This is a somewhat extreme example.  I took this with an F2 RASA in Bortle 7 skies, so my optimal subexposure is very short, 10 seconds.  But it's correct.   Below is what a single sub (the data has been stretched) looks like.  Not much.  But stack 650 of them and process, and see what you get.  That's a great deal of dim detail.

 

This is how DSO astrophotography works.  You build an image with subs slowly.  The key to getting dim detail is _not_ subexposure length, it's total imaging time.

 

Exactly what ISO and subexposure are best is a complicated deal, often discussed here.  I just wanted to make the basic point.

 

attachicon.gifPleiadies subv2.jpg

 

attachicon.gifPleadies 2019 V3 smaller.jpg

Bob, I'd like to explore the concept you brought up.  I'm trying to further my understanding since I'm new to astrophotography, not argue with you. smile.gif 

Putting aside tracking for a minute (yeah, I know that's part of the total picture and needs to be taken into account, but for this discussion let's assume you have a theoretically perfect way of tracking and perfect skies that would allow any length exposure you wanted).  You bring up the issue of dynamic range which is an important aspect of DSLR sensors.  Essentially it can be boiled down to difference between the brightest details that can be captured without blowing out the pixel on the sensor (or blowing out one of the color channels on the sensor since DSLRs are bayer sensors) vs. the darkest details that can be captured without losing information in black, yes?  Well, there's also the issue of smoother gradients between similar pixels that higher dynamic range allows. For most cameras if you look at the Photons to Photos site you can see than dynamic range decreases as ISO increases ( https://www.photonst.../Charts/PDR.htm ).  So shooting at a lower ISO gives you the ability to capture more dynamic range.  But, the example the OP posted of the Owl Nebula has no blown out pixels within the nebula itself.  Yes, there are a number of stars around the nebula that have a couple blown out channels but in that regards the question becomes how important is the star color to the image you're processing.  And yes, I also see some blockiness when zooming the nebula in the OP's picture, but I'm not sure how much of that is JPEG artifacts from posting to this site vs. lack of separation in the detail captured by the sensor.

Thanks for any insights you care to share - like I said, I'm trying to learn and increase my understanding!



#8 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 24,180
  • Joined: 27 Oct 2014

Posted 12 April 2021 - 10:11 AM

Bob, I'd like to explore the concept you brought up.  I'm trying to further my understanding since I'm new to astrophotography, not argue with you. smile.gif 

Putting aside tracking for a minute (yeah, I know that's part of the total picture and needs to be taken into account, but for this discussion let's assume you have a theoretically perfect way of tracking and perfect skies that would allow any length exposure you wanted).  You bring up the issue of dynamic range which is an important aspect of DSLR sensors.  Essentially it can be boiled down to difference between the brightest details that can be captured without blowing out the pixel on the sensor (or blowing out one of the color channels on the sensor since DSLRs are bayer sensors) vs. the darkest details that can be captured without losing information in black, yes?  Well, there's also the issue of smoother gradients between similar pixels that higher dynamic range allows. For most cameras if you look at the Photons to Photos site you can see than dynamic range decreases as ISO increases ( https://www.photonst.../Charts/PDR.htm ).  So shooting at a lower ISO gives you the ability to capture more dynamic range.  But, the example the OP posted of the Owl Nebula has no blown out pixels within the nebula itself.  Yes, there are a number of stars around the nebula that have a couple blown out channels but in that regards the question becomes how important is the star color to the image you're processing.  And yes, I also see some blockiness when zooming the nebula in the OP's picture, but I'm not sure how much of that is JPEG artifacts from posting to this site vs. lack of separation in the detail captured by the sensor.

Thanks for any insights you care to share - like I said, I'm trying to learn and increase my understanding!

There are two main factors.  Dynamic range and read noise.  Consider a fixed total imaging time, say 1 hour.  (See below for more about that).  Lower ISO, shorter exposures, more subs, maximize dynamic range, but increase the effects of read noise, which add with every sub,

 

You set subexposure to balance out the effects of dynamic range and read noise.  A complicated topic needing study.  Some concepts and pointers below, but it's a research project.

