cgardner
Offline
• • • • •
Upload & Sell: Off
|
 p.1 #12 · CANON: HSS off camera discussion no triggers | |
BrianO wrote:
The OP never said "Master," he said only "Group A."
The Master flash defaults to Group A when it placed into Master mode so it is a given. Also if the Master is on or over the camera it acts as fill. If also you like to do the Master is placed on a 25' long TTL cord A can be used as key or fill. But again the OP appeared to be using the Master as I do so my advice was correct in the context of the question.
Some new users don't realize the Master is A by default because the manual, at least for the 580ex, never explicitly states that fact. But it becomes obvious when A:B ratios are used with the Master in the hotshoe or bracket because the Master Group A acts as Fill and the the Key requiring an A:B= 1:2 ratio to produce the same actual ratio as a 3:1 H:S ratio set manually.
H:S
1:1 Fill (Group A on axis)
2:0 Key (Group B off axis)
===
3:1 Reflected ratio
What the A:B ratio numbers express is not the reflected ratio as with the H:S convention but the relative incident strengths where they hit the subject and reflect when fired separately. During the pre-flash phase the A:B groups are not fired together to meter the ratio, likely because the metering zones on the viewfinder are not small enough to it accurately. It's because the evaluative zones the metering uses are so large that any TTL exposure will have a margin for error in small highlight areas.
When the Master is used on-axis (as required for the metering notation convention to work) the A:B ratios wind up being the reverse of the convention That is very confusing for many and one of the reasons I wrote my tutorials back in 2005 when I got mine and figured it out.
There's no rule that says fill must be centered but the ratio convention dates back to when that as normal practice. That's why two equal lights 1:1 incident, create a 2:1 H:S reflected ratio. The ratio math only works when fill is the same everywhere on the face and the key light overlaps:
H:S
1:1 Fill (Group A on axis)
1:0 Key (Group B off axis)
===
2:1 Reflected ratio
The "key" by definition is the source creating the modeling over the "fill", not in opposition to it. When two equal incident lights overlap the one creating the highlights is the key and the other the fill.
When two lights are placed on opposite sides this occurs:
H:S
1:0 light on left
0:1 light on right
===
1:1
In that situation there are two "key" lights and no fill. Both sides of the face are the same brightness and if the lights are moved far enough to the side you get a dark stripe down the middle of a 3D face or object. On a face there will be no fill in places like smile lines, the mouth and other places the cheeks shade both lights (cast shadows) and the tip nose will often have shadows going two different directions. It's great strategy to copy flat artwork without reflections back into the camera (one I used at National Geographic on a copy camera) but not one that is natural, flattering.
When lights are place that way with unequal power the brighter become the key and weaker the fill and they work in opposition to each other creating a split pattern but there is always darker stripe in the middle..
http://super.nova.org/DPR/Canon/images/ABtest2.jpg
In terms of cause and effect split / crossed lighting can be effective on white background because it makes the darker front of the face contrast more. The eye is pulled from lighter edges to faces and held there. It is also a way to create a ying/yang or dark/light dual personality vibe in a portrait. But when using that strategy one must watch for low spot in the shade of both lights because they will be very dark (e.g., smile lines, in and around the mouth where shaded by the cheeks).
I find that for the type of PJ style location shooting I do with my speedlights the centered fill on bracket strategy works best because I never need worry about fill placement or unfilled harsh shadows. Can simply dial in a ratio to set the mood, move my key light as needed to meet the creative goals for the shot pattern-wise, and the only adjustment required is FEC to keep the highlights under clipping.
In situations like those where you probably use your Master on a 25' leash for static setups I will instead be using my studio lights at home, or in the field I would use three speedlights, keeping one in centered to ensure I don't have unfilled dark gaps in my lighting pattern and two off of the sides as in this shot I did for a marketing piece at work for our copier management section which manages the 1,000 or so copiers at the US State Dept.
http://super.nova.org/TP/Three580ex.jpg
The challenge there was "short" lighting the two opposite oblique subjects and the full face subject behind in a flattering way with them different distances from the same. Part of the solution was to put the guy in the middle against a white background making the resulting crossed key lights work better on his face. The fill was centered split with my DIY diffuser on my bracket standing on a chair with the top of pressed to the ceiling for maximum spill fill into the background.
