Lighting for Still Photography
by Paul C. Buff
The behavior of light is not well understood by many photographers. This applies both from a technical as well as compositional standpoint. It must, however, be comprehended, at least in principle, if one is to master the art. This paper will attempt to bring forth some understanding of both the physics of light propagation, as well as methods of applying what is learned in the photography studio. The evolution and state of the art of studio flash systems will be covered in particular detail, as will be their use. It is hoped that what is presented will provide useful tools and incentives to amateurs and professionals alike.
LIGHT AND FILM
The first thing which must be made clear is the differentiation between light intensity and an amount of light. Think of a frame of film as a bucket of water. It takes a certain amount of light to expose the film, like it takes a certain amount of water to fill the bucket. We can fill the bucket by letting a trickle (low intensity) of water flow for a long time, or we can fill it almost instantly from a fire hydrant (high intensity). By the same token, the low intensity flame from a candle can expose any film if we give it enough time. It is light intensity vs. time that results in the correct amount of light reaching the film to make a proper exposure. Whether a high intensity light reaches the film for a short time, or vice versa, the amount of light for proper exposure is always the same (except for the extremes...see Reciprocity Effect). All statements defining an amount of light will contain both illumination intensity and a time of measurement, such as in ”Footcandle Seconds.”
In the practical world of photography, however, there are certain limitations regarding which combinations of intensity and time can be successfully used. In the darkroom or graphics studio, we think nothing of using rather low light intensities and long exposure times, even a minute or two. Most photographic objectives, though, are not afforded this luxury since neither the subjects nor the equipment will hold still that long. As we all know, if anything moves significantly during the exposure, blur results on film. Therefore, in most cases, the closer we can come to the high intensity/short duration end of the spectrum, the better the resulting picture. How far we need to go in this direction, obviously, is a function of the types of subjects we choose, the techniques and equipment we employ, and the degree of perfection we strive for. While we might get by with a relatively low light intensity and, say a 1/15 second exposure on a tripod portrait shot, anything longer than 1/60 second is sure to degrade pictures of moving subjects, hence higher illumination intensities will be required.
Now, there are two more variables that determine how much light intensity is needed: lens aperture and film speed. While it is practical to obtain short exposure times in low light levels using fast film and ”fast” (large aperture) lenses, we pay for the privilege on both fronts, in the form of inferior results. Simply stated, the faster the film (higher ASA/ISO numbers), the more grainy and poorly resolved the result. Similarly, the larger the lens aperture (lower f stop numbers), the smaller the depth of field or zone of sharp focus, regardless of the cost of the lens.
There are no two ways about it. If we want sharp and brilliant pictures with good depth of field on moving subjects, we must use slow or moderate speed film, moderate or high apertures, fast exposure times, and lots of light.
Ok, we can define film speed to the range of, say, ASA/ISO 25 to 160, lens apertures from, perhaps, f8 to f22, and exposure times in the range of 1/60 and upward, but how do we define ”lots of light”? This is where most of us start guessing. Let’s, then, get to the basics of light specification and propagation. First, though, we’ll touch on one more film related subject...Reciprocity.
As stated, films all require a specific amount of light (intensity times exposure time) for correct results. At the extremes of exposure time, however, the amount of light changes. A greater amount of light might be required for correct exposure of a given film if the exposure time is, say, longer than one second, or shorter than 1/1000 sec. In other words, the film speed actually changes at the extremes of exposure time. In black and white work, this does not pose too much of a problem since the net result is a slight under or over exposure which can be easily corrected in the darkroom. With color film, a more potentially serious result can occur, when it is considered that color film is actually three films in one. If the three color layers exhibit reciprocity effects that are different from each other, the result of overly short or long exposure times can be a shift in the color balance and unfaithful reproduction of the color spectrum. This effect is much harder to correct for. Little information is given on this subject by film manufacturers. Consequently, most color-conscious professionals will try to avoid extremely short exposure times (less than 1/1000 sec.) as well as inordinately long ones. Of particular concern are the ultra-short flash durations, which may be exhibited by poorly designed or mismatched flash components.
DEFINING LIGHT INTENSITY AND AMOUNT
While there are many systems of defining light strength, such as lumens, lux, lamberts and others, the photographic community has standardized on the original ”candle” family of specification terms. Thus, these are what our attention will be directed to. If we are to progress beyond terms such as bright, dim and lottalight, a basic understanding of these terms is in order.
CANDLEPOWER: This term defines, as one might suspect, the intensity of light emitted by a standardized form of candle. One Candlepower (CP) relates to the light of one candle.
CANDLEPOWER SECOND: An amount of light emitted. One Candlepower Second (CPS) is the amount of light emitted by a candle, which is lighted for one second.
FOOTCANDLE: The intensity to which a surface is illuminated. One Footcandle (FC) is the intensity of illumination that will fall upon a surface, placed one foot from a candle.
FOOTCANDLE SECOND: The amount of light falling on a surface. One Footcandle Second (FCS) is the amount of light received by a surface one foot away from a candle during a one second interval. One Footcandle Second would also result on a surface placed one foot from 10 candles if a 1/10 second measurement period were used. However, one Footcandle Second will not result is two candles are placed two feet from the surface...see Inverse Square Law Propagation.
BEAM CANDLEPOWER: The Beam Candlepower term measures the effective intensity of a light source when it is focused into a beam by a reflector or lens. A ”naked” candle in the center of a room, for instance, would emit 1 CP in all directions. If a mirror were placed behind the candle, however, the light traveling backwards would be reflected to the front, thus aiding the front emission and doubling the intensity of light projected front wards. Thus, the effective, or ”Beam” Candlepower in front of the source/reflector combination would be 2 BCP. No more light is actually produced...it is simply concentrated into a smaller area or emission angle. BCP specifications and measurements are made along the axis of the projected beam.
