February 1, 2012

bounce flash photography & the inverse square law

After you’re done noticing the decorated candles that the bride is holding while dancing with the groom (a tradition in Palestinian weddings), you may well notice how evenly lit this photograph is – from foreground to background.

The people visible in the background seen there between the bride and the groom, are nearly as well lit as the bride and groom. Because this was on-camera bounce flash, the background will be brighter than may have been anticipated. If I had used direct flash, or flash with a diffuser cup or bounce card, my background would’ve been much darker. This is because when we bounce flash behind us, the Inverse Square Law works for us.

This gets interesting, but hopefully we can make it less complicated than the topic often appears. So hang in there …

As comparison, here is an image taken shortly before, where I shot too fast for my on-camera flash to keep up. So this misfire will give you an idea of the effect the single flash had.

My camera settings:
the image without flash:  1/100 @ f3.5 @ 3200 ISO
the image with flash: 1/80 @ f3.2 @ 3200 ISO … TTL flash set to 0 FEC; gelled with 1/2 CTS
Nikon D3S  (B&H); Nikon 24-70mm f2.8G ED AF-S  (B&H)
Nikon SB-910 Speedlight (B&H);  Nikon SD-9 battery pack (B&H)

Looking at this image again compared to the image where the flash didn’t go off, we can see just how much the flash lit up the entire room. Now, it helps to keep in mind that it was the flash riding ‘on top of’ ambient light, and not quite the flash doing all the work here in getting to the final correct exposure.

Even though this reception venue is quite large, I was able to use only an on-camera bounce flash. I bounced it over my shoulder, using a black foamie thing to flag it so I don’t hurt people behind me with the blindingly powerful flash. I was able to do it because of the incredible high-ISO performance of the Nikon D3S  (B&H) which allowed me to comfortably shoot at 3200 ISO.

Now, when we bounce flash like this without a plastic diffuser or bounce card, we don’t think of our flash as our light source anymore. Instead, the surface that we bounce off, becomes our light source. In this case it was the ceiling and walls above and behind me. The walls behind me were about as far away as the walls there in the back of the photograph. In other words, the de facto light source is now quite some distance away.  And this is where the Inverse Square Law kicks in to help explain why the background is relatively bright in comparison to the subject.

 

the inverse square law

The Inverse Square Law implies that as we move away from a light source, the light becomes less bright in proportion to the square of the distance.

This video clip by Mark Wallace as posted on the Fstoppers site, is the best explanation of the Inverse Square Law that I’ve seen to date. Mark Wallace really breaks the daunting topic down to something which can be visually understood, and those 12 minutes will be a solid investment in your photography.

In a similar way, I want to make it visually understandable why bouncing flash behind you would open up the background of the image.  It might sound counter-intuitive, but here is what happens. And it relates directly to the Inverse Square Law.

Let’s work through an example:

If we measure a (manual) flash at 12 feet and let’s say we measure f8
then if we measure the same flash at twice that distance (24 feet),
we will lose 2 stops of light. Our light will measure now measure as f4

If we double our distance from the light, then we will lose another 2 stops of light.
So if we move from 24 feet, ie, where we measure f4.0
then at 48 feet, we will measure f2.0

Every halving or doubling of distance means a 2 stop difference in light.

This is where the gradient of light fall off becomes important.

If you move from 6 feet (which would measure f16 in this scenario)
to only 12 feet away (where we measured f8)
then a mere 6 feet change in distance implied a massive 2 stops drop in light.

But moving from 24 feet to 48 feet – a huge jump in distance,
would mean the same drop in light levels … 2 stops.

Those are the numbers. And the theory works in practice too. The clear implication is that the gradient of light fall-off is less steep, the further we move from the light source. This is hugely important!

 

how all this adds up with bounce flash photography on location

In a real shooting situation, we might not get as technical as that. We just need an innate understanding of what is going on.  The further our subject+background moves from the light source, the brighter our background will appear, relative to when our subject+background is much closer to the light source.

Now if we look at the image at the top … the people in the background are not twice the distance from my light source compared to the bride and groom. They are closer than that. Therefore, they will be only a stop or so darker than my subjects. Or even less than a stop. The fall-off becomes much slower now at this distance from our light.

If I had used direct flash, then the people in the background would comparatively have been much further from my light source than the bride and groom .. hence, the people in the background would’ve been much more under-exposed with direct flash …. than they would’ve been with flash bounced over our shoulder behind us.

