2. Continuous vs. Instantaneous light - vs. Shutter Speed
3. Soft light (and diffusion domes?)
4. Flash pictures are double exposures
5. Extras (Rear Curtain Sync, White Balance with Color Gels)
2. Continuous vs. Instantaneous light - vs. Shutter Speed
Continuous light is daylight, or incandescent, or fluorescent, etc. - always on, longer than shutter speed duration.
Flash is near-instantaneous light - only on for an extremely brief pulse, shorter than shutter speed duration.
This is a huge difference, and flash exposure works very differently than continuous light. Both light sources are often present, so we need to pay attention to both types of light, both sets of rules. Any flash picture often involves these two separate exposures (Part 4).
Continuous light works like we learned exposure works... a shutter speed two times faster (shorter duration) gives a result half as bright, requiring opening the aperture one stop to compensate. So shutter speed definitely affects continuous light, like we always understood it does. Continuous light lasts from long before the shutter, to long after the shutter, and the shutter speed simply takes a tiny time sample of its intensity. The amount of continuous light exposure we see depends on the shutter duration. The camera can only use the amount of light seen while the shutter is open.
But flash is not affected by shutter speed. Flash is a near-instantaneous pulse, possibly 1/1000 second duration or perhaps much shorter, but much faster than our shutter speed duration. Therefore, regardless if our shutter speed is 1/200 second, 1/60 second, or 1 full second, the flash does all it can do in perhaps 1/1000 second. The shutter merely must be open when the instantaneous flash pulse happens. It simply does not matter to the flash exposure how much longer the shutter might stay open after the flash finishes - the flash already finished long ago. The significance of this is that shutter speed is simply NOT any factor for normal flash exposure. Aperture is the factor affecting flash power level and exposure - and ISO and subject distance too of course, but not shutter speed. One big plus is that the very fast flash duration will stop motion extremely well. But any longer shutter duration does allow more continuous ambient room light to be seen, affecting total exposure (and which can still blur motion if the light is significant). We need to be aware that the rules are different for continuous light and flash, in this way. Shutter speed not affecting flash exposure is indeed something big to know. Part 4 here is more about using this fact.
Changing aperture or ISO changes flash exposure for Manual flash, but not for TTL automatic flash. Instead, TTL automatic flash simply changes power level to give the same metered exposure for the new situation. The way that digital TTL flash automation works is that we first set some aperture, maybe f/5.6. Then the automation triggers a low level preflash which the camera meters to judge the requirement, and then from that metered preflash, the flash power level adjusts automatically to be appropriate for that f/5.6 that we set previously. If we are using manual flash mode, then we set the flash power level manually, for the same aperture goal. Aperture is key for flash. The aperture affects flash TTL power level and maximum range and recycle time - and shutter speed does not affect flash - shutter speed only affects the continuous ambient light, room light or daylight.
Power differences (why continuous lamps are woefully underpowered for photography)
Sunlight is intensely strong, but continuous lamps indoors are relatively weak. Flash is typically strong - it may be a struggle, but at close range (like ten feet), flash can match the sun for fill in bright sunlight. Studio flash units are rated in watt seconds, which is the total of the input electrical energy (same concept as kilowatt hours on our electric bill). Continuous lamps are rated in watts (which is a rate), with no time period defined to determine total energy. So by definition, if we use a shutter speed time sample of say 1/100 second, then the total energy available to us from a 500 watt continuous lamp is (500 watts x 1/100 second) = 5 watt seconds. Which is not very much, because our 1/100 second shutter speed can only use a tiny fraction of the total power being expended (only while the shutter is actually open).
However in this case, a one second shutter speed could use (500 watts x 1 second) = 500 watt seconds, or 100 times more of the continuous energy. Which is a lot of power then, but the one second shutter is normally too slow to be useful for many of our pictures (but could be suitable for still life photos). Note that watt seconds is a measure of the electrical input energy. The amount of output light depends on the efficiency of the light when converting electricity to light. We tend to ignore this, but flash and fluorescent are a few times more efficient at making light than are tungsten filaments (incandescent).
