Perspective (yet again!)

One word…, “DISTANCE”

Here we go again! Let’s start with the lesson drilled in my my college photography professor, Ed Sculley:

 
Pick the camera to subject distance that gives the perspective you want, then select the focal length lens to fill the frame appropriately
— Ed Sculley
 

Wise words! They harken back to a favorite Ansel Adams quote:

 
A good photograph is knowing where to stand
— Ansel Adams
 

For this post I define perspective as the relationship between the elements in a photograph. In a portrait that would be the size of the nose compared to the eyes compared to the ears, compared to the background. In an outdoor scene it might be the tree in the front yard compared to the house compared to the mountains in the distance.

Let’s start with the portrait. All too often I see articles claiming to show how lenses affect the look of a face. You’ve seen them, too. They show a photo made with a short wide angle lens where the face narrow with a big nose and then a series of photos with longer lenses where the face is flatter and flatter as the lenses get longer and longer. But what they don’t tell you is that not only did they change the lenses, but they also moved the camera. And moving the camera is what caused the changes in the looks. The lens just determines what is in the frame at the various distances.

Think of changing lenses as cropping your image in camera. All lenses show the same perspective at the same camera to subject distance. By putting on a longer lens you are magnifying the central area of the frame. If you don’t move the camera everything in the photo scales equally—the perspective remains the same. Let’s imagine a scene where you have a person in front of a window. with a tree outside. If you switch from a 50mm lens to a 100mm lens everything in the scene (the person, the window and the tree) all become twice as large in the frame, but their sizes relative to each other remain the same. If, on the other hand, you move the camera closer a more natural change occurs. The person in the front gets larger while the elements in the background appear smaller. The act of moving in might require us to change to a shorter lens to fit everything in the frame. But it isn’t the changing of the lens that altered the perspective, the move did that. The lens just determined what would fit into the frame.

 Start with our view of our subject by a window with a tree outside as seen with a normal lens.

Start with our view of our subject by a window with a tree outside as seen with a normal lens.

 Leaving the camera in the same position, a longer lens magnifies everything in the scene equally. But the relationship between objects remains the same.

Leaving the camera in the same position, a longer lens magnifies everything in the scene equally. But the relationship between objects remains the same.

 If instead of changing lenses we move the camera closer the subject in front gets larger while the background elements appear smaller. The size relationships all change.

If instead of changing lenses we move the camera closer the subject in front gets larger while the background elements appear smaller. The size relationships all change.

 
The closer the camera is to your subject the smaller the background elements will appear
— John Cornicello
 

As you move back away from your subject the objects in the background get larger in relation to the size of the subject—compressing the scene. Longer lenses force you to move back, leading one to think that it is the lens that is doing the compression. In reality it was moving back that compressed the scene. The longer lens magnified the subject to fill the frame better.

This is why you cannot “zoom with your feet.” Cinematographers tend to use prime lenses and move the camera (dolly in or out) to change the size relations in the frame. A zoom tempts you to keep the camera in one position and zoom to change size. But this is an unnatural look. In our 3-D world perspective changes as we move closer to and further from the objects around us. In a moving image a zoom instead of a dolly move just doesn’t quite look or feel right.

Don’t take this to mean I am against zoom lenses. Quite the opposite. I think they are very useful, especially in still photography, if used correctly. With a zoom lens you can position your camera exactly where you want it and then fine tune the framing to avoid or limit the cropping that you would need to do in post-processing.

And I do encourage you to move your feet. Step to the left or the right. Take a step in or back. Get down on the ground or up on a ladder. These all give you a different perspective. And you can do this with a prime lens or a zoom lens.

This all started a few days ago when my wife showed me this “great demonstration of lenses” in an Instagram post. I felt bad that I had to tell her it was all wrong. So I quickly made these crude illustrations to clarify things. The video on the left shows photos made with the same 16mm lens at various distances and resized to show how the drawing of the face is affected by the camera to subject distance. In the video on the right the camera remained stationary while the lenses were changed to show that the lens doesn’t alter the drawing of the face.

Here are side-by-side comparisons to help get the point across. The images in the first set were all made with a 24mm lens at different camera to subject distances and resized to maintain the same head size.

 24mm lens at various distances

24mm lens at various distances

This images in this next set were all taken from the same distance with different focal length lenses and resized to maintain the same head size.

 Different lenses from the same distance show the same perspective

Different lenses from the same distance show the same perspective

One more example to show that telephoto compression is a fallacy. Here we have a scene from a railroad station in Kyoto, Japan photographed with four different focal lengths from 18mm to 400mm. As you can see, the area of the scene that is common to all four images (in the magenta box) is exactly the same. The longer lens just isolates the central area, but the perspective is the same as in the wide angle photo. If you were able to enlarge the 18mm photo to match the size of the 400mm photo the quality would be terrible, but the compression (perspective) would be the same.

For more entertainment, please visit my post about the Incredible Shrinking Space Needle.

Well, enough early morning rambling for now. I hope you got something useful out of this.

Until next time, Cheers!
John
(Oh, yeah, please buy my book! THANKS!)

Learning to use your light meter

You are smarter than your light meter!