 

another example

I’ve used the next two images in the article on flash photography during the wedding ceremony in church.

1/125 @ f3.5 @ 1600 ISO
on-camera TTL bounce flash (gelled with 1/2 CTS)

The shot without flash is uniformly dark. The shot with flash bounced behind me into the front of the church … the background brightens up considerably more than we might’ve anticipated. People behind the bride and her parents, are well lit by the on-camera bounce flash.

Again, it happens this way because the people in the background are not that much further from my light source, than the bride and her parents are from the light source.  The Inverse Square Law implies that the further our subject+background is from our light source, the less dramatic the light fall-off between our subject and background.

 

summary

Hopefully this article helps easing in a subject which can be daunting from the outset.  And hopefully this article helps with the “feel” if not the immediate math. The part we need to intuitively grasp here is that if we bounce our flash behind us, our background opens up.

And for those already familiar with the top of the Inverse Square Law, I hope that this article brings it in again from a different direction.

Let me know if there are any questions, and let’s work through this. Understanding all of this (to some degree at least), will help our understanding of flash photography.

 

related articles

 

{ 46 comments… read them below or add one }

1 Paul Pearce February 2, 2012 at 3:19 am

I can relate to this and it does make sense. In a way you are increasing the light source to one big umbrella and hence the background comes into focus and depth is given. Love it

Reply

2 Neil vN February 2, 2012 at 4:42 am

Paul .. just to be sure we’re on track here:
The size of the light source doesn’t affect depth-of-field.

Reply

3 Neil Marks February 2, 2012 at 4:55 am

Hi Neil,

Thank you for posting this article, and the accompanying video from Mark Wallace. I am an avid reader of your blog and this is one of the most eye-opening and instructive items I have read on flash photography. I was aware of the inverse square law prior to this, but the implications of this article have never before sunk-in!. It has been a genuine “aha” moment for me. Understanding how the light source to subject distance impacts on the light falloff for the background area is something I had never understood or even considered. The ISL maths should have made this obvious, but as usual it takes someone like you to point out the blindingly obvious!. Thanks again for a great site. I have your on camera flash book and will soon be buying the second book.
Neil

Reply

4 John Woods February 2, 2012 at 8:43 am

Love to see examples of how good on-camera flash can be! Quick question, when you are shooting like this, with the flash pointed backwards, your BFT is “under” the flash (relative to horizon), correct?

Reply

5 Roel Van Ranst February 2, 2012 at 10:09 am

Neil,
What about the following statement:

The background is lit up almost evenly because – by bouncing – the relative lightsource (I.e. the massive wall you bounced off) becomes so big that the relatively very small difference in distance hardly makes a noticeable difference in lighting.

(freely adapted from one of Strobist’s 101 or 102′s where David manages to evenly light a matchbox toycar with a single speedlight just by limiting the distance between the two. If you allow the link: http://strobist.blogspot.com/2007/07/lighting-102-unit-21-apparent-light.html

Regards,

Roel

Reply

6 Neville Stringer February 2, 2012 at 10:47 am

Thanks for the great explanation. Conversely, if the shot was taken in a smaller hall with very low ceilings (say 10ft), then the background could not be as evenly lit because there is greater distance between the light source (the ceiling/wall) & the B&G and the people in the background? Just trying to solidify my understanding of the theory.

Reply

7 Neil vN February 2, 2012 at 11:58 am

That is exactly what happens in a reception room with lower ceilings where the room is longer and less boxy.

Reply

8 Peter Salo February 2, 2012 at 12:06 pm

Got it!!! Thanks Neil, as always!

Reply

9 H.Dilrosun February 2, 2012 at 12:12 pm

This was absolutely a big Ahaaaa !
Best tut I’ve read.If there was a price for “best tut” I would give you a big vote.
Cheerio from Curacao

Reply

10 Yasmeen Anderson February 2, 2012 at 12:28 pm

oooooo, ahhhhhhhh. I love your simple explanations of seemingly complicated photography topics. Thanks as always!

Reply

11 naftoli February 2, 2012 at 3:20 pm

hey beat me to it! i was just thinking of starting a thread about this that would be called something like another benefit to bouncing! one more plus is that when using manual flash if u set ur flash to near full power the chances of getting a gross over or underexposer is much slimmer because the falloff is so slow, b/c u have a much larger sweet spot of correctly exposed area

Reply

12 Neil vN February 2, 2012 at 3:22 pm

Just so we stay on track here …

Manual flash has nothing to do with light fall-off falloff.
Neither does and any possible under / over-exposure have an effect on light fall-off.