In contrast to continuous lights, a small studio flash unit rated 160 watt seconds will give us the full 160 watt seconds of energy instantly, regardless of what our shutter speed may be. Note that 160 watt seconds is greatly more energy than the 5 watt seconds above. The overwhelming result is that if using the flash, we probably must turn its power level way down to use it for portraits (perhaps 1/8 power level at f/8 at ISO 200). But if using the 500 watt continuous light, we must place it up as close as the heat will allow, probably at ISO 400, f/2.8, and 1/30 second (as far as we can go in every direction), and hope for the best. This is a huge difference.
To better illustrate the point: Studio lights have incandescent modeling lights built in. 160 watt second lights may have 150 watt incandescent modeling lights, in same reflector, at the same distance. But if we unplug the sync cord to disable the flash, and take the same picture at the same f/8 aperture with just the modeling lights, we get a pitch black picture. There is no significant contribution from the modeling lights at f/8.
Even our camera speedlights (the SB-800 is 75 watt seconds equivalent if in an umbrella) will run circles around continuous lights for photography. As much as f/8 ISO 100, or f/11 ISO 200, is possible for portraits with speedlights in white umbrellas (full power with main light's reflected fabric at four feet). Battery recycle time will be slow at full power level however.
If shopping for studio lights, you should be aware that ISO 200 with 160 watt seconds will give the same exposure combinations that ISO 100 users see with 320 watt seconds. Both cases likely shoot f/8 portraits at about 1/8 power level. Or full power on both should allow f/8 for groups with white umbrellas at ten feet. That seems plenty of power, and more power can be an issue in the living room. What needs more power is greater distances (school gym maybe), or small apertures like f/22, or trying to overpower the sun outdoors.
Metering differences
Our camera lightmeters easily meter continuous light, but they do not meter flash. One exception is TTL flash, where we set aperture and the automatic TTL flash responds with the correct power level for that specified aperture. However, even then, the exposure shown to us by the camera light meter in the viewfinder is only about the continuous ambient light, and this reading is NOT about the flash at all (which has its own system - More in Part 4). The camera systems with the fancy TTL wireless remote flash features can operate a couple of their own flash units in an automatic way. For example, Nikons CLS Commander/Remote flash system is quite awesome, a point&shoot remote wireless multiple flash system, which instantly and automatically does the equivalent of several minutes of manual setup to equalize the two lights at the subject. Here is a quick look at it... We can also specify the lighting ratio between main and fill by simply specifying it, and it does it. It has many fans, but automation always gives up some control of course, and there are associated downsides, like the preflash that makes our subjects blink in the picture (using FV Lock is one solution).
But for manual flash, the only good solution is a handheld flash meter, like Sekonic. In a studio situation with multiple lights which work in manual flash mode, we meter each light individually to manually set its power level so that it does in fact meter what we want it to meter. This is big plus, a huge advantage called "control". We can set them exactly like we want them. For example, maybe we meter the main light (alone) at the subject to give f/8. We meter the fill light (alone) at the subject to give our lighting ratio, maybe to be one stop less if desired, or to meter f/5.6. We meter the background light at the background to give f/8, or whatever effect we want there. We meter the hair light at the subjects head, to give maybe f/11 for black hair (more light) or f/5.6 for light hair (less light), whatever we know we want. Then we also meter the main and fill light together to get the lens aperture setting (both together will be a fraction of one stop more than the main light, depending on lighting ratio). This is full control, consistent and repeatable, which we can easily do again when we setup next time.
The incident flash meter for manual flash has another advantage: The automatic camera TTL meter necessarily uses reflected light, which is dependent on the light reflected from the bright or dark color of the subject. For example, we get different readings if the subject's dress is black or white. But the incident manual flash meter points at the light instead of at the subject, and it measures the actual light intensity itself, which is totally independent of the subject, and frankly, is pretty awesome. Any subject will come out about right then.
Shutter Speed Sync differences
We can of course use any shutter speed with continuous light, like sunlight. Faster shutter speed does limit the amount of light seen, but we simply open the aperture for any other equivalent exposure.
