ISO 1600 v ISO 800

[allcart]

The test subs were only 1sec each. It was an experiment I did before I got my CG5-GT. I thought I would push the boundaries to find out if I could take subs at over 1100mm on a basic camera tripod without getting star trails. I could see some stars, but the image was useless. I never tried reducing the tso to find out what the limits were as I decided that I needed to change my way of thinking and follow the tried and tested methods.

I only did this due to my lack of understanding of ap. I naively thought I could apply the same rules to ap as I use with terrestial photographty. A lesson learned, but at least it only took 120secs to learn it.

[dph1nm]

I still believe that single-sub exposure length effects the depth of an image and long exposure lengths are needed to get extremely faint detail in an image that would not be attainable with shorter length subs. Or that you would need much much more short subs and longer total time.

This is quite true IF read-noise makes a significant contribution to the total noise in each sub-exposure. If it doesn't, then there really is no difference between long and short subs if they add to the same total exposure time. However, whether read-noise is significant depends on so many things (sky brightness, camera pixel size, focal length of scope, QE of the camera etc) that is it almost impossible to judge without making proper measurements on the night.

NIgelM

[dph1nm]

I can't help feeling his maths must be wrong, because the faint parts of a 15 second sub would quantize to zero, so nothing would emerge from the stack.

Not if you have the gain set to at least 1 ADU per photo-electron (actually it is even better than this - I think you only need to Nyquist-sample the noise level to this accuracy, so if you have, say 10e- rms noise, you only need 1ADU ~ 5e- to recover the faint detail).

NigelM

[nmoushon]

Ian - I finally found the time to read through (or at least skim) all the links you posted. Very interesting and informative links. Thank you for them. But really only the first and the last links can be applied to DSLR AP. The rest are regarding CCD imaging and can not be used to help determine settings for DSLR cameras. Well you might be able to but you would have to have a very good understanding of all the math and everything else involved with all those equations to be able to translate that over to work with DSLR. Plus it's a lot easier to get the need information about a CCD camera than a DSLR camera. A lot of the need information on a DSLR just isn't published my the manufacture or extremely hard to find and is not common knowledge. If you have a CCD camera the are GREAT LINKS!

The first link was a particularly good read in regards to DLSR cameras. I dont have the knowledge on cameras and the software on images that he has so some of his explanations are a bit hard to grasp at times. But basically his paper is saying that anything over ISO400 is actually hurting your image. I still find that hard to believe. Not saying it wrong, just hard to grasp that ISO800 hurts your image. Maybe its just that I've had this iso misconception drilled into my brain for so long that the truth is hard to grasp or that I need to test this out and prove it through my own testing is getting in the way. Again not that I dont believe him thats just my personality type.

[Ags]

The way I see it, ISO800 or 1600 only lose data if your exposures are long. If you can keep the stars as points for five minutes then you don't need to up the gain/ISO to quantize the faint bits.

[nmoushon]

The way I see it, ISO800 or 1600 only lose data if your exposures are long. If you can keep the stars as points for five minutes then you don't need to up the gain/ISO to quantize the faint bits.

Thats a good way to look at it. As long your your not burning out your stars higher ISO is ok. That makes since to me.

But his writing still suggest that anything over ISO400 is a waist. Which is whats getting my brain churning and wanting to do experiments for myself.

[Ags]

On the subject of super-short subs, look at this thread where FOUR SECOND subs have actually caught Barnard's Loop:

I think this is a shot that definitely is not hurt by ISO1600 :-)

[nmoushon]

It wasnt just that it was shot at iso1600. Its the combo of iso1600 and f/2.8. Where f/2.8 is the really kicker there. I have nothing against short subs. I've capture some great images on M42 with 15-60sec subs. Short subs can produce great images if you have the right combo of gear.

[Sp@ce_d]

Thats a good way to look at it. As long your your not burning out your stars higher ISO is ok. That makes since to me.

But his writing still suggest that anything over ISO400 is a waist. Which is whats getting my brain churning and wanting to do experiments for myself.

There was a discussion on here at the end of this thread about Craigs article..

