Using hot shoe flashes in softboxes is becoming more and more popular and we're seeing an explosion of brackets and gizmos for mounting flashes inside softboxes.
Whenever I thought about placing a flash with a rectangular front lens (such as a Nikon or Canon Speedlight, not a flash with a round flash tube like a Sunpak 120J) into a rectangular softbox I made what I thought was a logical assumption by rotating the flash head so the orientation matched that of the softbox. I was pretty confident in that conclusion and thought it made the most sense and would produce the most even light.
As it turns out, that was a bad and incorrect assumption.
I spent some time doing a series of tests with what I consider to be average and commonly sized softboxes used with hot shoe flashes. The softboxes tested were a Photoflex 24" x 32" and 36" x 48" LiteDome Q39. Both had white interiors with both layers of diffusion installed. The flash is entering the back of the softbox and firing forward with the flash zoom setting on 24mm. I tested with a Nikon SB-900, which is roughly the equivalent of a Canon 580EX II in size and power.
Using a Sekonic L-758DR at 4 feet from the front face of the softbox, I took readings at the center, the long edge and short edge (about 2 inches from the actual edge) with both orientations. I no longer do tests or evaluate readings taken on the front face of the modifier. I really don't care what's happening there. My concern is what happens at real world working distances from the lightbox. I also didn't take corner readings. I don't care what happens in the corners.
Bottom line....
Center reading is as expected, and the same for both orientations.
Long edge reading is the same as the center for both orientations.
Short edge reading was down 1/10 stop (- .1EV).
These results were encouraging and I'm no longer worrying about orientation of the flash within a rectangular softbox.
I hope others find this snippet of info helpful.