Flash is different. Shutter speed may not affect flash exposure, but our camera has a maximum shutter sync speed for flash, in the ballpark of about 1/200 second for the focal plane shutters used on most DSLR. The shutter must be fully open when the flash fires, to expose the entire area of the photo frame in that instant. At faster shutter speeds, the focal plane shutter is never fully open all at once, it is only a narrow open slit moving across the frame. This means faster shutter speeds cannot be used for flash, or else we would get a dark unexposed band in our picture, where the total frame area was not open. The fastest shutter speed when the shutter is in fact 100% open all at one time to allow flash to go through it is the definition of "the maximum shutter sync speed" (see camera specs). It is a hard limit, but see the Electronic shutter and FP HSS sections below.
See info possibly useful to give the idea about the minimal hardware needed to use your camera speedlights in umbrellas (scroll down lower there).
Summary:
1. Flash is greatly affected by its inverse square distance, but sunlight is not affected by distance (on Earth).
2. Flash is not affected by shutter speed, but shutter speed only briefly samples continuous light.
Shutter speed simply does not affect the exposure from the flash. Aperture and ISO and flash power level affects flash exposure. Aperture and ISO and shutter speed affect continuous light exposure. See Part 4 here about using this difference. Note that we can turn the flash power up or down, but we have to seek some shade to affect the sunlight.
Ifs and buts - Exceeding maximum shutter sync speed
A couple of confusion factors which do not change the basics above:
If we had a faster type of shutter, like an electronic CCD chip shutter, which could sync flash at say 1/4000 second shutter speed (electronically enables and disables the CCD sensor, used as a shutter), then yes, it could be fast enough to truncate the quick flash duration, and then of course, a shorter shutter speed could cut off a longer flash exposure. However, this is simply not the normal case. The Nikon D40, D50, D70 cameras do use the CCD shutter, but these also limit the maximum shutter sync speed to 1/500 second. But if you use a PC sync cord (which is Manual flash mode only) to break communication so the camera does not know the flash is present (so firmware does not limit shutter speed), you can use any faster shutter second (on these models with this type of electronic shutter, indicated with 1/500 flash sync speed), which will still sync flash, which in some cases might truncate the flash duration and reduce the useful flash exposure. So things do change then - this case of flash is affected by shutter speed. It is just NOT a normal case.
It is very difficult to define how to measure actual flash duration. It is a fast pulse which then decays relatively slowly to zero, and it is very hard to agree when it effectively finishes. When does the gradual trail-off stop being effective? (10%? 5%? 1%?) The standard method for published flash duration is called t.5, which measures the time that the flash is stronger than 50% of its peak intensity. This is an engineering convention, convenient for engineers, but is not a photographer's convention. Because 50% intensity is only one stop down, and is still rather bright, so photographically, this means the t.5 specification number is actually about three times faster than we realistically see in our pictures - the useful 10% limit of the flash (called t.1, measuring 90% of the power) is about 3 times longer than the conventional 50% spec number for duration - BUT this limit of usability is really difficult to specify.
Camera speedlight units are a big exception, and are "different", being extremely fast at lower power levels (the name "speedlight"). Their 1/32 power level may have an actual duration of 1/20,000 second, because speedights reduce power by truncating the flash duration. This truncated pulse has very steep sides going up and down, and so t.5 really hardly applies, EXCEPT when at their full power level when t.5 very much does apply (just saying, the truncated lower power speed specs are more unequivocally precise regarding what actually happens). But otherwise, speaking of the speedlight full power level, then in general, a t.1 time is more meaningful for photographers, which duration is mathematically about three times slower than the t.5 stated by the spec, and so the full power duration may approach normal maximum shutter sync speeds closer than we may realize. If that may be not be clear, please see the the actual description at High Speed Flash.
Most cameras with interchangeable lenses, like film SLR, and more expensive DSLR models (like the Nikon D60, D80, D90, D200, D300) have a mechanical focal plane shutter, which is like an open slit moving across the sensor frame (to expose the frame). Not all of the frame is open to be exposed at any one instant, except at shutter speeds not exceeding maximum sync shutter speed, which by definition, is when the frame is fully open, to be able to be exposed by the instantaneous flash. If the camera incorrectly allowed using a shutter faster than maximum sync speed, we would see a dark unexposed band, where the moving slit was not fully open at the time. A shutter speed faster than its maximum sync speed simply cannot work right with flash. The realistic focal plane maximum shutter sync speed is typically around 1/200 second (a few are 1/250 second). But technically, the flash is even faster duration, with more motion-stopping ability than the shutter has, so long as there is little continuous ambient light to let the slower shutter blur the motion.