Looks like he revised his opinion upwards after the subsequent discussion on CN.. to be honest I start clouding over trying to get my head around this. My brain still thinks in the terms of ASA & ISO in relation to film

[Merlin66]

All you need is to be able to record at least one "good" photon/ ADU in each of your shortest exposure. These will accumulate over the stacked exposures to present the image at the end....

No reprocity failure or photons being stored and not used as is the case with conventional film.

[nmoushon]

There was a discussion on here at the end of this thread about Craigs article.. http://stargazerslou...efore-boosting/

Looks like he revised his opinion upwards after the subsequent discussion on CN.. to be honest I start clouding over trying to get my head around this. My brain still thinks in the terms of ASA & ISO in relation to film

I'm sure films a bit different than digital. lol Dont think I could get myself to go to film.

I read that he revised his thought but he wasnt clear as to what he revised it too.

[Ags]

It wasnt just that it was shot at iso1600. Its the combo of iso1600 and f/2.8. Where f/2.8 is the really kicker there. I have nothing against short subs. I've capture some great images on M42 with 15-60sec subs. Short subs can produce great images if you have the right combo of gear.

Yes I agree f2.8 is important - 16 seconds would be required at f5.6, and that would be 48 seconds at f5.6 with an unmodded camera. I didn't mention the f ratio because I luckily own the same lens, but you are right it is critical, along with moddedness (a new word!).

[nmoushon]

Yes I agree f2.8 is important - 16 seconds would be required at f5.6, and that would be 48 seconds at f5.6 with an unmodded camera. I didn't mention the f ratio because I luckily own the same lens, but you are right it is critical, along with moddedness (a new word!).

MODDEDNESS! Love it!

And ya modding a camera can shorten needed exposure time on the Ha objects. Not so much on others but there are a lot of good Ha objects that would benefit from it. To bad modding is difficult and 'dangerous' to be done DIY. It would be very helpful for those who cant afford high priced mounts. You could get filters I guess but I would recon they arent as good as modding.

[Merlin66]

Mods to a Canon DSLR are pretty straight forward.....

Filters block some light (always!) whereas the mod improves the sensitivity of the CCD chip by removing the colour balance filter..

Re exposure times...I don't wholely disagree, but.....If the photons are arriving from your favourite galaxy (far far away) at a steady rate - then the only variable is - how many do you collect? This must, with a CCD, be directly related to the length of exposure???

[Ags]

Mods to a Canon DSLR are pretty straight forward.....

Filters block some light (always!) whereas the mod improves the sensitivity of the CCD chip by removing the colour balance filter..

Re exposure times...I don't wholely disagree, but.....If the photons are arriving from your favourite galaxy (far far away) at a steady rate - then the only variable is - how many do you collect? This must, with a CCD, be directly related to the length of exposure???

Exposure length... And F ratio.

[Merlin66]

Ags,

All the f ratio does is alter the "landing position" of the photons onto the CCD chip. Sure, there may be more "hits" on some pixels, but the total amount received only depends on the exposure???

(This is why larger f ratios are OK for the stars, and lower fratios used with extended objects (to get more "hits" into the individual pixel. I agree.)

[nmoushon]

Mods to a Canon DSLR are pretty straight forward.....

Filters block some light (always!) whereas the mod improves the sensitivity of the CCD chip by removing the colour balance filter..

Re exposure times...I don't wholely disagree, but.....If the photons are arriving from your favourite galaxy (far far away) at a steady rate - then the only variable is - how many do you collect? This must, with a CCD, be directly related to the length of exposure???

I would think and agree with what you and Agnes said but you also have to consider LP. I think that adds another factor...at least for DSLR and iso it does. Not sure for CCD. Obviously from the links Ian put up there are some different views on that.

[Ags]

I do agree that an f10 200mm SCT gathers four times as much light as a 200mm F2 camera lens (mixing astro and photo terminology here) and if you have a tiny galaxy the SCT would gather more data in a given time. But that ignores the point that these two systems would be used to photograph completely different objects, and it would be a waste pointing the f2 lens at the tiny galaxy. Pointed at an object scaled for the lens, the f2 lens gathers 25 times as much light per pixel per second.

[nmoushon]

I do agree that an f10 200mm SCT gathers four times as much light as a 200mm F2 camera lens

A bit confused...are you saying a 200mm aperture SCT vs a 200mm FL lens?

[Ags]

Yes, a 200mm f2 lens should have 100mm aperture and 1/4 the light gathering of the 200mm aperture SCT.

[nmoushon]

I know this is a whole nother bag of worms and deserves its own thread (because theres people on both sides of this) but in DSO AP aperture does not effect the amount of light to your image. Only the f/ratio does. And that is why the SCT is not a good scope for imaging DSO. On planetary imaging aperture does matter (from my understanding) as aperture increases so does the contrast which in turn helps get more detail.

[Luke]

The way I understand it (which I hope is right, as it took two years to get here, LOL) is:

Suppose you have an 80mm aperture scope and a 200mm aperture scope, both F5.

You take a single image of a galaxy in both scopes for the same amount of exposure time.

The galaxy looks about the same brightness in both scopes.

So what's the difference? What's the point of using the 200mm scope?

The 200mm scope has a 1000mm focal length, wheras the refractor has a 400mm focal length. The 200mm shows the galaxy at a much larger scale. If you want to get closer in to the galaxy, the 200mm is the way to go.

The 200mm scope has collected more photons of light from the galaxy than the 80mm scope (it collects 625% as much from the galaxy as the 80mm).

But the galaxy is 2.5 the size in the 200mm scope.

It is spreading the 625% as many photons over a bigger area, 2.5 x 2.5 times as many pixels. Which, guess what, means it spreads the light over 6.25 times as many pixels (NOT 2.5 times as many pixels, as both the width and height get 2.5 times bigger, so it's 2.5 * 2.5 times as many pixels = 6.25x - if you say double an image size from 50x50 pixels to 100x100 pixels, that's FOUR times as many pixels, not twice as many as might seem to be the case)

So the average number of photons hitting each pixel ends up being about the same in both telescopes and the images are equal brightness.

As you spread light over more pixels, hopefully it makes sense that each pixel is then getting less light.

As said this does not mean there is no advantage to the bigger aperture, it gets you in closer. If you put a 2.5 times barlow on the 80mm scope it would end up imaging the galaxy at the same scale, but you would have to image for 6.25x the amount of the 200mm scope to end up with the same brightness.

To sum it up, provided two telescopes have the same focal ratio, the boost in brightness that you get from the bigger aperture telescope is cancelled out but the loss in brightness caused by spreading the light from a target over more pixels. Bigger aperture at the same focal ratio does not buy you more brightness, it buys you more scale.

I hope that makes sense. I'd love to hear if I'm barking up the wrong tree, as this subject baffled me for ages, and that's how I understand it these days!

I know this is a whole nother bag of worms and deserves its own thread (because theres people on both sides of this) but in DSO AP aperture does not effect the amount of light to your image. Only the f/ratio does. And that is why the SCT is not a good scope for imaging DSO. On planetary imaging aperture does matter (from my understanding) as aperture increases so does the contrast which in turn helps get more detail.

[nmoushon]

You're right and wrong.

You're right that with the 80mm and 200mm they will both produce the same brightness of said galaxy because the brightness of that galaxy is a constant. Given that both scopes are at the same f/5 for example.

You're wrong in thinking that the aperture is whats effecting the scape of the image. The ONLY reason the aperture of an imaging telescope increases is to compensate for its focal length which in turn keeps the f/ratio to as small as possible. (Again this only applies for DSO imaging. Planetary imaging and visual have completely different set of 'rules') Think of it this was - If there was some magical device that could increase your focal length of your 80mm from 400mmFL to 1000mmFL but still keep it at f/5 to match your SCT at 200mm and 1000mmFL at f/5 you would see identical images and both would collect the same amount of light. This is completely impossible in our world so you might have to stretch you mind a bit to grasp it but it is a way to show that aperture doesnt effect light gather strength when it comes to imaging DSO.