Reply

13 Stephen February 2, 2012 at 3:26 pm

This is a good blog post for everybody, since a number of us were all trying to make sense of this in the forums. Thanks!

Reply

14 Arnold February 2, 2012 at 4:02 pm

Neil, I am trying to analyze the light direction in the first shot and by looking at it, the light direction seems to me different to what you described. Looking at the brides shadow on the face (shadow on left-short lit) and her color bone (shadow casting down) and also the grooms back, the light seems to be coming from the top and camera right. The light bounce you described from behind would make the image flatter with the grooms back lit more from the back and the brides face would be lit more even with less shadow on her color bone etc. The light is nice and directional, I am just surprised that you can get that kind of direction from bouncing light behind you. Great job as always.

Reply

15 Neil vN February 2, 2012 at 4:08 pm

My description mentioned “I bounced it over my shoulder, …” clearly implying I didn’t bounce my flash directly backwards over my head. Hence, directional bounce flash like I nearly always do when I bounce flash.

Reply

16 Russ February 2, 2012 at 4:13 pm

When you say the flash was pointing over your shoulder, relative to the camera that’s about 45° to the right and 45° up?

Reply

17 Neil vN February 2, 2012 at 4:15 pm

Russ, somewhere around there, yes.

Reply

18 Blake February 2, 2012 at 4:30 pm

Neil, thanks for the great explanation (again!). Did you do any noise reduction to the image? ISO 3200 is pretty high… and that image is noise-free at least at web resolution. Thanks.

Reply

19 Neil vN February 2, 2012 at 4:40 pm

Blake .. just the default noise reduction in Bridge / ACR. Not extreme at all.

The D3s is pretty remarkable for the high-ISO noise, and I think *this* is the reason the next generation professional Canon DSLR, the Canon EOS 1-Dx sports less megapixels (18 Mp) compared the the 24 megapixels of the 1Ds mk3.

This previous article shows a 100% crop at 2000 ISO where NO noise reduction was used.

Reply

20 Ty Freeman February 2, 2012 at 5:25 pm

Neil,

Can you explain the effect of YOUR relative position to the wall/ceiling has on the bounce flash output produced? Is it better to be as close as possible to the surface you plan to use for bounce to maximize the “reach” of the flash?

Reply

21 Neil vN February 2, 2012 at 5:30 pm

That’s not really the thought-process. I just want the photographs.

I rely on TTL flash to give me correct exposure (or close enough), and I need to make sure thatI work with an aperture / ISO combination where the TTL flash can give me correct exposure.

Then I just have to take care that my direction of light is good, and that my framing and timing of the photographs are good.

Reply

22 MvH February 2, 2012 at 5:31 pm

The crux mentioned above directly on youtube for easier viewing:

Digital Photography One on One: Episode 59: Inverse Square Law

Reply

23 Winston Mattis February 2, 2012 at 6:26 pm

You said you flag the flash, was the black foamie thing under or on top of the flash head?

Reply

24 Neil vN February 2, 2012 at 6:33 pm

There have been two questions so far about whether the flag goes “under” or “above” the flash head.

The answer is in the video clip on the black foamie thing, as well as the related links.
There’s a certain thought-process in using this. And it’s all there.

Reply

25 Trina Cheney February 3, 2012 at 2:59 am

Hi Neil,

Thanks for the insight and haha moment. But,I have noticed that when my 24-70mm 2.8 lens is racked out to 70mm, the lighting on my subject(S) is nice, even, and directional. At 60mm or less, I have to ride the FEC and the lighting is sometimes uneven. I think it’s the inverse square law at work…

Trina

Reply

26 Neil vN February 3, 2012 at 3:27 am

Trina .. this won’t be an effect of the Inverse Square Law.
I think what you’re observing there, is simply when the lens is zoomed wider, you’re seeing the deeper darker corners of the venue / room too.

Reply

27 Jeff February 3, 2012 at 12:04 pm

Neil, great article (even if it’s technical enough that I had to read it a few times!). But I noticed on your equipment list, you were using the SB-910. Whats your impressions of the new flash so far? Thanks Neil!