These focal plane cameras often offer a FP High Speed Sync mode (often named Auto FP), which in fact does allow shutter speed faster than the maximum sync speed possible. BTW, the terms FP High Speed Sync, or HSS, or Auto FP, are all the same in my vocabulary here. HSS is the feature, and Auto FP is the camera option that automatically switches it on for flash, but only when the shutter speed actually passes the maximum shutter sync speed.
This FP HSS mode does not modify the shutter action or the maximum sync speed, but instead it changes the physical flash unit, from being a "flash" to instead emit a continuous stream of flash pulses which mimic continuous light, continuously lasting for the duration of the shutter slit travel time (around 1/250 second for any fast shutter speed). So no matter where the 1/2000 second narrow shutter slit is in its travel over the frame, the flash is still outputting light then, and no dark bands occur. The flash has literally become a continuous light source, same as the sun, same as a table lamp, now on all the time during the exposure duration. But the flash unit only has finite power, so this continuous mode necessarily reduces maximum flash power to less than 1/4 power, reduced by at least two stops (which can only compete with sun at close distances, so possibly is insufficient power for some purposes then). But, this FP mode will allow a faster shutter speed with flash, which is used to achieve a wider aperture for fill flash in sunshine.
HSS is NOT High Speed Flash, it is merely High Speed Sync (simply because continuous light has no sync issues.) HSS is fully the opposite of High Speed Flash. HSS is like the sun, continuous, no motion stopping ability at all. In contrast, a SB-600 in real flash mode at 1/2 power level has a flash duration of only 1/1600 second (page 88, rear spec chart), and is much faster at lower power levels.
The Nikon SB-800 and SB-900 speedlight flash shows a distance range on its LCD (TTL Mode, flash head pointed straight ahead) which shows the maximum distance at the current settings, for the FP High Speed Sync option too.
Tricky: There are two descriptions here. My emphasis is NOT on this first part. This first part is camera Manual mode with aperture fixed at f/4 (as if indoors in insignificant ambient, when not concerned about matching ambient exposure) ... just for the dramatics of it. :) Note that we would be crazy to use FP HSS flash mode indoors this way (use regular flash mode indoors).
The SB-800 flash LCD shows the following maximum ranges, using settings of 50mm zoom, ISO 200 at f/4 and camera 1/250 Auto FP option, reported below: (f/4 is an arbitrary fixed constant here)
52 feet - ANY shutter speed not exceeding D300 1/250 maximum shutter sync speed (full power)
22 feet - 1/320 second (continuous FP mode kicks in above 1/250 second - reduced power)
20 feet - 1/400 second
14 feet - 1/800 second
10 feet - 1/1600 second
7.0 feet - 1/3200 second (still SAME flash power, but shutter speed decimates continuous light)
4.9 feet - 1/6400 second
True enough, but this approach seriously misguides our thinking. It seems terrible at first glance, but we really ought not to react with horror that 1/6400 shutter decimates the power of this light when at the same fixed aperture. Yes, sure, it does this, but so does sunlight, or any indoor continuous lights, which work EXACTLY the same way (continuous), and produce exactly the same kind of reduced numbers (with fixed aperture). 1/6400 second shutter always sees only 1/16 of the continuous light that 1/400 second samples (four stops down). What else did you expect? :)
It is true that instantaneous flash is not affected by shutter speed, but HSS is not flash anymore, this is a continuous light now. HSS appears to flash like flash, but it is continuous for the entire longer duration of the shutter travel time, allowing this flash with any fast shutter speed. For a fast shutter in sunlight, we would open the aperture to exactly compensate the light called equivalent exposure. FP HSS is the same as daylight in that regard. FP is not like real flash anymore. It is continuous, just like sunlight (the difference is that we do not notice sunlight being affected by the inverse square law.) There is an initial power loss, yes, but there is no further effect of faster shutter speed - not when at equivalent exposures.