 

The bottom line of the post below is still the bottom line.  Subexposure is approximate, and less important than total imaging time.
 


  • Astrolamb and jpengstrom like this

#9 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 24,180
  • Joined: 27 Oct 2014

Posted 12 April 2021 - 10:26 AM

 

The purpose of the longer exposures was to get fainter details of the nebula

That's a common misconception.  Things don't work like that.

 

I've been trying to explain that's not the role of subexposure.  The role of subexposure is to balance dynamic range and read noise, as described above.  The way you get more dim detail is more total imaging time.

 

The more photons in the stack the better.  That drives the train, provided subexposures are anywhere near right.  How you break total imaging time up into subexposures is substantially less important.

 

There are many methods of setting subexposure.  I like setting the average value of a frame to 5-10 X the read noise squared.  With a DSLR, setting the back of the camera histogram to 1/3 over from the left is popular.  With a Nikon D5300+, and it's low read noise, 1/4 over is probably better.

 

All this is better discussed (it takes several pages) in this superb book.

 

https://www.amazon.c...h/dp/1138055360

 

Bottom line.  We could argue about which is better; 60 60 second subs or 120 30 second subs.  But what's _important_ is that _either_ 120 60 second subs or 240 30 second subs will beat _either_ of the first alternatives, handily.  Subexposure is just not that important.

 

To capture more detail, get subexposure in the ballpark, and shoot more subs.  <smile>  My rule of thumb is that one hour is the absolute minimum total imaging time, 2 is better, 4 good.  I haven't shot more than one target in a night for years.  My best images (and most people's) come from imaging one target over multiple nights.  More total imaging time.


Edited by bobzeq25, 12 April 2021 - 10:28 AM.

  • deansjc, Kevin_A, asanmax and 2 others like this

#10 DubbelDerp

DubbelDerp

    Surveyor 1

  • *****
  • Posts: 1,587
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 12 April 2021 - 10:35 AM

I find this thread to be a good reference in how to pull out really dim stuff, especially response #10.

 

https://www.cloudyni...nd-dust-clouds/

 

This is essentially what OP was doing - sometimes the details you want to capture are just buried down in the bottom of the histogram. And by sacrificing your dynamic range, it can be helpful to bump up your ISO and increase your exposure time so that you're sampling those really dim details with a bit more of your sensor's range. And you can always blend in some shorter exposures to recover your star colors. Besides, that dual-band filter ruins star colors anyways, so there's not much point in trying to optimize for star colors when you won't catch them anyways with the filter.

 

It's certainly not for everyone, though.


  • 2ghouls likes this

#11 Challenger75

Challenger75

    Explorer 1

  • *****
  • Posts: 64
  • Joined: 29 Sep 2020
  • Loc: Rhode Island

Posted 12 April 2021 - 10:39 AM

That's a common misconception.  Things don't work like that.

 

I've been trying to explain that's not the role of subexposure.  The role of subexposure is to balance dynamic range and read noise, as described above.  The way you get more dim detail is more total imaging time.

 

The more photons in the stack the better.  That drives the train, provided subexposures are anywhere near right.  How you break total imaging time up into subexposures is substantially less important.

 

There are many methods of setting subexposure.  I like setting the average value of a frame to 5-10 X the read noise squared.  With a DSLR, setting the back of the camera histogram to 1/3 over from the left is popular.  With a Nikon D5300+, and it's low read noise, 1/4 over is probably better.

 

All this is better discussed (it takes several pages) in this superb book.

 

https://www.amazon.c...h/dp/1138055360

 

Bottom line.  We could argue about which is better; 60 60 second subs or 120 30 second subs.  But what's _important_ is that _either_ 120 60 second subs or 240 30 second subs will beat _either_ of the first alternatives, handily.  Subexposure is just not that important.

 

To capture more detail, get subexposure in the ballpark, and shoot more subs.  <smile>  My rule of thumb is that one hour is the absolute minimum total imaging time, 2 is better, 4 good.  I haven't shot more than one target in a night for years.  My best images (and most people's) come from imaging one target over multiple nights.  More total imaging time.