I light perceptually based on an understanding of how things like ratio affect perception. I start from a A:B = 1:2 baseline, see what that ratio which normally fits scene to sensor produces and adjust the ratio if it isn't revealing the shadow detail I want.
The second day I owned my Canon flashes I set up this this target and did a test to visualize what the different ratios actually produced and how accurately the metering would control them
http://super.nova.org/TP/RatioTestShot.jpg
The Group A fill on bracket was illuminating everything evenly except for front>back fall. Group B was to the right and only hitting the right side of the binder. Both lights overlapped on the flat card. I started with A:B = 1:1 and adjusted FEC to make the highlights white but allow some headroom for measuring fluctuation, then "ran the table" on the ratios from 8:1 to 1:8 without changing anything other than ratio.
http://super.nova.org/TP/CanonRatioTest.jpg
As you can see from 8:1 to 1:1 there is very little difference because in that configuration the Fill is overpower the Key with those ratios. At 1:1 the same 1K+1F:1F = 2:1 reflected ratio overlaps similar to the appearance of a manually set H:S = 2:1 , a A:B=1:2 looks similar to the conventionally metered H:S = 3:1 and so on. More or less.
In practice factors such as the modifiers used and how much bounced spill fill they contribute will change appearance of even conventionally incident metered and ETTL controlled ratios so you shouldn't get hung up on the numbers. The same H:S ratio set with a meter will even look different on light and dark backgrounds because the pupil of the eye reacts to overall brightness level in the field of vision making the same shadow tone that looks "normal" on a dark background appear darker when viewed surrounded by a white one.
http://super.nova.org/TP/RatiosPerception.jpg
There are two somewhat conflicting agendas with lighting. Technically it is desirable to fit the scene range to the sensor because that is what allows the camera to record detail everywhere, but the ratio needed to accomplish that often is not the one desired for conveying the mood of the subject or convey that the environment the subject is in.
I find that a 3:1 conventional or A:B = 1:2 ratio fits an indoor flash lit scene like my test example to the sensor. If I use 2:1 / A:B = 1:1 I get lighter shadows which perceptually makes the objects seem rounder and less angular — "softer" in lighting parlance. Why? The highlights are the same after exposure is adjusted but the lighter shadows lower the overall contrast gradient. Even of you use direct flash or very small modifiers and simply change the lighting ratio to create lighter shadows the overall perception will seem softer even with sharply defined shadows because the contrast gradient is less steep between highlight and shadows.
Choice of modification strategy matters with HSS because of the limited power. Modification strategy impacts light placement and key / fill strategies. The ones that work best with HHS aren't necessarily the best in general for lighting, but the best solutions to work around the power limitation successfully. Bringing a knife to a gunfight isn't wise, but having one available for close quarters combat when the bullets run out isn't a bad idea either.
The mood in lighting is created by controlling the shadows: placement with the key light, tone with the fill power, and transitions with fill placement.
Fill placed at camera falls off front > back on a face making the nose shadow the lightest one with a gradient in an oblique view that fades darker towards the ears helping to hide them. Fill placed to over on the shadow side of in the same oblique pose will create the opposite gradient in the shadows. The light> dark gradient is the reverse of frontal fill. The ear, being closer, to the fill will be brighter than the nose shadow furthest away. The nose often winds up with the darkest shadow on the face and if the fill is placed further back than the tip of the nose the nose and cheeks begin to shade it.
It's not a matter of one strategy being better than the other in absolute terms, but rather if shadows created by the strategy are a good match to the intended mood or impression one is trying to create with the lighting pattern.
For beginners fill is an afterthought. They are so focused on getting that flash off the camera axis that all they can see is what the key light is doing. Understanding what is happening in the shadows is a bit like tasting wine. If all you have ever tasted is $5/jug table wine you might not appreciate the subtle differences in flavor notes between vintage years of a more expensive non-generic wine. Not because you don't have good taste in general, but because your taste buds and brain haven't been conditioned by drinking hundreds of type and educating your palette regarding the different flavor notes in wine. With wine the difference is discerned in part by the "nose". The same is true for shadows in lighting. It is only by comparing different fill strategies that you can come to discern the difference it makes.
