Darks, or not?

[russp]

Did a sequence a week or so ago and spent some time examining different options and unusually for me took darks too to see if my contention that at best they don't help with modern dlsr with dark current suppression technology and that at worse they're counter productive held true. Well here's the results from PixInsight - same stack one without and one with darks, both with an ABE applied and screenshotted with an auto stretch applied. Conclusion: my previous experiments held true ... I'm not doing darks any more!

[russp]

The final image is here 

 

[Adreneline]

A stunning image. May I ask which dslr you are using? I tried a few images with my son's new Sony A7 last night but it was just a quick trial. I would be over the moon if I could get this type of result.

Thank you for sharing.

Adrian

[MARS1960]

Very nice image, think i'll try a similar composition with 150-600 tonight if the skies stay clear. 

Personally i'm in your camp, i have only ever done flats and bias when using my DSLR.

[russp]

55 minutes ago, Adreneline said:

A stunning image. May I ask which dslr you are using? I tried a few images with my son's new Sony A7 last night but it was just a quick trial. I would be over the moon if I could get this type of result.

Thank you for sharing.

Adrian

Thanks - It's a Nikon D500, stack of 52 30sec exposures at f3.2 ISO800

 

[Droogie 2001]

Interesting outcome. I rarely use ABE as I find it creates too much of a dramatic change in brightness within the image (this is when using a CCD without Darks). I know some people swear by it but I am increasingly avoiding the process.
I always use DBE as this gives me more control. 

Not suggesting the outcome would be any different but maybe worth a quick test to see if you get similar results. Your example with ABE & Darks appears to be dramactically different.

If no darks help you then thats a bonus I'd say.

[russp]

9 hours ago, Droogie 2001 said:

Interesting outcome. I rarely use ABE as I find it creates too much of a dramatic change in brightness within the image (this is when using a CCD without Darks). I know some people swear by it but I am increasingly avoiding the process.
I always use DBE as this gives me more control. 

Not suggesting the outcome would be any different but maybe worth a quick test to see if you get similar results. Your example with ABE & Darks appears to be dramactically different.

If no darks help you then thats a bonus I'd say.

I usually use DBE but reading Warren Kellers book he suggests that for simple even gradients ABE can do an excellent job and shouldn't be discounted. With this image it was so good I didn't even bother going to DBE. The stacked files showed an even gradient for the without darks and a very uneven one for the with (the green splodges are just a manifestation of this). DBE might do a better job on the with darks stack but the point is the darks introduced a gradient that wasn't there before. As you say it's a bonus

 

[ollypenrice]

ABE, in my opinion, breaks down when a target like a galaxy emerges out of a dark sky. It fails to recognize outer parts of the galaxy as real signal and, therefore, pulls them down creating a dark ring around the target. It doesn't have to be a galaxy, a PN will suffer the same fate. I think it's far better to use DBE and accept that it will do a great job with 5 markers. (That's my personal record. Harry Page says it will work with even fewer. I haven't tried.)

Darks? A blunt instrument. Several years ago I stopped using them even with set point cooled CCD data on noisy Kodak chips. I don't use them on my Sony CCD chip either. In all cases I use bias-as-dark.

Olly