Raw workflow what is best practice?

[GaryMarsh]

Just a question about processing my camera raw files, I have a Nikon slr and always shoot in raw mode and I find that using Nikon Capture NX2 for editing my NEF files is much more responsive than using Adobe Camera Raw's version.
But where is the best starting point for editing? I always start off by cropping, removing dust spots and any signs of chromatic abberation and then if need be set my white balance, but after that I would like to know what is best procedure. Should I be using levels and curves next to adjust contrast e.g setting my black and white point or would it be better to use the shadows and highlight protection sliders beforehand.
If anyone could give me any suggestions for the steps I should take in my raw editing software it would be greatly appreciated.

Regards
Gary

[revi]

I would correct chromatic abberation before cropping: This correction boils down to slightly scaling
the blue and red channel images (leaving the green part alone). This scaling is relative to the current centre
if the image, but should be done relative to the original centre of the image; so after a crop you might not
be able to correct the whole image. Also, doing it as soon as possible might allow it to be done even before
demosaicing, when you still have the raw sensor data (possible in theory, not sure which programs, if any,
work that way)

The same reasoning holds for distortion correction (and in a lesser degree for perspective correction and rotation).
Next step for me would be cropping and capture sharpening. Then black and white points, and then curves etc.

Only then I go to dust removal etc, but that's because I have to switch programs to do that, and switch to
8 bit/channel processing. I prefer to do as much as possible in 16 bit/channel.

I finish with creative sharpening, and, if needed, resampling and output sharpening.

One thing that complicates advice about workflow is the number of different programs available:
several steps can be done either before/during demosaicing or after. White balance and chromatic
aberration correction are the most evident, but black/white point and curves can also be done at
both stages. And there are advantages to do such steps before demosaicing. But not all programs
allow that for all possible steps.

Also note that chromatic abberation and lens distortion are fairly easy to correct automatically, as they
are constant for a given lens setting (focal length and F-stop).

[Polar01]

Gary: played with Capture NX2, liked it however I have CS6, NX2 does do colours better than CS, but why have two programs. As to workflow, I do to lens correction than chromatic abberation, next increase contrast and hunt for the dust spots. Once that is done, return contrast to starting point (I find that the increase helps me find them) and adjust the controls until I believe I have the image I want. I would then got into Photoshop to do any final adjustments. I never crop an image until I am ready to print and then never save that cropped image. This is because in the future I may want to print it another way and if I have cropped away the rest of the image I am not going to get it back. However that is just me.

Cheers:

Allan

[Manfred M]

My workflow tends to be somewhat dependent on the type of image I am working on, and with a few exceptions, there is little difference in the way that I handle RAW, jpegs, TIFF or any other image formats.

When I work RAW, I will one of three tools quite commonly;

1. As I shoot Nikon, I will use ViewNX2 as a RAW converter for run-off the mill conversions;

2. I will use DxO Optics Pro 8 manage any lens corrections (it does a better job than the Adobe software and includes some of the camera / lens combinations that I use that are not in the Adobe database). This is primarily distortion, CA (chromatic aberrations) and vignetting; and

3. Adobe Camera RAW. At times I will also use Lightroom 4. I find that there is little difference between the two pieces of software, but I find the way that Lightroom ties everything into a database a bit limiting.

The important output from any of this software is getting an image with proper white balance as a starting point. After I get this done, I usually do all my other changes using Bridge and Photoshop CS 6.

My next step is to get the image into good shape for PP work. With RAW files I will do import sharpening (I use Nik Sharpener Pro plugin). I then do all of the “stuff” that is relatively fast to do; contrast, exposure, clarity, hightlights, shadows, etc. Generally I will do this non-destructively though the use of adjustment layers. That is I try to see if I can get the image into good shape before getting into time consuming retouching work. If the image looks promising at this point, I will continue to the next level, otherwise I will stop and delete the image.

The next phase I get into is what I would refer to as fast retouching; getting rid of sensor dust, obvious defects that I can eliminate with the healing brush, spot healing brush, patch tool, content aware tools and cloning. I’m really talking about minor things here that I can get done within about 10 minutes.

