I’m asked it quite a lot, so I thought it might be useful to write something about my workflow, such as it is.
I’ll be honest and say that for the longest time I felt that it was wildly overstating my way of doing things to describe it as a “workflow” – it was far too basic a process for a name like that, I felt. But now that it has developed a bit I feel a bit more comfortable about talking about it.
Getting the images off the camera
I use a card reader to get the files onto my PC and use a simple Windows Explorer-based file structure to control the files: a top-level folder for each camera; a sub-folder for the year; further sub-folders for each session, named by date. So CR2s from today’s shoot would be saved in the following folder:
7D > 2010 > 23 March, 2010.
It’s simple and it works really well.
Once the files are on the PC, I do my initial culling using Irfanview.
I simply browse to the folder and open the first file in Irfanview (set up as the default program to open CR2 files) and very quickly and easily work through the shots. I delete files from within Irfanview using the keyboard’s Delete key, and navigate through the files using the left and right arrow keys. By setting Irfanview to open files maximised to “fit images to desktop” and to “centre image in window” I get a very good idea of what’s coming off the camera, making this stage of the workflow really straightforward – and fast.
Once the cull is done, I open Capture One.
Cap One is my converter of choice these days. Much as I like Bibble 5 for my 40D files, it doesn’t really hack it for the 7D, whereas Cap One is every bit as good in terms of basic clean IQ on conversion as DPP, but with the added benefits of the kind of sharpness and detail Bibble provides, and much, much more in the way of adjustment options than DPP.
Using Cap One’s Library tool, I navigate to the image folder I’ll be working on and create an “output” sub-folder, always called Converted.
Then it’s down to actual processing.
Cap One supports the 7D and provides fantastic quality results, but I actually find that I prefer the Canon 1D Mk III ICC profile – it’s somewhat less saturated and less contrasty than the 7D default – so I select that.
The variety and lack of control over subjects and conditions in bird photography means that we rarely get to batch process a bunch of files based on a single set of processing parameters, so there’s no single group of settings I can suggest.
I always start by roughly adjusting the Exposure and Levels – doing this first defines the edges of the histogram, which I think has to be the starting point for further adjustments like white balance.
I then make use of the High Dynamic Range (“Highlights and Shadows”) tool as required. If I’m honest, highlight recovery in Cap One is not quite as good as the equivalent in Lightroom or Raw Therapee, but it’s works just fine for the vast majority of files: it can cause a slight colour shift if used heavily, but the other reason I use the 1D Mk III profile is that there’s less of a colour shift using that profile.
Adjusting the white balance is done at this stage too, and then I’ll make changes as needed to saturation and contrast.
The key to Cap One’s performance is its demosaicing (which we have no say over – but which is very good) and noise handling.
Careful adjustment of the excellent chroma reduction and very good luma reduction provides images which – seriously – look extremely like the files that people (quite rightly) rave over from Nikon full frame cameras, especially the look of the noise, which is fine-grained, very low in chroma noise and easy on the eye: I’ve honestly wondered before now if some of Nikon’s noise “magic” might be explained by them licencing Cap One’s chroma noise reduction.
Typical noise settings (assuming a reasonably well exposed 400 ISO file) are around 15 to 20 on the Luma slider and 35 to 40. I fine-tune the noise reduction using the very useful “Focus Tool”, which is a zoomable window view of a section of the image: click into the image and the Focus Tool will show that section – and the effects of your NR and sharpening choices – at your choice of zoom.
I usually use a sharpening setting between 260 and 300 – “capture” sharpening which brings out detail nicely without generating artefacts.
Although I mention above that we don’t often get to use a single set of parameters for a whole batch, I do make a great deal of use of the copy and paste settings option in Cap One, because it’s usual to have a run of a few similar images in a batch of files, and it is handy to be able to simply paste the same settings onto them with a single click.
I convert files to full-sized jpeg at 100% (no compression) – I see no real point in saving them as TIFFs because I have the RAWs and the one extra save that the jpegs will get after I’ve done the next stage of the workflow has no visible effect on them whatsoever.
I do my finishing work in PaintShop Pro (PSP) X2. This is a deliberate choice, based on the simple fact that I prefer the results I get from PSP to Photoshop – I particularly like the sharpening tool in PSP called “Focus”, which I actually think is better than any sharpening tool in Photoshop.
All of what follows can be done in Photoshop of course, and I think Elements too.
A typical trip to PSP involves the following steps, more or less in order:
cropping and resizing;
Neat Image if needed: sometimes the resize algorithm will bring out a little bit of noise, and so light NR (and – perversely, perhaps – light sharpening) can be helpful;
further levels/black point/white point/saturation adjustment if needed;
additional highlights/shadows adjustment, again if needed.
These days I always apply sharpening selectively. The easiest way I’ve found to do this is to:
create a Duplicate layer;
apply sharpening (with the “Focus” sharpening tool) to that layer;
use the Eraser brush to remove sharpening from everything but the bird and any other part of the image that needs to be sharp.
It’s really quick and easy to do this, and you generally don’t need to be as accurate with the brush as you’d need to be to brush sharpness onto the subject.
I often use a Duplicate layer to blur the background too, for a “shallow DOF” effect. Here I apply (say) a low level of Gaussian Blur to the image then erase the blur from the bird and whatever else that should be sharp. Again, it’s easy and straightforward to do this, especially if you make the effect of the brush visible on the layer.
Once all the major tweaks have been done I might try the Clarity (Local Contrast Adjustment) tool and play with the white balance, just to see what happens.
When I’m happy with the image, I stick my name on and save (including the Exif), and that’s that.
I always save images with a name that identifies the species and the location. Doing so means that I can use the search in Windows Explorer to easily find all of my Purple Sandpiper images, or every picture I’ve taken in Hartlepool, and so on. This means that for me there’s no real need to get immersed in DAM, cataloguing, tagging, keywording and whatnot.
As I say above, it’s not a particularly sophisticated way of doing things, but it works, and I’m usually pretty happy with results. And I’m happy to report that while I rarely print, this workflow works really well for prints too. I’ve had big prints made (just to see) that look, if I might say so myself, fantastic – a version of this picture is currently on a friend’s wall at A3, printed and matted as a birthday present, and she’s delighted (the owner, not the dog!) and in fact this really isn’t a patch on the printed version.
Rebel the German Shepherd