Looking at colour (part 2)

In the first post, we looked at how we see colour and what it means to us as modellers and photographers of scale models striving to replicate colour as accurately as possible.

In this second post we’ll expand the discussion about how we see colour to look at how we think we can remember a specific colour and how this often falls short of our expectations and can lead to confusion.

Although most people can distinguish between millions of viewed colours, (there are currently 17.5 million combinations of colour capable of being produced by the digital red/green/ blue combination screen pixel method) we have trouble remembering specific tones and shades because our brains tend to store what we’ve seen as one of just a few basic colours. Before we go on, lets look at some universal terms:

Colour terms:

  • Hue is pure colour: red, blue etc.

  • Tint is adding white to a hue

  • Tone is adding black and white to a hue (you could say adding grey)

  • Shade is adding only black to a hue.

Let's use a popular colour such as RLM 04 Gelb as an example of a colour most modellers have come across and think we can reliably remember what it looks like. . .

Does your remembered version of the colour look like this? To my eyes (remember I have a colour vision defect) it looks a little too gold, too vibrant, saturated like a good egg yolk, yet it is a properly calibrated display screen value.

Perhaps my initial memories and opinion that this is an overly saturated false value are based on period colour images which have been subjected to erosion, fading and atmospheric pollution? Or maybe it's also coloured (pardon the pun) by seeing the real colour on various models and real life preserved warbirds?

Researchers dispute standard assumptions about memory, suggesting that people’s memory for colours are biased in favour of 'best' versions of basic colours over the colours they actually see. For example, there’s azure, navy, cobalt and ultramarine; all strong versions of blue. The human brain is sensitive to the differences between these hues—we can tell them apart when they are in front of our eyes. But when we store them in our memory we label all of these various colours as “blue”.

To make this task of recalling colour values harder, there is a difference in the memorability of certain colours and this is actually the result of the brain’s tendency to categorise colour. We remember colours more accurately when the colours are good examples of their respective categories.

We can differentiate millions of colours, but to store this information, our brain has a trick to tag the colour with a coarse label. When faced with a multitude of ‘somethings’ people tend to remember them later as more prototypical. It’s not that the brain 'doesn’t have enough space' to remember the millions of options, it’s that the brain tries to reconcile those precise details with more limited, language-driven categories. So a late war Luftwaffe airframe that’s coated with the controversial RLM 84 might be 'remembered' as being more 'green' while the under surface Medium Sea Grey of an RAF aircraft object might be remembered as more of a 'blue' depending on our own vision.

Our perceptions are further confused by the adjacent colours and the quality of light falling on the 84 or MSG. We'll come back to colour temperature in a later post.

So that's just one aspect of colour which influences our perception and preconception of what a colour should be; what about the period images we use as black and white reference points. Is it possible to determine colour from a monochromatic image?

Any colour you like as long as it’s grey - a quick look at interpreting greyscale images

The fact that many of our visual reference points are black and white (greyscale) period images, only serves to further ‘subvert’ our colour memory and our visual translation of colour values. I deal with this tricky subject and how colour works in more detail in my manual.

Here's a picture to interrogate.

For arguments sake, what might the Typhoon's various colour values be? Let's think about the chin radiator, that Medium Sea Grey underside colour. It's very close to the tonal value of the gear door. So how can we be certain that the chin lower colour value isn't the same as the inner gear door? For reference, here's a calibrated screen reference swatch.

Doesn't look like the same value as the monochromatic photo right? What the heck?

What is actually going on here is that the film stock used to record the scene has a ‘latitude’, a limited response to a range of light frequencies and has a fixed range of contrast between black and absolute white.

But maybe the photographer got it wrong?

Even if we assume the exposure in camera was correct, we can never know what the photographic paper used to create the print was. Photographic print paper also has a fixed contrast range (modern papers have a multigrade coating) or may have had a red or pink filter used during the darkroom printing process to boost the contrast in a flat image.

This means that certain colour values which translate as fixed grey values can be altered without any intent to do so (look at the MSG on the chin and then look at the natural metal inner gear door tonal similarity) and film type, exposure and the skill of the photographer in the darkroom will alter the greyscale values. Of course a skilled darkroom operator will know that the image must have true blacks and silky whites to be considered as a properly exposed photograph, but the limitations of the photo paper means that it will not record all colour values accurately.

This all plays with our visual perception and idea of what a colour should be. We know its Dark Green, Ocean Grey and Medium Sea Grey but is that mono image right? Maybe we are seeing a faded version or a weathered version or. . .

So basically we can never quite be certain that our black and white references are accurate representations of a certain colour value and translating these back into a model colour value is complicated and fraught with dangers of misinterpretation.

What we can be certain of is the colour demarcation and variances in values in a greyscale image.

To further complicate the issue, certain colours have very similar grey value when converted from colour to grey scale. If you don’t believe me get a green apple and a red apple, and then either photograph them side by side in your cameras monotone setting, or shoot in colour and then desaturate them in your favourite editing software.

Same shade of grey?

Confused much?

So that spinner back plate on the Tiffy might even be red, or blue or green. . . In the next post, we’ll look at what this means for our appreciation and interpretation of colour on the model, and how the quality of light we view it under alters our perceptions. All of this leads to a better ability to understand how we go about photographing our models.

Photo credits:


35 views0 comments