Am I playing it safe with My ISO in Dslr imaging?

[VikN46]

Hi All,

A quick question, if I may, when imaging using a DSLR. When I image using BYEOS at ISO 800 the image seems washed out and light so I tend to drop to ISO 400 but then not get the luminance and dataneeded. Should I go for the ISO 800 and not be put off by the lightness of the frames? Below is last nights image of NGC 6960 Western Veil at ISO 400. 10 x 300s LIGHTS, 5 x 300s Darks at 400ISO found it difficult to process, guiding I think was ok. Thoughts please

NGC 6960 - Western Veil by VikN46, on Flickr

Guiding graph:

20140721_231416 by VikN46, on Flickr

Thank you for any guidance

[Uplooker]

I would stick to ISO800 as this generally maximises the signal to noise ratio.

I am sure others will be along shortly to give a more detailed explanation of the science behind it.

Ian

Sent from ma fone using Tapatalk

[VikN46]

Thanks, I have imaged at ISO 800 with results but when I see twhe light image at ISO 800 I tend to back off and knock it down to 400.

[Alien 13]

It can depend on the camera but i find an ISO of 800 or1600 works fine for me even with the histogram to the far right and i do find it far easier to process from a bright stacked image than trying to pull data fron a dark one but thats my personal view.

Some of my subs are very bright but they still work i have gone much brighter than this one.

Alan

[ronin]

If you are not getting the data at 400 then try 800.

[Pete Presland]

i would say 800 would be fine, i would not be concerned too much about a washed out look.my last Rosetta was washed out but nothing that i could not sort in processing.

your image does look very noisy though, there looks plenty of data in the image, perhaps you have been a little harsh with your processing?

[Leveye]

Depending on the camera's sensor even 1200-1600 will work. I tend to go with ISO 800 for DSO's of this nature like the others here suggest. Your camera has a lot of  color noise but with DSS using darks and bias frames and then some tweaking in PS you will be good. Your going to want to use a gradient exterminator filter on it as well from the looks of the right hand side of your image.

[nmoushon]

From my personal tests with my DSLR I found that the washing out isnt something to be too concerned about....BUT it depends if the washing out is from just lots of data from a dark site or your local LP being picked up. If its the latter then you'll have problems and you will need to make some adjustments.

Now what I learned from my personal tests is that ISO does not plays as big a factor in the end when it comes to data collection for long exposure imaging. ISO is a setting that is only on DSLRs and in a very basic description all it does is pre-processes the image to pull out more data. This allows the user to see results better on the image that pops up on the LCD screen and is brighter and easier to see. From what I've found is that you can not pull out any more data from an image shot at ISO800 vs ISO 1600. That amount of data is the same in both cases and can be brought out equally the same in post processing. Now the biggest factor I found that ISO played into was that of noise. Higher ISOs will introduce more noise than lower ISOs because it increase the heat on the chip more. Now this can be taken care of with darks of course but on to an extent. It gets very hard because it is very hard to regulate the temp of the camera and match each light frame with a perfectly matching dark frame. Now its hard to get a perfectly matching dark frame that why we take multiple and stack them together to get an average. Though you can control a good portion of the noise in the end with darks and processing you will always reach a limit as to what noise you can remove and what you can't. Where I found the most benefit of keeping the ISO lower was that it allowed me to take longer exposures before both noise increased to much and my local LP creeping in.

[VikN46]

Thanks all, looks like it is clear again tonight so will try again using ISO 800, and see what happens my NGC 6888 was fine with ISO 800 so perhaps just too cautious. All a learning curve, perseverance is key as is working out what is best living just 5 mins from Leicester City Centre.

[Owmuchonomy]

Now then. ISO800 should be ok for this DSO. Your main enemy is S/N ratio. 10 subs is not sufficient in my experience. I aim for a minimum of 60 lights (preferably more) 20 darks, 20 flats and 20 bias. Your first image also appears to have a gradient L to R so you could try GTX if you haven't done so already. If you could get more light frames it may help your S/N but at this time of year that means several nights of data. Clear skies!

[laser_jock99]

Don't immediately assume a washed out looking sub is devoid of useful data. If you are shooting RAW then there is potentially a lot of information there still if it is processed with care.

This washed out looking 1200 second sub straight from the camera.....

....was actually hiding this image

A lot depends on how much your local light pollution swamps the faint signal- but it's worth experimenting to see where your site limits are.

[Astrosurf]

Ditto above. Do try and get a clip-in light pollution filter. They're brilliant. Astronomik do some for the EOS range, but I don't know if they do one for yours.

Alexxx

[lensman57]

Hi,

I don't know what you mean by washed out? If it is the case that your sky is so light polluted that your images show little detail then the only solution as well using a first class filter is to reduce the sub lenghts and increase the number of the subs in the inverse square relation, 1/2 the exposure length 4 times the number of the subs. The other factor is your sky back ground  value and even for CCD imaging ( NB filters excluded ) there is an optimum exposure length beyond which no increase in the signal is recorded, the darker the site and the sky the longer the sub length can be, for my sky and an OSC CCD this is about 1200s per sub at best, for DSLR I keep to 600s max @ iso 800 or 1600 depending.

A.G

[wxsatuser]

There is really only one way to get the correct exposure with a dslr, that suits your sky conditions, at that particular time.

Correct exposure is when the in camera histogram mountain peak is at least at the 25% mark but no more than 40%.

I always aim for the peak at 25% then I know that all my exposures are the same and correct for the sky conditions.

Does'nt matter what iso you use but get that peak at around 25%.

Some imagers are advocating shooting to the right, as laserjock shows you can get an image but you will lose the detail.

Leave shooting to the right for your daylight pics it's not a good idea for astrophotos.

For faint objects with the latest dslrs I would crank the iso to 1600, for bright objects where you want star colour knock it

down to around 400, you may be ok at 800 as well but bring the histogram back a bit to around 15>20%.

There will always be the brighter stars saturated to white so don't get to worried about this.

Whatever iso you use the total exposure is what counts.

For most dso aim for a minimum of an hour and upto 4hours, the longer the better.

[frugal]

Remember, all the ISO setting on a DSLR does is provide a multiplication factor to the signal read off of the chip. Typically iso400 is a multiplier of about 1. Anything higher than that is a multiplier of greater than 1, and any iso lower than 400 gives a multiplier of less than one. It is not like film where there were emulsions with physically different properties for the different isos. Really, anything above ISO 400 is going to give you the same data, just scaled up.

You would be better off sticking to ISO 400 and extending the exposure time