How dark is really the dark DSO background?

Hello,

My two cents…

I would deflect this discussion a bit. What I do not unterstand about the background “blackness” discussion is:

Gradients, mist, clouds… every thing what we consider “noise” resp. “light pollution” from the sky, we try to filter out and remove it in post processing. BUT: Sky glow we keep and say “oh well, the background of DSO images should not be pitch black”. To be honest, I have never understood this argument, because space is physically pitch black! I know, some photons roam around due to straying, but they are actually neglectable). Everything what emits photons, we try to image: Nebula, galaxies, stars… but why suddenly empty space emits photons?

The only argument is pure esthetics, since pitch black background looks strange. And hence, the OP’s question can be answered in this way from my point of few: How you like it!

CS
Rüdiger

andrea tasselli · Dec 30, 2025, 10:31 PM

Deep sky background isn't "pitch black", as seen from Earth hence the value given above. End of. Yet it is far darker than the darkest skies we can hold of onto, hence in that context it is indeed the blackest black but this is pure fiction since we can't image away from the Earth. As the answer to the OP then it's as you like it. RGB has no meaning to me as there is no way I can either quantify it or measure it in any objective way.

If you refer to me, you missed my point. I asked why one light pollution is removed, and the other pollution not. Is there an accepted and one not accepted pollution? I have a different opinion on this topic to yours. That’s all.

Hello Rainer,

my advice is purely pragmatic: Don’t make it pure black and not so bright that faint structures disappear. Do not obsess about it. It is commonly accepted that the background is “some how” dark grey and this fine.

I only have an issue with this pseudo scientific argumentation. Just to push it to top: Is a moon gradient in the image less natural than a sky glow? A non black background can only be argued based on esthetic aspects, but not science, as long as you not create a class “less natural” and “more natural” light pollution. 🤨

To bring to the point: There is no formula to calculate “the” correct background values. At least I do not know one. But I am pretty sure, If there would be one: there would be an script or process in PI easily calculating it for each and every image. Since there is none… guess why…

As said: keep it not black and take care that tiny things are still visible. The rest is personal test and the judgment of of the individual viewer of your image. 20 viewer, at least 20 opinions. 🤣

And the most important: You have to like your own image! Dare to have your own taste!
(except you want to go for batches. In this case you have to follow other’s taste).

For my personal taste, your image looks perfect background wise. But this is of no importance - you have to like it. 😉

PS: Many people work with un-calibrated displays or even mobiles. There the black point of the displays differ so much, that any discussion about correct background is completely wasted time.

Ruediger · Dec 30, 2025, 08:43 PM

To be honest, I have never understood this argument, because space is physically pitch black! I know, some photons roam around due to straying, but they are actually neglectable). Everything what emits photons, we try to image: Nebula, galaxies, stars… but why suddenly empty space emits photons?

Ruediger,

The empty sky is actually not black as in the absence of photons. First, the black body afterglow from the big bang is red-shifted all the way down to about 2.7K and that radiation is remarkably uniform to about 1 part in 25,000 over the entire sky. The Gegenschein, which is backscattered light from interplanetary dust in the solar system is bright enough to see with the naked eye under a very dark sky. It’s pretty directional but regardless, it’s a region that stand out as brighter than the surrounding sky. Olbers' paradox argues that in a homogeneous, transparent space filled with stars, the number of stars grows as the square of the distance to perfectly cancel the 1/r² radiometric light fall off with distance so the sky should be uniformly as bright as the sun. The problem is resolved by realizing that the universe is not homogeneous, it is full of absorbing material, and it is not infinite. However, space is full of emitters and light is scattered from interstellar dust. From space, the empty sky is indeed very black and much darker that what we see under the atmosphere, but it is certainly not absolutely black.

As for processing, we generally don’t want to present the sky as absolutely black to avoid a couple of problems. First, stellar profiles look unnatural when the low-end part of the Moffat brightness profile gets truncated. Second, when we display faint extended objects, the regions with the lowest brightness may sit only a couple of ADU above zero. When you make the background too black, a lot of that low brightness stuff become invisible and appears to be truncated, which looks terrible. Setting the background level a little bit above zero fixes this problems and if it’s done right, it still provides the appearance of an inky-black background.

John

Do not take it too serious…

Deducing from John’s answer we should then make our background with a subtle but very subtle touch in Red

🤣

We managed to get another year done and let us see what does us bring 2026, hopefully 365.25 days of clear night skies. Let it rain during the day but not at night 🥴

Wish you all a Happy and Prosperous Year 2026