Just about every modern camera has a light meter built in. Plus digital cameras have a screen on the back where you can check your exposures and/or histogram. Do you really need an external meter, too?

The answer is a definite “Maybe.”

But before we try to answer that, let’s go through the basics of how a light meter works, the options available, and how to use the meter. We will look at both the meter in your camera and at external (usually hand-held) meters. And now there is a new meter that wirelessly connects to your mobile phone.

Light meters are color blind and unable to think. They make bit assumptions. They don’t always get things right. You have to do a lot of the work for them. Their readings are open to interpretation. You area much smarter than your meter, especially the one in your camera. Most cameras offer a few different metering modes. For most people the “evauative” or “matrix” mode can handle most situations. In this mode the camera meter reads different segments of the scene and quickly compares them to scenes programmed into the camera to try to make the best guess as to what you might be pointing your camera at. For general every day photography you should probably be using this mode. They usually also offer a center-weighted average and a partial metering pattern that only reads the center of the frame. Higher end cameras may also add a “spot” metering mode that reads an even tighter area. How big a spot varies by camera. For example, the Canon 5D Mkiii has a 1.5% area coverage, the Canon 6D Mkii uses a larger 3.2% area for its spot meter. It is the spot meter that I want to concentrate on here.

The meter in your camera basically assumes that everything you point it at is a medium tone of gray. And it wants to suggest an exposure setting that renders your subject a medium tone of gray. Luckily, many scenes that we photograph have a wide range of tones from darks to highlights that when averaged together are a mid-tone of gray. Whether that gray is 18% or 13% or 12.75% is an argument for another day. For now let’s just accept that average scenes average out to a gray tone that our meters are looking for.This averaging out is what makes the meter work. But it is also what makes us have to think about it and sometimes override what the meter tells us to do.

Let’s take some extreme examples—the proverbial polar bear in a snow storm and the black cat in a coal mine. If we relied on our in-camera meters both of these photos would be a mid-tone gray with a wildly overexposed gray cat and an underexposed gray polar bear. I am writing this at 7am in my PJs, so instead of going out to find the cat or the polar bear (or the coal mine or the snow storm) I am going to simulate the situations using a target with wide black, gray, and white stripes. Using the spot metering mode in my camera I metered each stripe separately (shown by the red targets) and made three photographs with the meter readings. I also made one more image using a hand-held incident meter, which I will get to later.

spot-meter-test.jpg

In the upper left I placed took the spot meter reading off of the gray stripe and all three stripes fall into line of black, gray, and white. In the upper right I made a spot meter reading of the black stripe (our black cat in the coal mine) and you can see that the meter made its expected assumption that the scene would average out to gray and by following along with the camera settings, that is just what we got. The black stripe turned gray and the white stripe overexposed. In the lower left I took the meter reading off of the white stripe (our polar bear) and the meter again assumed it was gray and turned the white patch to a muddy gray. The lower right shows a hand-held incident meter which reads the light falling onto the subject instead of the light reflected from the subject. More about that in the section below on flash meters.

This is where things can go wildly wrong if you don’t know what you are doing with the spot meter. Most of my photos are of people, so let’s look at what might happen in these situations. Caucasian skin tones are usually about twice as bright as the average gray (maybe 36% reflective as opposed to 18%). Darker skin tones might range from 18% reflectance down to 6% or so. If you point your spot meter at the different skin tones you will get very different readings that will lead to incorrect exposures with the caucasian skin being underexposed and darker skins being correctly exposed or overexposed. If you use the spot meter in your camera or if you use a hand-held spot meter you need to think things through more than when you use a matrix or center weighted mode and be ready to use your exposure compensation dial on your camera or work in manual exposure mode to get things just where you want them. Let’s look at some examples of where a spot meter can lead us if we blindly follow it.

spot-meter-scene.jpg

Here we have a still life set with objects of different brightness. From left to right we have a Color Checker Passport, a white index card, a Kodak gray card, a black card, a Gnomicello*, and a set of gray patches. In the upper left I made made a spot meter reading off of the gray card, giving a good overall exposure. Upper right has the spot reading made off of the black card, leading to an extreme over exposure of the scene. Lower left has the spot reading on the white index card, and the expected under exposure of the scene. Lower right is the exposure suggested by the evaluative meter mode in the camera. I actually think that the evaluative meter did the best in this series. The spot meter is not always our friend.

Extending this outside the studio, where should you point your spot meter? Point it at the white clouds and you get an underexposure. Point it at deep green foliage, get overexposure. You need to figure out something in the scene that you want to have render at a medium gray and point the spot meter only at that element, then lock in the exposure and recompose to make the photo. Bottom line, I think that spot meters have their place for photographers working with film and following the Zone System to control the tonal range of individual images. For the rest of us spot meters are useful if you want to figure out the contrast range of a scene, but you still have to work out the proper exposure overall.

For kicks, let’s compare the evaluative meter reading with the center weighted mode.

scene-meter.jpg

I think you can make a case for either exposure here. This all, of course, assuming that the monitor you are viewing this on is calibrated and set to a decently low brightness setting of around 100 cd/m2. Most monitors are set much too bright from the factory (for gaming or for watching movies), so make sure your calibrator has an option to set brightness levels.

