In previous articles, I covered equipment and techniques for photographing concerts. But even if you followed all the advice I gave, you probably came home, looked at your images, and didn’t quite see what you hoped for based on the examples you have seen in my articles or elsewhere. There are a few fundamental things I always do my concert photographs in post-processing (PP) that make all the difference in the world with relatively little effort. In this article, I will explain some of these techniques.
First, I should mention that the software I use for PP is ACDSee Pro. This application provides integrated image management and non-destructive editing much like Lightroom or Aperture or a few other programs. The techniques I describe are mostly things that can be done in any of these or similar programs, as well as other popular editing programs like Photoshop Elements, Paint Shop Pro, or even Picasa. Since I expect people reading this will be using all sorts of different programs, I will try to be as general as I can rather than deal with the specifics of ACDSee Pro.
To give you an idea of the extent of the difference PP can make, I find it useful to look at a whole screenful of thumbnails first before focusing on what might be done to images individually. Here is what I might typically see when I first download my images:
As with the other examples in this article, you can click on it to see it larger.
There are a couple of things that usually end up being disappointing when we first look at our images. One is that we typically have a lot of very similar-looking shots. Part of this is the fact that concert lighting is usually fairly strongly colored. Even plain “white” spotlights are usually pretty orange. The color of the lighting usually dominates any other color in the scene. But another reason we often have a bunch of similar-looking images at first is because of how we typically shoot. We often take a bunch of pictures of the same musician or scene in hopes of one turning out well before moving on to the next.
One of the most important but often overlooked principles of PP is that editing is as much a matter of selecting images as it is of actually adjusting the images themselves. Anyone can look like a much better photographer if we only look at their best images. So as soon as possible, I start the process of rating my images in ACDSee. My scheme is fairly simple: 1-star for images that are out of focus blurry or are otherwise not worth keeping; 2-star for images that are competent and not worthy of deletion but which I don’t particularly need to keep looking at; 3-star for the images I intend to keep copies of after the originals have been should been archived to an external archive, 4-star for images that I think are worth sharing with others. Just by grouping my 4-star images together, already I look like a better photographer.
If I’m going to try to look like a good photographer, however, I am better off doing my PP first. And the other benefit of doing my rating first is that I can focus my efforts on the 4-star images. While I might process those individually, I often just copy the settings I used on my 4-star images to any similar 3-star images.
The following screen shot shows just my 4-star images from the same shoot as the previous example, so there is more variety in the shots themselves. And since I have done my PP – including white balance – there is more variety in color as well. Instead of the screen shot I showed previously, I’d much rather see this:
White balance isn’t the only thing I changed in PP, of course, but it’s the thing that made the most noticeable difference in this screen shot. The other changes I made would be more noticeable in larger views.
So now let’s turn to the specific PP techniques I use. There’s actually not a lot to it, but as the above examples demonstrate, they really make a difference.
The first thing I usually do with a concert shot is to apply a preset I created some time ago that sets the white balance to something that works for basic tungsten lighting, sets noise reduction to levels that often work well with my camera at ISO 1600 (my usual setting for concerts), and adds what I think of as a typically appropriate level of sharpening. With ACDSee Pro, I can apply this preset as a batch to all my 3-star and 4-star images if I like, giving me a good starting place for further processing in just seconds. Many other RAW processing program provide a similar sort of capability, but there are some older applications using an older paradigm that don’t work this way: you can’t apply a preset other than while converting to JPEG, which means no further RAW processing would be possible. I would not like to use such a program, but if that’s what you have, then you’ll have to do all your work on files one at a time.
Here is an image from a recent shoot loaded into ACDSee Pro as it came from my camera (before applying the preset):
After applying my preset, the colors look more natural:
The orange color cast is lessened, but now it is too magenta. The lights on the stage at moment probably used a colored gel. So I further adjust the WB by hand (using an eyedropper on the hair then fine tuning from there) to yield something I liked:
It is common for WB changes like this to affect our perception of exposure. The original image was mostly red light. By removing much of the red light from the image, I’m left with better color balance, but less light overall. So even though the original image actually showed clipping on the histogram – all in the red channel, which is almost inevitable when shooting under red lighting – the result is now underexposed. On the positive side, most of the clipping is gone. Also, I should mention that because my camera is limited to ISO 1600 and sometimes that is not enough to avoid blur, I will often deliberately shoot underexposed and expect to push exposure in PP. Beyond that, I would also say that metering in concert photography is tricky, and there is no shame in needing exposure adjustments in PP.
In this case, I added 0.75EV compensation to bring the overall brightness where I wanted it. This actually blew a few more highlights in the hair and the reflection in the bell of the trumpet, so I used a little highlight recovery to lessen that. While I was at it, I made small changes to two other controls on this same panel. I added a small amount of fill light to lighten shadows further, and I slightly reduced vibrance (similar to saturation, but “smarter”) to make the skin tones a little more neutral still.