Could you explain to me how you are going to get a histogram  1/4-1/3 off the left by taking short exposures without jacking the ISO through the roof? Are you stacking in all the sub exposures and then looking at where the finally Histogram is? Thank you. 



#12 asanmax

asanmax

    Vendor - DSLR Modifications

  • *****
  • Vendors
  • topic starter
  • Posts: 954
  • Joined: 17 Sep 2018
  • Loc: Vancouver BC

Posted 12 April 2021 - 11:01 AM

That's a common misconception.  Things don't work like that.

 

I've been trying to explain that's not the role of subexposure.  The role of subexposure is to balance dynamic range and read noise, as described above.  The way you get more dim detail is more total imaging time.

 

The more photons in the stack the better.  That drives the train, provided subexposures are anywhere near right.  How you break total imaging time up into subexposures is substantially less important.

 

There are many methods of setting subexposure.  I like setting the average value of a frame to 5-10 X the read noise squared.  With a DSLR, setting the back of the camera histogram to 1/3 over from the left is popular.  With a Nikon D5300+, and it's low read noise, 1/4 over is probably better.

 

All this is better discussed (it takes several pages) in this superb book.

 

https://www.amazon.c...h/dp/1138055360

 

Bottom line.  We could argue about which is better; 60 60 second subs or 120 30 second subs.  But what's _important_ is that _either_ 120 60 second subs or 240 30 second subs will beat _either_ of the first alternatives, handily.  Subexposure is just not that important.

 

To capture more detail, get subexposure in the ballpark, and shoot more subs.  <smile>  My rule of thumb is that one hour is the absolute minimum total imaging time, 2 is better, 4 good.  I haven't shot more than one target in a night for years.  My best images (and most people's) come from imaging one target over multiple nights.  More total imaging time.

Thank you, your feedback is always appreciated. 

Interestingly enough, the histogram was sitting at 1/3 in a 15 min sub.

I know that for my camera and the equipment the best exposure time to keep the star colors  is around 60sec-120sec.

But, my 4+ hour stack of 480sec subs did not reveal enough fainter details in the outskirts of the nebula. 

It's interesting to see that a single 15min sub showed some good detail.

Not arguing about the best exposure time considering the camera I used, but for the sake of getting more detail, the winner is the 15min sub.


  • calypsob and 2ghouls like this

#13 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 24,180
  • Joined: 27 Oct 2014

Posted 12 April 2021 - 11:06 AM

Could you explain to me how you are going to get a histogram  1/4-1/3 off the left by taking short exposures without jacking the ISO through the roof? Are you stacking in all the sub exposures and then looking at where the finally Histogram is? Thank you. 

Back of the camera histogram (which is stretched).  Just one sub.  The proper ISO, a topic all to itself.

 

It's somewhat crude, but subexposure doesn't need to be precise.


  • asanmax likes this

#14 deansjc

deansjc

    Mariner 2

  • -----
  • Posts: 211
  • Joined: 14 Jan 2018

Posted 12 April 2021 - 11:12 AM

Fascinating thread - great contributions.

 

Right on point:  Robin Glover discusses maximum exposure length.  It is a great 53 minutes.  Highly recommended:

 

https://www.youtube....93UvP358&t=179s


  • coblr, bobzeq25 and jpengstrom like this

#15 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 24,180
  • Joined: 27 Oct 2014

Posted 12 April 2021 - 11:15 AM

Thank you, your feedback is always appreciated. 

Interestingly enough, the histogram was sitting at 1/3 in a 15 min sub.

I know that for my camera and the equipment the best exposure time to keep the star colors  is around 60sec-120sec.

But, my 4+ hour stack of 480sec subs did not reveal enough fainter details in the outskirts of the nebula. 

It's interesting to see that a single 15min sub showed some good detail.

Not arguing about the best exposure time considering the camera I used, but for the sake of getting more detail, the winner is the 15min sub.

Ah.  If your back of the camera histogram (which is stretched) is 1/3 over, you're fine.  And I envy you your extremely dark skies.

 

If the linear histogram is 1/3 over, you're horribly overexposed.  Here's what a proper linear histogram and the accompanying linear sub looks like.  Click on it for visibility.  I know, you can't see it, except for a few stars.  <smile>  That's correct, your eyes don't see linear.

