Lighting Q and A, 09-26-08
Christine, in Virginia (USA), asks:"Is it possible for a specular highlight to be anything other than blown out? If so, how do you meter and control its intensity?"
Yes it is, Christine.
Keep reading for how to control specular intensity -- and for your second question, which I liked even better...
Christine's question went into more detail, and was asking about the process of using one light as both a key light and as a background element via the light's specular reflection.
The short answer is yes, the specular can be just about any tone you choose. It depends on two things: The reflective efficiency of the object creating the reflection and the intensity of the light source, on a per-square-inch basis.
The more efficiently your object reflects, the brighter your specular will be. But you might not have much control over the object, so you can also control the specular intensity by increasing the size of the light source.
Let's say you have two lights, each set to illuminate your subject to, say, f/5.6. One is a hard light, and the other is a big, soft shoot-through umbrella.
They will both correctly expose the subject at f/5.6. But in addition to creating very different qualities of light on the subject, they also will create very different specular highlights whatever is reflecting the specular -- your subject and/or your background.
The specular of the hard light will be tiny and very bright -- and almost certainly blown out. This is because the light is very small -- and very intense over those few square inches of light size.
The specular of the big, soft light will be large and manageable. The size of the light source is a determining factor in the intensity of the specular. And that highlight could be a variety of tones based on how close the light is to is to the subject and background.
This is because, while both light sources are putting out the same amount of light, they look very different when you measure them on an intensity-per-square-inch basis. And since a specular is basically a reflection of the light source, the intensity of each of those square inches is the main factor in the brightness of the specular.
As for metering those highlights, you can use a flashmeter in reflective mode to check how bright a specular is (vs. your shooting aperture.) Or simply zoom into only the specular highlight and shoot a photo. Then look at your image and histogram on the back. The spike will tell you the tone where the specular is, as compared to medium grey.
Now, Christine is Thinking:
She goes on to describe a second setup:
"Possible Scenario 2: The specular is caused by a light placed for the sole purpose of adding a specular--i.e. a rim light on the edge of someone's face or a background light on a shiny wall--and changing the power and position of that light will not affect the overall exposure."
Absolutely, but I would do it in a slightly different way. Why not position the light so that it is very easy to line up the specular highlight? And yes, you could squeeze some double-duty out of it, too.
And that is exactly what we did in the photo at left. This was one of a recent series of photos for a software company. My assistant Patrick Smith and I were working very quickly in a hotel conference room to do eight full-page portraits in just two hours.My main light is coming from camera left, a few feet out of the frame and pretty close to my subject. It is warmed up with a 1/4 CTO and is lighting his face with soft light via a shoot-through Westcott double-fold.
The second flash is also firing through an umbrella, but that flash is directly behind me and a tad high.
That is basically a "buy one, get one free" flash, as it not only provides the perfect angle to create a soft specular on the dark wood background, but also serves as soft, on-axis fill for my subject.
As mentioned in the SB-III post, on-axis fill is something I am using more and more, and in several different ways. More to come on that soon.
______________
:: Read More: Using Specular Highlights ::
__________
New to Strobist, or lighting? Start here.
Now shipping, in DVD box set or download: Lighting in Layers
Connect: Discussion Threads | Reader Photos | Twitter
Posted at 12:00 AM.
16 Comments:
Love it, thanks David! The on-axis fill looks great and takes the guess work out of fill light!
Just a quick comment as a designer of imaging systems. This is a very interesting and astute technical question.
By definition a specular highlight has a reflectance great than 100% reflector. It is common for designers of digital cameras to design the metering system to place the 100% reflector at a code value of less than the maximum available. Where exactly the 100% reflector is placed defines how much headroom the system has. Most pro cameras have about 50% headroom. So, if you're specular is less than half a stop brighter than a 100% reflector in the scene, your specular won't clip. If it is, than it will.
This one still baffles me. The best or easiest way I’ve understood is that the speccular highlight stays constant in relation to the diffused . When you put something over the light, say an umbrella, you loose light. Say your reading now says f2.8. When you raise the power of the flash back up to read f5.6 the diffused value gets brighter in relation to the specular highlight which stood constant hence appearing to diminish in intensity.
Same concept for moving a light (softbox/umbrella) closer. From what I understand of still trying to. Of course there is much more to it.
Cesar T Sanchez
Looking forward to an on-axis fill article from you David. The first thing you taught me was to get the lighting OFF-camera. Now I guess we'll work on balancing off-camera with on-axis....
Hey Dave,
A few of us are putting together a Strobist meet in Maine (Portland-ish area). Not sure if you've ever been up this way, but you are, of course, welcome and invited.
Let me know if you'd like more details on the event.
Houser
info@houserdesign.com
http://www.flickr.com/photos/houser
Great article and I love the portrait you shot to illustrate your point. It'll be very interesting to read your article on fill-flashing from the camera axis.
I notice you seem to use a 1/4 CTO quite often in shooting portraits. I have two questions about that: 1st how do you determine what percentage a gel is, i.e. 1/4, 1/2, 1/1 and #2 do you always set your WB to incandescent when you add that CTO gell to your light?
As always, thanks for the informative post!
I maybe asking a basic question here but from my limited knowledge, difference between an Umbrella and Softbox is that Umbrella has more spill. If you are using two umbrellas (both for main light and key light) why is it that lower part of the model is not lit by the Key light?
