Post processing: how much is too much?

[Eric Hughes]

Ok so you guys know I'm new at this. When I first process my photos, I try to squeeze out as much color and detail as possible. Then, I'll go back the next day and it looks totally over-processed. Sometimes almost cartoonish!

So a couple of my questions are these:

How do you know when to stop with the processing?
How do I know when I have achieved accurate colors of what I'm shooting?

I'm currently only using my camera's software to process, which seems to work good, but photoshop is in my future. I know the power of photoshop to create something that isn't really there, so is there some "gentleman's agreement" in processing? You guys know enough about this to know if someone is fudging on a photo from the scope the say they use, is this why you guys put the scope/exposure stats on your pics, so in case someone tries to call you out? OK I know that last question is goofy, I just wanna know, lol

So lemme know what you guys think...

[Eric Hughes]

Nice! I knew I could count on you guys. These are all possibilities I had considered when I asked these questions, I kinda figured it was a coin flip between art and science. As for me, I just wanna take rad pics!! Something I've wanted to do since Mom got me Cosmos for my 8th birthday. So, if there are clear skies tonight, I shall go once more into the breach, dear friends… :)

PS: we're gonna need a bigger boat…

[Greg Marshall]

What? Someone makes money at this?!?!? It must be from teaching others how to do it. Sure, there are people who sell AP prints on a regular basis - I'm one of them. But I'd have to stop buying stuff and spend the rest of my life selling the images I already have to even dream of breaking even!

Anyway, Marc, thanks for sharing those threads. It underscores the idea that one person's "stunning" is another person's "eyesore". I try to appeal to the largest possible number of people and find that most like a bit more saturation and contrast than is really called for, but over-the-top processing definitely does not sell.

[Marc Basti]

If you want to see good examples of over-processing check out my website. One of these days I'll figure it out, but I'm having fun doing it and that's what is important to me. What are there about 5 people who make money at AP and for the rest of us it's just a hobby that costs us a lot of money. It's still amazing to me that I can sit here on Earth and take a pic of something x-millions of miles away and it looks like "something". I digressed from the point I came on to make-there was a good thread(s) going in CN about processing/presentation, just a little food for thought. The first one puts the 2nd in context
http://www.cloudynights.com/ubbthreads/showflat.php/Cat/0/Number/55...
http://www.cloudynights.com/ubbthreads/showflat.php/Cat/0/Number/55...
On posting the pic stats, I find those very useful. When you look around you'll see there are a number of people who have the same (or pretty similar) setups as your own, so instead of reinventing the wheel check out their stats for a planned target. Can I stuff m31 into the fov or which way should I rotate the camera to fit m45 or he used 2mn exposures but, said he should have gone 5mn etc. Saves a lot of trial and error.
Now I got to get back to putting in another 6hrs of processing on NGC281 lol. Marc

[Russ Ruggles]

Is your head buzzing yet Eric??

I told ya... more questions.

[Greg Marshall]

I find that often the best time to stop processing is an hour ago - or yesterday. That is, when you notice that it looks over-processed, you want to be able to go back. So the important thing is to save intermediate work.

Although I try to make colors more-or-less correct, I'm more interested in making them pleasing and therefore dislike the Hubble palette. Using H-alpha, S-2, and O-3 filters allows me to capture some kinds of objects from my home observatory - even when there is a fair amount of Moon. However, H-alpha and S-2 are very nearly the same wavelength (both are deep red), so adding them together to form the red channel doesn't make much sense. More importantly, with just H-alpha and O-3 (which is blue-green) you can get accurate color, but it lacks richness because everything lies somewhere on the straight line between red and blue-green. So I use the S-2 data to provide a third vertex to the color space, thus trading color accuracy for color richness. I also sometimes shift the hue a bit because it's a good and easy way to add clarity to the differences in neighboring regions.

As has been pointed out already, imaging for scientific research is different and there are ways to "calibrate" the color balance (at least to an accepted standard). But for pretty pictures, I think any kind of color manipulation is valid. A little trickier is manipulating spatial relationships. Most astronomy types frown on that. Of course, many people insist on a consistent "north is up" orientation, while I am a strong believer in rotating the camera for optimal composition.

Virtually ALL astrophotography involves wild manipulation of the dynamic range, including wildly non-linear curves and, sometimes, regional non-uniformity. That is, the curve is different in some regions of the image to enhance details. I find the extremes of this kind of processing unattractive. The classic example of this is an image of the Orion Nebula in which every part shows detail, despite ridiculously large variations in the true brightness. A not-too-extreme example can be found here: http://apod.nasa.gov/apod/ap061120.html. But moderate levels of such manipulation can be quite pleasing. My own Orion image simply combines a shorter exposure of the core so that it is not just a big white blob: http://wa-chur-ed.com/OriNeb.html.

I've also done at least one image that is, by conventional metrics, totally "unethical": http://wa-chur-ed.com/VenTran.html. This image of the Venus transit last year combines 2 completely separate exposures to provide a terrestrial reference and balance the composition. I tell people that this is what it would look like if our eyes could handle a billion-to-one dynamic range and there's some truth in that. Except that I also used different magnifications for the 2 shots. Just between you and me...

[Russ Ruggles]

Well Eric you've come to the proverbial crossroads of questioning your own methods and work. This my friend can only lead to more questions and no "real" answers.
First and foremost - why are you doing this? For science or because you like to take and display pretty pictures?
The science is as deep and you want to go and as broad as the universe. For me I choose 'pretty pictures' however many of mine are not so pretty. It's real easy to compare your own work against others here and elsewhere. The thing or things to keep in mind is if you choose the pretty picture path then it's all subjective. The extent of process, the colors, the target, the amount of total imaging time, the method, the gear. Everything is subjective and nothing is the same.

Like Steve mentioned, it's all about one's preference. If YOU feel it's overdone, then it's overdone. If YOU don't like the colors then change them. Compare DSLR to CCD LRGB to CCD L,H-a,S-II,O-III to the Hubble Pallet of any image you can goggle. The colors seem endless and the choice is the option of the artist. Yes, it is art.

Now as far as Photoshop goes. If you own a copy - great! If you can get a copy cheap - great! If you're paying full price.. then not so great because you'll need to buy some "actions" that make PS more Astro friendly. There may be better options. More cost effect like Images Plus or Pixinsight. Both very complete astro imaging processing tools that are considerably cheaper.
The other thing that will "add" color to you images and make processing easier is taking longer sub-frames and more of them. That means some type of autoguiding or a very expensive quality mount. Taking Flats and Bias frames tend to clean up the image as well. LP filters, better scope, better camera..... ahhhhhhhhhhhhhhhhhh IT'S ENDLESS!!! Well you get my drift.
It is what you make of it. In the endless struggle to get better at this don't forget to enjoy it and have fun.

[Eric Hughes]

Yes! Excellent.

[Steve Coates]

Here is another article that may help you understand colors in astronomical targets.

[Steve Coates]

I am so guilty of posting early. Once I do my stacking and post-processing I post the image and over the coarse of a week I tweak my image probably 4-5 times. I do this as I look at how others have processed their image. As time goes on you develop a style that you like. I personally like the "natural" look, not heavy on saturation. That is my personal taste. I do this for aesthetics and not science. This article may help you understand the "gentleman's agreement" in processing.
As for the last question, I post how I got my images as a way of letting others know how I got to get this image. Some may appreciate the added info, some may think I am going at this all wrong. In the end, if you are happy capturing photons from several to million of light years away and processing them your way and you enjoy it, as long as it is your work you are claiming, no one will give you grief.