 

a linear histogram and sub.jpg


Edited by bobzeq25, 12 April 2021 - 11:16 AM.

  • asanmax, Astrolamb and galacticinsomnia like this

#16 asanmax

asanmax

    Vendor - DSLR Modifications

  • *****
  • Vendors
  • topic starter
  • Posts: 954
  • Joined: 17 Sep 2018
  • Loc: Vancouver BC

Posted 12 April 2021 - 11:16 AM

Fascinating thread - great contributions.

 

Right on point:  Robin Glover discusses maximum exposure length.  It is a great 53 minutes.  Highly recommended:

 

https://www.youtube....93UvP358&t=179s

That's one of the best videos I've seen on the exposure time topic.

I've watched it twice.



#17 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 24,180
  • Joined: 27 Oct 2014

Posted 12 April 2021 - 11:31 AM

Better illustration, in some regards.  My Nikon D5500.  Linear sub.  Linear histogram at the bottom.  Stretched histogram on top, similar (ignore the spikes, irrelevant) to what one would see on the back of the camera.  30 seconds at ISO 400, Nikon D5500 with 200mm F4 lens, Bortle 7 skies.

 

A touch underexposed (this was _very_ early days), 60 seconds would have been better.  Maybe even 120.

 

But it doesn't matter much.  See post #9.  My big mistake was shooting only 24 subs, 12 minutes is ridiculous.  I needed more like 200.

 

a linear sub and histogram V2.jpg


Edited by bobzeq25, 12 April 2021 - 11:50 AM.

  • Astrolamb likes this

#18 coblr

coblr

    Mariner 2

  • *****
  • Posts: 261
  • Joined: 09 Jan 2009

Posted 13 April 2021 - 12:24 AM

Fascinating thread - great contributions.

 

Right on point:  Robin Glover discusses maximum exposure length.  It is a great 53 minutes.  Highly recommended:

 

https://www.youtube....93UvP358&t=179s

Thanks for this link. I liked being able to get the equations as well as the explanations. I'm bummed that he had to jump through the gain stuff. So close!

 

I haven't had quite the right opportunity to try out the shorter exposure thing since getting new equipment, but the whole "it's about total integration time" thing seems to more and more sense with the newer technology.

 

I'm also glad to see the comparison between the sub and the stack. Most people only show their stacks so when you take a sub and don't see much, it's hard to imagine the final thing having much more than that. There is a strong tendency to take subs long enough so that you actually see something in it because then you feel safe that the final thing you have will have something in it.



#19 Astrolamb

Astrolamb

    Vostok 1

  • -----
  • Posts: 155
  • Joined: 26 Feb 2020
  • Loc: Arlington, Texas

Posted 13 April 2021 - 02:58 AM

Better illustration, in some regards.  My Nikon D5500.  Linear sub.  Linear histogram at the bottom.  Stretched histogram on top, similar (ignore the spikes, irrelevant) to what one would see on the back of the camera.  30 seconds at ISO 400, Nikon D5500 with 200mm F4 lens, Bortle 7 skies.

 

A touch underexposed (this was _very_ early days), 60 seconds would have been better.  Maybe even 120.

 

But it doesn't matter much.  See post #9.  My big mistake was shooting only 24 subs, 12 minutes is ridiculous.  I needed more like 200.

 

attachicon.gifa linear sub and histogram V2.jpg

 

I referenced you and this wisdom you continue to shout yesterday on a post I made about having trouble with DSS.

In case you didn't see it there thank you for sharing the information and understanding of what a sub should actually be exposed to.

Since most of what I was doing was troubleshooting a problem I didnt have a lot of subs to stack but it was still enough to see the huge difference after backing off the exposure and aiming for about 1000 ADU. I also learned I have been severely under exposing my flats thanks to your advice! 

It was incredible to see just how much more detail actually came out and how much better the star color was overall. 


  • bobzeq25 likes this

#20 deansjc

deansjc

    Mariner 2

  • -----
  • Posts: 211
  • Joined: 14 Jan 2018

Posted 13 April 2021 - 08:38 AM

Bob, I went back over this thread - an excellent one - and did not find what led you to determine that you were underexposing your flats.  Can you expand a bit?  What is your target exposure for them now?

