ISO and exposure time. Bit confused

[Simon Dunsmore]

Hi Folks, 

I hope you're all staying safe and well. 

Loving the clear skies we're having here in Kent right now. So, I've been getting into imaging and just about to start guiding and dithering but before I do, I have a question. 

I've had some success with stacking images of 90 seconds mostly shot at ISO 1600. The question I have is, if I shoot for longer exposures, say 2-3mins, will I need to reduce the ISO and if I do, will this reduce the amount detail? 

I'm using a Skywatcher ED72 on Heq5 pro goto mount, polar aligned with sharpcap and  Canon 60d dslr. 

Also I imaged M61 last night, obviously I'm not going to get a large image, but I have a 130pds on order and was wondering if the focal length would give me much more magnification. 

Thanks for your help 

Simon 

[richard ashbee]

Yes reduce the ISO which will reduce noise , detail will be better

[vlaiv]

I'm not sure about your particular model - but best ISO for most DSLRs is about 800 - 1600.

ISO is just numerical multiplier and as such it has no impact on SNR. It does have some impact on read noise and best ratio of read noise and pixel well depth tends to be around said ISO figure (each camera has sweet spot - so you might want to look it up for 60d).

Using longer exposure - you don't need to reduce ISO nor will it impact detail. If you have saturation areas (it happens with very bright nebulae like M42) - use a few short exposures to blend in saturated parts in post processing (or you can even stack them together in DSS - just use appropriate stacking method - I think entropy based average one deals with different exposure lengths efficiently).

 

[Simon Dunsmore]

Hi Vlaiv, 

Thanks for your help. I'm guessing that if the images are over exposed I can amend in the processing stage? 

[FaDG]

It depends on what you mean by "overexposed". 

Once you saturare the well depth, information is definitely lost for those pixels and no kind of post processing can get it back. So, take care to avoid saturation by reducing ISO (or exposure time) in order to better exploit the dynamic range. 

The other way to get around that is to compensate with shorter exposure and blend them into the saturated areas, but I don't find it a simple task, still to improve

Edited May 26, 2020 by FaDG

[vlaiv]

1 hour ago, Simon Dunsmore said:

Hi Vlaiv, 

Thanks for your help. I'm guessing that if the images are over exposed I can amend in the processing stage? 

Only if you also have short exposures of the same target - then you can replace saturated parts with short exposure stack.

If something is saturating in long exposure - that means signal is strong there and there will be enough signal in short exposure in those parts to get good SNR. You don't really need many of short exposures - get just a couple of minutes worth of short exposures - like 15x10s or similar to use for saturated areas.

I think that deep sky stacker can do that all for you if you use particular stacking method:

Quote

Entropy Weighted Average (High Dynamic Range)
This method is based on the work of German, Jenkin and Lesperance (see Entropy-Based image merging - 2005) and is used to stack the picture while keeping for each pixel the best dynamic.
It is particularly useful when stacking pictures taken with different exposure times and ISO speeds, and it creates an averaged picture with the best possible dynamic. To put it simply it avoids burning galaxies and nebula centers.
Note: this method is very CPU and memory intensive.

This was copied from DSS technical page on stacking algorithms found here: http://deepskystacker.free.fr/english/technical.htm#Stacking

Other than that, there are at least two other methods for combining short and long exposure subs.

1. Linear combination - requires software where you can do sort of pixel math - PixInsight has this capability and ImageJ as open source alternative. You need to take both stack while linear and aligned and then scale shorter stack to make it compatible (multiply it with ratio of exposure lengths). After that - just do pixel math - if long stack pixel value is larger than some threshold (like 95% - it could be saturated - don't go for 100% as calibration can mess up this if you have vignetting and such and make it smaller than 100%) just simply replace it with scaled pixel value from shorter stack

2. Non linear combination - this one is rather simple - again stack both and align and process each as you normally would - use layers in PS or Gimp to replace saturated parts of image with unsaturated version.