Opinions on two different processing results please.

[Brent]

For a couple of months now I've been flitting between 2 packages for processing my images and failing to settle on one.

I need to make a definitive decision I think, as I get different results with each - but I'm going round in circles trying to decide

It would be MUCH appreciated if fellow members of the forum would help me decide

So, one set of data, 2 images (please discount the lower resolution of the first if at all possible).

Which do you prefer, (and reasons why if you would care to say would also be much appreciated!)

Thanks.

Image 1

Image 2

[Stub Mandrel]

1 has better colour balance, 2 has a red cast right through, including the background. The better noise control in 2 might be simply because tit hasn't been stretched as much.

If they were mine I would use 1 as an RGB layer and desaturate 2 and use it as a Luminance layer and get the best of both worlds by combining the two :-)

[happy-kat]

I am a beginner with a camera and lens.

On opening image 1 and zooming in there is I think a blotchy background though I prefer the stars.

Image 2 is more precise but somehow has less vibrancy.

Why not process the image both times again to see if they come out the same.

[Brent]

1 has better colour balance, 2 has a red cast right through, including the background. The better noise control in 2 might be simply because tit hasn't been stretched as much.

If they were mine I would use 1 as an RGB layer and desaturate 2 and use it as a Luminance layer and get the best of both worlds by combining the two :-)

Thanks Neil.

You're right. I just took image 2 into PS and had a look, can't believe I got one of the basics so wrong, and I keep doing it! 

I think I'm going to have to stop processing when I'm tired, lol.

The layers approach will be an interesting experiment!

I am a beginner with a camera and lens.

On opening image 1 and zooming in there is I think a blotchy background though I prefer the stars.

Image 2 is more precise but somehow has less vibrancy.

Why not process the image both times again to see if they come out the same.

Thanks happy-kat.

That's the way forward I think. I'm going to wait 'til tomorrow when I can start from scratch with fresh eyes

[Brent]

Just had a quick look at the colour balance in PS, hopefully it's a lot better now. The image is the same except for that and a little levels and saturation tweaking.

So, below, (again!  ), are the 2 images I'd like opinions on please.

Thanks.

Image 1

Image 2

[laudropb]

Much prefer the first.

[Brent]

Thanks laudropb.

Just to be clear (as I've messed the thread up now!), are you talking about from the set of images in the first post, or the revised set just above your post?

[happy-kat]

Surely image 1 is the same both times?

Scrub that i see what you are asking.

[Garethr]

+1 for the 1st image

2nd one looks clipped now 

Gareth  

[Brent]

+1 for the 1st image

2nd one looks clipped now 

Gareth  

Thanks Gareth.

You've reminded me, I meant to say when I posted it that the 2nd one is very close to being clipped now, but it isn't quite 

[happy-kat]

Did you start the processing again from scratch?

The first image I still prefer the number of stars but the background is still very blotchy.

[ollypenrice]

The first one has an admirably colour-neutral background sky but has invented (I think!) irregular patches of background brightness.

The second one has an overall red bias. The second one is also slightly clipped but has smaller stars.

I don't believe that the strengths and weaknesses of each are inherent to the programmes, whatever they might be.

Olly

[Brent]

Did you start the processing again from scratch?

The first image I still prefer the number of stars but the background is still very blotchy.

No, I didn't have time but I do need to.

The first image was stretched with the stars, for the second image I masked the nebula and just stretched that.

[Brent]

The first one has an admirably colour-neutral background sky but has invented (I think!) irregular patches of background brightness.

The second one has an overall red bias. The second one is also slightly clipped but has smaller stars.

I don't believe that the strengths and weaknesses of each are inherent to the programmes, whatever they might be.

Olly

Thanks Olly.

I think you're correct in that the weaknesses are all down to my (lack of!) skill

The first image was processed in StarTools with some extra NR in PS which I think caused the blotchiness).

The second image was processed purely in Photoshop.

Do you mind if I ask a question with regards to this specific circumstance? Should I stretch the stars/background at all if there is no nebulosity present?

[ollypenrice]

Thanks Olly.

I think you're correct in that the weaknesses are all down to my (lack of!) skill

The first image was processed in StarTools with some extra NR in PS which I think caused the blotchiness).

The second image was processed purely in Photoshop.

Do you mind if I ask a question with regards to this specific circumstance? Should I stretch the stars/background at all if there is no nebulosity present?

Awe, I didn't say that!!!    When anyone re-does a processing job from scratch they'll get a different result. I see this regularly because I often do demo processing sessions when giving tutorials and I never get an identical result however often I use the same raw data.

With regard to your background level, I'd aim to get it to about 26 or 28 per channel at the initial stretch. (Meausre it in Ps with the colour sampler eyedropper at several points.) After that I wouldn't stretch it harder. You can pin a curve at and below the background level and stretch above that to avoid dragging your background above the noise floor. 28 is a bit bright, yes, but some processing routines eat into the black point and that 28 can get nibbled down remarkably quickly. When it's gone it's gone and you can only have it back by restarting. The very last thing I do in a process is bring the background down to 23/23/23.

(I did this quick bit of fun graphics while helping a friend with one of his first attempts... He didn't like school so I gave it the 'red ink' look as a wind-up.)

Olly

[Brent]

Awe, I didn't say that!!!    When anyone re-does a processing job from scratch they'll get a different result. I see this regularly because I often do demo processing sessions when giving tutorials and I never get an identical result however often I use the same raw data.

With regard to your background level, I'd aim to get it to about 26 or 28 per channel at the initial stretch. (Meausre it in Ps with the colour sampler eyedropper at several points.) After that I wouldn't stretch it harder. You can pin a curve at and below the background level and stretch above that to avoid dragging your background above the noise floor. 28 is a bit bright, yes, but some processing routines eat into the black point and that 28 can get nibbled down remarkably quickly. When it's gone it's gone and you can only have it back by restarting. The very last thing I do in a process is bring the background down to 23/23/23.

(I did this quick bit of fun graphics while helping a friend with one of his first attempts... He didn't like school so I gave it the 'red ink' look as a wind-up.)

clipped histo.JPG

Olly

Thanks for that Olly, I really appreciate it. I love the illustration as well, I've found my new desktop wallpaper!     

I was thinking along the right lines then, it's just the execution I need to work on. (Just! )

I hear the other half of the battle to making the processing easier is getting enough data - this was just an hours worth due to a short 'window' of opportunity. First job this spring is to cut my tree down to size.

[ollypenrice]

If I had to offer only two things I've learned over the last ten years they would be these.

1 Learn to look at your image to see what's right and wrong with it.

2 Learn your way around the processing tools in your graphics programme(s).

Olly