I can understand if you have used a grid or a snoot for Key. But using an Umbrella and getting such resulting baffles me.
In addition to adjusting the size of the source or the reflectivity of the background, another approach would be to move the single light source closer to the subject, and dial it down so that the illuminance on the subject remains the same -- thanks to the inverse square law, the illuminance on the background will be lower.
On the subject of on-axis fill, have you seen the blog post from videographer Art Adams where he suggests, "Go craaaaazy: fill from the key side!" I'm gonna experiment with this...
@Anil,
I'm guessing the spill from the key is control by having the key quite close to the subject. Goes back to Dave's lesson on the "depth of field" of a light. If a light is close to the subject, it's intensity falls of faster at greater distances.
Thanks Dave, great stuff as always.
John: I don't know how David does it, but once I was familiar with the different gel percentages, it was pretty easy to tell just by looking at them what they were.
Anil: I am pretty certain David shot the on-axis fill/specular SB straight through the umbrella at the wall, and set it a couple stops below the main light. This means that it will only minimally effect the overal exposure of the subject, while bouncing off of the much more reflective hardwood behind, creating a large, diffuse specular. Depending on the hight & angle of the on axis light, the person shooting will also cut off some of the spill on the lower part of the frame, further increasing fall-off. I don't really see that much fall-off from the main light.
David, please elaborate and or correct me if I am wrong.
Thank you so much for the detailed answer to my question!
David~ Whats the benefit of on-axis lighting vs off-axis. The studio I work for only uses off-axis lighting when on the road. Does ON allow for more shadows??
Doug Holcomb
John:
I know your questions were directed towards Mr. Hobby, but I'll answer them (in backwards order) since I shoot with a similar technique to his.
In general, I set my camera's white balance to a daylight setting. In an earlier post that I read, Mr. Hobby said that he uses a daylight setting, too.
Personally, I use the Kelvin white balance presets on my cameras. And I typically set mine between 5500K and 6200K depending on how warm or how cool I want the image to look. The closer I set it towards 6200K, the warmer the image will be, especially with CTO gel. And the closer I set it towards 5500K, the cooler the image will be.
My white balance is always determined how warm or cool I want the image to look and feel. And sometimes, I just want the image to look exactly how I see it in real life. So, in that case, it wouldn't make sense for me to try to "white balance" the image "correctly."
For example, if I was shooting a sunset, and if I wanted to bring out the deep oranges in that scene, then I would set my white balance to daylight or to maybe even somewhere in the 6000K-7000K range to bring out those oranges. If I wanted to "white balance" it correctly, then I would set it around 3000K.
And since I shoot RAW/NEF, I can always change my mind once I get to the post-processing phase.
As far as how strong of a gel to use, I find that 1/4 is a great place to start [for portraits] since it's the most subtle looking CTO gel. You can always experiment to see which ones work best for you, but it does depend on the subject, the setting, and the overall situation.
The only time I've ever used 1/1 was when I shot in a banquet room with incandescent lighting that seemed to be burning at about 2800K-3000K. I used 1/1 CTO gel on my flash to match the color temperature of the incandescent lighting, and the balance was nearly perfect.
After you've had a little bit of experience/testing with the gels, you begin to develop an intuition about which ones are best to use in certain situations.
RE on axis fill
Many moons ago I used to work as a theatrical lighting designer, and am aware that we used to tackle lighting completely differently to photogs/ film guys.
Having long since left that arena; getting into flash photographic lighting has been great, and given me the chance to learn about light all over again. “plus ca change”. The big distinction of “soft light/ hard light” simply isn’t there in theatre for lots of reasons: but principally because the lights by and large are always out of eye line of the audience and hence a long way from the illuminated people/ objects: i.e. hard.
Theatrical lighting rigs always have a lot of “lanterns” in them: partly because every lighting state must be up there before the show starts; and you often have 100+ states in the show; but also, I can now see that in photographic terms, we were continuously simulating virtual large soft light sources using multiple lights, and the use of the lights becomes very fluid: you often use a “touch of fill” from another part of the rig; that wasn’t originally put up there for that purpose. We also use a lot of reflected/ scattered light: when you come to “plot” the lighting states; the nature of the objects on stage (texture, colour, size incl. costumes) really matter... so perhaps I’m beginning to see my two lighting brains converge after all this time. Great ! I always felt that you sculpted with light for each scene, with a mix of light from the available palette, often coming from almost every direction at varying intensities—and then in the next scene you’d find yourself illuminating a scene with one single back light: allowing the fill being the back scatter off the floor… (mind you; we had the advantage the dynamic range of the human eye is far greater than that of the camera, which explains why whenever journalists came to take pictures they’d grizle, get their strobes out, and nuke our beautiful lighting Ha !)
I’m always amazed how little kit is used by photogs and film guys, as they only ever set up for one “scene” before striking and going onto the next one: but clearly often preduce mesmerising images: David’s slogan is so apt. In theatre you get lazy about being focussed (ha ha).
Looking forward to the forthcoming post.
I very much appreciate your generous efforts to educate me, and mostly you are successful.
If I can offer one perspective though: I think that your biggest obstacle to effective communication is the loose approach to wording, eg "zero out a manual exposure" in a section that aims to demystify some other piece of jargon.
Thanks again.
Caroline.
Post a Comment
<< Home