OK, it's been a long time since I've posted a new entry here. Life has not been all copacetic for a while, and this blog has been pretty far down the priority list. For those who've noticed, thank you and I'm sorry. For the rest of you ... well, take my word for it - it's been a while.
Anyway ... post processing. It's not uncommon for someone to ask me "do you ever share you processing secrets?" My answer is "no, I can't because I don't have any." Then they usually call me a smartass and assume it's just a clever way to avoid the question. It's not, and to prove it, I'm going to lay out, step by step, the post processing steps I use most often.
For the most part I'm only going to cover the what, not the why. That would be way too long and boring, even for a geek like me. If you want to know why I've adopted the methods I have, you'll have to settle for this for now: I've taken several technical college courses on digital image processing, spent a lot of time listening to some very experienced, intelligent, and talented people, and done a whole lot of experimenting, and the method I'll lay out here gets the result I want most of the time. That said, I'm always experimenting and testing out new options as well. There is no single "right" approach, and this may not be exactly the same as my standard process 6 months from now.
So, here we go. You can click any of the images to see a larger version.
Step 1. Open the RAW format image Canon's Digital Photo Professional (DPP) software. Adjust white balance and exposure.
Step 2. Transfer to Photoshop as a 16 bit tiff image and crop to desired format.
Step 5. Sharpening. I use a couple different sharpening strategies, depending on the look I want to achieve, but the one I use most often is a Frequency Sharpening approach (developed by Sean Baker on the MM forums - for the geeks, you can read all the details and tons of useful derivatives Here with more useful info Here)
First, duplicate the main layer and perform a Gaussian blur, with the radius just enough to soften blemishes. In this case I used 1.4.
Turn off the blurred layer. Duplicate the main layer again (not the blurred layer). On the new layer copy, use the Apply Image tool (under the Image menu). Choose the blurred layer for your Layer, check the Invert box, use Add for the blending mode and 2 for the scale, and choose OK. Don't panic when you get a gray mask over the whole image. Go to the top of the Layers pallet and change the blending mode for the layer to Linear Light and you'll see your image again. You can delete the Gaussian Blurred layer at this point, but DO NOT flatten the image.
Here's a before and after at actual pixels scale.
After sharpening, I move on to tonal adjustments. Some people say never sharpen until your last step, and there is some logic in that, but the high-frequency layer from the sharpening routine is a very valuable tool for tonal adjustments as well. With the high-frequency layer active, open the Curves tool. You'll see that the histogram for that layer is heavily center-weighted. It's useful to experiment with this for different effects, but in this case I want to exaggerate the contrast a bit so I'm just going to choose the Medium Contrast preset.
Next I'll flatten the image then open the Levels tool. Usually when using this approach, I'll have just a very thin tail on the right side of the histogram, with a notable point where there is more substance to the graph. I'll drag the high end pointer back toward that point, being careful not to blow out important highlights. In this case I'm watching the reflection on the helmet and the white shirt to make sure I don't loose the detail visible in them.
Now I want to bring up the detail a bit in the shadows. I'll switch to the Channels pallet then Ctrl+click on the thumbnail of the RGB channel to select the high end of the luminosity in the image, then invert the selection so I have the lower end of the luminosity. In some cases I'll make adjustments from there, but on this particular image I want to dodge just the deepest shadows so I'll save that selection as a channel then Ctrl+Shift+Alt+Click on the thumbnail of the new channel. Once I get that selection done I'll delete that Alapha 1 channel, just so I don't forget about it, but make sure not to loose the current selection.
And that's it. Of course, there are exceptions - nothing is set in stone, and there are dozens of different ways I deal with special circumstances - but for most images that's the general process I use for post. It takes between 7 and 12 minutes per image, so I can do 5 to 9 images per hour. In a set where I have 30 keepers, that's 4 to 5 hours of post processing.