There is a belief by some that bigger is better with modifiers. It is if one needs a huge footprint for a close in light, want to maintain the character of the light as the key is move from 3ft to 10ft. or in situations where the fact as much light is bouncing off the ceiling as aimed directly is an effective lighting strategy....
But want to guess how big my SBs were for this shot?
http://super.nova.org/TP/table/Table04.jpg
You might be surprised.
http://super.nova.org/TP/table/Setup.jpg
I just used bare bulb and the room as the SB with a black drape over the right wall to create a subtractive lighting effect. That was to show a furniture maker who e-mailed me for advice how to do shots on a white background with just the two AB lights he owned.
Big modifers create the same cause and effect (less contrast between highlight and shadow tone) because they do two things: increase the family of angles hitting an object and in a small space the larger footprint the create at the same distance (relative to a smaller source) bounces more "spill fill" into the shadows.
If you compare a single large source 45°V/45°H to the nose in a full face shot of a person in a dark sweater vs. two direct flashes, one the same 45°/45° and the other chin level near the camera as fill where you will notice the subtle differences in strategies is the shadow on the side of the nose.
The single larger source light will "wrap" around the rounder forehead and chin and the top of the nose, not be able to put light on the side if the nose: it will remain very dark and between the highlighted forehead, cheeks and chin be very distracting when viewing the portrait.
If take off the big modifier and use it direct from the same spot as the big source it will create the same pattern on the face with highlights on forehead, cheeks and chin defining it's 3D shape with the same mask of highlights. The sided of the nose will also be equally dark. The differences you will see will be: 1) more distinct shadow borders, 2) more specular reflections and smaller catchlights, 3) you need far less flash power to shoot at the same f/stop.
When you add the second direct flash from near the camera it changes the appearance of the first direct source perceptually. Because it is even as it is raised overlapping the key light already adjusted for correct exposure for every stop you increase the fill you will must reduce the power of the key light to keep the highlights the same, thus the more fill that is added the less key light power is needed for exposure at the same f/stop. What you will notice about the nose shadow is that it gets lighter. What you will notice about the borders is that while they are still more distinct than the larger source they are not as sharp and when the direct flash is used without fill. Because the fill source is physically closest to the nose its inverse-square fall off front > back on the face will create a light > dark transition on the shaded side of the face not seen in either single light shots. When the fill is raised to the point where there is detail is revealed in the shadows of the dark shirt the overall appearance on the face will be softer than that of a large source if that large source isn't bouncing spill fill onto the front and sides of the face.
Perceptually the two direct flash strategy can be made to look as soft or softer than a single larger source. Soft/hard and the perception of 3D shape in 2D photos are all optical illusions in terms of cause and effect and there is more than one way to fool a viewer's brain into thinking and object has 3D shape and how angular or rounded that shape is.
The reason this is germane to speedlights and HHS in particular is that speedlights have limited power and it is reduced when HHS is used. Thus using a lighting strategy that employs direct flash will result in greater effective range than one where modifiers cut flash power by 1-2 stops. The strategy I have found works best with direct speedlights —having learned that way and tried everything else since — is to use centered fill with an overlapping key light.
A few years ago in the middle of a thread on how big a modifier is needed outdoors I grabbed the camera and flashes, went outside in the early evening on a summer day and dragged my wife out of the garden to take the shot below....
http://super.nova.org/TP/DualSunBacklight.jpg
It was shot in Av mode, -2 EC to get the sunlit skin on the shoulder and hair below clipping. I used no diffusers, just one 580ex on a bracket and the other 45° from the nose (90° from camera) above and to the right. Flash was in ETTL A:B mode. I don't recall the A:B ratio but it was before I started using a 1:2 starting baseline and I think it was 1:4 in part because I wanted to show the role soft skylight play in that situation. This is not how I would shoot a portrait session nor am I suggesting shooting one that way. But this is typical of how I light most of my candid shots outdoors with speedlights — without diffusers because speedlights just don't have the power to used them. Look at the nose shadow and the transition on the cheeks on the shaded near side. The shadows only get really dark in areas neither the flash fill on the bracket or skylight fill reached. The shadow is dark under the chin because the gray card used to monitor tonal range was shading it.
It's not the best lighting strategy or the most flattering results, but due to the low power and range in HSS mode it is often the only ones that are practical. If only shooting with single flash outdoors in sun I stick with full face poses and "butterfly lighting" because it minimizes shadows. No shadows, no fill needed because the loss of detail, if any, occurs where it is not noticed.