BEAM CANDLEPOWER SECONDS (BCPS): If this one isn’t obvious by now, I haven’t made things clear or you haven’t read them. For the record, one BCPS is an amount of light emitted by a source/reflector combination, as measured along the axis of the beam, which is equivalent to that amount of light which would be emitted by a ”naked,” or weakened candle, lighted for a period of one second.
INVERSE SQUARE LAW PROPAGATION: This basic law of physics probably gives photographers more headaches than any other. What it says is that the intensity of light falling upon a surface is inversely proportional to the distance between the surface and the light source. If the distance is doubled, for instance, the surface illumination decreases by a factor of four (2 squared). If the distance is increased tenfold, the surface receives only 1/100 as much light. Thus, one candle placed two feet from a surface provides only 1/4 Footcandle of illumination.
SPEAKING IN F STOPS
The term ”f-stop,” in actuality, defines the ratio of the lens aperture diameter to the lens focal length. Thus, a 100mm lens whose aperture is set at 25mm diameter has an f-number (factor) of f4. As the aperture is made larger or smaller, more or less light is transmitted by the lens, onto the film or digital file. Modern aperture settings are marked in the following sequence: f1, 1.4, 2, 2.8, 4, 5.6, 8, 11, 16, 22, 32, 45, 64 and so on. While the numbers may seem random, the fact is that each step is 1.414 times the previous step (1.414 is the square root of two). Each increment, or f-stop, represents a doubling or halving of the amount of light reaching the film...example: if the film/digital file receives, say, 10 Footcandles of light intensity at a lens setting of f8, it will receive 5 fc at f11, or 20 fc at f5.6. Photographers have come to use the f-stop notation, not only for defining lens aperture, but, indeed, for specifying any parameter which affects the amount of light reaching the film. For instance, if a 500 watt photoflood lamp were illuminating a scene, doubling its power to 1000 watts would be stated as ”increasing the light by one f.” If a filter were used which passed only one half the light, it would be referred to as having ”one f less.” The same applies to film speed. Each time the ”speed” (ASA/ISO#) of the film is doubled, it means it is twice as sensitive...it is ”one f-stop faster.” Thus, we know if we double the film speed, we will need ”one f-stop less light,” or one half the amount of light. We could obtain one half the light intensity by halving the wattage of the bulb (or halving the BCPS of the flash), or we could leave the light intensity alone, and halve the exposure time. It is seen, then, that each halving or doubling of exposure time can also be referred to as a ”one f increase or decrease.”
Remembering the Inverse Square Law, however, we must consider that doubling the light to subject distance will result in not a ”one f loss,” but rather, a ”two f loss.” To achieve a ”one f” change in intensity by moving the light’s, we must remember the square root of two (1.414) or its reciprocal the square root of one half (.707). Thus, increasing the light to subject distance by 1.4 (say from 10 feet to 14 feet) will incur a ”one f loss,” while moving it in to 7 feet would increase subject illumination by ”one f.”
CALCULATING LIGHT REQUIREMENTS
Having touched upon the rudiments of light behavior and specification, let us set up a hypothetical lighting situation and determine what is needed. While we will get into multiple light sources, diffusion and other techniques later, we presume for the moment our scene will be illuminated from a single incandescent lamp in a reflector. The subject will be human (prone to a certain amount of movement), while our objective is a sharp, highly resolved and generally high quality photograph. In order to meet these objectives we might choose an ASA/ISO 100 speed film, a lens aperture of F11 and a shutter speed of 1/60 second. For even lighting, we might choose a minimum light to subject distance of 6 feet. By referring to appropriate charts, or to an exposure meter, we find that approximately 32 Footcandle Seconds are needed to expose ASA100 film at f11. In order to make the exposure in 1/60 second, we will need 60 times 32 FCS, or 2000 Footcandles of light intensity. Since our light is to be placed 6 feet from the subject, we will need 6 squared, or 36 times 2000 BCP, or 72000 BCP from the light source. We can typically expect a reflector gain (concentration factor) of around 6, so the required 72,000BCP can be achieved with around a 12,000 CP bulb. Now for the big question...what wattage light bulb will produce 12,000 Candlepower? Typical incandescent photolamps produce on the order of 1.5 Candlepower per watt. Thus, (Gulp!) an 8000 watt bulb is indicated. Now, an 8000 watt incandescent bulb...even just one, will require heavy duty 230V circuits, and will produce as much room heating as a moderate sized home furnace turned up full blast. We will need an industrial air conditioning system and large studio environment if we are to be at all comfortable in the room.
Unfortunately, the facts of life are that if we are going to meet our initial objectives, using incandescent lighting, we will need a number of lights in the 5 to 10kw region... the type used in TV and film studios, as well as a similar shooting environment. Since these requirements are out of reach for the average photographer, the only direction he (she) may take is to begin to degrade the initial objectives...faster film, lower apertures, and slower shutter speeds.
The still photographer, however, has one big ace in the hole, with respect to the TV/movie studio. Instead of requiring continuous lighting, we are only concerned with catching instantaneous slices of the action, and need light only during the brief periods of exposure. Let us, then, suppose we went back to our hypothetical set up and turned on the 8000 watt lamp only for the 1/60 second we actually needed it for exposure. We would now need an amount of power equal to 8000/60, or 133 watt/seconds to make each exposure. By thinking in terms of amounts, rather than intensities, we could retrace our mathematics and see that 133 watt/seconds of electrical power fed to the incandescent bulb will produce the 32 Footcandle Seconds of required scene illumination. Now, we could achieve the 133 watt/second amount in a variety of ways: A 133 watt bulb could be turned on for one second, for instance, but we would be back to a situation of movement blur due to the long exposure time. More ideally, we could use a 133,000 watt bulb, and turn it on for a scant 1/1000 second. Any way we combine wattage vs. time to come up with 133 watt/seconds, we find that a relatively small amount of actual power is needed...133 watts per exposure per second. There is really no need for special air conditioning and all that. The fly in the ointment, however, is that it is impractical to turn incandescent bulbs on and off quickly...they simply don’t respond, taking a second or more to warm up.
ELECTRONIC XENON FLASH SYSTEMS: Xenon flashtubes are small glass or quartz cavities filled with Xenon gas. In operation, a charge of electrical energy is ”trickled” into a storage capacitor(s), creating a high voltage ”chunk,” or amount, of energy. The capacitor terminals are connected directly across electrodes in the flashtube. Nothing happens until a ”trigger” pulse is applied to a third flashtube terminal. What happens then is exactly the same as what occurs in a bolt of lightning...the gas ionizes, forming nearly a short circuit across the capacitor. An avalanche of current results and the energy stored in the capacitor is dissipated in a brief flash of extreme intensity light. The duration of the flash discharge is a function of several design parameters, and may be made to be from around 1/100 to 1/100,000 second or so. After each flash the capacitor is recharged, the unit is ”recycled.”
There are several advantages to the Xenon flash system, over incandescent lighting, in addition to its ability to produce brief pulses of extreme intensity. Firstly, its efficiency in converting electrical energy to light can be nearly three times as good...to around 4 Candlepower Seconds per watt/second. Consequently, a 50ws flash can produce as much light as would require 133ws in an incandescent lamp. Secondly, the color spectrum produced is very close to the natural sunlight spectrum, is more consistent over the life of the tube, and varies little with power line voltages or power settings. Flashtube life can be made quite well, typically from 2,500 to 100,000 flashes.
In short, a rather smallish Xenon flash of around 50ws can fully meet all the hypothetical requirements set forth above, with the added advantage of exposure times in the ideal 1/500 to 1/1000 second range for essential elimination of movement blur. The disadvantages are very small when compared to the tremendous advantages. They are:
1. the photographer must wait for recycle between shots
2. visual monitoring of the lighting pattern is difficult since no light is present, except during actual exposure
BATTERY POWERED, ON-CAMERA FLASHES: We are all somewhat familiar with these small, camera-mounted units. Their effectiveness is obvious from their proliferation. The better flashes are typically rated in the general region of 50ws and are thus quite capable of meeting the criteria set forth. Most units contain computational circuitry to automatically adjust flashpower as required for automatic control of exposure with varying distances and apertures. In most cases, the flash duration is in the general area of 1/1000 second (shorter when operated at less than full power). It is usual to set the camera exposure time to 1/60 or similar ”flash synchronization” setting provided by the camera manufacturer. Unless there are extreme amounts of ambient light present, such as sunlight, the actual exposure time and action stopping capability are determined by flash duration, not shutter speed. The reason the camera shutter is set to a slower speed is to assure that mechanical lags in the shutter mechanism do not allow a situation where the shutter is only partially open at the instant of flash. Cameras with ”leaf” shutters generally allow for shorter shutter times than do focal plane shutters, when used to synchronize to electronic flash.
Battery powered units are usually designed to mount on the camera, and will concentrate with a ”normal” lens. Outside this area, light output will usually fall off rapidly, requiring a diffuser or wide-angle attachment when wide-angle lenses are used. Typical flashes will recycle to full power in from 8 to 15 seconds...faster at reduced power settings and slower when batteries become low. No provision is made for previewing the light pattern for visual observation of lighting patterns and effect. Units of this sort are extremely popular, and can provide the rough equivalent of some 10,000 watts of incandescent light when used under the conditions for which they were designed. There are, however, certain limitations that crop up whenever the photographer takes such a flash off the camera, and begins to explore studio lighting techniques. The most pervasive problem relates to the lack of preview capability. When the flash is mounted on-camera, there is little need for previewing, since the flash will invariably illuminate the entire frame (with ”normal” lens). About all the photographer needs to be concerned with is the ratio of light at the front of the scene vs. the back. The ”straight-on” lighting produced by on-axis lighting, however, does not lead to very creative or warm pictures, for reasons we will discuss later. As soon as the flash is taken off the camera for purposes of more pleasing lighting effects, the user has only his imagination, and experience, to tell him what the illumination pattern is going to look like on film. This problem is worse when the pattern projected by most portable units is considered. The rapid light fall-off outside the primary light angle may now very well fall within the scene, causing spotty lighting and unexpected dark areas. Again, preview capability is needed. Finally, there is the near-point-source nature of light emanating from a very small reflector. As soon as the flash is taken off-camera, shadows begin to form on the subject. While it is, in fact, these shadows that break up the stark effect of head-on lighting, the shadows produced by point-source lighting are extremely well defined and the result is a hard ”mug-shot” appearance.
The experienced photographer can overcome many of these problems by the careful use of diffusion techniques such as bounce lighting and umbrellas. Such techniques, however, tend to reduce the available light on the scene, and somewhat higher flashpower is indicated. Even with the best of techniques, the problem of previewing remains, rendering battery portables a rather hit or miss way of composing and shooting creative pictures. Of course, the relatively long recycle rates and the need to replace batteries regularly cause additional inconvenience.
STUDIO FLASH SYSTEMS: The design of flash systems, intended primarily for studio usage, considers all of the factors mentioned above, in addition to others. One initial consideration is the inclusion of some form of continuous light source for previewing, or ”modeling” the illumination pattern of the actual flash section. Larger reflectors are generally used, and they are usually less directional and do not cut off the light pattern as sharply outside the angle of primary illumination. Because of these ”softer” reflectors, the BCPS rating of a good studio flash may well be lower than for a similarly powered portable unit. This does not indicate less efficiency or less output, it simply indicates the light output is made to cover a wide area. It should be noted that the more concentrated the reflector, the greater the loss incurred when diffusion techniques are used. Thus, an exceedingly high BCPS rating (or Guide Number) may be deceiving, as diffusion techniques may reveal a much lower actual flashpower.