Reply

28 Ted Felsberg February 3, 2012 at 4:10 pm

Oh my goodness. Thank you so much Neil. This was yet another aha moment for me. This makes perfect sense after you explained it. It’s also one more reason to not use direct flash with modifiers in dark venues. because the background will always be more dark than the bounce flash. I love it!! I want to go practice now. One question though. Do you ever find that your speedlight just isnt powerfull enough to hit that back wall or a tall ceiling and make it back to light subject enough?? And what do you do if the walls and ceiling aren’t white?

Reply

29 Gene February 3, 2012 at 5:07 pm

From the colour of the background, I assume you either shot this in auto white balance and used the 1/2 CTS gel to add warmth to the flash. Would this be correct?

The light fall off or lack of it is something that I have experienced but never gave it consideration, But aha, of course it is so simple. its the ratio of the background source to the subject, and the subject to the subject background.

Reply

30 Trev February 3, 2012 at 8:23 pm

Gene,

Auto WB would not have been used: a] Neil never uses it; b] Too unpredictable since gelling then setting to auto, the camera will merely still try to ‘correct’ overall WB.

The flash once gelled would then mean the camera’s WB being set to around the 3700K mark, manually, or if full gel would be around 2800K, and since Neil stated, and uses 95% of the time; that he used a 1/2 CTS when gelling, it would have been around 3700K tweaked in RAW to taste.

If you just leave Auto WB to do it’s own thing, the color temp can fluctuate even just on zoom, as the camera would be trying to optimise an even temp for entire scene, something you do not want, since if the exact same color temp has been set in camera, you adjust 1 image in the set to taste and very quickly in around 2-3 clicks of mouse can adjust dozens of images instantly with the same settings.

If Auto WB, you would be adjusting each and every one of them almost.

Trev.

Reply

31 Neil vN February 3, 2012 at 10:43 pm

I didn’t use the gel to add warmth, but rather to help the flash blend with the incandescent light.

related links:
gelling your flash for incandescent light
flash in low incandescent light
dealing with the videographer’s light

Also, I don’t use Auto WB, since the camera would have no idea I have a piece of gel taped to the top of the flash.

(If you use the filter kit that comes with the SB-900 / SB-910, it does affect AWB.)

Reply

32 Andrew February 4, 2012 at 1:24 am

Very helpful. I wish to see if I have grasped your concept. The nugget here is that since you lose 2 stops each time you double the distance from the light source, go ahead a put the source far enough away to take advantage. As you say, let’s call the 2-stop distances 3, 6, 12, 24, and 48 feet. With direct flash a subject 3 feet from the source with a background 12 feet from the source (9 feet further) will yield a 4 stop difference. But move the source far back by bouncing, so the subject is 24 feet from it and the background is 33 feet (still 9 feet further) and now the drop is only a fraction of one stop. Wow! Ahaaa! This is great. BTW I love your first book, have been using since it came out. Many thanks for this post and the blog.

Reply

33 Neil vN February 4, 2012 at 1:30 am

Andrew .. that’s exactly it. Beautiful, isn’t it?

Reply

34 Gene February 4, 2012 at 12:52 pm

Thanks very much for the help Neil. I am a very big fan of your work and this blog. Have learned a lot from it and your books.

Reply

35 Glenn K. February 5, 2012 at 2:02 pm

Neil — Don’t laugh, but I kind of feel like a junior high science student who read an exam question, answered it based on intuition, and realized the teacher had tricked the class with a genuine brain-teaser.

When I read the title of the article, I thought the article would just remind us flash-bouncers not to expect the light to reach all parts of a room, because of the light-eating demon known as the inverse square law. What did our teacher do? He moved the effective light source on us, which changed the relation between the subject/surrounding people/background to the light source, as Andrew clearly stated. I had to read and re-read the article several times to “get it”. Instead of a demon, the inverse square law became a friend.

What a wonderful article and teaching technique! Many thanks again, Neil.

Glenn

Reply

36 rené February 5, 2012 at 2:13 pm

i noticed you have been using sporadicaly the nikon 24-120 f4 lens in previuos tangents
what is your experience with this lens .

René

Reply

37 Neil vN February 6, 2012 at 10:09 am

Gene .. I like the Nikon 24-120mm f4 for general use, but for the critical stuff like weddings in low light, the f2.8 optics are better … and slightly sharper.

Reply

38 Paul February 6, 2012 at 12:46 pm

Neil,

Good article. My question is: how did you determine using 1/2 CTS gel? Did you test at the venue beforehand? Or is it a David Hobby thing where he just likes the look of a 1/4 CTO (IIRC) for people?

