Orion wide-field w/ Canon 28-300mm L

mtbdudex wrote in post #11467459
I'm trying to hone my PP skills, learn and grasp "overclipped" here.

When I look at both picts the RGB values in "black" are not zero, they are 8-13 over what appears just black space, meaning not true black (0,0,0) therefore some data still there, correct?

HTTP response: 404 | MIME changed to 'image/png' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'image/png' | Byte size: ZERO

There is plenty there, adjustments need to be made carefully so not to lower the levels as shown. Keeping the histogram(3 channel) palette up while stretching the file in PS will help with this as well as monitoring color balance. The numbers should read close to or at R=25, G=25, B=25.

Nighthound: I use DSS all the time. I don't think it is clipping the dark points. It's too easy to clip away unintentionally in PP, especially when you're trying to over compensate for noise and or LP. I'm guilty of the same.

mtbdudex wrote in post #11467459
I'm trying to hone my PP skills, learn and grasp "overclipped" here.

When I look at both picts the RGB values in "black" are not zero, they are 8-13 over what appears just black space, meaning not true black (0,0,0) therefore some data still there, correct?

That shows there is some image color left in there, at least on the pixel you have selected.
Take a look at the histogram, especially as you're processing.

I hope you don't mind, According to your profile you allow image editing. Unfortunately there isn't much you can do with a jpeg that has already been cooked. What original image data that was there is no longer there, at least not like it was.

But here is a copy of your first post that I edited. It's exaggerated and probably too bright for some peoples taste, and, the quality sucks because I had to reduce it so much to get it down to 150kb or less to attach to this message.
Nevertheless, I think it demonstrates where the image data was clipped revealing where some of the Dust and nebulosity is throughout that whole area. Such as the dark dust and nebulosity that can be seen below the Orion and Running Man Nebulae, as well as the blue reflection nebula at the very top above the running man that was completely clipped away.
I can only imagine what could be done with the unedited stack.

HOSTED PHOTO
please log in to view hosted photos in full size.

WOW

tkerr wrote in post #11470502
Nighthound: I use DSS all the time. I don't think it is clipping the dark points. It's too easy to clip away unintentionally in PP, especially when you're trying to over compensate for noise and or LP. I'm guilty of the same.

Great demonstration Tim, there's a lot there for sure. Oh, and after all these years please call me Steve.

Wouldn't that indicate that it's not black clipped, but that the histogram has the potential to be stretched more aggressively ?
If I take the black point and set it higher (eg. 25), the histogram looks better (less black clipping ?) but there's no more information showing in the image. However, if I take the middle levels slider and move it towards the black point, I get more information out of the image. (It really needs to be done in curves, but levels is easier to describe).
It seems to me, that to describe this as being black clipped doesn't describe the situation at all. Are we just short of phrases in our repertoire ?

Steve, while some data can be revealed by what you've described considerable data was lost in the file posted when the black point was clipped. You'll notice that the left edge of the black point peak can only be adjusted to show a flat left side. It should appear a bit more like the right side with less of a sudden drop. It's that more gradual slope that's lost in the clip and that's the faint stuff that has to be protected throughout the stretching. I think that's probably the most difficult part of this is working slowly enough to watch all the details. I used to rush to get to the finished image but over the years I realized slow and steady was best. I've spent as long as 24 hours on one stretch. I like to stop and start through the final process since seeing the file with fresh eyes along the way seems to help.

Perhaps the term "black points" is not the best description and may be a little misleading because you're not clipping the color black, but rather clipping the color data making the color or colors darker(blacker).

This might help a little.
http://www.astropix.co​m …Set%20the%20Bla​ck%20Point
Note that he is not pushing the slider all the way snug against the graph indicating where the image data is.

This one might help as well.
http://astrochat.co.uk​/forum/viewtopic.php?t​=13241

That's why I ask. These are the histograms from the original post and from Tims adjustments. I see a gap and a slope. What am I missing ?

Steve.

HOSTED PHOTO
please log in to view hosted photos in full size.

HOSTED PHOTO
please log in to view hosted photos in full size.

SteveInNZ wrote in post #11471200
That's why I ask. These are the histograms from the original post and from Tims adjustments. I see a gap and a slope. What am I missing ?

Steve.

Those histograms, especially the one from my edit, are misleading because they are only representing the image data after all the adjustments, editing and conversion to jpeg. Anything that may have been in their prior to the edits was destroyed.
In Photoshop look at the image "Info" dialog panel. Point to a dark point on the image and read the color values. Like Steve said above they should be close to 25-25-25. Obviously I couldn't accomplish that working with a pre-edited jpeg, and the original is showing a values of 10-8-9.

I would like to see the original stacked image data before any adjustments or conversion to jpeg. Because from what I can tell, he had a lot of good data there.

16-bit raw stack (TIF) on Megaupload: http://www.megaupload.​com/?d=AJXEPZ0U

I think your images are extraordinary . Good colour and detail.

Wow! Nice job! Here's what you did well IMHO:
- almost no noise (some noise is normal, and you nailed it)...this is the biggest challenge for DSLR astrophotography, and you conquered it
- color saturation looks great
- nebula structure in both emission and reflection nebulae look fantastic
- shooting at f/8 was the right thing to do since f/5.6 is the fastest f/stop for that lens
- guiding was well done since the stars are nice and round

Area for improvement:
- background is too black

Congrats on a spectacular image.

I'll play around with the TIF. =)

-Phil

Played with the raw tif in PS CS5. Found that some of the deeper data cannot be separated from a gradient.

You've got a lot of good clean image data hidden away in there with great SNR. I can only imagine what someone like Noel Carboni with much great PS skills than I could do with this.
I could spend all day playing with this one.
This is what I got with only spending a short time on it, and I didn't even do any kind of NR. Just kept playing around with levels and curves and then a little sharpening.

PHOTOBUCKET EMBEDDING IS DISABLED BY THIS MEMBER. Photobucket sends ads instead of embedding photos from their free galleries. Click the link (if available) below to see the image in a gallery page. http://i68.photobucket​.com …nHH-full-01-tk-edited.jpg

PHOTOBUCKET EMBEDDING IS DISABLED BY THIS MEMBER. Photobucket sends ads instead of embedding photos from their free galleries. Click the link (if available) below to see the image in a gallery page. http://i68.photobucket​.com …HH-full-01-tk-edited2.jpg

fantastic...Nice.