Flash Exposures

hand-held-meter.jpg

This is all well and good for photography lit by continuous light sources like the sun, incandescent lights, LED lights, etc. What happens when you need to figure out the exposure for studio flash units that are not TTL** compatible? The meter in your camera has no concept of flash, so you need to rely on an external meter. Typically these have been hand-held devices such as the one shown here, my trusty old Sekonic L-358 (that is no longer in production). There is also a new meter, the Illuminati Meter (which I have not yet had a chance to try), that can be placed in the scene and connects with your mobile phone so you don’t have to hold it in your hand.

These meters can either be connected to your flash via a sync cable, a wireless transmitter, or they can be set to wait for a flash of light to trigger the reading. As noted, these meters are placed in the scene to read the light that is reaching them, the “incident” light. This provides a more accurate reading because it only reads the light falling on the subject and is not swayed by the brightness or reflectance of the subject or anything else in the scene. You set the ISO and shutter speed on the meter to match that on your camera, fire off the flash, and the meter tells you the aperture (F/stop) to set the lens to. For the most accurate readings the dome of the meter should be pointed at the main light illuminating the scene.

Some hand-held meters can also be set to read reflective light and/or be used as a spot meter, but then you run into the same issues that you have with your in-camera meters. And you don’t have the computer “smarts” of Evaluative or Matrix metering in the hand-held meter.

Color meters

The Illuminati Meter mentioned above and another relatively new device, the Lumu meter (iPhone only at this time and also which I have not had the opportunity to work with yet) extend the function of the meter by adding in a color meter mode. Normally color meters cost around $1,000 and only do their one thing. Having this available in a light meter costing around $300 - $500 is a big deal for those who need critical control over color.

Back to the beginning

So, back to the question of “do I need an external, hand-held light meter?” The answer remains a definite “maybe.” It depends on how you work, what type of lighting you use, how comfortable and sure you are about the display on the back of your camera, how much you know about reading the histogram on the camera, and on and on. I have friends like Tony Corbell and Matthew Jordan Smith or take multiple meter readings of every scene and friends like Joel Grimes who never touches a meter. The photographer makes the image, not the tools.

Wishing you good exposures!
John Cornicello

* Gnomicello is a birthday gift I received from artist Mike Oncley a few years ago.

** TTL means Through The Lens. Some camera and dedicated flash combinations offer a metering mode where the flash fires twice. Once at low power just before the shutter opens letting the camera meter read the flash and then again with the shutter open to make the exposure.

Color My World

So your light provides accurate color, what about your modifiers

I recently posted a self portrait I made while testing the new Interfit Studio Essentials LED monolight. I got some comments about how good the color looked and asking if I did much color correction on the image. I actually did very little color work on the image (shown here). I made one frame with an X-Rite Color Checker Passport and the second frame without. Using Adobe Lightroom Classic's white balance picker, I clicked on the middle gray swatch on the Color Checker and that made a small change, -100K and -5 tint. I thought the image was a bit too saturated, so I also pulled down the vibrance slider -11 points. That was it. 

 Image straight out of camera with no color adjustment

Image straight out of camera with no color adjustment

 Image straight out of camera, no color adjustments

Image straight out of camera, no color adjustments

 Image with minor color corrections (temp -100, tint +5, vibrance -11)

Image with minor color corrections (temp -100, tint +5, vibrance -11)

Or was it????

That got me to thinking about the modifiers I have available. I see so much concern over the color accuracy of strobes, fluorescent lamps, and LEDs. But not so much talk about the modifiers. With so many types and brands of modifiers I figured there must be differences. And I was right. Sometimes big differences.

For the above photos I used two LED monolights on me each with an Interfit 1x3-foot strip box. The light on the background was a Honey Badger studio strobe at its lowest power setting. I was quite happy with the results. But what about my other modifiers.

I grabbed a few modifiers: 7-inch silver metal dish, 24-inch pop-up that comes with the LED monolights (and also with the Honey Badger), a 20-inch metal dish reflector, the same reflector with two different diffusion socks, a 22-inch white beauty dish, a Photoflex 42-inch umbrella, a Photek 46" umbrella (a Softlighter without the diffuser), a Westcott 40-inch shoot-through umbrella, a Westcott Apollo Orb, an Interfit 1x3-foot strip box, and a 32" Photoflex white translucent pop-up diffuser panel. I set the camera (this time a Canon EOS M5 mirrorless with the 70-200 f/4 IS lens. AV mode at f/4.5. 

The first set of images shows have no corrections, just as they came out of the camera. For the second set I used the White Balance eyedropper to balance on the middle gray patch. I also attempted to get the exposure evened out between all of the modifier images (click on each image for a larger version). Quite a variety of colors!!