Thanks,
Paul

Reply

39 Neil vN February 7, 2012 at 12:11 pm

I use either a full CTS or 1/2 CTS gel, depending on whether I need to completely correct my flash for Incandescent … or just bring my flash closer to the Incandescent spectrum. But the 1/2 CTS gel is my default usually.

Reply

40 Sri February 7, 2012 at 5:35 pm

Hi Neil, I was referred to your site by someone in DPS couple of days back. Since then I am glued, you have so much information to share and you did it gracefully. I have my almost first event to shoot this weekend – it’s a small Valentine’s party shoot though but I want to impress those who will buy pictures from me. It will be a low light event, I have Sigma 530 DG Super TTL flash (off camera), Nikon D80, 18-135mm lens. Also, bought umbrella to hold my off camera flash. I would be clicking some family portraits with white backdrop and some random party pictures. I am sure your wonderful suggestions in the site will help me with decent pictures. I have couple of questions:
1. When you mentioned “bounce flash behind us”, what does it mean? Is it, turn or point the flash head backward with no angle?
2. …and “bounced it over my shoulder”, what does that mean?
(sorry for asking so stupid questions)
3. Given my setup, white backdrop, group of people, one umbrella/off-camera flash that triggers from my popup flash (on camera). What suggestion do you give for the light setup?
I would appreciate if you can answer this before the weekend. And BIG thanks for sharing your vast knowledge with us. God bless you.

Reply

41 Sarah February 9, 2012 at 1:34 pm

To Sri:

In case Neil doesn’t get back here in time to answer your questions for you I thought I’d help. When he says bounce flash behind us he means the flash is on the camera pointed backwards to bounce off of the wall. I believe he rarely bounces it directly behind him as that would give more of a flat light, but generally “bounces it over the shoulder” which wouuld mean the flash head was turned backward but with a slight turn to go over the right shoulder and a little up so it will bounce off of the wall and celing some giving more direction to your light and therefore a nicer photo. If you’re shooting the event, I’d suggest using that method. If you are then going to set up for some portraits later, then you may want to use the light and umbrella off camera as you mentioned. Good luck to you!

To Neil:
Thanks for that info, it made me think of my wedding formals also as at times when I get the whole family up there sometimes the people in the front are much brighter than the ones in the back and I hadn’t thought about just moving my light back! Thanks again!

Sarah

Reply

42 Gwen February 13, 2012 at 3:36 am

Hey Neil!

As someone who is just starting out in the journey of wedding and portrait photograpy – thank you, thank you, thank you! I’m hooked on your site!

Reply

43 Stephen February 14, 2012 at 1:21 pm

When you bounce flash off a surface, that surface becomes the “softbox” that shines on your subject. Therefore, how you swivel your flash head is determined by where you want your virtual softbox to be located at.

See: http://neilvn.com/tangents/directional-bounce-flash/

There is not a single formula that works in all cases. Sometimes, you might want to point the flash head over your shoulder and directly behind you. Other times, you want to bounce the flash against a wall that is behind you but off to the side at a 45-degree angle. Other times, you want to bounce off the ceiling behind you at a 45-degree angle. Just imagine where you want that virtual softbox is located and that is where you would point your flash head.

Reply

44 Todd July 13, 2012 at 5:20 am

Hi Neil,
great stuff as always, thanks a lot!!

A perhaps silly question:
If we calculate with bounce flash and inverse square law — do we have to *double* the distances (or divide it by half)? — I am totally confused :-)))

Example: I shoot a model standing in front of me, distance to the bounce wall: 2 meters, on-cam flash on full power, directed to the wall.

Now I double the wall-to-subject distance (and in parallel also the flash-to-wall distance).

Do I have to multiply my ISO settings times 4? or more (given the fact, that the light travel distance is now 8 meters)?

If I double it again:
light travel is now 16 meters …a.s.o.

does this “doubling” through bouncing has any influence at all?
any help welcome,
blue skies — Todd

Reply

45 Neil vN July 13, 2012 at 11:39 am

That makes my brain hurt. Shoot in TTL. Let the camera sort it out for you.

Reply

46 Pete (Barnet, uk) October 1, 2012 at 7:43 pm

Just read the article. I must say it’s a revelation – I got it! It makes so much sense from your article. Thanks Neil

Reply

Leave a Comment

Previous post:

Next post: