Andy Ciddor /
05.05.2004 12:00 AM
Picking a Winner in the Color Temperature Wars
Every day we are forced to pick a winner in the color temperature battle that rages all around us. Even at night, when we're shooting in a totally unlit situation and can choose our own sources, we are still forced to select one color temperature from the arsenal in our lighting kit.
Unlike our forebears, who shot on film that was locked inflexibly to a single color temperature, we have camera gear that will allow us to shoot at pretty much any point, on or near the curve, followed by a black-body object.
With a little bit of 21st century jiggery-pokery, we can even get something vaguely resembling a white- balance under such non-black body sources as Cool Daylight fluorescent tubes, or mercury vapor and high-pressure sodium street lighting. Despite this, I have noticed a tendency to shoot "actuality," of both the real and pseudo-real varieties, with uncorrected or under-corrected color balances. I suspect that this means that we have reached the point where urban-night footage must have an orange cast to be credible, just as all-night scenes by convention have to have a blue cast.
Out in the big wide world, we are almost never faced with just a single type of light source to which we can easily white-balance. Almost every situation calls for mixed sources, and unless it's out of doors, there's a very good chance that one of those sources is not of the "black-body" variety.
If the term black-body source is not familiar to you, we should look at a few basic concepts of color temperature. A color temperature is the aggregate mix of the colors of light that are produced when an object is heated to incandescence. The hotter the object, the whiter the light it emits. However, in the world of real objects, as things get hotter, they melt, vaporize, catch fire, oxidize, etc, which makes it really hard to plot any simple curve relating actual physical temperature to light colors.
To get around this problem, a fictional object named the "Black Body Radiator" was devised. It is a combination of the light-emitting properties of a wide range of materials that can be heated until they glow. This theoretical object is black, so it has no reflective properties, only emissive ones.
Physicists have plotted the colors of light that would be emitted from this object at temperatures ranging from the theoretical Absolute Zero point (-459.67°F or -273.15°C) to infinity. The temperature scale used for these measurements is Kelvin, which has Absolute Zero as its starting point and uses centigrade degrees as its steps. On the Kelvin scale, water freezes at 273.15 degrees (32°F or 0°C) and boils at 373.15 degrees (212°F or 100°C).
Color temperature describes the color of a light source by comparing it to the color of light emitted by the Black Body Radiator heated to a particular Kelvin temperature. The unit for measuring color temperature is not degrees, however, it's just "Kelvin."
For example, the color temperature of the lamp in a Readhead is 3200 Kelvin. This doesn't mean that the Tungsten wire in the lamp is necessarily running at a temperature of 3200°K. What it actually means is that the color of the light coming from the Readhead's lamp matches what would be emitted by the Black Body Radiator at 3200°K.
A major failing of the color temperature system, as it applies to making television pictures, is that it deals only with incandescent sources (ones that are heated until they glow) such as the sun, a candle flame, a filament lamp or the mantle on a gas lamp. All these objects produce the entire visible spectrum when heated to incandescence, although their emissions contain progressively more of the higher energy wavelengths (blue and violet) as the object gets hotter and looks whiter.
Light sources that produce their light by methods other than heating frequently do not emit the entire visible spectrum. Such non-incandescent or nonblack-body sources include fluorescent tubes, LEDs, mercury vapor and sodium vapor lamps, as well as the entire family of metal halides, such as HMI, MSR, HTI, CDM and MSD.
As the mix of colors produced by these sources is often in different proportions to the light from the Black Body Radiator, it cannot be directly compared using color temperature. Instead, by measuring the relative amounts of red, green and blue light being emitted by the source, an approximation of a color temperature, known as a "Correlated Color Temperature," can be reached. However, relying on the correlated color temperature for setting up your pictures is fraught with dangers for the unwary.
A standard commercial "Cool Daylight" fluorescent tube may have a correlated color temperature of 6200K, but mixing its light with real daylight at 6200K, or the output of an HMI lamp with a similar correlated color temperature, can produce quite startling results. As green happens to be the easiest color to produce from the phosphor coating of a fluorescent lamp, almost all fluorescent lamps have way more green in their output than their correlated color temperature would lead you to expect. This is the source of the green cast that is often associated with shooting under fluorescent lighting. Similar, but less pronounced, magenta and blue-green casts are associated with various commercial metal halide light sources that are often found in shopping malls and other large public spaces.
It is these nonblack-body sources that are the cause of many of our woes. If there is enough available light to get the pictures you want, then it doesn't really matter what the source is, because a push of the white-balance button will generally give you very acceptable pictures. The more common situation is to have a mix of different sources, usually of different color temperatures and frequently also of different source families. This is where picking a winning light source becomes important.
If you choose a source color that either dominates the important parts of your shot or can easily be matched through the use of correction filters, then color temperature differences may be manageable and may not even be obvious in the finished product.
If the shot you need is in a fluorescent-lit shop, with windows looking out on to a sunny daytime scene, you immediately have a serious dilemma as to which color temperature should be the winner. Should you choose to go with the fluorescents and their green cast, the exterior views will look slightly magenta and may become overexposed with the movement of the sun. If you choose the daylight color and decide to light the foreground action with daylight matching HMI, then your foreground and exterior elements will look fine, but the majority of the shop interior will have a green tinge.
This is where we get to the art of choosing a color matching strategy. We can either filter the daylight to match the fluorescents, as it comes through the windows, or filter the fluorescents to match the daylight. Both are major tasks and may be completely ruled out by the resources available. In a future column, we will look at the process of bringing these color temperatures together through the use of some of the less widely known and understood correction filters.
In the meantime, my personal strategy has always been to look at the color temperature that has the most effect on the shot in question and work to that, even if it may mean rebalancing to a different source for the very next shot in the same location. I pick my winners on a match-by-match basis, rather than sticking with one team for an entire season.