If I am working on a portrait, I will work on skin defect, wrinkles as part of the previous work. I will work to reduce dark spots under the eyes, do some dodging and burning to enhance the lighting. I’ll probably reduce the red in the eyes and lighten the eyes, might brighten the teeth, enhance the lips. Fix up eyebrows, hair, eyelashes, etc. This too takes very little time.

In landscapes, cityscapes, wildlife shots, I will do “parallel operations”. Remove stray blades of grass, telephone wires, fix perspective issues, local sharpening, etc. Again, I do the fast stuff first, because if I ever get to a point where I am going to have to do “major surgery”, I want to do that knowing that the image is worth the effort.

Major surgery is usually not required, if I managed to get things right in the camera. When this is not possible I will do these activities as one of my last activities because of the effort involved. This could include contouring the body in portraits or removing major defects in other types of shots. The content aware tools that were first introduced in CS5 really work well and reduce the time it takes to do things significantly.

At this point I save my file as a Photoshop document and it is my “new” master file. I want the data stored in such a way that I can output to a web page or do a super-sized print. This means I will crop, output sharpen, resize and produce a final product as an image file. I may not save this version as it only takes a few moments to recreate and output image from the PSD.

[pnodrog]

Cropping is the second to last thing I do (final sharpening is the last and amount depends on use). The area beyond where you are going to crop can sometimes provide a source for cloning and some filters produce minor artifacts at edges. I always retain the worked on uncropped version in case I need to crop for a different use. I use Lightroom so this approach works well.

I suppose a hangover from programming is that I tend to start with global (overall) changes such as exposure, levels and WB first and then progress to the more specific localized adjustments such as cloning out defects or dodging/burning etc. Usually I do the highlight/shadow before using curves but as I tend to use adjustment layers in Photoshop or use Lightroom I find that I sometimes revise the adjustment before I am happy with the result.

[Geoff F]

Cropping is the first thing I do in Raw editing. That means I don't have to worry about areas which I don't intend to use. But I do retain the Raw files for any future re edits.

Then tweak the white balance and exposure although the exposure options are used in different orders depending on any obvious defects.

Sometimes I then go directly to the saturation settings and return to exposure options for a further tweak.

Last part of Raw edit is capture sharpening and any noise reduction.

Although I sometimes create a smart object from the Raw files and return to ACR for further editing on a smart object copy. The two versions are then combined with masking.

[Mike Buckley]

Gary,

I strongly recommend Jason Odell's The Photographer's Guide to Capture NX2. It will answer all of your questions and many more.

To pique your interest, if you ever decide to use a crop in the future that is different from your current crop, all cropping should be done last (unlike your current workflow). That's due to certain characteristics of NX2's various masking tools.

Another tip is that most of my adjustments are made using the LCH Editor. As an example, the tool built into the LCH Editor that adjusts levels and curves does not alter the color, whereas the standard Levels & Curve tool and other sliders found elsewhere does even though you would not want that to happen.

[nimitzbenedicto]

Hi Gary,

If I recall, CaptureNX has a suggested workflow in its manual.
[ something like start with the whole image, then local areas, apply effects, lastly sharpening and resizing. ]

Considering you have Capture NX2, maybe it's best to follow and learn their workflow. first.
Later on, change it as you gain experience and/or if you get Lightroom or some other software.

Hope this helps.

[Mike Buckley]

If I recall, CaptureNX has a suggested workflow in its manual. [ something like start with the whole image, then local areas, apply effects, lastly sharpening and resizing. ]

Yes, but it's very basic and doesn't even mention the LCH Editor, instead mentioning the Levels & Curves tool. In fact, the information provides very little more than in your post. To find the information, go to the Help screens and conduct a search of "workflow."

[nimitzbenedicto]

Yes, but it's very basic and doesn't even mention the LCH Editor, instead mentioning the Levels & Curves tool. In fact, the information provides very little more than in your post. To find the information, go to the Help screens and conduct a search of "workflow."

Yes, you are correct. Perhaps, he should go to the help screens and read up on its "workflow".