You can watch your flash LCD scale, and this is what it says as you change shutter speed (again, always at fixed aperture here). Photo tests confirm there is about 2.3 stops of light loss initially (when the shutter exceeds maximum sync speed and the flash converts to continuous mode). And then each two times that shutter speed is doubled gives half the distance at the maximum available flash power again, which is the inverse square law. The flash power is not changing, simply the shutter duration seeing it is changing, but that is effectively weaker. So everything makes reasonable sense, so long as we realize Auto FP flash mode becomes Continuous light, and is subject to the rules of continuous light now (no longer acts like a real flash).
I doubt we would ever consider using Auto FP flash mode this way indoors, FP HSS is for fill in bright sun, to use an aperture wider than f/16. So this description above, while correct, is very misleading. It misguides our thinking down the wrong road, confusing us, and preventing our realizing the big picture. Yes, shutter speed affects the FP flash now. Yes, regular flash is unaffected by shutter speed, however, the problem is that the shutter cannot sync flash faster than maximum shutter sync speed. So FP HSS mode converts the flash to be a continuous light, and then shutter sync speed is no longer any issue. Then yes, a shutter speed of 4 EV faster reduces the light by 4 EV. Said that way, yes, it can sound like a terrible problem, but it is still exactly the same as sunshine works, same as all continuous light works, same as we learned exposure works.
So we would never use FP HSS mode that way (not as sole light source with a fixed aperture) - if we want speed, we would use regular flash then. We use FP HSS flash for fill light in bright sun. So we simply also open the aperture 4 stops to compensate, to achieve equal EV exposure again, exactly like sunshine has always worked. FP HSS mode is used for fill flash in bright sun, and the sun itself requires the equivalent exposure too. This point being that the FP flash is now continuous light, like sunshine is continuous. It shows flash rules no longer apply to FP HSS mode, continuous light rules apply now.
Remember, regular flash normally has big advantage over FP HSS mode (power and speed), until the maximum shutter sync speed is reached. This is little issue indoors in dimmer light (the flash is much faster than any shutter, and is unaffected by shutter speed). But outdoors in bright sunshine, the sunlight requires Sunny 16, which is ballpark f/16 at maximum shutter sync speed (at ISO 200). In bright sunlight, the maximum shutter sync speed limit locks us in near f/16 (which also requires a lot of regular flash power). But FP HSS mode was invented for this bright sun situation. It allows using fast shutter to achieve large aperture and minimal depth of field, if we want it (but yes, the power is limited).
So here's the HSS deal
Outdoors in bright sunlight is where we might use Auto FP mode and HSS for fill flash. The emphasis is here, in this section. The problem with regular flash mode is that ISO 200 and Sunny 16 says sunny daylight exposure will be f/16 at 1/200 second. We are unable to reselect equivalent exposures of f/11 at 1/400, or f/8 at 1/800, because the maximum shutter sync speed for flash is limited to ballpark of 1/200 second. But Auto FP mode can be used to allow a faster shutter speed, for the purpose of a wider aperture. When we do, we obviously must keep the ambient bright sunshine exposure correct too. So in sunshine, when using camera A, P, or S mode, the aperture and shutter speed will automatically track together and compensate each other, so that the EV exposure always remains the same (the ordinary way exposure works in continuous light).
1/400 second at f/11
1/800 second at f/8
1/1600 second at f/5.6
1/3200 second at f/4
1/6400 second at f/2.8
These are all the same equivalent EV exposure, all equal for continuous light like sunshine, or for Auto FP flash which is continuous too. The HSS flash power level does not care which equivalent you use, because then aperture compensates for shutter speed, like it always does, like this chart shows. This is how HSS FP flash should be contemplated, as continuous light. Each of these equivalent exposures use the same power of the sun, and the same power of the flash. The 1/6400 second is no actual additional light loss, because the f/2.8 fully compensates for it (did I mention, just like sunlight?). This equivalence is how exposure has always worked. However those fast shutter speed numbers are simply not possible with regular flash in bright sunshine, due to the maximum shutter sync speed. But in FP mode, they are possible. There is still the initial 2.3 stops loss with FP, but no additional light loss with shutter speed - not while aperture is compensating with shutter speed for equivalent EV (speaking of FP mode, which is continuous light). The flash LCD will show the FP maximum distance range is not changing with shutter speed now - however it probably only has about 5 to 7 feet of range too (fill flash should be one stop or more down, so 10 to 12 feet ought to work for fill). You could play Joe McNally, and use four SB-900 in parallel. :) Using four flashes gives 2 stops more light than one flash (and 4x more range).