In regards to spreading the light over pixels - You would be correct that it gathers more light but that only matters for visual purposes. I couldnt begin to explain the details about how the camera sensor works so bear with me as I try my best to explain what i've read/learned. The 625% more light can't be spread over more pixels because the camera senor has a set number of pixels that absorbs the same amount of light. The galaxy is a point source light. Though you 'zoom' in on this galaxy with greater FL the galaxy itself doesn't actually get any bigger. (No brainer there. Everyone knows thats.) So the amount of light reaching the senor doesnt increase. It might seem that way because your galaxy is more detailed in the 200mm vs the 80mm but that again is only because of the FL not the aperture.

[Ags]

The amount of light gathered by an optical system is simply the size of the opening at the front of the tube. The F ratio does not affect that. The focal length determines the size of the focussed image at the image plane. The ratio between image scale (focal length) and aperture (light gathered) is the F ratio.

A long focal length forms a bigger image, so for the same aperture the image will be dimmer (more stretched).

Merlin66 is perfectly correct that the SCT would gather 4 times as much light/data from the galaxy compared to the lens - although the image is faint the image is also much bigger and is formed by 4 times as many photons. Whether you want to try image a galaxy at f10 is another matter.

[Luke]

I agree aperture doesn't control image scale, focal length (and I guess pixel size of the sensor) does. I meant the advantage of the bigger aperture telescope of the two scopes mentioned is that it gets you in closer (as you say, this is down to the difference in focal length, not aperture!)

I was trying to say the benefit of the larger aperture is not wasted, it is going into helping maintain the brightness of the galaxy, even though the galaxy is a bigger scale than when imaged with the 80mm.

I disagree that the galaxy is a point light source, though maybe I'm missing your point? If it was point source, wouldn't that mean it would look the same size when doing visual astronomy no matter what magnification is used?

I'm not sure I understand what you are saying about the amount of light. I would say that the amount of light / number of photons from the galaxy reaching the sensor IS increased with the bigger aperture telescope (assuming the galaxy fits within the sensor). Those extra photons help to maintain image brightness because the photons from the galaxy cover a larger area of the sensor - they are being spread over more sensor pixels, but there are more of the galaxy photons, so each pixel that the galaxy covers ends up the same average brightness as with the 80mm scope.

The extra aperture is not wasted in terms of light gathering power. If it wasn't for the scope collecting more light from the galaxy, then the galaxy would be fainter with the increased image scale, just like if you barlow the 80mm scope. Get the smaller aperture scope using a barlow to image the galaxy at the same scale, then the bigger aperture scope wins, that's where its extra light gathering power counts.

That's my thoughts anyhow, or am I wrong? Quite a confusing area!! :-o

You're wrong in thinking that the aperture is whats effecting the scape of the image. The ONLY reason the aperture of an imaging telescope increases is to compensate for its focal length which in turn keeps the f/ratio to as small as possible. (Again this only applies for DSO imaging. Planetary imaging and visual have completely different set of 'rules') Think of it this was - If there was some magical device that could increase your focal length of your 80mm from 400mmFL to 1000mmFL but still keep it at f/5 to match your SCT at 200mm and 1000mmFL at f/5 you would see identical images and both would collect the same amount of light. This is completely impossible in our world so you might have to stretch you mind a bit to grasp it but it is a way to show that aperture doesnt effect light gather strength when it comes to imaging DSO.

In regards to spreading the light over pixels - You would be correct that it gathers more light but that only matters for visual purposes. I couldnt begin to explain the details about how the camera sensor works so bear with me as I try my best to explain what i've read/learned. The 625% more light can't be spread over more pixels because the camera senor has a set number of pixels that absorbs the same amount of light. The galaxy is a point source light. Though you 'zoom' in on this galaxy with greater FL the galaxy itself doesn't actually get any bigger. (No brainer there. Everyone knows thats.) So the amount of light reaching the senor doesnt increase. It might seem that way because your galaxy is more detailed in the 200mm vs the 80mm but that again is only because of the FL not the aperture.