Here is the result of those changes. The green dots in the background show where I have deliberately allowed the background to clip to black, and the red dots show where I have intentionally allowed highlights to clip to white:
The next step for me is usually to use the Lighting tool in ACDSee to further balance the highlights, midtones, and shadows, while also increasing local contrast and bringing out detail appropriate. This is a tool that may have have an exact analogue in your software, but similar effects might be achieved using curves, local contrast enhancement, or shadow/highlight tools that you may have access to. It is kind of hard to explain exactly how ACDSee’s Lighting tool works, but the effect is to allow me to lighten the shadow side of face without reducing contrast in that area, and similarly increasing contrast and bringing out detail on the light side of the face. I am not sure if this will come across well here, but here the effect of the tool on this image:
At this size, the image now looks pretty much the way I want it to. But I realize than in larger views, noise will be more apparent. The amount of NR I applied in my preset is pretty conservative, and after exposure adjustments, I often need to apply more. taking an exposure made at ISO 1600 and pushing it by 0.75EV is like shooting at the equivalent of about ISO 2600. Now, I should say that I am not as noise-averse as some are – some amount of noise is pretty much expected in concert photography. Too much NR can smooth away detail, and I prefer a somewhat “grainy” look over a “plastic” one. I should also say that while NR is best done while looking at a 100% view, this is not very representative of how most people will ever see your images. Viewed on the web, they will be much smaller than 100%, and even an 11×14″ print won’t show noise to the same degree as a 100% view on screen. So don’t get too discouraged by how your images may look at 100%. Chances are they will be fine on the web and in smaller prints.
The camera I used here is the Pentax K200D, which is fairly average in terms of noise. It is sometimes characterized as having more chroma (color) noise than some cameras but less luminance noise, with an overall blotchiness due to a small amount of in-camera NR that is performed at ISO 1600 even for RAW files. Here is a 100% view of my images with my other changes intact, but the NR I originally applied in my preset turned off:
Here is the NR applied by my preset (50% chroma, 5% luminance):
While it might not look fantastic at 100% I know from experience it should look good enough at “normal” sizes. But since if I look closely I do see some larger purple splotches that I know I can control with a bit more NR, I turned the chroma NR up to 75% and luminance to 10%:
That’s as far as I’d want to take this. Folks who are really allergic to noise might want to investigate dedicated NR programs like Neat Image, but when I have tried them, I find I do not usually prefer the results I get. I could also try turning up ACDSee’s controls all the way, leading to the sort of “plastic” look I referred to earlier:
Actually, while this is noticeably softer if you click on the image to see it full size, I admit that at typical viewing sizes, the softness is no more likely to be noticed than the noise is. Still, I prefer the image the way I had it before.
The final thing I would consider doing here is a slight crop. I don’t crop very often, and when I do, it is usually because of something really distracting like another musicians’ elbow in the picture or a stray microphone stand. But sometimes it is for purely aesthetic reasons, to achieve what I think of as a better balance in the composition by moving a face a little off center or trying to get the right proportion of body to instrument to background in the shot.
Some have an almost religious belief in not cropping, insisting on getting it right in camera. While I respect this, I feel I don’t have that luxury. Concert photography is too fast paced for me to feel comfortable limiting myself in that way, and when shooting with primes as I usually do, it’s often the case that the lens I have is not exactly the focal length I would choose if using a zoom. I do, however, maintain an equally irrational desire to preserve my camera’s 2:3 aspect ratio when I do crop.
Anyhow, here is the image as it stands before the crop:
All I would want to do here is remove some of the empty space above the head, thus making the face more dominant element in the composition, and paying attention to where I place it. this is what I came up with:
Had I performed a more significant crop that might have affected my perception of the color or exposure, I would have done this much earlier in the process – perhaps right after applying the preset. To be honest, I went out of my way to find an example of an image I had cropped for artistic reasons as opposed to simply cutting out something I didn’t wanted, but I don’t feel very strongly about the need for the crop here.
This leads to a related topic. I virtually never resort to cloning out stands or other distracting items that cannot be cropped away. I suppose this is like some people never cropping. Actually, though, in my case it’s more because cloning is a rather more complex operation, and ACDSee cannot do it within the scope of its non-destructive Develop mode. Instead I need to switch to its traditional Edit mode, which in turn requires me to convert to another format in order to preserve my changes, and a separate copy of my file is created as a result. This just seldom seems worth the effort to me as it might be if I using were a program that aloowed cloning in the non-destructive RAW processing, but even then, I still have reluctance to want to mess with brushes and selections and so forth.
Instead, I mostly just try to be conscious of my backgrounds when I shoot. I try to position myself so there are no distracting elements that would require cloning. In the case of this image, I suppose some might try to remove the bright line to the left of the trumpet (a reflection off the raised piano lid behind him), but I kind of like it.
There are more operations I could employ, and occasional do when necessary, but the steps I have outlined here are pretty much my usual routine. Preset to get WB and NR in the right ballpark, fine tune WB if necessary, correct exposure, play with lighting, further adjust NR if I had to correct exposure by much, and perhaps crop.
Here then, is the finished image:
Because I shoot in manual mode, it is pretty much guaranteed that all other shots I took of the same musician from the same vantage point will have the same issues in terms of color, exposure, and noise. So I can simply copy these settings (everything but the crop, which will rarely match from shot to shot) to all my other similar images in one operation, thus giving me a very useful starting point for further processing should I wish to do more. Often, though, I find no further processing is necessary at all after copying settings. But if the color of the lights changed, I can adjust WB while leaving everything set. Or if I had changed my shutter speed while shooting, I can adjust the brightness to compensate, again while leaving my other settings intact. This allows me to process a large number of pictures relatively quickly.