But in an oblique view the darker sky filled shadows alone wouldn't look "normal". That were I will use the two flash strategy with HHS. The added benefit of two lights vs. one is more range — 10ft vs. 7ft with single flash.
I use my fill on a bracket vs a stand because it's simpler logistically. That's a trade off because the bracket puts it higher than chin level which is more ideal for fill, creating a unfilled shadow under the chin. But then so does natural light so it doesn't bother me much.
I also use the same strategy of centered fill for my studio work. There power is not a concern and I can use larger modiifers when needed to make highlights less specular, catchlights bigger, get more "spill fill" when needed, etc. I use centered fill because it makes the nose shadow the lightest and least distracting on the face in my lighting patterns. Others might not notice the difference, but I do. So will anyone who objectively compares the strategies as I suggest.
http://super.nova.org/TP/DIYvsSB.jpg
Above when friends dropped by with their kids I started shooting with speedlights and switched to my studio lights with larger modifiers. I wasn't a "portrait" session, just have my lights set up all time. I didn't reshoot the boy because the speedlight shot was OK. Comparing the results afterwards I noticed, as you will if you look critically, that the girl's shot while using larger modifiers and a bit more and lower placed fill than with the speedlights looked "harder" to me. It puzzled me why until I realized I was reacting perceptually to the specularity in the highlights. The kids had been out in the sun all day sweating. The boy washed his face, the girl wearing make-up didn't. One of the things larger modifiers do is make the skin highlights less specular. but shiny skin will trump a big SB. I could have had her wash her face or use powder to cut the shine but they were just snap shots with decent lighting in t-shirts so I didn't bother.
What matters to me isn't the numerical ratio but how it looks and how I want the viewer to react to it. That's more about understanding the psychological aspects of lighting — how and why the viewer will react to the clues the lighting direction and contrast provided. That's the "bigger picture" beyond learning to control lighting with a meter to create a 2:1 or 4:1 — understanding they trigger different emotional reactions.
The challenge in a flash assisted shot is making the result look natural and not fake. That requires understanding how the brain decides lighting is fake and then avoiding creating those clues in your photos.
The two qualities that help make a viewer thing a photo is natural and real are:
1) the same tonal range they would expect to see in person by eye - a full range of detail.
2) the same directional clues of shadow angle and highlight placement they would see in person within the context of that setting. If the context of the photo is outdoors in mid-day a long sideways shadow hanging off the nose created by a flash placed too low will not seem natural. But if shooting in the afternoon with the natural light creating shadows like than in the background the same low placement of the flash would be in context. Natural light is a moving target because the sun moved in the sky. The flash needs to move similarly in sync and create the same perceptual clues to look "real" outdoors.
Indoors I approach flash as more like lighting a stage. If the ambient indoor lighting alone is doing that in a way that works for the goals of the shot I'll use it. If not I will supplement it and sometimes overpower it. More often than not with color photography the problem isn't the amount or direction of the ambient light but the fact is a different color temp. When practical I may gel the flash to blend it seamlessly. When that isn't possible I will usually take my wide shots where faces are not the focal point with the ambient light only with WB set to match, then crop out the off color background and try to overpower it with the flash for the people shots. When I want more background ambience in mixed lighting I will either bounce either fill, key or both or light the background with flash.
The playback is invaluable. Your eyes can see the ambient, but not how the flash changes it. Back in the days of film the reason technique didn't vary much is because using systematic technique was the only way to predict results and produce a photo without blown highlights or no shadow detail. The playback, histogram, and clipping warning are what allows me the freedom to use my lights a dozen different ways to create lighting patterns, but regardless of the pattern and how I created it I know I need the lighting ratio to fit scene to sensor. I know most of the time A:B = 1:2 used directly does that. But if bouncing and mixing flash with ambient that might not be the case. I start from the baseline of A:B = 1:2 but don't always wind up there in the end result. The "fit scene to sensor" baseline just makes it easier to compare what works better or worse.
|
?) you will not be able to use HSS with "regular triggers", so your only other option is to use your camera manufacturers built in wireless system assuming you have one flash or camera body capable of being a "Master" and another flash capable of being a "slave". 