Samples from each modifier straight out of camera with no color adjustments

Samples from each modifier with color and exposure corrections 

With the camera set to Daylight White Balance, Lightroom read it as 5000K and +9 tint. Here is a list of the color corrections for each modifier:

7-inch silver dish: 5050 (+50K) +1 (-8)
24-inch popup: 4900 (-50K) +13 (+4)
20" dish: 4800 (-150K) +15 (+6)
20-inch + speedotron diffuser: 4450 (-550K) +6 (-3)
20-inch = mola diffuser: 4250 (-750K) +14 (+5)
Beauty dish: 5200 (+200K) +10 (+1)
Photoflex umbrella: 4900 (-50K) +4 (-5)
Photek umbrella: 4820 (-180K) -1 (-11)
Shoot through umbrella: 4800 (-150K) +8 (-1)
Apollo Orb: 5000 (0) +14 (+5)
Interfit strip bank: 4900 (-100K) +14 (+5)
Photoflex disc: 4650 (-350K) +4 (-5)

For reference, a +K value added yellow, -K added blue and +tint added magenta and -tint added green.

Have you done a comparison of the effects your various modifiers have on your lights? 

I have.

Thanks for playing along...
John

 

 

Continuing the Diffusion Discussion

Size still matters

In recent posts about diffusion I have been concentrating on small light sources such as speed lights or the "standard" 7-inch metal dish reflectors on studio strobes. But what about larger sources? I was recently gifted a Speedotron 20-inch metal dish reflector. As I am no longer using Speedotrons as my main lights I adapted the reflector to an S-mount for use with my Interfit Photographic lights. This post contains affiliate links and I will be compensated if you make a purchase after clicking on my links.

As I started testing this franken-reflector I noticed that the light was harder than I expected. A look inside the bowl gave some cues as to what was going on. Speedotron lamp heads have large protruding flash tubes and/or flash tube covers. This pushes the light source out in front of the mount on the reflector, allowing the light to completely fill the bowl. The bulbs on the Interfit, though they do protrude, are not quite as long as the Speedo tubes. So most of the light was coming straight out of the flash going forward and not spreading out quickly enough to fill the bowl. 

I did some cutting and drilling to change the position of the S-mount to try to push the flash tube a bit further into the bowl. That still wasn't enough. A quick experiment using a 12x12-inch sheet of diffusion gel in the shape of a dome over the flash tube confirmed what I needed. My first thought was to find a Tupperware bowl around the size of the pyrex flash tube cover, but a quick look through kitchen cabinets didn't turn one up. That got me to thinking that people used to call some speed light flash diffusers Tupperware. And I remembered that I had a very old one of those made of a hard plastic that was specific to a speed light I no longer had. I quickly found that diffuser and a hack saw and went to work.

After cutting off the bottom of the diffuser I was relieved to find it to be exactly the right size to fit in the S-mount. A bit of hot glue and clear caulk and it was fixed in place, ready for testing. And some disappointment. The light was still harder than I expected. Even with the diffuser in place, most of the light was concentrated in the middle and not filling the bowl. Then I remembered that there was a second piece to the diffuser, a dome that snaps onto it. I quickly located that and voila, the second test did show a noticeable change in the shadows and more light "wrapping" around to the side of the subject (notice the ear is brighter). Now the diffuser was doing its job--making the light spread wider to fill the bowl of the reflector. The diffuser itself didn't make the light softer, but it did make the light source larger which did soften the light. 

The next option was to add a diffusion sock over the entire reflector. This has the effect of spreading out the light across the full 20-inches of the bowl making the light source even larger and more even across the surface. Again the diffuser spread the light and evened it out. The last test was to try the sock with and without the dome inside the reflector. This showed a subtle change. With the dome the light source appeared a bit larger, as seen by more light reaching the ear. But the shadow transition was pretty much the same with and without the dome. It was just that with the dome more of the bowl was lit and the light wasn't as directional.

Let's look at the tests (click on the image for a larger version)...

Here are side views of the interior of the modified 20-inch bowl reflector with and without the dome on the diffuser. I made one more modification, not shown here. I drilled a series of small holes around the edge of the snap on dome to help the fans in the strobe head be more efficient.

180827-cornicello-0503.jpg
180827-cornicello-0502.jpg

Safety First
IMPORTANT: DO NOT TRY THIS WITH A STROBE THAT USES TUNGSTEN OR QUARTZ MODELING LAMPS!! I was able to do this with the Interfit strobes because they use relatively cool running LED modeling lamps. For comparison, the 60-watt LED in the Honey Badger is brighter than the 250-watt quartz-halogen modeling lamp in my older strobes. But the LED's temperature after being on for hours remains around 85-degrees F while the quartz bulbs reach temperatures over 300-degrees F in a matter of seconds.


So, you might ask, what is the point of all this today. It is all part of trying to dispel myths around diffusion. It is easy to look at the photos and decide that the various diffusers changed the quality of the light. But it still comes back to the fact that the quality of the light is determined by the size of the light. In these examples the diffusion spread the light to better fill the bowl of the reflector. Without any diffusion the 20-inch reflector is bright in the middle and darker around the edges, making it appear smaller than its actual size. As diffusion is added the light spreads out to fill the bowl, making it appear to be its full size.

 

 

Inconsistent exposures at "in between" ISO settings on Canon 6D mkII

Over exposure at some ISO settings

I started out this evening trying to figure out how to compare noise levels at various ISO settings on my Canon 6D Mark II camera. But instead, I noticed a pattern of overexposure at "in-between" ISO settings. By "in-between" I am referring to the settings between the main 100, 200, 400, etc. These would be 125, 160, 250, and 320.