NOTE: As the only light source (indoors), that LCD distance is the range. You always want to use regular flash mode indoors, as it is both powerful and fast, and maximum shutter sync speed is not an issue in the dim indoor light. But the fill flash range is greater than the LCD says, since as fill, it is not the only light source. We are in bright sun, which provides most of the light. The flash is fill, which is normally intentionally underexposed a stop or two. So if you need f/2.8 or 1/6400 second with fill flash outdoors, then on the theory that any added fill is good, the LCD showing 6 feet probably works well for fill flash at up to double that range. WATCH YOUR READY LED warning of insufficient flash power. Changing ISO will shift the EV combinations, but ISO will not change the FP flash range from the continuous light (because TTL flash readjusts flash power for ISO, back to the same intensity and same range).
FP HSS flash is NOT like regular flash. It takes much flash power to match the sun, difficult in any flash mode. Any subject distance of several feet can be a big issue at sunlight exposure levels. The best chance is with a larger flash and staying within maximum shutter sync speed, however Sunny 16 and maximum shutter sync speed says you will be near f/16 then (at ISO 200). The case for real flash mode is this: A wider aperture makes the flash power more effective (wider aperture requires less flash power), because real flash is not affected or limited by the corresponding shorter shutter speed. When shutter speed does not exceed maximum sync speed (real flash mode), we see the flash LCD scale showing 41% greater flash range from every wider aperture stop (equivalent of double power), even while the continuous sunlight exposure stays constant due to the equivalent aperture/shutter EV combinations. FP flash mode does not share that aperture advantage, because FP is continuous light now. The only problem with real flash mode is that in bright sunshine, Sunny 16 and the maximum shutter sync speed will force that aperture to be about f/16 for this very bright EV.
(off topic) A Neutral Density filter (ND), or a lower ISO too, can help to widen that aperture, but they require more flash power too, so the only change then is depth of field, and NOT effective flash power or change of flash range. ND and lower ISO do cause a wider aperture, but both also reduce the flash intensity (whereas shutter speed does not). Of course, we can turn the flash power back up (TTL or manual flash, assuming sufficient flash power exists), but the sunshine does not get turned up, so ND or lower ISO can increase the ratio of flash to sun (turning flash up makes fill brighter), and also depth of field is less at wider aperture, but the flash range is not changed by ND or ISO (we readjust flash power to keep it equal). The ND filter can cause darker viewfinder and AF inability to focus. The point is, neither helps the low FP power situation either (because we need to turn the speedlight power level up to compensate them).
So the case of bright sunlight is when we might use Auto FP to bypass sync speed limits. While the first example is true with aperture fixed at f/4, indoors may not be a practical use. In a dim situation where ambient is insignificant and can be ignored, we would be crazy to consider continuous Auto FP mode, not just because of flash power, not just because maximum sync speed is likely not any issue, but because the regular non-FP speedlight real flash mode makes the flash duration be so much faster than any shutter speed can ever be (around 1/11,000 second at 1/16 power, from spec chart in rear of flash manual). High speed photography is done that way. Auto FP mode would be very counterproductive then. Auto FP is instead for sunshine.
FP flash mode is still automatic exposure in TTL mode, within the limits of its range capability. At a fixed shutter speed, say 1/1000 second (manual mode, fixed and constant, no longer tracking and compensating again), you can change the aperture from say f/4 to f/5.6 to f/8 to f/11, and the flash power level still adjusts to give you exactly the same picture exposure - if - the subject distance is still within the maximum distance range for all apertures, at the bright EV value at which it must compete. This last part is not that easy at these reduced power levels.