While suffering somewhat less light loss through diffusion, most studio flash systems will offer higher actual flashpower than do battery portables. Studio flashes are usually used in multiple: two, three, four or more units, depending on the objectives and budget. A typical setup for the type of work described here may have three or so light units with a total flashpower around 400 ws. Much larger systems, even 20,000ws or more, are available for large scene commercial requirements covered in this paper.
Typical recycle rates for studio systems may range from around two to four seconds or so, at full power. Reduced power settings are normally provided so that the balance of light and shadow in the scene may be adjusted by varying the power of the individual light units.
MODELING LAMP SYSTEMS: Ideally, the modeling, or preview, lamp system will provide the user with an absolute preview of the lighting effect which will expose the film under flash illumination. In order to accomplish this, there must be an exacting correlation between flash and model light in a number of parameters. The modeling lamp must be mounted in the same reflector as the flashtube, and must be positioned such that the projected pattern from both sources is nearly identical. The degree of pattern correlation varies from design to design, and is a difficult parameter to control since two separate light sources should, ideally, be placed at the exact focal point in the reflector. Most units will project a somewhat broader beam during modeling, with respect to the flash pattern. With careful design, the two patterns may be made close enough to allow confident previewing. If gross errors exist in pattern correlation, which they do in certain systems, the result can be hot spots or dark areas on film, which were not seen under modeling light.
When multiple light units are used, it is of prime concern that a constant ratio of modeling intensity vs. flashpower exist on all units, and that modeling lamp intensity is proportional to the various flashpowers which might be selected on each unit. Only then can the light ratios be effectively judged under modeling lamp illumination.
The new user of studio flash systems may be concerned that the much lower color temperature (yellow cast) of the incandescent modeling lamps vs. flash might influence the color balance on the exposure. This is not a real concern, since the amount of light emitted by the flash is usually 1000 or more times that of the modeling lamps. Thus, the modeling lamp is effectively not seen by the film, and its color temperature difference has no effect on film.
CENTRAL POWER SUPPLY VS. SELF CONTAINED TYPES: There are two basic philosophies in the structure of studio flash systems. Each has its own advantages and disadvantages. In a central supply type system, the charging circuitry and storage capacitors are contained in a separate, central power supply. Several lightheads may be attached to the power supply with relatively heavy cables. Each lighthead contains only the flashtube and triggering circuitry, and a modeling lamp. The advantage to this sort of system is that the lightheads may be made more compact and lighterweight. This is a particularly valid concern when extreme amounts of flashpower are to be produced, as the weight and mass of high power charging circuits and capacitors would prove unwieldy if located in the lightheads. There are several disadvantages to the central approach though, not the least of which is the possibility of failure of the entire system should the power supply fail. There are restrictions on interconnect cable length and extendibility, since these cables have to carry the extremely high currents encountered in the flash discharge process. There is a certain amount of flashpower lost in these cables, thus system efficiency is somewhat impaired. Since a central system is a ”mix and match” sort of structure, with the user deciding which types of lightheads to use and how many to connect, the system invariably operates without an optimum match between flashtubes and power supply. This lack of optimum matching can invite excessively short flash duration, lower efficiency and possible early failure of the components. The establishment of a variety of power ratios between the multiple lightheads is often limited, while the maintenance of an exact ratio of model intensity to flashpower is nearly impossible.
Recent trends appear to favor the self-contained studio flash, particularly in the moderate power ranges of up to 400ws or so. With effective power supply engineering, fully self-contained units can be made quite compact and lightweight. In a self-contained studio flash, the integral power supply is specifically matched to the flashtube. The result can be higher efficiencies (more light per watt/second), controlled flash durations, greater accuracy in the correlation of flash to model system, and a more versatile selection of power ratios. Since each unit simply plugs into the nearest AC power outlet, there is no restraint on how far apart the individual units may be placed. Only the relatively low charging currents and model lamp currents flow through the power cord, so extension is easily made via ordinary household extension cords. Of course, each unit needs to be synchronized to the camera shutter so they all fire at the same time. With self-contained units, this is normally accomplished by using a sync cord on only one unit, and light sensitive ”slave trippers” on the remaining units. From the standpoint of reliability, a system made up of a number of self-contained units is essentially failsafe since, at worst, one unit may fail, leaving the remainder of the system operational.
DETERMINING EFFECTIVE FLASHPOWER AND EXPOSURE
There are three primary methods manufacturers of flash equipment use to specify power. They are: wattsecond ratings, Guide Numbers and BCPS. Unfortunately, none of the three methods definitely describes how much actual flashpower is present. Taking them one at a time, we begin with wattsecond ratings. Wattseconds defines the amount of electrical power supplied to the flashtube(s)...not the light energy which the tubes emit. There are several design factors that affect how efficiently the flashtube converts electrical energy into useable light. These same factors also affect the flash duration. The same parameters which tend to decrease system efficiency also work to shorten flash duration...potentially to the point of inviting reciprocity induced color errors. Specifically, the use of high values of flashtube current with lower flash voltages results in the longer flash durations (1/500 to 1/1000 second) and the higher conversion efficiencies (approaching 4cps per ws). Inversely, systems using lower currents but higher voltages tend to be less efficient and produce shorter flash durations. A second major consideration is the ”loading” of the flashtube. “Loading” simply describes how much power is fed to a flashtube, relative to the amount of power it was designed to dissipate. Thus, when a 200ws flashtube is operated at 200ws, it is ”optimally loaded,” and produces relatively long flash durations and high efficiency. If it is fed only 20ws, it is ”lightly loaded” and will produce very short flash durations (perhaps 1/5000 sec.) and fewer cps per ws (less efficiency). If it were operated at 400 ws, it would be overloaded, and subject to early failure. A third potential cause of lowered efficiency is a loss of power in the connecting wires between power supply and flashtube. Since cable losses are a function of cable length and the flash current (not voltage) most all central power supply type systems employ high voltage/low current circuitry to allow reasonable lengths of interconnect cable, of manageable diameter, at minimum power loss in the cables. A further consideration in central systems is the fact that flashtubes are generally rather ”lightly loaded”; to allow various sorts of interconnection without danger of overload. Consider an extreme case, with a 400ws power supply fitted with four 2400ws flashtubes, operated at 1/2 power (200ws). Each tube, though designed for 2400ws, receives only 50ws. While the tubes will practically last forever, the system will be very inefficient, and will produce exceedingly short durations, possibly less than 1/8000 sec.