The exposures at ISO 100, 200, and 400 were consistent with the camera in AV (aperture priority), as were those at 125 and 250. However the photos at 160 and 320 were overexposed by about 1/3 of a stop (the pattern continued throughout the ISO range up to ISO 40000). Of course the first thing I checked was the shutter speed to see if the metering was somehow off in those two photos. No, the shutter speeds were correct at 1/20 and 1/40 (see the progression of shutter speeds below). And I was using continuous light, so it wasn't inconsistent flash power.

How about an inconsistent shutter speed? No. I changed the light a bit and tested again. Same pattern of overexposure, but this time the shutter speeds were the correct 1/25, 1/50, and 1/100. The overexposure is happening due to something else going on in the processing of the image in-camera.

SHUTTER SPEED PROGRESSION
1/10   1/13   1 /15   1/20   1/25   1/30   1/40   1/50   1/60   1/80   1/100

Here is the set of image where you can clearly see that the photos at ISO 160 and ISO 320 are overexposed compared to the other images.

I tested this on two other Canon cameras, a 5D Mark III and a M5 mirrorless, and did not see this issue with either of those. Exposures at all ISO settings were equal.

This morning I added in another set of images lit with flash and see similar results. I took my meter reading at ISO 100 (f/4) then raised the ISO in 1/3 steps while stopping down 1/3 stop for each image in the progression. Again, ISO 160 and 320 were overall a little bit brighter than the other images.

Have you done a comparison of exposure or noise at various ISO settings on your camera? If you have a 6D Mk II please let me know if you see the same exposure issues at the +2/3 ISO settings. 

Oh, here is a set of closer crops in case you want to see the noise comparisons...

At these low ISO settings I cannot see much, if any practical difference. But I will stick with ISO 100 for most of my studio work and bump it up a bit if I need a little extra aperture for more depth of field.

Thanks!
John

Diffusion Confusion (take 3)

Diffusing the situation

Time for my yearly update on the topic of diffusion and how it affects your lighting. As usual, let's start with a definition of diffusion:
• the spreading of something more widely
• the action of spreading the light from a light source evenly so as to reduce glare and harsh shadows.

I will start of by taking some exception with the second definition. The first part about the spreading of light is correct. Reducing glare earns a "maybe," but changing the angle of the light will be more effective (remember the angle of incidence equals the angle of reflection). Now about those shadows... diffusion on its own (and I am talking here about putting some diffusion material right on the light, not using an umbrella or softbox or bouncing off a wall) doesn't reduce shadows. It doesn't soften shadows. It hardly does anything to the shadows. It mostly reduces the output of the light as it makes the light spread out wider to cover a larger area. Diffusion on the light is used to allow the light to cover a wider area when using a wider angle lens. Or it might be used to even out the spread of light across the surface of the light (useful for evenly lighting a background, for instance).

Shadow edges, which we use to describe the quality of light, are determined by the size of the light in relation to the subject. Adding a sheet of diffusion material to the front of a light does not change its size, and doesn't change the shadow transition. The diffusion will make the light disperse in a wider pattern which makes the light seem less powerful as it covers a wider area (which might be a concern for battery-powered lights). It will even the pattern of illumination to make the light more uniform on the subject (more noticeable with larger light sources). But it won't make the shadows any softer. To do that you need to make the light bigger. Do that by using a scrim that is placed a distance in front of the light and that is much larger than the light. 

Here is a set of images lit with a 7-inch dish reflector and with a 20-inch dish reflector. The first image in each series is the bare reflector. The middle image adds one sheet of diffusion material over the reflector. The third image has two sheets of diffusion material on the front of the light. Look at the nose shadow in each as you go across. 

Comparing 7-inch and 20-inch reflectors with and without diffusion on them (click on the image to enlarge it)

light-wrap.jpg

The shadow edge across each row does not change as diffusion is added. To change the shadow edge and make it softer you need to make the light larger. Notice that the 20-inch reflector with no diffusion is softer than the 7-inch reflector with double diffusion. Diffusion material on the light spreads the light to cover a wider area, it doesn't soften the light. No matter how close you bring the 7-inch reflector to your subject, or how much diffusion material you put right on the reflector that light is never going to be larger than your subject's head, and hence will always be a small/hard light source. Soft light comes from light being able to reach your subject from a variety of angles instead of just straight on. Look at the diagram here. You see that from the small light source all the rays are concentrated leading to hard shadows. A bigger source from the same position provides its own fill light to soften the shadow. Some people say that the light "wraps" around the subject, but it is all traveling in straight lines. If you were to take that larger light source, though, and back it up a ways it will become relatively smaller compared to your subject, and hence more directional causing harder shadow edges. So what about the scrim I mentioned above?

Make it bigger!
Yes, you can make the light bigger by using a diffusion scrim, but the scrim has to be separated from the lamp, not used right up against it. Here we have the same 7-inch dish at 36-inches from the face with "double diffusion" and next to it the dish with a diffusion scrim added about 17-inches from the face. Notice the big difference in the shadow edges. The strobe head (the origin of the light) remained in the same place. But the source of the light changed from the strobe head (small) to the scrim (large). The scrim made the light larger and closer, both of which help soften the light.

