Over-processing

This is really a strange term when I think about it, and would have had no meaning at all before digital. With film you could push-process, or over-develop, or cross-process, but you couldn’t meaningfully over-process. But now you certainly can. Caffeinated processing has become the new standard for some photographers.

A short while ago someone wrote to me asking “I notice that your photography is not enhanced much. I see so much photography that is processed with high saturation levels. How do you feel about that?” How I feel is that pumping up the colour and the contrast quickly becomes very tiresome. But I’m jumping ahead; that’s supposed to come at the end of a considered view.

Let me first explain what I mean by caffeinated processing. Assuming that we’re all dealing with Raw image files, which I think is likely, the software engine for creating usable images is some form of Raw processor, and there are many of them. Photoshop’s ACR is probably the most widely used, but what they all have in common is that they’ve been been adding features constantly, as software engineers and manufacturers like to do. Who could argue against being given more and better ways of treating an image? Cost aside, that is. A moderately recent example is what Photoshop calls Clarity (mid-tone contrast enhancement), and another is Vibrance (clipping-protected saturation). Ah, I see I’ve already started the descent into tech-land, and I’ll stop right there. The point is that processing is something we all have to do for ourselves (well, some photographers have their assistants do it), and the choices get wider.

A processing control panel for digital photographs is coming to resemble a pilot’s console. Now, faced with all this choice, at least a dozen sliders, you could do nothing, which isn’t very thoughtful. Or you could accept the software manufacturer’s Auto setting — autopilot, in other words, and not a bad idea if the image is fairly predictable in its tone, contrast, colour and so on. Or you could work it out, slider by slider, to make it look exactly as you would like.

And this is exactly the issue. It’s as you like, with very few limits. Dark to light, flat to contrasty, monochrome to fluorescent, you name it, there’s a way of fixing it at some point on a very long scale. So how do you decide which point is right for you on, say, the scale of contrast, or saturation, or even brightness? Is there a way of defining normal, standard, ‘as it was’? Up to a point there is, but by no means is it cut and dried. Colour, as the Bauhaus teacher and artist Josef Albers said, “deceives continually”. He also wrote,

“If one says ‘Red’ (the name of a color)
And there are 50 people listening,
It can be expected that there are 50 reds in their minds
And one can be sure that all these reds will be very different”.

Sliding across this image from left to right, adjustments in Contrast, Clarity and Vibrance from subdued to intense, all in raw conversion.

The standard reference, as much as you can say one exists, is a colour target, which for all practical purposes means the Gretag Macbeth ColorChecker. This is based on the Munsell system, widely used in colour testing, and each of the patches is made to rigorous standards, hence its quite high cost. If colour accuracy really matters to you, this is the thing to use. Photograph it in the same lighting as the image you are shooting, and you’ll have an objective record — and there’s software that can work with it. It’s also a good way to make your own camera profile (see here).

ColorChecker

However, we don’t see colours in a standard ‘accurate’ way, and we remember them even less well. Should colours appear warmer in the light of a setting sun, or ‘accurate’? Commonsense and common taste suggest the former. Ultimately, when people have a choice about how an image should look, they exercise their own taste. It’s a matter of preference — and there have been enough color-preference tests in recent years to confirm that when people don’t have to think about using a colour (such as in furniture or as paint on a living room wall), and at first glance, they go for more saturated, lighter and more contrasty. So, “most adults tend to choose colors of strong hue over neutralized and gray tones, and they choose light tones over dark.” (Kopacz, 2003), and it seems to be related to the way our neural cells work: “A model of color contrast…based on double opponent cells…predicts that preference should increase with increasing saturation (Katz, 1999).

And how about this recent paper to raise a groan (well, from me at least): “Reproduction of more pleasing colors for natural objects is one of the available methods to improve image quality. This paper deals with the saturation enhancement of blue sky to increase the preference of scenery images”  (You and Chien, 2008). So, it’s official; we’re morons when it comes to choosing how intense an image should look. Or at least, partly pre-conditioned.

Over-processing with a (rare) purpose: cherry blossom viewing time in Tokyo, with the subtle flowers inelegantly seen against garish store signs, that say “Really cheap!”. There’s nothing false about the colours, but the processing needs to maintain their original, almost fluorescent quality

But, like strong flavours and really loud music, it’s hard to keep going with turbo-charged colour and contrast. Albers again: “[The] initially exciting effect feels aggressive and often even uncomfortable to our eyes”. Because, I think, there’s another kind of contrast that we respond to — the contrast between strong, rich and garish on the one hand, and subdued, restrained and quiet on the other. Martin Parr’s series of garish British holiday pleasures make a subject out of saturation (and what some people would think was bad taste), so those saturated images had a reason. But otherwise, turning up the dial on every image seems to me self-defeating. What’s most likely to get saturated is other people’s tolerance for looking at caffeinated photographs. Unfortunately, a trawl through Flickr recently suggested that it’s a growing trend (for example, try www.flickr.com/groups/colourful). Or perhaps I just took a few wrong turnings.

Remember this...

Hide Sites

Tags: colour, contrast, processing, saturation