 

Thanks.

 

John



#21 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 24,180
  • Joined: 27 Oct 2014

Posted 13 April 2021 - 01:05 PM

Bob, I went back over this thread - an excellent one - and did not find what led you to determine that you were underexposing your flats.  Can you expand a bit?  What is your target exposure for them now?

 

Thanks.

 

John

Did you mean _this_ thread?

 

That wasn't a flat.    The DSLR light above was underexposed based on its ADU, which I glanced at.  The calculation is a bit complicated, and I didn't do it (no point), but my experience told me that ADU was too low.  Back then I had very little experience, was pretty much just winging subexposure.  

 

Flats exposure should be easier than most people make it.  You want the _linear_ histogram to be about in the middle.  Exact doesn't buy you anything, you just need to keep away from the edges, where sensors can get non linear.  With a color camera, the 3 channels will generally be separated some, also not a problem.

 

I have a problem, I can't not answer a question  <smile>  The ADU of the light was too low, because I generally set my subexposure so that the bias corrected ADU of a light, converted to electrons, is 5-10 X the read noise squared.  If you want to know why, the book I cited in #9 has a terrific (but lengthy) explanation.  Warning - math ahead.

 

That's very similar to the method Glover uses.  We're both trying to be sure read noise is minimal compared to sky noise.


Edited by bobzeq25, 13 April 2021 - 01:20 PM.

  • Astrolamb likes this

#22 deansjc

deansjc

    Mariner 2

  • -----
  • Posts: 211
  • Joined: 14 Jan 2018

Posted 13 April 2021 - 02:07 PM

Hi again.  Sorry, I messed up a bit, it was

 

Astrolamb's post, not yours....confusion from embedding - my bad.

 

"I also learned I have been severely under exposing my flats thanks to your advice!

It was incredible to see just how much more detail actually came out and how much better the star color was overall."



#23 Astrolamb

Astrolamb

    Vostok 1

  • -----
  • Posts: 155
  • Joined: 26 Feb 2020
  • Loc: Arlington, Texas

Posted 13 April 2021 - 03:56 PM

Bob, I went back over this thread - an excellent one - and did not find what led you to determine that you were underexposing your flats.  Can you expand a bit?  What is your target exposure for them now?

 

Thanks.

 

John

 

Did you mean _this_ thread?

 

That wasn't a flat.    The DSLR light above was underexposed based on its ADU, which I glanced at.  The calculation is a bit complicated, and I didn't do it (no point), but my experience told me that ADU was too low.  Back then I had very little experience, was pretty much just winging subexposure.  

 

Flats exposure should be easier than most people make it.  You want the _linear_ histogram to be about in the middle.  Exact doesn't buy you anything, you just need to keep away from the edges, where sensors can get non linear.  With a color camera, the 3 channels will generally be separated some, also not a problem.

 

I have a problem, I can't not answer a question  <smile>  The ADU of the light was too low, because I generally set my subexposure so that the bias corrected ADU of a light, converted to electrons, is 5-10 X the read noise squared.  If you want to know why, the book I cited in #9 has a terrific (but lengthy) explanation.  Warning - math ahead.

 

That's very similar to the method Glover uses.  We're both trying to be sure read noise is minimal compared to sky noise.

 

Hi again.  Sorry, I messed up a bit, it was

 

Astrolamb's post, not yours....confusion from embedding - my bad.

 

"I also learned I have been severely under exposing my flats thanks to your advice!

It was incredible to see just how much more detail actually came out and how much better the star color was overall."

 

 

You will have to bear with me because this is a relatively new concept for me, but explaining it in my own words and having bobzeq25 correct any mistakes I might make will definitely help solidify the concept in my mind.  

 

What led me to the realization was finding empirical data for what a proper exposure is for subs.

Because all the data is captured linearly, and any sub you see on the back of the camera or within an acquisition software is actually stretched and no longer linear, you can use the average values of the pixels to estimate a proper exposure. This value is measured in ADU.

 

If this is your first introduction to ADU, imagine your histogram as an x, y chart(which it is). The horizontal axis is measured in ADU and the vertical axis is measured in Density. So in other words ADU would be the pixel value and the density would be the number of pixels with that value. 

 

If a proper exposure should be capturing just enough data that the read noise in a bias frame can be subtracted away and leave true data across the entire frame, then my flat my flat frames as I saw them in image previews were actually extremely dark on the edges if left in the linear format.  

 

Stacking software only works with the linear data, it does not apply any stretch or alteration to linear data, if it did it would alter our photos far too much and it would make post processing much more difficult. Since it only works with the linear data and my flat frames were so dark linearly, the correction it applies would be clipped at the edges of the frame and possibly less accurate across the entire frame as well. 

 

I happened to find another thread as I was thinking about all of this that was discussing the proper exposure for a flat frame in a linear format measured in average ADU across the frame. So I tested this with my own setup (happened to be really good timing since I just received my new led tracing board to take my flat frames with). 

 

Since my Nikon D5300 is natively a 12bit camera, it can capture up to ~16000 ADU as a maximum. A bias frame registers as ~600 ADU. To capture a data across the entire frame with my camera I should aim for about 1000 ADU, This prevents me from washing out the stars as I stack my data together and it preserves the level of details in the shadows through the highlights of nebula and galaxies.

With my flat frames that I thought were properly exposed (1/2way through the histogram on BYN, and showed a nice even gradient across the entire frame) viewed linearly they are entirely black. I measured that the data starts below 1000ADU and peaks near 1400ADU. For effective flat frame calibration you should be utilizing as much of the linear data as possible right?

 

Since you want your flats to have the best possible effect across the entire frame, you should be aiming for an average value right in the middle of the possible range. Since that range is 0-16000ADU for my camera I aimed for 8000 ADU. This means that the brightest sections still have a perfect gradient without being washed out and the darkest sections have a perfect gradient without being clipped. 

 

On BYN the preview photo looks completely washed out and the histogram is almost all the way to the right without being clipped off(probably close to 5/6 if I gave it an estimate). But the result of using the flats that were exposed to this level was blatantly apparent. 

 

As far as numbers go, my flats before only took up a range of 800 ADU out of the available 0-16000ADU where my new flats take up a range of 7200 ADU out of the available 0-16000ADU. That alone should indicate how much more accurate the gradients in the flat are and how much better they actually correct my sub exposures. 

 

I hope that was somewhat clear and bobzeq25 please correct me anywhere I am wrong or too ambiguous in my explanation.

 

**I must give credit where credit is due and thank bobzeq25 again for sharing this information on so many different threads, every time I searched for more information about this topic bob was in the comments sharing a wealth of information on the subject. I can really credit my understanding of the subject almost entirely to him.**

 

I almost forgot to suggest a free program called Iris to examine your ADU if you are interested. I also found this because of bob. lol bow.gif


Edited by Astrolamb, 13 April 2021 - 03:59 PM.

  • limeyx likes this

#24 limeyx

limeyx

    Apollo

  • -----
  • Posts: 1,221
  • Joined: 23 May 2020
  • Loc: Seattle, WA

Posted 14 April 2021 - 01:39 AM

This is a timely thread for me also. I have been shooting this target from a Bortle 6 with a D5300 and no filters (night 3 now in a row)

Nights 1 & 2 are about 2.5 hours of 180s subs at ISO400

On Facebook today I saw someone who got a lot more detail from a similar location with shorter subs

 

Tonight I will capture around 3 hours of 60s subs of the same target and except for the moon being a little different, everything else will be as similar as possible

 

Tomorrow I am going to do some stacking and basic processing and see how things are different

 

I am definitely clipping a LOT fewer pixels tonight (~70 compared to ~900) and my subs are in the 900 ADU range compared to I think around 1500 ADU at 180s

 

It's going to be a super interesting test


  • asanmax and Astrolamb like this

#25 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 24,180
  • Joined: 27 Oct 2014

Posted 14 April 2021 - 12:56 PM

You will have to bear with me because this is a relatively new concept for me, but explaining it in my own words and having bobzeq25 correct any mistakes I might make will definitely help solidify the concept in my mind.  

 

What led me to the realization was finding empirical data for what a proper exposure is for subs.

Because all the data is captured linearly, and any sub you see on the back of the camera or within an acquisition software is actually stretched and no longer linear, you can use the average values of the pixels to estimate a proper exposure. This value is measured in ADU.

 

If this is your first introduction to ADU, imagine your histogram as an x, y chart(which it is). The horizontal axis is measured in ADU and the vertical axis is measured in Density. So in other words ADU would be the pixel value and the density would be the number of pixels with that value. 

 

If a proper exposure should be capturing just enough data that the read noise in a bias frame can be subtracted away and leave true data across the entire frame, then my flat my flat frames as I saw them in image previews were actually extremely dark on the edges if left in the linear format.  

 

Stacking software only works with the linear data, it does not apply any stretch or alteration to linear data, if it did it would alter our photos far too much and it would make post processing much more difficult. Since it only works with the linear data and my flat frames were so dark linearly, the correction it applies would be clipped at the edges of the frame and possibly less accurate across the entire frame as well. 

 

I happened to find another thread as I was thinking about all of this that was discussing the proper exposure for a flat frame in a linear format measured in average ADU across the frame. So I tested this with my own setup (happened to be really good timing since I just received my new led tracing board to take my flat frames with). 

 

Since my Nikon D5300 is natively a 12bit camera, it can capture up to ~16000 ADU as a maximum. A bias frame registers as ~600 ADU. To capture a data across the entire frame with my camera I should aim for about 1000 ADU, This prevents me from washing out the stars as I stack my data together and it preserves the level of details in the shadows through the highlights of nebula and galaxies.

With my flat frames that I thought were properly exposed (1/2way through the histogram on BYN, and showed a nice even gradient across the entire frame) viewed linearly they are entirely black. I measured that the data starts below 1000ADU and peaks near 1400ADU. For effective flat frame calibration you should be utilizing as much of the linear data as possible right?

 

Since you want your flats to have the best possible effect across the entire frame, you should be aiming for an average value right in the middle of the possible range. Since that range is 0-16000ADU for my camera I aimed for 8000 ADU. This means that the brightest sections still have a perfect gradient without being washed out and the darkest sections have a perfect gradient without being clipped. 

 

On BYN the preview photo looks completely washed out and the histogram is almost all the way to the right without being clipped off(probably close to 5/6 if I gave it an estimate). But the result of using the flats that were exposed to this level was blatantly apparent. 

 

As far as numbers go, my flats before only took up a range of 800 ADU out of the available 0-16000ADU where my new flats take up a range of 7200 ADU out of the available 0-16000ADU. That alone should indicate how much more accurate the gradients in the flat are and how much better they actually correct my sub exposures. 

 

I hope that was somewhat clear and bobzeq25 please correct me anywhere I am wrong or too ambiguous in my explanation.

 

**I must give credit where credit is due and thank bobzeq25 again for sharing this information on so many different threads, every time I searched for more information about this topic bob was in the comments sharing a wealth of information on the subject. I can really credit my understanding of the subject almost entirely to him.**

 

I almost forgot to suggest a free program called Iris to examine your ADU if you are interested. I also found this because of bob. lol bow.gif

Here's your only real mistake.  It's pretty serious.

 

Flats ADU should be _much_ larger than lights ADU.  For example, take the D5300 (yours) and the D5500 (mine, identical for this purpose).

 

Lights ADU should be in the vicinity of 1000 ADU (16 bits), raw.  Bias subtracted, 400 ADU (16bits).  ISO 200.  Getting it exactly at 1000 has no value whatsoever.  The linear histogram is a thin spike on the far left.

 

Flats should be about 32000 ADU (16 bits, which has 64000 as the maximum possible ADU).  The linear histogram should be about in the middle of the graph.  Getting it exactly at 32000 has no value.

 

If you shot flats the same ADU as your lights, it's no wonder your corners were too dark.   I'd expect them to be.

 

Needless to say, the calibration software makes this all work.  BUT, you must include either bias or dark flats.  Or it won't work.


Edited by bobzeq25, 14 April 2021 - 12:59 PM.

  • Astrolamb and Skysmacker like 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