The design of most self-contained systems, be they battery ”on-camera” portables, or studio flash units, will employ low voltage/high current type power supplies, and will operate the flashtubes somewhere near the point of optimum loading. There are not any long interconnect cables (the AC power cable does not carry actual flash current, only the much lower charging currents, and is not a factor in efficiency or generally optimum flash durations). In the case of units with very small fractional power setting, such as 1/8 or 1/16 power efficiencies and flash durations will be reduced.
Thus, it is seen that the wattsecond specification serves only as a general guide to real useable power. It can be expected that most self-contained systems will produce light energy at the rate of near 4cps per ws, while central powered system can vary substantially with configuration, typically in the range of 2.5 to 3.5 cps/ws.
Even considering the variables mentioned, it will be seen momentarily that the wattsecond rating (when specified) is by far a better determinant of useable flashpower than either the BCPS rating or Guide Number.
BCPS RATINGS: While this rating, indeed, defines the amount of light actually projected by the system, it does not relate with any definable accuracy to the amount of light produced by the system. As you will remember, the BCPS rating includes the magnifying effect of the reflector. The reflector does not actually increase the amount of light; it simply concentrates it in a smaller field. Conceivably, a designer could place an extremely high gain reflector on a very low power flashtube and come up with an extraordinarily high BCPS rating by concentrating all the available light into a dime sized spot on the wall. The spot would be very intense, but would serve little photographic purpose. If this hypothetical unit were ”bounced” or fed into or through an umbrella, the light would spread, covering a wider field, and reveal the true low power nature of the system. When you consider that the primary use mode of battery portable flash units is direct...without diffusion, the BCPS rating makes perfect sense: it indicates exactly how much light will fall on the subject. In the studio, where the norm is the use of diffusion and light pattern modification, the BCPS rating means little, except when coupled with accurate data regarding projected beam width. Potential buyers of studio flash systems will sometimes compare BCPS ratings (or Guide Numbers) with those of their portable equipment and say ”Gee, my little XYZ has as much BCPS or as high a GN.” It must be realized that portables generally project a narrower pattern (have higher gain reflectors) than do most studio units. Consequently, they will usually suffer considerably more loss when used with diffusion, as you cannot compare the BCPS or GN when bounce or diffusion is to be used. By the same token, look out for studio units which specify inordinately high BCPS or GN, relative to other units of similar watt/second rating...this is a sure sign they have narrow beam reflectors and may produce hot spots as well as a considerable loss when used with diffusion.
GUIDE NUMBERS: Actually the ”Guide Number” is no more, or less, than another way of stating BCPS. The Guide number tells you what aperture setting you will need on the camera for a given light to subject distance at a given film speed. For example, if you are using an ASA/ISO 100 film, you would use the ASA 100 GN, and divide this number by the distance from light to subject to find the correct lens aperture. If, for example, the ASA 100 GN for your flash was 110, and your subject was positioned 10 feet away from the flash, you would use an f11 aperture. By looking at an appropriate chart, you could also find that an ASA 100 GN of 110 is simply another way of stating a rating of 2800BCPS. Like the BCPS rating, the Guide Number only has relevance when the flash is used direct, without diffusion.
STUDIO FLASH EXPOSURE DETERMINATION: It is impractical to use automatic exposure methods with studio flash setups, because of the myriad of parameters involved. Essentially all professional studio flash systems are entirely manual for this reason. It is also impractical to rely on BCPS ratings or Guide Numbers for the aforementioned reasons. To be sure, you can use one studio flash direct and pretty much rely on its BCPS or GN, but when you use three or four units with umbrellas and other devices, you would have to do an awful lot of calculating. For these reasons, the most accepted method of determining correct exposure aperture settings, in a studio setup, is by the use of a flash meter. Once the lights have been set up and the scene composed, the photographer will ”test flash” the system and use the flash meter to measure the total amount of light reaching the subject or the camera from the aggregate of all flash units. The flash meter will either directly, or through dial settings, indicate the aperture to which the camera lens should be set. Flash meters often allow the user the choice of measuring either incident light or reflected light. When measuring incident light, a hemispherical diffuser dome is normally placed in front of the meter’s sensing element, where it gathers light from a very broad angle, as will the subject. The meter is then placed at the primary position of the subject, and aimed toward the camera. The incident method thus measures the amount of light that falls upon the subject. In the reflective measurement mode, the hemispherical diffuser is removed, and the meter measures light reflected off the subject in the same manner that the exposure meter built into most SLR cameras does. In using the flash meter in the reflected light mode, the photographer can make spot measurements of the amount of light reflected off various parts of the subject by aiming the meter at the various spots instead of at the camera. In this manner, he may determine the contrast, or range of light intensities contained in the scene to be photographed. Whether one uses incident or reflective measurements is purely a matter of personal taste and objective. For most uses, the incident method is faster and simpler, but requires the photographer to make mental notes as to the presence of any very light or very dark areas in the scene which might suggest a somewhat different exposure than indicated. For example, if you make an incident measurement of a wedding scene involving a white silk dress, you would want to close the lens to a somewhat higher aperture than indicated by the meter to avoid overexposure and loss of detail in the white expanse. For reasons to be discussed later, you are liable to find a significant disagreement in reading in flash meters (and also exposure meters) from one manufacturer to another. It is not unusual to walk into a camera store and try a number of meters, and come up with as much as two f-stops of difference in reading the same scene. Therefore, it is paramount that, for excellent results, you test your equipment with trial exposures and bracketing before tackling an important assignment.
METERING MODELING LAMPS TO DETERMINE FLASH EXPOSURE: On a very few studio flash systems, there exists a sufficient degree of accuracy in the correlation of modeling system to flash system to a low accuracy setting of the aperture based on reading the intensity of the modeling lamps. In order to do this, there must be an exact system ratio of modeling intensity to flash amount (BCP to BCPS, or CP to CPS), and the projected patterns from the two light sources must be essentially the same. If these criteria are met and specified by the manufacturer, it is then possible to use either the exposure meter in the camera, or a hand held exposure meter, to measure the modeling illumination intensity, then add in a specified correction figure to establish correct flash aperture. While this method relies heavily on the accuracy of the lighting equipment, it is possible to get very good results when conditions are right. Once more, it is wise to pretest the equipment and setup, and to bracket exposures.
FILM RELATED EXPOSURE ADJUSTMENTS: Most photographers will, from time to time, use both negative film for prints and positive film for transparencies. Unfortunately, there is a fundamental difference in how the two films respond to over and underexposure. With negative film, the result of overexposure is far less consequential than that of underexposure. When a negative film is underexposed, the shadow areas can turn the film completely clear, thus saturating and losing all detail. Inversely, overexposing a positive film can clear, or saturate, on highlights. The result of approaching saturation at the other end of the film spectrum (overexposing negative film or underexposing positive film) is far less damaging to the end product. Therefore, the rule of thumb for best results is to lean toward overexposure when using negative film (expose for shadow detail) and toward underexposure when using positive film (expose for highlight detail). This philosophy holds true whether continuous light or flash is employed. It is, no doubt, out of this relationship that the vast majority of exposure meters and flash meters are found to be inaccurate, as they will under-read the light, leading to overexposure of the film. The logic of this statement can be found by thinking strictly in amateur ”snapshot” terms: slight overexposure will insure the best overall results when negative films are used, while it will produce ”bright,” though often burned out, slides. The average tourist type slide show gets more praise for the nice bright slides than it gets boos for the loss of highlight detail. The photographer who invests his time and money on studio equipment and shooting, however, cannot tolerate the washout of overexposed slides, or the uncertainty of inaccurate metering equipment. Thus, it is recommended that, in addition to doing tests to determine the final results on film and bracketing exposures, the user attempt to verify the accuracy of his metering equipment by comparison with other equipment, and by seeking out equipment whose manufacturer has specified the accuracy.
Once you are sure of your equipment, and have learned to ”read” how a scene will go on whatever type of film you are using, you’ll be well on the way toward great pictures with less bracketing.
CONTROLLING DIFFUSION, SPECULARITY, SHADOW AND CONTRAST
There are fundamental properties of lighting that form the basis of our value judgments concerning its effectiveness. Many of these properties are not understood fully by the novice photographer. Let’s begin with contrast.
Contrast is the degree of separation between light tones and dark tones. In the basic context of defining contrast, we might think of a picture of a black and white striped surface as being ”high contrast.” While this is the correct line of thinking, there is a far more complex relationship to lighting and subject matter and contrast than initially comes to mind. Consider as a model, for instance, a brown skinned person in a brown suit with a brown dog against a brown background. The amateur might think the contrast of this scene is established by the subject coloration, while the professional realizes the same scene may be shot with almost any degree of contrast desired, by manipulation of the lighting equipment. Besides the obvious placement of lights to form actual gradients, or variations of light intensity across selected parts of the scene, the surface textures of the various scene components can play an extremely important part in how the scene looks under certain lighting conditions. In order that we fully understand lighting and texture and contrast, let us conduct a mental experiment:
We shall use two light sources, each producing, say, 100 candlepower. Light source #1 is essentially a point source of light, measuring, say, 1/4” in diameter. When we look at #1, it appears extremely bright and uncomfortable to look at. It appears bright since all 100CP are emanating from a very small surface area, having a very high ”spot intensity.” For light source #2, we will place a 100 CP lamp inside a huge sphere of diffusing material (white cloth, frosted glass, etc.), thus producing a very large light source, perhaps 10 feet in diameter. When we look at #2, even though it too produces 100 CP, it does not appear nearly as bright as #1, nor is it uncomfortable to look at. Its spot intensity is extremely low with respect to #1, though its actual intensity is the same. It is a large ”soft” light source. Now, imagine a room. One wall is covered with silver colored felt while the opposite wall is silver surfaced glass (a mirror). If we look at the felt wall, it looks exactly the same regardless of which light source is used. In either case it is illuminated by 100CP and there are no surface reflections to give us a clue to the size or shape of the light source. If we look at the mirrored wall, we see something entirely different. If we turn on light #1, we see the same image as we would by looking directly at the light: a very intense spot of light. If we photographed this, we would have an extreme contrast film, with a bright (probably overexposed) spot, surrounded by a dark background. On the other hand, if we viewed or photographed the mirrored wall using light #2, we would see a much lower contrast with a large soft glow of much lower spot intensity.
From this experiment, we can see that the surface texture has a great deal of bearing on things. While both walls were silver in color, the flat textured felt appeared the same under either point-source or highly diffused light, while the mirrored (specular) surface looked entirely different under the two lighting conditions.