FP High Speed Sync mode converts the flash to be continuous light, and then the shutter speed affects it the same as any other continuous light. So note that FP High Speed Sync mode is NOT to be confused with flash. It technically is simply no longer flash. It is continuous light from the flash tube, and the rules of continuous light apply. It also is certainly NOT high speed flash. The flash is continuous, the slowest possible. It is only high speed sync.
HSS flash has advantage of allowing a fast shutter and a wide aperture in bright sun. It has two disadvantages. It is continuous light, and so it is affected by shutter speed, and it has no motion stopping ability like flash does. And basically, it is around -2.3 stops less power than regular flash.
Is it actually 2.3 stops loss? Here are four ways to verify it.
1. The Nikon SB-900 prints Guide Numbers for FP mode (manual page F-20), and FP mode reduces GN by a factor of 2.18x of guide number (for 1/500 second shutter), which computes to be 2.24 stops loss.
2. The ratio of 52/22 feet on the SB-800 LCD above (round off errors) computes 2.48 stops loss (ISO 200, f4, 50mm zoom).
3. Setup a non-FP TTL flash with the distance such that you get a -1/3 EV underexposure warning at maximimum shutter sync speed (perhaps via bounce). Then increase shutter speed 1/3 stop to enable FP mode and take another. Now the SB-800 shows a -2 2/3 stop warning, which is -2.3 EV difference (this warning only goes to -3EV).
4. Take a test picture at maximum sync speed, and also 1/3 stop faster to trigger FP mode, and adjust aperture to give equal exposure to match. My SB-800 results show -2.3 stops, others say they think -2.6 stops. It is this ballpark. Result is something less than 1/4 power, maximum.
These methods all introduce an additional 1/3 stop shutter speed increase, as the only way to enable FP mode, which are not quite equal situations. But 2.3 stops is ballpark.
Some FP Flash examples
Pictures below show a garage door, looking west at 11:20 AM, in partial shade from a roof shadow at top, and a tree shadow lower. Nikon D300 in 1/250 second Auto FP mode with hot shoe SB-800. ISO 200 and Center metering, Aperture priority. 24-70mm lens at 24mm. Subject distance (garage door) was carefully measured to be at 12 feet (3.66 meters), which is about the limit for FP fill flash to help.
Garage door in shade, at 12 feet, ISO 200, hot shoe flash
Above: No flash. 1/250 second f/16 (dark shadows are the problem).
Above: Regular TTL flash 1/250 second f/16, flash LCD range says 8.8 feet (if we had been at 8.8 feet, the flash power would have lighted the shadow to full expected exposure. At 12 feet, we still have shadow, but a ligher shadow).
SB-800 Guide Number at 24mm is 98x1.414 = 138 ISO 200, divided by f/16 is 8.7 feet range (assuming the sole light souce, but the sun is helping considerably here). Maximum shutter sync speed is 1/250 second with flash, so 1/400 at f/11 is simply not possible (with flash in sunlight).
Above: Regular TTL flash -1 EV flash compensation, 1/200 second f/16, LCD range says 12 feet
Above: TTL FP HSS flash, 1/400 f/11, 0EV flash compensation
Above: TTL FP HSS flash, 1/6400 f/2.8, 0EV flash compensation, LCD range says 4.6 feet, yet the flash seems helpful for fill at 12 feet. Note that all equivalent exposures show the same flash range in FP HSS flash mode (works same as sun). It is just not much range. But fill does not not need as much power as a sole light source would need. And note that the regular flash mode LCD above reported only 8.7 feet range, and was still usable (at f/16, which allowed maximum sync speed to be honored).
Above: TTL FP HSS flash, -1 EV flash compensation, 1/6400 f/2.8, LCD range says 6.6 feet.
This is TTL FP mode, and it is NOT showing warning about exceeding maximum power, so the TTL metering must think it is accurate for -1 EV fill.
So, is FP HSS fill flash usable in bright sun? Yes, for the purpose of a wider aperture. Is it powerful? No. And the regular speedlight is not so strong either (not at the necessary f/16). So perhaps HSS is not optimum power for fill at 12 feet, but we still get considerable helpful flash fill, often usable for f/2.8 in bright sun if desired (within these range limits).
Above: Again, no flash. 1/250 second f/16. Notice the bricks in upper right.
