Best amount of images for stacking

I came across a video of a Astro-photographer saying 25 light frames and 2 dark frames are the best and anything more yields little improvement to overall milky way IQ.

Is this true? And what are your experiences with stacking images?

There's no straightforward answer to your question as it depends on your skies, your equipment, your target.

Generally speaking for light frames the more data the less noise, the noise goes down with the square root of the number of images roughly, so 4 times the images is half the noise, 100 times is 1/10th the noise.

As to darks the same sort of impact occurs with many images, to average out the noise. Two is a tiny, tiny number.  Also, flats are quite important in astrophotography, some people think of them as only helping with dust motes, but that is not true, it adjusts the field for both small scale and large scale (e.g. vignette) effects.

What most people look for is a point of diminishing returns, and experience helps if you try to predict it rather than wait for it.  I do a trial stack after each night's runs, and when my image stops getting substantially better I know it is time to move on.

For more casual work where you may be doing it all in one night, and anxious to call it an evening, it is more difficult.  Though there are programs that will do live stacking as you shoot, so you can see it "developing" as you go and judge.

But as a general statement -- more is always better.  Just a LOT more is only a little better.

Linwood

Ferguson wrote:

<snip>

Generally speaking for light frames the more data the less noise, the noise goes down with the square root of the number of images roughly, so 4 times the images is half the noise, 100 times is 1/10th the noise.

<snip>

Linwood

Part of the reason that people suggest little improvement after 25 lights is that the perceived reward drops rapidly because of the square root factor that Linwood points out.

If you take one light frame and compare it to a 4 image composite, you are delighted. The noise is reduced by a factor if two. And, wow! But to get the next factor of two you need to take 16 images (4X4). Wow again (maybe). And the next factor of 2 requires 64 images (4X4X4). By this time your reaction may be "ouch", not "wow".

Or more likely, after taking 16 images you only take another 16 images and look at the results. They are not impressive because it isn't a factor of 2 better. And you get discouraged.

We feel the pain linearly with the number of images, but the final product only gets better as the square root of the (total) number of images.

So more is always better, but you have to realize that each increment requires a lot more work.

There are other factors, such as read noise, which does not improve with the number of images, so the above explanation is a simplified one. But it captures what you need to know to answer your question.

imperial wrote:

I came across a video of a Astro-photographer saying 25 light frames and 2 dark frames are the best and anything more yields little improvement to overall milky way IQ.

Is this true? And what are your experiences with stacking images?

No, it's 100% incorrect and a completely arbitrary statement to be making, I would definitely ignore whatever advice that astrophotographer is given because they are making completely baseless claims.

As the others have said, the reduction in noise is proportional to the square root of the number of images stacked. A LOT of factors will go into what kind of visual result comes from stacking 10, 25, 250, or 1000 subs so saying "there's no reason to stack more than 25 exposures" is a completely pointless number. And no matter what 2 dark frames is an abhorrent number of dark frames to stack, you should not stack anything less than 30 dark frames IMO due to the nature of frame to frame variation as well as slight temperature shifts which can impact noise distribution. It is very easy to screw up dark frames and have a negative impact on your final image quality/calibration, using only 2 dark frames would probably guarantee screwing up your calibration and actually adding noise to your image. There are a lot of deep discussions about the number of frames to stack for each type, a lot of great comparisons showing the increase in SNR and detail with more sub exposures stacked. The takeaway is simple for beginners, the more integration time and the more high quality exposures you can stack the better. For ultra wide angle imaging (say, wider than 100mm focal length) with a very fast lens you probably could get away with stacking an hour's worth of exposures. I don't recommend stacking anything less than 3 hours unless you only stacking for Milky Way/Nightscape purposes. For DSO imaging I tell people to aim for 6 hours+ of integration time.

Alen K wrote:

You probably missed imperial's other thread asking how to fix a problem with blur in the foreground of a nightscape that was caused by wind-induced motion. Maybe not coincidentally, the number of 10-second light frames that imperial took for that one was 25.

So, it would seem imperial's immediate interest is nightscapes, and untracked ones at that. Imperial will be quite surprised to hear that for one of those, he, she or they could "get away with" only a mere hour's worth of exposures.

In fact, haven't most of your nightscapes used a lot less integration time or total time on the sky than that? I am perusing your utahastrophotography website now and I'm not seeing anything that high. Of course, you mostly shoot in very dark locations and the OP may indeed require an hour (or likely much longer!) in more typical skies to get the kind of SNR you regularly achieve. But shooting for an hour or more means a lot of change in the MW's position and that brings its own problems, of course.

When it comes to nightscape/ultra wide angle stuff I don't even bother stacking, just single tracked exposures that are long enough for good SNR without clipping highlights. I do a lot of mosaics/panos though. When it comes to longer or even intermediate focal lengths where I start stacking, like 85mm and up, that's when integration matters. With my fast lenses I can get away with an hour, maybe two on many targets (unless I'm shooting something like dark nebula or things with deep dusty details), with my scopes I'm shooting for at least 6 hours, usually 10 or more.

Alen K wrote:

EricTheAstroJunkie wrote:

When it comes to nightscape/ultra wide angle stuff I don't even bother stacking, just single tracked exposures that are long enough for good SNR without clipping highlights.

Not hard to do in Bortle 1 or 2. Much tougher in Bortle 4. I know: I should move closer to someplace really dark!

It's all about positioning and post processing, here are a bunch of images I've shot from Bortle 4/5 skies, I position myself so I'm shooting away from the most intense sources of light pollution and do a lot of background subtraction in post processing:

https://flic.kr/p/VPGEAh

https://flic.kr/p/KuBEU6

https://flic.kr/p/K4kDWm

https://flic.kr/p/SovnCH

Is it tougher? Sure. Is the detail, color, and overall quality a little worse? Ya, but not horribly so. Bortle 4 is plenty dark to get some great shots with not that much effort, that last shot is at a place that's one of our favorites for testing out gear, it's a ~25-30 minute drive for us and even though it's bortle 4 (almost 5 now) we can shoot away from the light pollution and with the Milky Way low enough we can shoot into Bortle 2 skies.

Ferguson wrote:

There's no straightforward answer to your question as it depends on your skies, your equipment, your target.

Generally speaking for light frames the more data the less noise, the noise goes down with the square root of the number of images roughly, so 4 times the images is half the noise, 100 times is 1/10th the noise.

As to darks the same sort of impact occurs with many images, to average out the noise. Two is a tiny, tiny number. Also, flats are quite important in astrophotography, some people think of them as only helping with dust motes, but that is not true, it adjusts the field for both small scale and large scale (e.g. vignette) effects.

What most people look for is a point of diminishing returns, and experience helps if you try to predict it rather than wait for it. I do a trial stack after each night's runs, and when my image stops getting substantially better I know it is time to move on.

For more casual work where you may be doing it all in one night, and anxious to call it an evening, it is more difficult. Though there are programs that will do live stacking as you shoot, so you can see it "developing" as you go and judge.

But as a general statement -- more is always better. Just a LOT more is only a little better.

Linwood

okay, thanks!

EricTheAstroJunkie wrote:

imperial wrote:

I came across a video of a Astro-photographer saying 25 light frames and 2 dark frames are the best and anything more yields little improvement to overall milky way IQ.

Is this true? And what are your experiences with stacking images?

No, it's 100% incorrect and a completely arbitrary statement to be making, I would definitely ignore whatever advice that astrophotographer is given because they are making completely baseless claims.

As the others have said, the reduction in noise is proportional to the square root of the number of images stacked. A LOT of factors will go into what kind of visual result comes from stacking 10, 25, 250, or 1000 subs so saying "there's no reason to stack more than 25 exposures" is a completely pointless number. And no matter what 2 dark frames is an abhorrent number of dark frames to stack, you should not stack anything less than 30 dark frames IMO due to the nature of frame to frame variation as well as slight temperature shifts which can impact noise distribution. It is very easy to screw up dark frames and have a negative impact on your final image quality/calibration, using only 2 dark frames would probably guarantee screwing up your calibration and actually adding noise to your image. There are a lot of deep discussions about the number of frames to stack for each type, a lot of great comparisons showing the increase in SNR and detail with more sub exposures stacked. The takeaway is simple for beginners, the more integration time and the more high quality exposures you can stack the better. For ultra wide angle imaging (say, wider than 100mm focal length) with a very fast lens you probably could get away with stacking an hour's worth of exposures. I don't recommend stacking anything less than 3 hours unless you only stacking for Milky Way/Nightscape purposes. For DSO imaging I tell people to aim for 6 hours+ of integration time.

Wow, thanks. I've learned something but I did not expect the depth of it.

Alen K wrote:

You probably missed imperial's other thread asking how to fix a problem with blur in the foreground of a nightscape that was caused by wind-induced motion. Maybe not coincidentally, the number of 10-second light frames that imperial took for that one was 25.

So, it would seem imperial's immediate interest is nightscapes, and untracked ones at that. Imperial will be quite surprised to hear that for one of those, he, she or they could "get away with" only a mere hour's worth of exposures.

Yeah, I've also learned milkyway is considered nightscape and not astro. I do know about those star trackers but man anything beyond 1 min would be insane to me already.

In fact, haven't most of your nightscapes used a lot less integration time or total time on the sky than that? I am perusing your utahastrophotography website now and I'm not seeing anything that high. Of course, you mostly shoot in very dark locations and the OP may indeed require an hour (or likely much longer!) in more typical skies to get the kind of SNR you regularly achieve. But shooting for an hour or more means a lot of change in the MW's position and that brings its own problems, of course.