How many hours integration for a decent veil?

[irtuk]

I appear to be blessed with another clear night, 3 in a row, and I am debating whether to gather more veil data or move onto a new target. I am up to 8 hours or thereabouts on narrowband and I have these two images, slightly different processing, not entirely sure which I prefer, to be honest I quite like both.

Anyhow, if I rack up another four or five hours do you think it would be worth the time investment or am I chasing diminishing returns?

Would you add some RGB subs in? more narrowband data or call it a day and move on to a new target?

I am also debating a multi-panel job using telescopius to work out better coordinates and get more of the region, but that's a different question.

Ed.

 

 

 

 

 

[Clarkey]

Adding more data is always diminishing returns - only you can decide the cut-off point. Personally, I rarely do less than about 8 hours per image and often more - especially for NB. You could add some RGB for the stars which will help with the overall 'feel' of the target.

Edited October 13, 2022 by Clarkey
Missed number of hours. D'oh!

[Bluemoonjim]

 

I would recommend having a look at www.urbanastrophotography.com.
Lee, who created the website, has some clear thoughts on hours of data to collect.

Edited October 14, 2022 by Bluemoonjim

[tomato]

A lot depends on what equipment you are using and the quality of the sky at your location. The very best images posted on here rarely have <5 hrs integration and usually more like twice that amount. If you take more than that required to give you an optimum result you can then be discerning about the quality of the subs to use w.r.t focus, guiding, high cloud, moon interference etc.

Having said all that, probably 95% of the images I post have insufficient integration time...🙄

[tooth_dr]

I find 8 hours is the right amount for an acceptable image at ~F5.  If I image at F2.8 then 4 hours seems to be sufficient for an ok image.

[ollypenrice]

It would be illogical to collect RGB for stars if the narrowband pallet to which you added them were not a close approximation of RGB. What I would do if wanting to create a natural (ish) colour image with RGB stars would be shoot a quick RGB for the stars and then add deep Ha to its red channel and deep OIII to its green and blue channels, all in blend mode lighten. This would be, in essence, an RGB-HOO image. I wouldn't add RGB stars to the colour pallet in the images you posted because they are not remotely similar to RGB.

Also, if you do shoot more data (and I would) be careful not to throw it away in processing. Your first image has had its faint signal discarded by black clipping (ie bringing in the black point too far.)

From a very dark site at F5 I generally shoot 10 to 20 hours per image.

Olly

[vlaiv]

1 hour ago, ollypenrice said:

From a very dark site at F5 I generally shoot 10 to 20 hours per image.

I thought we are past F/ratio as indicator of speed :D

What is that in aperture @ resolution units?

[ONIKKINEN]

I dont think i have taken a single shot with enough integration to be fully happy yet. I hope to remedy that now and have some long integrations in the works, but who knows how many years it takes. Im aiming for at least 20 hours from decent skies with a couple broadband targets and if that's not enough ill throw everything off a cliff and pretend i took a shot i was happy with 👍.

[ollypenrice]

10 hours ago, vlaiv said:

I thought we are past F/ratio as indicator of speed :D

What is that in aperture @ resolution units?

I knew you'd say that! 🤣

We really do need to popularize the aperture@resolution unit somehow.

Olly