Working on Backgrounds

M is for Mantis
Originally uploaded by Dalantech.

Still experimenting with different materials and techniques. First up is a mantis that I shot in front of a piece of light blue construction paper. II though that since the paper had a matte finish it would make a good background, but oddly enough it looked kind of grey in some of the images that I took. Also notice that the pseudo pupil is almost missing in the critter's left eye -that effect is caused by the way the flash reflects off of the compound eyes (a mantis doesn't really have a pupil). The second flash that I was using to illuminate the background all but washed it out, and what's left is in the wrong position. In the future if I use a second flash when shooting a mantis I'll place it directly behind it.

For this next shot I used a vinyl table cloth, and shooting against a reflective surface works a lot better than shooting against one that has a texture.



There's no natural light in those first two shots, but this next one is a mix of natural light and flash. I set the shutter to 1/20 of a second to expose the reeds in the background, and then let E-TTL metering expose the subject with the flash. This is my preferred method for keeping the background from being black, especially when I can get a good mix of natural light and flash. Even though the sunlight was pretty harsh (taken at about 3PM on a cloudless day) the color and saturation in the background look good because it's out of focus and a little under exposed. A digital sensor reacts to under exposure in the same way as color positive slide film -colors saturate. The downside to this method is that there can be enough natural light to partially expose the subject which makes freezing motion difficult.



I've also noticed that when creating artificial backgrounds that smooth backgrounds do not look as good as ones that have some color or texture variation (especially if the background is not blue). The brain sort of expects vegetation to keep the background from looking smooth. Compare the dragonfly in the previous image with the bee in the next shot -this one is all flash:

Macro, the 1/f Rule, and Flash Durations

Many, many, moons ago Fred Vachss made a really interesting post over at DP Review concerning the 1/f rule and flash durations as they pertained to macro photography. Basically we were all wondering what shutter speed (or flash duration) was necessary for freezing motion at different magnifications. The mistake that a lot of people made back then was jumping all over the lowest flash duration (flash at minimum power) and claiming that the flash could freeze anything. Unfortunately you'll almost never be shooting with a flash at minimum power, so you'll rarely see those lightning fast bursts of light. The only time I've been able to expose a brightly colored subject at life size magnification and 1/64th power with the MT-24EX is when the flash heads are bare. Not exactly the best light quality... ;)

 If you boil Fred's post down here's the 1/f rule for macro photography: Shutter speed = 1/f((1+m)2) where f = focal length of the lens and m = the magnification you're shooting at. So at life size magnification you need a shutter speed that's four times faster than using the same focal length lens for a non macro photo, at twice life size your shutter needs to be six times faster, etc. Substitute flash duration for shutter speed if you're using a flash as the primary, or only, light source. Also keep in mind that the 1/f rule, even modified for macro photography, is a minimum requirement. To "be safe" you'd never want to get close to 1/f rule speeds. Now let's add insult to injury and explain why: Noticeable diffraction is defined as light rays spreading out so that they strike half way into adjacent pixels. The 1/f rule assumes that you're trying to keep the motion in the scene to less than one pixel, and not to less than one half of a pixel. So how easy is it to amplify diffraction with very small amounts of motion, less than the width of half a pixel, while the flash is firing? It's always been my contention that diffraction isn't the monster that most people have made it out to be, and that there is a very synergistic effect between motion and diffraction. You won't necessarily see motion blur in your images but if you're not doing everything you can to minimize movement, and to keep your flash durations as short as possible, you'll have difficulty getting a lot of detail when you shoot at high magnification and high Fstops.

One of these days, just for the focus stacking crowd, I'm going to make a post about how your photos may be diffraction limited even if you're shooting with a lens that's wide open. It has to do with the diffraction limit of green light being four micro meters at F2.8, at the center of a digital sensor (as you move away from the center of the sensor diffraction gets worse). So if your camera has pixels that are four micro meters wide (like the 50D, 60D, and 70D), or smaller (like most of the Power Shot models), you're already recording diffraction at F2.8...

 P.S. Not that I have anything against focus stacking -it's just another tool in your tool bag, and some day I might actually stack an image or two. But I do think that a lot of the people who stack their images are actually doing more to defeat macro motion blur than diffraction...