If we replace the two walls with more common subjects such as the ”all brown” scene, we will find that each surface, though brown, has its own degree of specularity, or surface reflectivity. The man’s face may have small beads of sweat, for instance, which form tiny mirrors. The brown shoes may be highly polished, the laces flat, the suit may have some shiny polyester threads, etc., etc. Given a certain level of scene illumination, the picture will appear entirely different depending upon the spot intensity and source size of the light sources...The smaller the sources (higher spot intensity) the more intense the reflections from the ”specular,” or reflective, bits of the scene. Where the specular areas are relatively large and flat, such as the polished shoes, we will actually see a reflection of the light. Where the specularity resides in the much smaller textures, such as in the polyester threads, we will simply see glistening highlights. In any events, the highest possible scene contrast will appear as the lights approach point-source, while lower contrast and less surface definition will prevail under conditions of high light diffusion. The degree of contrast we wish to convey via lighting is entirely governed by the subject matter and artistic goals. Too high a contrast can result in pictures that are impossible to print, perhaps with glare and washed out areas, or emphasized undesirable features such as wrinkles. Too low a contrast, caused by excessive diffusion, can lead to dull, uninteresting and undefined pictures. The answer is to first understand the principles involved, and learn the range of contrast that will ”print” well, to know what it is you’re after. You can then use various combinations of diffused and direct light to achieve your goals.
SHADOW STRUCTURE AND CONTRAST: The degree of light diffusion affects scene contrast in another way...via the introduction of shadows. Again using hypothetical light sources #1 and #2, as well as the silver colored felt wall, let us conduct another experiment: This time, let’s place a new object, say a person, in front of the wall, four feet away. Using light #1, all light that falls on the scene comes from a single point. Wherever there is an obstruction, a distinct black shadow is formed. The person’s shadow appears on the wall, and dark shadows form under the chin, behind the nose, etc. Bright specular highlights also appear wherever shiny areas exist on the subject. The contrast is very high, when the brightness of the highlights is compared to the blackness of the shadows. Wrinkles once more become pronounced, this time due to the dark shadows which form with each ”hill and valley” of the skin. If we now substitute light #2, the shadows nearly disappear, since light is coming from many angles and ”wraps around” any surface obstructions. Since there are now no deep dark shadows, the ratio of highlight intensity to shadow intensity is much lower...lower contrast. Again, but this time because of shadow structure, too small a light source produces excess contrast (a hard look), while too large a source produces a bland featureless picture.
Thus, while there are two distinctly different mechanisms working to associate contrast with size of the light source(s), they both work in the same general direction. To illustrate that there are, in fact, two separate parameters, and to show that they can be controlled independently, let us set up a third hypothetical lighting situation:
This time, assume there are 10 point-source lights, each producing 10CP, arranged in a broad array about 10 feet square, in front of the subject. From the standpoint of shadow formation, there will be essentially as much ”wrap around” effect as would be obtained with a 10’ diffusion panel. Since the light will come at the subject from a broad range of angles, there will be no deep black shadows, rather a multiplicity of distinct shadows which will blend to form a soft overall shadow structure. As far as specularity is concerned, while each 10CP light has only 1/10 the spot intensity of a single 100CP light, the spot intensities are still 100 times or more greater than would be evident with a 10 foot diffuser. Each light is quite capable of producing strong reflections from shiny components of the subject. What’s more, when we are dealing with surfaces containing a large number of tiny reflective particles (perspiration, sequins, metallic threads, or generally reflective surfaces of angular form such as lips, eyes, small objects), the multiplicity of small ”pin” lights will produce 10 times as many specular ”glints” as would one point-source light. Thus, while the reflections are not as intense, there will be more of them. The net result of this lighting scheme should be a picture which has a high apparent contrast and specularity, yet which is much easier to print than a picture made with a single point-source light. Notice, the parameter, which causes the relatively high contrast, has been isolated to the reflective component, while the shadow structure has been dealt with separately. This example is quite theoretical and does not represent a very common studio light setup. It would seem, however, to be an interesting lighting method to explore, particularly in lighting for high impact presentations of fashions and products.
TRIED AND TRUE LIGHTING SETUPS: While the author advocates individual approaches to lighting, there are certain techniques which are commonly used, and which have passed the test of time. These ”standard” setups are particularly ubiquitous in portrait photography.
The objectives are usually to present the subject in a light and setting which brings out the best qualities of the subject’s appearance. The first general rule is to avoid drawing the viewer’s interest to scene components secondary to the subject, such as background and miscellaneous objects. The rank amateur’s first urge invariably seems to be a desire to place something ”real exciting” behind the subject, like a red car, a patchwork quilt, or some zany wallpaper. Presumably, the reader is above all this and is well aware of the result. The most universal background for ”people photography” by professionals is seamless background paper. It comes in 6’ or 9’ rolls in a variety of colors. While certain photographers might say ”Oh no, not seamless paper again,” the truth is it makes a damn good background, especially with good lighting technique. If you look through a good fashion catalog, you will see shot after shot using the same neutral colored seamless paper background, and never tire of it. You are too busy looking at the models to notice the background. In another catalog, you may see a more contrived set used over and over. This is the one you’ll remember, because you’ll remember how boring it was, when used over again. Selected props, used sparingly, can help convey the mood you’re after without distracting. You could occasionally use tables, houseplants, or a bar. The secret is to look at it and think, ”Is this really doing something for the picture, or is it just filling up space?” If it has nothing important to say, get rid of it. You’ll be glad later.
Ok, we’ve got our background and the scene components, what about lighting? There are four basic light techniques that are usually mixed to light the scene. By name, they are the ”Key light,” ”Fill light,” ”Background light” and ”Back light.” The key light is normally placed in such a position that it is, visually, the predominant source of light. It is commonly placed at about 45 degrees to one side or the other of the subject, rarely head on. This off-angle placement encourages shadows to form, shadows that define the shape of the subject. Any degree of diffusion that suits the objective may be used on the key light (or ”Main light). In general, if the subject is young and fashionable, and you are after a dramatic presentation, small amounts of diffusion (or none at all) might be in order. On the other hand, if you are after a softer, more romantic, or more serious feel, greater amounts of diffusion are in order. The old standby white umbrella, either the ”bounce” type or the ”shoot through” type, makes an excellent large source diffuser for these purposes. Remember, though, the closer the diffuser is to the subject, the greater the effective amount of diffusion and light ”wrap around.” Don’t be afraid to place a shoot-through umbrella extremely close to the subject...even a few inches, to get the effect you’re after. Remembering that light falls off according to the square of the light to subject distance, you will find that close light placements result in rather strong gradients of intensity across the subject. On a single subject, this effect can be a point of interest and can create a very personal sort of mood. With groups of people, more distant lighting is indicated, in order to achieve a more or less uniform illumination of all subjects.