Comparing 7-inch reflector with diffusion vs. adding a diffusion scrim at a distance to make the light source larger

This is where I think the confusion comes in. In both cases the light has been diffused by translucent material. The difference is in where the diffusion is applied and how it affects the size of the light and thereby the quality of the light. 

scrimlight.jpg

Size Matters!
It comes down to this... Use diffusion to spread the light wider, use size to control the quality of the light. A large undiffused light is going to be softer than a small diffused light.

There can be an advantage to using a small light with a scrim over using an umbrella or a softbox--more control. You can place the scrim at the position you want to control the falloff of light from the subject to the background and then you can move the light in closer to the scrim making it smaller to make it harder or back it away to light up more of the scrim making the light softer.

Previous blog posts about diffusion:
2016
2017

More about light size and distance is covered in by book Anatomy of a Studio Portrait. You can help support this blog by purchasing the book at Amazon.com 

Thanks!

John

 

High Speed Sync vs Neutral Density Filters

HSS vs Neutral Density

I was reading online forums again (yeah, I know...)... I saw a discussion where a photographer was trying to figure out exposure settings for outdoor flash vs ambient light using neutral density filters. He was adamant that he did NOT want to use high speed sync (HSS) because HSS robs the flash of a lot of its power.

This got me thinking... Doesn't using ND filters also rob the flash of power? If you put a 6-stop ND filter over the lens you are effectively lowering the power of the flash 6 stops and also lowering the amount of ambient light 6 stops. If you use high speed sync to raise the shutter speed by six stops you lower the ambient light 6 stops. You also lose power in the flash, about the same 6 stops. Seems pretty much equivalent.


NOTE: Some of the links in this post are affiliate links and I might be compensated if you purchase equipment using the links. But you will also get a discount by using the link or the code CORNICELLO10 on the Interfit Photographic web site. Win/Win situation!


Testing:
So, time to do some testing. Camera is a Canon 5D mkIII with an 85mm f/1.8 lens. All of the photos were made with the white balance set to Daylight. Flash is an Interfit Photographic S1 with high speed sync capabilities modified with a 24-inch collapsible beauty dish. Ambient light meter reading was 1/125 at f/9 and the flash was a bit brighter (f/13). I wanted to make the photographs at f/1.8. 

I needed about 5 and 2/3 stops of neutral density. I had two 0.9 (3-stop) neutral density filters handy, so I stacked them on top of each other and made the first series of photos below with normal sync. Then I removed the filters and switched the Interfit S1 to high speed sync mode and made a similar set of images above the normal sync speed of the camera. The third set as the Interfit S1 in HSS mode plus TTL metering. All the photos were imported into Adobe Lightroom Classic with no adjustments made to them and then I made the following groups via the Print module. The power level on the strobe was not changed between shots in the first two sets of photos. In the third set the strobe was in TTL mode, so it did vary the power (raising it as the shutter speed increased) to maintain the proper exposure.

While this test isn't super scientific, the things I notice are a definite color cast in the photos made with the ND filter, the ND photos do not appear to be quite as sharp as the HSS photos (a complaint I often hear about variable density ND filters, but these were two single density filters), and the ND filters I used required me to take off the lens shade (something I rarely if ever do) to attach them. The ND filters make it more difficult to see through the viewfinder (6 stops more difficult, I don't want to do the math to figure out how many times darker that is). I could/should have bumped the flash power up a third of a stop or so because of the extra density. The background would have remained the same with the subject being a little bit brighter. But it is what it is. All in all a bit of a pain to work with, but it gets the job done if you have a strobe unit that doesn't do HSS.

In the middle set, using the flash in manual exposure mode with high speed sync. Again, I should have/could have bumped up the power of the flash as I went above 1/3200 sec. on the shutter speed. But the last three are still OK and salvageable. 

The set with the S1 in high speed sync mode and TTL exposure seems to be the winner to me. I was able to adjust the shutter speed to make the trees in the background lighter and darker while at the same time keeping the exposure on the subject pretty consistent. I did not make any adjustments on the flash.

Until fairly recently, I hadn't used HSS or TTL all that much with my flashes. I've been a pretty strict manual mode and in the studio type of photographer. Working with the Interfit S1 strobes has changed my mind about this. What are your thoughts about using neutral density filters to balance flash and daylight versus using high speed sync? 

For those interested in how to figure out the exposure using ND filters, here is how I do it.

