DPChallenge: A Digital Photography Contest You are not logged in. (log in or register) 
 

DPChallenge Forums >> Tips, Tricks, and Q&A >> The order of processing
Pages:  
Showing posts 1 - 17 of 17, (reverse)
AuthorThread
 12/22/2010 09:50:22 PM · #1
Does it make a difference what you do first? Denoise then adjust tones, levels, colors, etc. Or do all the adjustments first, then denoise? And in those adjustments, is there a preferred order, or is that just personal preference.

Thanks.
 12/22/2010 10:02:18 PM · #2
I normally let the image itself determine the order of processing steps. It depends on what I want to accomplish, and how the image lends itself to my plan..I shoot everything in Raw, so my first step is to tweek exposure and white balance if needed. After that it's anyone's guess.
 12/22/2010 10:07:08 PM · #3
The philosophy in general I follow is big moves first followed by small moves. No use making a fine adjustment on a saturation when you next jack the contrast by +30. And of course most people would sharpen very last, although my latest CS5 book actually shirks that conventional wisdom and says that sharpening out of RAW is beneficial (followed by a smaller sharpening at the end).

Message edited by author 2010-12-22 22:07:46.
 12/22/2010 10:07:56 PM · #4
IF you are denoising, then that should be your first step. Much of what you do in processing has the effect of increasing noise somewhere in the image, so it's best to start out with as smooth an image as possible.

That's my approach, anyway... But I don't have much need of denoising, with the 5D. Now that I am P&Sing, though, I do it more often, and always first. Topaz really accentuates noise, and I do a lot of Topaz.

R.
 12/22/2010 10:12:12 PM · #5
My workflow is similar to Roberts. I prefer do noise reduction early on, as other adjustments will work better on a cleaner image.

I usually do levels first, if needed. Once that is adjusted, I can see how much noise reduction is really needed, so NR is the second step. From there it varies by the image and what I want to do with it.

Message edited by author 2010-12-22 22:14:09.
 12/22/2010 10:38:15 PM · #6
1. Set workflow options and make general adjustments in Camara Raw including light to medium noise reduction
2. Levels (may emphasize sensor dust and reveal additional areas of excessive noise)
3. Heal sensor dust, if necessary
4. Run NIK Dfine in manual mode to finalize noise reduction and/or address banding issues
5. Make various global adjustments
6. Local adjustments (often followed by additional global adjustments, toning, that sort of thing)
7. (Flatten and) convert to 8 bit
8. Check destination space (Convert to Profile)
8. Sharpening, if appropriate

12/23/2010 12:58:43 AM · #7
This is fascinating. Thank you all.

I usually left denoising until the end, but have found that I'm getting better results doing that first and adjusting off that. Then if necessary, running again after whatever adjustments the image required. I tried sharpening in RAW and I haven't quite gotten the hang of it. I've recently tried and seem to get better results with the multi-step image size reduction version (Adams?).

ZZ, you bring up another question, which is "flattening". What is that. How is that different from merging the layers, which I understand is to take several different layers and combined them into one.
 12/23/2010 01:03:10 AM · #8
Originally posted by tanguera:
I've recently tried and seem to get better results with the multi-step image size reduction version (Adams?).


There lies the path to madness...

Well, at least it's a bad habit because it may look a web shot look nice, but it's not going to be a good way to get a print looking nice and then one day you are going to want to print a DPC entry and it will be infuriating to try to reproduce your results...

And it's Adamus. A local...

Message edited by author 2010-12-23 01:03:29.
 12/23/2010 01:20:24 AM · #9
Originally posted by tanguera:
ZZ, you bring up another question, which is "flattening". What is that. How is that different from merging the layers, which I understand is to take several different layers and combined them into one.

Same thing.
 12/23/2010 01:37:07 AM · #10
Originally posted by DrAchoo:
Originally posted by tanguera:
I've recently tried and seem to get better results with the multi-step image size reduction version (Adams?).