When making mental judgments of the effectiveness of the main lighting, as viewed with modeling lamps, you must remember that the tonal range which can be captured on film and printed is much narrower than what you see with your eyes. This particularly applies when using positive transparency film. What appears to the eyes as a nice contrast of light and shadow is apt to come out as over exposed highlights and formless black shadows. Therefore, when previewing the scene, look for the visual effect of light and shadow, but think in terms of a significant increase in contrast in the final pictures. This is where the second basic light source...the fill light, comes into play. In most cases, the fill light will be quite diffuse, possibly using a large white umbrella or a ”bounce” off an adjacent wall. Its historic purpose is to introduce sufficient light from the side opposite the key light, to lighten the shadows to pleasing (and printable) proportions. It will usually be less powerful than the key light so as to not appear as a primary source of perceived illumination. The term ”fill” means what is implied...to fill in.
In most portraiture, the subject will be placed sufficiently away from the background to preclude subject shadows from being cast on the background paper. With front lights at an angle towards, and slightly above the subject, the shadows will fall below and to the sides. In any event, the front lights (key and fill) will not cause much illumination of the background, nor should they. Without some form of rear lighting, though, the subject can tend to appear as being in front of a black nothingness, and to have little visual dimension of depth. The common use of rear lights, then, is to illuminate the background to the degree desired, and to separate the subject from the background, giving the sensation of depth. The ”background light,” as its name implies, is used to light up the background. Using a seamless paper, the photographer has a wonderful opportunity here to use lighting to create a background that does not actually exist in the room. Instead of just lighting the paper, good photographers will ”paint” a scene upon the paper with background lights. The simplest way of doing this is to place one light between subject and background, to one side of the subject, with the light aimed at the background. By altering the position, elevation, angle, and degree of diffusion, a strong or slight gradient of light to dark may be impressed on the background, forming a frame or halo, which enhances the subject. Careful visual observation of the overall effect can lead you to proper positioning and structure of the background lighting. For instance, if some part of the subject is dark, or weakly lit by the front lights, this is a good area to have substantial illumination of the background. By contrast (and for contrast), light colored or brightly lit portions of the subject can be brought out by lower light levels on the corresponding portion of the background. Many other interesting effects can be obtained with background lighting. Colored filters may be used to project background hues. Multiple lights and filters can produce very interesting backgrounds. Special projection lenses may be fitted such that actual artwork is projected (soft, unfocused color swirls and the like can be very effective). As with actual objects in the set, use your eyes to decide what is effective and what is contrived.
The second type of rear lighting commonly used is the ”back light.” This light will be used at some position behind the subject, but will be pointed at the subject, to light it from behind. Back lighting can work wonders to emphasize the subject and to visually separate it from the background, particularly on hair and fibrous clothing. When back lighting is specifically directed to the hair, it is called ”hair lighting,” and can offer a spectacular effect. For example, a dark haired model against a dark, or even neutral, background can appear dull and ill defined with only front lighting. Once the hair is lit from the back, the individual strands sparkle with light. The definition is enhanced, as is the sensation of depth. Back lighting can also be very effective as an apparent main source of light...an example being the sort of silhouette effect that occurs when sunlight falls over a person’s shoulder. In the studio, interesting shots may be made by placing a backlight at around a 45-degree angle behind the subject and then adjusting a front fill light to achieve not a silhouette, but the hint of one. Very dramatic effects can be obtained by using rather pronounced coloration on back lights, such as an intense blue filtered light streaming through black hair. ”Party” colors from behind can introduce a feeling of nightlife. Yellows and ambers can hint of sundown. It is often useful to employ partial light blocking on back lights to prevent spill, or unwanted light, onto the foreground, or into the camera lens where it can cause flare, or ”fogging.” Snoots, barndoors, gobos or other light shaping obstructions can be used.
Certainly, what has been presented here is not a complete study of lighting techniques. If you have interest in learning more, there are any number of good books available on these subjects. Hopefully, what has been presented will instill a rudimentary understanding of how light behaves, and how it affects the visual impact of the scene it illuminates. Most of all, the intention of this paper is to convey the point that desire and practice are instrumental in success. The better a photographer understands the mechanics of the medium, and the more he (she) is willing to put that understanding in practice through trial and error and educated experimentation, the more likely it is the world will see photographic art, instead of just a bunch of pictures.
Lighting for Digital addendum
There are common misconceptions that digital photography requires less light than film, and that continuous lighting is preferred over flash for digital cameras. The fact is, digital cameras have the same lighting requirements as film cameras, unless one is willing to settle for lower image quality from the digital medium. For example, if you wish to shoot a scene using flash with a digital camera set for an aperture of f8 and a film speed setting of ISO 100, you will need exactly the same amount of flashpower that you would for a film camera with these settings.
Of course you can easily reduce the amount of light required by the digital camera by increasing the sensitivity to ISO 200, 400 or even 3200. But just like in a film camera, using a higher ISO results in lower quality exposures, with more noise and grain.
As for using continuous light sources instead of flash, the trade-offs are the same. Low light levels require long exposure times and low aperture numbers.
The cost is also the same for digital or film: subject movement blur, camera movement blur and shallow depth of field.
The real root of the digital lighting myth is the fact that most digital cameras simply do not provide the higher aperture settings and manual adjustability needed for high resolution professional shooting, nor can they typically handle the high lighting levels required for such shots. But this is changing rapidly as "high end" digital cameras with SLR style lenses and full manual capability come down into the price range of similarly equipped film cameras.