  • Take a normal meter reading at 1/125 sec. shutter speed (this gives me leeway to raise the shutter speed slightly to darken the background without going over the sync speed of the camera)
  • Example: ISO 100, 1/125 Sec. at f/11
  • Adjust the output of your flash to read the same f/stop (f/11 in this case) or a bit higher WITHOUT THE ND FILTER (I went for f/13 above)
  • Decide on the aperture you want to use for depth of field. In this case I wanted f/1.8
  • Figure that from f/13 to f/1.8 is 5 and 2/3 stops
  • Find a 5 and 2/3 stop neutral density filter or adjust your aperture to match the ND filter(s) you have. If you have a 3 stop filter you can go from f/11 to f/4, with a 4 stop filter you can go to f/2.8, with a 6 stop filter you can go to f/1.4 (I opted for stacking two 3 stop filters above)
  • Take the photo with the ND filter(s) in place
  • Adjust the shutter speed up/down to darken or lighten the background
  • Adjust the power of the flash up/down to get the proper exposure on the subject
  • Deal with focus and color issues

To figure out the exposure with high speed sync

  • Decide on the aperture you want to use
  • Set the camera to that aperture
  • Make sure your flash is in HSS mode
  • Adjust your shutter speed up/down to darken or lighten the background to how you want it to look
  • Adjust the power of the flash to give proper exposure on the subject or use TTL if available

Do note that HSS on speed lights will run through batteries quicker. The folks at Interfit, though, tell me that the S1 battery actually lasts longer in HSS mode. HSS may also shorten the life of the flash tube. But everything is a tradeoff in photography.

Revisiting Shutter Speed In Relation to Flash Photography

How shutter speed affects flash in and out of the studio

 Does this ever happen to you?

Does this ever happen to you?

From in-person discussions, q&a sessions in CreativeLive classes I’ve been involved in, and from online discussion groups I have come to the conclusion that shutter speed when using flash is a difficult concept for photographers to grasp when first introduced. We’re all familiar with the three legs of the exposure triangle: ISO (sensitivity), Aperture (amount of light passing through the lens), and Shutter Speed (how long the light is allowed to reach the sensor. For a more in depth look at how shutter speed affects your photos please visit this guide on the PhotographyTalk site). When we add flash into the situation there are another two legs added, the amount of light and the duration of the flash. And at the same time, the Shutter Speed leg’s effect is somewhat tossed away. This makes it sound more complicated than it actually is.

exposure_variables

It is a double exposure
I find that it helps to think of flash photography as making a double exposure with one click of the shutter button. We have an exposure for the existing ambient light that is controlled by the big three (ISO, f/stop, shutter speed). And we have a second simultaneous exposure for the flash that is controlled by ISO, f/stop, and flash power. As ISO and f/stop affect both exposures we can eliminate them from the exposure equation. That leaves us with shutter speed to control the ambient light and flash power to control the flash light. More about that when I talk about working with flash outdoors in sunlight. Let’s start with studio lighting.

In The Studio

What I am saying is that in studio flash photography in the studio the shutter speed doesn’t matter all that much within a certain range, usually around 1/30th of a second to 1/200th of a second. That’s around a 3-stop range. How can that be? What about motion blur at 1/30? 

In the studio we have control over the ambient light situation. We can make the studio completely dark so that even a full 1-second shutter speed at a typical aperture of f/8 or f/11 will not record anything on the sensor. Then we add the flash. The flash provides the powerful light that does record on the sensor. But it only records for that split second that the flash is firing (the flash duration). So out of that 1-second the shutter is opened, the flash is only firing for a fraction of the time. Maybe 1/300 of a second for a big old powerful studio power pack and head system or only 1/9000 of a second for a newer flash unit. The flash duration has effectively replaced the shutter speed in terms of both providing the light for the exposure and for providing the speed necessary to stop motion. Now I am not advocating a shutter speed of 1/30, just using that as an example. Go ahead and test this yourself. Set the ISO and aperture you would be using in the studio (ISO 100 and f/8 is a good starting point) and turn down the ambient light to have a darkened studio. Before connecting and turning on your flash take a photo with those settings to see what gets recorded. In many cases the frame will be black, or show just a faint image. If you see too much image raise your shutter speed to 1/60 and try again.

So why set 1/200 as the other end of the range for studio flash photography? That comes down to how the focal plane shutter in our camera works. With a focal plane shutter (which is what we have in most dSLRs and many mirrorless cameras) there is basically a set of two curtains (simplified, as some have multiple blades, but the effect is the same). When you press the button to take a photograph the first curtain slides out of the way to allow light to reach the sensor. Then the second curtain slides across covering the sensor to end the exposure. To synchronize this with a flash the flash has to fire when neither curtain is covering the sensor. The fastest shutter speed at which there is no curtain in the way is the sync speed for that camera.

Here is an illustration of what happens. On the left we see the first curtain open up to expose the entire sensor in the camera. Then the flash fires, exposing the entire scene. Then the second curtain closes to end the exposure. On the right we see that above the sync speed the first curtain opens, then the second curtain starts to close before the first curtain clears the sensor. Then the flash fires and gets blocked by one of the curtains causing a black band along one edge of the photograph.

 Syncronized shutter

Syncronized shutter

 Out of sync shutter

Out of sync shutter

Your camera specifications might tell you that the sync speed is 1/180, or 1/200, or even 1/250, so why not go that high with the shutter speed? That number is usually in relation to using a dedicated flash on your camera in the hot shoe. As an example, look at the user manual for the Canon EOS 6D mark II. On page 280 it says:

Non-Canon Flash Units
Sync Speed

The camera can synchronize with non-Canon compact flash units at 1/180 sec. and slower speeds. With large studio flash units, the flash duration is longer than that of a compact flash unit and varies depending on the model. Be sure to check before shooting if flash sync is properly performed by test shooting at a sync speed of approx. 1/60 sec. to 1/30 sec.