There lies the path to madness...

Well, at least it's a bad habit because it may look a web shot look nice, but it's not going to be a good way to get a print looking nice and then one day you are going to want to print a DPC entry and it will be infuriating to try to reproduce your results...

And it's Adamus. A local...


I only use it for the challenge entries, even save it as a separate document. It seems to work better with some images than others, for some reason. I'm still experimenting all over the place and don't have any one reliable way of doing anything. My work flow is convert from RAW, and then all bets are off. Have only been using PS for a few months, and most of it is still a mystery to me. In fact, am rather shocked I can produce anything of quality at all, LOL!

Thanks Steve.
 12/23/2010 01:40:08 AM · #11
I use a very simple order for almost all pictures.

1) If I didn't frame the picture correctly when I took it, then crop it
2) Resize the image to the desired size
3) Save the file as YYYY-MM-DD-IIII.jpg

12/23/2010 01:54:40 AM · #12
Originally posted by zeuszen:

7. (Flatten and) convert to 8 bit
8. Check destination space (Convert to Profile)


Most research I've done suggest inverting these two is a better option. Convert to the desired profile while still in 16-bit.
Also, resizing and sharpening while in 16-bit wide-gamut is less prone to create issues.

Message edited by author 2010-12-23 01:55:34.
 12/23/2010 02:14:53 AM · #13
I am an amateur, my first step in post processing is figuring out when to press the shutter button. ;-)
 12/23/2010 02:23:08 AM · #14
Originally posted by docpjv:
I am an amateur, my first step in post processing is figuring out when to press the shutter button. ;-)


That's a good start. A lot of pros should think about that one.
12/23/2010 07:14:54 PM · #15
Originally posted by fotomann_forever:
Originally posted by zeuszen:

7. (Flatten and) convert to 8 bit
8. Check destination space (Convert to Profile)


Most research I've done suggest inverting these two is a better option. Convert to the desired profile while still in 16-bit.
Also, resizing and sharpening while in 16-bit wide-gamut is less prone to create issues.


Note, I don't actually convert at this point (8.). I check before saving, is all. The file format, at this point, is likely TIFF, and I'm saving an edited, not yet proofed file.
Only time I convert (from Adobe98 to sRGB) here is in preparation to produce a web JPG (as for 1x or DPC), and I might amend this practice to Leroy's way of doing it from now on.

As to sharpening, I should add that I do not sharpen edited Tiffs at all. That would be part of the proofing workflow, and even there it wld have to be an image that actually benefits from sharpening.

12/23/2010 07:38:50 PM · #16
Originally posted by Yo_Spiff:
Originally posted by tanguera:
ZZ, you bring up another question, which is "flattening". What is that. How is that different from merging the layers, which I understand is to take several different layers and combined them into one.

Same thing.


Not exactly.

With flatten image you end up with one background layer.

With merge visible you still end up with one layer but it's not a background layer.

12/23/2010 08:47:25 PM · #17
Also you can merge just one layer with the one immediately below without affecting any other layers, regardless of their visibility status.

Note too that you can merge an Adjustment Layer onto one containing pixels to make the adjustment "permanent" ... useful if you want to apply an adjustment to a layer before merging it with another.

My typical processing order is:
-noise-reduction (if done; I don't very often)
-tone/contrast adjustment with one or more Curves layers
-save adjusted image as TIFF
-crop (if indicated)
-resize
-high-radius USM (if indicated)
-USM
-add border and caption (if needed, for print output)
-SaveAs JPEG

"Fortunately" my camera only captures JPEGs in sRGB, so I never mess withcolor spaces or profiles ...
Pages:  
Current Server Time: 08/26/2025 11:30:58 AM

Please log in or register to post to the forums.


Home - Challenges - Community - League - Photos - Cameras - Lenses - Learn - Help - Terms of Use - Privacy - Top ^
DPChallenge, and website content and design, Copyright © 2001-2025 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 08/26/2025 11:30:58 AM EDT.