One long exposure or lots of short ones?

[Hargo]

Which one will give better results?

[atmenterprises]

One long one because you are allowing more light to hit and potentially activate the pixels on the camera sensor.

[Digz]

One long one because you are allowing more light to hit and potentially activate the pixels on the camera sensor.

Ah yes but with DSLRs longer exposures can bring other undesirable effects into the image.......

The original question is quiet open and will no doubt generate a lot of debate

[Psychobilly]

It depends on a whole range of things - What kit are you going to be using - the gear in your sig?

[Photosbykev]

you need to handle any light pollution as well, if it's bad, then one long exposure is likely to be very overexposed

[cosmojaydee]

An open question that has so may more questions and a lot more answers

[Tinker1947]

I would say take lots of images at your optimum time setting, my 1100D works well pointing N/East couple of minute exposures, now if i point it to the South then a minute with the new LED lighting, so run some tests find your best setting and take lots at that....

[Stephen]

My unhelpful answer would be lots of long ones ... but only as long as your gear can handle Data, data, data!

[RobH]

Many factors are involved. This is a very brief synopsis (apologies if I missed anything!)

1. Sky background levels. At a certain exposure length your backround skyglow will overpower any faint data.

2. Saturation levels of target and stars. You shouldn't oversaturate your image....you will lose detail and colour information.

3. How long you can realistically expect to expose before something wrecks the exposure.

4. Tracking performance of mount.

5. Inherent noise of imaging system. Some systems build up noise over time and at a certain point your signal to noise ratio will get worse.

I tend to use, with a cooled CCD, between 10 and 15 minutes for my main target, and also shoot shorter (3-4 minutes) for colour data on my stars. Some targets (M31 and M42 for example) with a high dynamic range will require both longer and shorter exposures which are then blended later.

If I had darker skies I'd probably go to 20 minutes for some exposures, but 15 minutes is about the best I can do against the background.

Have a search on the forum...this is a subject that has been gone into in great detail many times in the past...you'll find lots of info.

Cheers

Rob

[laser_jock99]

Both- long exposures work well for faint objects but need darker skies. Short exposures are good for brighter objects in light polluted locations. Every situation is different, the trick is knowing what to do to get the most out the gear you have. A few test exposures should show you how long it's worth going for before sky fog and/or camera noise swamp the usefull signal.

[Tim]

Lots of long ones for me

[RobH]

Lots of long ones for me

But you shoot narrowband targets at F10 Tim.....the longer the better in this case, as you say

[Jannis]

Well, like everyone here have allready said, it depends. But in general, faint object needs long exposures, bright objects can be OK with shorter subs as long as you have enough of them.

If you get enough short subs on bright targets you can get OK-ish results too for sure.

The clue is that you need long enough exposures to be able to collect at least some of the light emitted by the targets in each picture, so it can later be averaged out into a more detailed picture. Like M42, M45 and so on, short exposures will be fine in most cases. But like the horsehead nebula and so without a modefied DSLR you'll need long exposures to be able to gather some of the IR light emitted. If you use shorter exposures on a faint target you'll just average out the noice, but you still won't be able to find any faint details.

[Cath]

Well, the more images you take the more stacking will even out the random noise (better signal-to-noise ratio), so allowing you to pull out those faint/detailed features in the stacked image that would otherwise have been lost in the random noise.

A combination of short and long exposures allows you to extract the detail in saturated/blown-out high brightness parts of the image and to also extract the faint dark features in the image at the same time .. this is often called HDR (High Dynamic Range).

[Luke]

I would think as said it depends on various things.

In my case with a Canon 450D DSLR, I found a lot of short exposures could give a suprising amount of detail.

But I found very short exposures definitely could not equal longer exposures on the target I shot with the same total exposure time. As far as I could tell, it was because some of the detail in the short exposures was lost down in the low level noise of the camera, whereas with longer exposures, those faint details got raised above the low level noise (I think that noise might be "read noise", and for me at least it is a no-go area for rescuing details).

[dph1nm]

So this is one of the most argued about (and misunderstood) topics on this forum!

Basically, stacking short exposures is less efficient than long ones (for the same total exposure) because of the read noise you get each time you take an image. How much less efficient depends on how much the read noise contributes to the total noise in your single sub - which can depend on many things, but in particular the noise from the sky background. If this dominates over the read noise then the advantage of longer subs is miniscule (in fact this is the point of longer subs - it is not the collecting of more photons from your faint DSO which is the advantage, it is collecting more from the sky!).

Notice I say less efficient - your faint DSO signal does not mysteriously vanish into the ether just because you have a short sub (unless you quantise it away, but that is another issue). It is still there, and will appear if you stack enough subs. You just need a longer total exposure time.

NigelM

[ollypenrice]

I've no experience of combating light pollution because I work from a dark site.

I'll say this, though; the idea that you burn out bright objects in longish exposures seems to me to be exaggerated. M42, yes. Give the Trapezium 10 to 20 seconds. But the notoroious star Alnitak, near the Horsehead, resolves as a double in 10 minute subs in RGB at F3.9 (fast!) for me. The trick is, how do you stretch it? The answer, in layers a bit at a time. Alnitak (see below) is not burned out in 10 min subs (RGB) at F3.9;

Rob is dead right on star colour. Stars show far more colour in short exposures and maybe in very short ones if using a DSLR. In LRGB imaging I try not to apply the L layer to the stars - based on Rob's observations on processing some time ago.

I very rarely use short subs to control bright objects because I find that different stretches of the data allow the whole dynamic range to be covered, 99 times out of 100. So I'm another 'Lots of long ones...'

Olly

[Tim]

When you get into narrowband imaging, as a general rule, the longer you expose, the more desirable signal you will acquire, up to a point. Because the filters are very strong and get rid of a lot of background noise you can get away with very long exposures.

The basic answer to the original question is that it depends on the target. M42 for instance is very very bright and lots of short exposures stacked will be better than one long one. But for galaxies, or nebulae, then one long exposure might give you a better result. But still a better result will come from lots of the longer exposures.

The best thing to do is experiment with your own kit and find out what works best for your own equipment.

Cheers

Tim

[Cath]

Depends on the cameras dynamic range as well, including it's noise level. A nice cooled 16-bit CCD will have much more dynamic range than say a 14-bit uncooled noisy DSLR.

[RobH]

I very rarely use short subs to control bright objects because I find that different stretches of the data allow the whole dynamic range to be covered, 99 times out of 100. So I'm another 'Lots of long ones...'

Olly

I'd agree with that Olly.....often you see mention of having to use short subs for the core of M31....I've never done this, I just use a lot of care stretching the data. The only objects I've ever used short subs combined with long ones on are M42 (30 seconds for the really bright trapezium region), and the Iris nebula, again for the very bright region close to the central star.

Rob

[Psychobilly]

Considering the OP's kit list I would have thought that tracking would be the limiting factor...and would suggest more shorter subs might give them an overall better result...

Peter.,.

[RobH]

Considering the OP's kit list I would have thought that tracking would be the limiting factor...and would suggest more shorter subs might give them an overall better result...

Peter.,.

Agreed...push it as far as you want, and then work back until you get a decent result.