I just read "Lessons Learned on The Flash Bus Tour" and saw this image. In it, an assistant is holding some sort of white shoot-through diffuser (I got a similar one in my 5-in-one reflector kit) right in front of a soft box, which is lighting a portrait from above.

This is Joe McNally and David Hobby teaching at a workshop, so I assume there's a good reason for this. However, I don't know what it could be, or why it works.

  • My understanding is that the larger the effective area of the light source, the softer the light.
  • The softbox accomplishes this by bouncing the point-source light from the strobe around inside the box. When it exits, it's coming from a much larger "hole" (the front of the softbox), and from multiple angles, so it's nice and diffuse.
  • The hand-held diffuser doesn't increase the surface area of the light much at all, since it's being held so close to the soft box.
  • It would definitely reduce the amount of light, and I can understand needing that, but I'd think you could also do that by reducing the flash power (and saving your batteries).

So, what is the intent, results, and mechanism behind adding in the additional diffuser?


3 Answers 3


There are a few reasons that I'm aware of to combine soft-box and an additional diffuser:

  1. With smaller softboxes and hotter lights (read: flashes) the softbox isn't always able to completely diffuse the light source, which causes a 'hotspot'... Essentially the center of the diffusion 'square' is brighter than the edges. It's better than a 'nekkid' flash, but often you can still see the effects of that hotspot on the subject. There are several solutions (lower flash power, bigger box, move the flash further back in the box, etc.) but often the most expedient way to deal with the problem during a 'live' shoot is simply to throw another layer of diffusion in front of the box to soften out the hotspot.

  2. To lower the power of the light without having to rip open the box and fiddle with the flash. If you're within a stop(ish) of where you want the light to be and your light source is a manual flash where the controls are not easy to get at, it's often more expedient to simply lower the light by throwing another layer of diffusion in there than to monkey with the controls on the flash.

  3. It is possible to make a small box somewhat bigger (at the expense of power) by hanging a diffuser in front of it... Now you're correct that this is not what's going on with this particular picture (the diffuser is too close to the box to have much effect in this regard), but for the sake of 'completeness' I thought I'd mention it. :-)

For me personally, adding a diffuser is mostly a 'hack' manuver (not hack as in 'unprofessional,' hack as in 'hacker') when I have actual clients in the studio and I don't have the time (or don't want to take the time) to be fussing with the lights while they sit there and wait for me to get it right by 'dialing it in.' If the light is already 'pretty close' but I just need to dial it down 'a bit more,' or 'soften it slightly' it's usually a big time savings just to hack the solution, rather than fiddle with making it 'really really right' by dialing it in at the source, or moving the lights back and forth... again all while the client sits there and waits for me (extra not good if the client is in the short-attention-span set... AKA a child). :-)

  • \$\begingroup\$ JoanneC: Where is yours? \$\endgroup\$
    – ysap
    Commented Apr 28, 2011 at 23:19
  • \$\begingroup\$ @ysap - I removed it, Jay was far more complete and expressive, so why keep it around? :) \$\endgroup\$
    – Joanne C
    Commented Apr 28, 2011 at 23:24
  • \$\begingroup\$ JoanneC: well, if everyone would do this, we'll end up with a concise version of Photo.SE ... Remember, this is a discussion forum, not an encyclopedia. Your answer could raise further discussions in the comments. \$\endgroup\$
    – ysap
    Commented Apr 29, 2011 at 0:41
  • \$\begingroup\$ @ysap - I generally agree, but it doesn't really add to the discussion in this case as it was a quick, off the cuff, answer and Jay's is better. Otherwise, I would have left it. \$\endgroup\$
    – Joanne C
    Commented Apr 29, 2011 at 1:11
  • 1
    \$\begingroup\$ @ysap: Actually, according to the FAQ, a discussion forum is specifically not what photo-SE is... With that said, I understand (and agree with) the underlying sentiment of "multiple answers are better than a lone answer" that you were trying to convey. Just wanted to make sure we weren't inadvertently propagating misinformation... \$\endgroup\$ Commented Apr 29, 2011 at 7:26

This question (and its answers) is wa-a-a-ay old, but it could stand to have another couple of good reasons thrown in for good measure. The existing answers are good, but they don't touch all of the bases.

The first addition is more pertinent to product photography, especially when photographing glossy surfaces (glass and polished metals, in particular). In those instances, a lot of what you're doing is taking a picture of the light sources as they are reflected from the subject, and not the subject itself. Frankly, the hard edges of a softbox's front surface aren't nearly as aesthetically pleasing as they might be, and placing a diffuser between the softbox and the subject allows a gradation of the edge, and thus a gradation of the reflection in the subject. Placing the softbox close to the diffuser and at an angle to it allows you to independently control the gradation on two sides of the light source. (With the softbox touching the diffuser, the cut-off is hard The further away the softbox is, the softer the edge will be.) You also get a controlled gradation across the light source on demand, something you can't get from a softbox.

The other missing reason is simply that the softbox's front panel might not offer as much diffusion as you want. The choice of diffusion materials is a bit of a compromise between diffusion quality and efficiency. It is possible to build a box (especially in the system-branded units) that produces a very even amount of light across the front panel and is still too efficient. That is, there might be an awful lot of light, but it's all headed in more-or-less the same direction, with relatively little spread from any given location on the front panel. That limits the ability of the softbox to fill its own shadows, not too terribly unlike using a fabric grid. Again, that depends on the individual box, and it's rarely the case with indirect sofboxes like the Westcott Apollo or the Elinchrom EL Octa, where the flash faces the back of the box and all of the light reaching the front surface has to bounce around first. Adding another diffusion panel in front of the softbox can significantly soften the light (at the cost of significantly reducing it, of course) when the box is too directional.


I can think of a couple of reasons why I might do that:

  1. The minimum power on the strobe is still more light than I want and so extra diffusion is needed. It can happen... :)

  2. The light duration is what I want, but the amount of it is too much. If I reduce the power on the strobe, I reduce the duration of the light, so this lets me reduce the amount of light without adjusting the power.

I suspect there may be others, but these two immediately come to mind.

  • 1
    \$\begingroup\$ Undeleted it under peer pressure... :) \$\endgroup\$
    – Joanne C
    Commented Apr 29, 2011 at 14:57

Not the answer you're looking for? Browse other questions tagged or ask your own question.