Additionally, the use of a remote radio signal flash trigger can add a little bit of a delay to the firing of the flash, requiring a slower shutter speed than expected.

So, what happens if you do set your shutter speed too high? You will notice a dark band along one edge of your photo or no image at all. Here is a set of test images I created with the Canon EOS 6D Mark II and studio strobes (two Interfit Honey Badger flashes, one on the subject and the other on the background). The shutter speed for each photo is shown above the mannequin head. You can clearly see that the flash synced at speeds up to 1/125, but at 1/250 there is the start of a black band along one edge and at 1/2000 the frame is completely black. With some lower cost remote flash triggers you would even see the black band at 1/125, so you would need to set your shutter speed to 1/80 sec. or 1/60. If you have had that black band along the edge of some of your flash photos, here is the explanation.

 Studio flash at various shutter speeds

Studio flash at various shutter speeds

High Speed Sync

For some time now dedicated camera flashes (I will refer to them as speed lights here) have offered a feature called High Speed Sync (HSS). This feature is now becoming available in studio flash units like the Interfit S1. It allows the flash unit to synchronize with faster shutter speeds. This comes in most handy when working outdoors in daylight and you want to make photos at wider apertures such as f/2.8 or wider and want to provide some fill light from the flash or even use the flash as your main light source. The issue there is that a typical mid-day exposure at ISO 100 and f/2.8 might call for a shutter speed of 1/2000. As we saw above, the flash won’t have any effect on the image at 1/2000 sec. 

Another situation is where you want to darken the ambient light exposure so that the flash is the main light on the subject. Changing your ISO or aperture will affect the flash exposure as well as the ambient exposure, so your control over the ambient light is your shutter speed. We’ve seen that you are pretty limited in available shutter speeds with normal flash sync. Only being able to speed up to 1/180 or 1/200 doesn’t give you all that much control to darken the ambient light.

Enter HSS
With HSS the flash fires in a stroboscopic fashion (pop, pop, pop, pop, pop) synchronized to the movement of the shutter curtains during the exposure. This happens very quickly, too fast for the human eye to see the multiple flashes, so it still looks like one flash burst. Here is a visual explanation of high speed sync...

 On the left there is too much available light to be able to work at f/1.8, everything is overexposed. On the right HSS has been enabled and the shutter speed set to 1/8000 sec. to take the background ambient exposure way down and have the Interfit S1 battery-powered flash provide the main exposure on the subject. Here the S1 is being modified by a 2x3-foot softbox.

On the left there is too much available light to be able to work at f/1.8, everything is overexposed. On the right HSS has been enabled and the shutter speed set to 1/8000 sec. to take the background ambient exposure way down and have the Interfit S1 battery-powered flash provide the main exposure on the subject. Here the S1 is being modified by a 2x3-foot softbox.

Trade-offs
While HSS allows the flash to sync at shutter speeds up to 1/8000 sec., there is a catch. Everything in photography is a trade-off. With HSS the trade-offs are power loss, battery life, and a shorter lifetime for the flash tube. If you have used flash before you are most likely familiar with recycle time. Each time the flash fires it releases energy stored in capacitors. Those capacitors need to charge up again to provide the power for the next flash. Depending on the flash, the time for the recharge can be from  a bit less than 1 second to 8 seconds or more at full power. The time to recycle goes down as the power of the strobe is turned down. Do you see where this is going? In order to have immediate recycle times to allow a number of rapid fire flashes within 1/8000 of a second (as per our example) the power of the flash has to be set way down--you get much less light out of the flash. Right now it seems that most HSS flash units have a maximum power output of around 500 to 600 watt seconds. Anything more powerful would not recycle fast enough or would require larger/heavier power units and be much more expensive. This limits us in that the flash has to be in pretty close to the subject, especially if using big modifiers like soft boxes or umbrellas. Or you end up using smaller/harder light sources and have to deal with the consequences of that. And all that rapid firing of the flash can shorten the lifetime of the flash tube (which is usually a user-replacable item).

 Without High Speed Sync you can see the shadow of the shutter curtain when the shutter speed is set higher than the sync speed of the camera

Without High Speed Sync you can see the shadow of the shutter curtain when the shutter speed is set higher than the sync speed of the camera

Do we need High Speed Sync?
It depends.
Yeah, I hate non-commital answers like that. But it does. If you want to totally blur the background on a portrait of one person yes, you might want to make the photo at f/1.4 or f/2.0. And for that you will need to use HSS to be able to work at the high shutter speeds you need in daylight with those apertures. But if you have multiple subjects like a family group you will often be working at f/8 or f/11 or even f/16 to get everyone in focus. At those apertures there is a good chance your shutter speed might drop down to 1/200 or slower and you won’t need HSS.  Even with an individual, if you can pose them someplace where the background is far enough away, you might be able to still get some separation at smaller f/stops and not need HSS. 

For more information about shutter speed in general and how it is used in non-flash photography please refer to this excellent article on the PhotographyTalk website

Thanks for following along! I cover all this and more in my book Anatomy of a Studio Portrait. I also teach lighting classes in the Seattle area at Glazer's Camera and at Photo Center Northwest. Private instruction is also available. Now please go out and light up the world!

John