Light balancing in CLS multi-flash photography

[chuck]

<p>Say I have two SB-800s in remote groups A and B, placed on either side of the subject. I set flash compensation to 0 for both groups on the command flash. Does this mean both SB-800s will put out light at the same rate, so that I can adjust how much illumination falls on the subject from either side by moving one flash closer to the subject and another further away? Or does this mean regardless of how far each flash is from the subject, the I-TTL system will try to make the illumination falling on the subject from either flash group be equal?<br /> in other words, in Nikon's remote flash control system, does adjusting each group's flash compensation value on the command flash adjust the output of each remote flash group? or does it adjust the resulting illumination from each remote flash group?</p>

[ilkka_nissila]

<p>The system analyses the impact of each flash group on the image separately by firing a set of preflashes prior to the exposure. It then selects such a flash energy for each group that achieves a balanced exposure (by default). If one group influences only a part of the image and another group affects another part of the image, then the system tries to bring all the parts of the image into balance with each other.</p>

<p>If you adjust flash compensation on a group to a nonzero value, it will change the impact of that flash group on the overall exposure or the region's exposure to which that group influences.</p>

<p>If you move the position of the flash closer or more distant position and have it in TTL mode then the overall exposure should not be impacted except shadows may change a bit and if you use a large diffuser then the apparent size of the light source will change (and it may also change the field of view of the illumination of each group). But the i-TTL system should keep the overall image brightness fixed as you make these position adjustments (but there can be subtle changes, "fixed" is not perhaps the right word with TTL flash ;-)).</p>

<p>This is useful as it compensates automatically for changes in the subject position relative to the flashes. Or tries to. TTL flash is tricksy and not entirely deterministic. ;-)</p>

[chuck]

Thank you Ilkka.

 

So I take it the flash exposure compensation for remote flashes really does adjust the actual flash illumination value, not

just the flash output level.

[ilkka_nissila]

<p>Yes.</p>

[craig_shearman1]

<p>IMHO, it's much easier to just put the flash units in manual. If I want them to have equal power I leave both on full power. If I want one to be a stop less than the other I put it on half power, or one quarter power if I want two stops less, etc. Set the exposure with a flash meter or chimp on the LCD and start shooting. Much easier IMHO than trying to figure out what TTL is doing.</p>

[rodeo_joe1]

<p>I'm with Craig. Use manual settings; it's easier in the long run. Then you know that your fill flash is, say, going to be two stops lower than the key light - instead of leaving it up to flakey I-TTL to decide how the lighting looks.</p>

[wayne_f1]

<p>>>Say I have two SB-800s in remote groups A and B, placed on either side of the subject. I set flash compensation to 0 for both groups on the command flash. Does this mean both SB-800s will put out light at the same rate, so that I can adjust how much illumination falls on the subject from either side by moving one flash closer to the subject and another further away? Or does this mean regardless of how far each flash is from the subject, the I-TTL system will try to make the illumination falling on the subject from either flash group be equal?</p>

<p>The latter. Don't bother trying to move the lights, that will be ignored. Speaking TTL with Commander, it does not matter how near or far one flash is, or if one is direct and one is bounced, or one is a big flash and one is a small flash, or whatever - the system meters the TTL preflashes individually, and (if possible) sets each power level to what is needed to give the same light on the subject from each group.</p>

<p>Normally portraits don't want both flashes the same, that would be a light too flat, uninteresting, too dull. However that idea is the equal base allowing us to compensation one (the fill light) to be say -1 EV to establish the known lighting ratio. This is the point. So set the fill light group to -1 EV compensation instead (in the first place), and the power level will be set up to do exactly that.</p>

<p>We can of course just throw two lights out there, the system does the same thing, but "either side of the camera" is not the best general plan for lighting. Normally the idea of portrait lighting is that the main light is placed high and wide from the subjects point of view, like 45 degrees high and wide, off camera so to speak, creating intentional shadows (on the face). Then the fill light is placed frontally, very near the lens axis (back near the lens so the lens can see around its umbrella), to be a frontal light. Frontal specifically means that it lights the shadows that the wide and high main light intentionally made, the same shadows that the lens sees, without making its own second set of shadows. So as compared to "either side of camera", the main is higher and wider (from subject), and the fill is closer and more central, at the camera.</p>

<p>The fill should be turned down about -1 EV, to partially fill and smooth the shadows, but still leaving smooth gradient tonal shading to show shape and curves of subject, and be pleasing lighting. The lighting ratio is very important. So the good plan is to set the Fill light group to about -1 EV compensation, so the system sets it to meter -1 EV at the subject. Some samples at http://www.scantips.com/lights/awl.html</p>