Affecting flash power with AF points?

Russ, unless I overlooked it, you failed to mention the camera's exposure setting (i.e. Spot, Center Weighted, or balanced). You also didn't mention which lens you were using (i.e. AIS, AFD, AFS); depending on the lens, it may or may not provide the camera distance information to be used in the exposure calculations. If metering is set to Spot, "Regardless of which focusing area is chosen the highlighted focus area becomes the active spot for calculating exposure." This clearly suggests the focus points on modern DSLRs can in fact play a direct role in exposure determination. There are also many other things which could have contributed to the exposure calculations. Without fully understanding how your tests were conducted, I will take your results with a grain of salt.

In image 7, the focus was turned to manual (as JRP says) and the focus point was changed. However, It is not a surprise to me that the statue is overexposed; as Alan said, the mistake is that Russ failed to re-focus along with the change in focus point (the focus is still on the hats). And as Hayo said, correct exposure is down to focus DISTANCE not focus point. Without the focus distance information provided by AF, the camera will have assumed that the subject is still at the (previously focussed) distance i.e. the hats! Correct exposure in TTL or even TTL-BL flash modes needs distance information from focus performed by AF - it does not work when in manual focus.

Learned again. Thank you this article, Russ.

J. Ramon Palacios (jrp) on September 20, 2013

Alan, The AF was turned off. Only the focus point was moved.

I think that some of the confusion may be due to a differences between camera models. I tried a similar experiment with black, gray and white pieces of paper taped the wall, so there was no difference in distance. I compared a D80 to a D7000 and found that the exposure of the D80 was more sensitive to whether the focus point was in the black, white, or gray area. The difference between the two cameras was a 1/2 stop or less, but it was consistent.

I dont want to sound picky, but the picture no 4 seems sharp at the bookcase, not hats with rope... It could be tricky, because the photos here are very tiny. Still, I did believe that TTL was partly calculating the flash power from the distance which should be pin sharp, not point of focus you choose......

Dear Russ, your images only show that (focus) DISTANCE is what determines flash output. This is something we already knew to be the case (and explains why the statue in some of your examples is over exposed). Given this fact, the focus point DOES actually matter.

Now you've given me reason to try TTL-BL. Thanks, good article.

Very surprised with your Back Lit results Ross. I had assumed that the metering system would look at what light was coming in from behind the subject, and adjust the frontal exposure to compensate. But there is nothing back light, so the meter just exposed as if there was not there. Interesting results! Many thanks for sharing your knowledge an test experience Russ. Cheers, Frederic in Montréal

I have a couple of points with regard to your interesting study. First, in your opening paragraph, you say ”different focus areas ....will change the place...that the FLASH meters from”. When a Nikon Speedlight is attached to a DSLR in anything other than “M” (Manual) mode, it is not the flash but the camera that does the metering. Secondly, you missed a couple of vital points in your conclusions about exposure relating to flash. The Metering Mode choice and what the camera itself does to "compute" correct exposure both pre and post-capture (based on the image brightness range) are crutial factors to take into consideration. Throw into the mix the fact that your D3 will use it’s “Scene Recognition System” to interpret your subject, and if you have it engaged Auto ISO will kick-in, and it is all a recipe for double nay even triple guessing as to what kind of exposure will come out of the shot in any auto flash mode! I think it is important to state however, that no matter what inconsistencies may arise with Speedlight use, it is still best practice to select a focus point that corresponds to the part of the subject that is most important - wherever in the frame that might be. I would recommend experimenting with FV (Flash Value) lock for optimum results.

Thank you Russ, apart from being very interesting, this helped me understand (finally) the difference between TTL and TTL-BL modes! Best regards from Bern, Switzerland

Thanks again for an excellent article. I knew that Nikon's meter was not set for 18% grey cards, but never knew how much. Now I know 1/2 stop difference. Thanks for the info. Does anyone make 12% grey cards?

An obvious slip on image 7: the statue is not in focus.

Please elaborate on "meter 12% gray reflectance". Thank you.

This has not been my experience- I have had the opposite results- When I have subjects in varying distances and select the closer subject in the frame I get an accurate exposure on the focal point selected subject but when switching focal point to the farthest subject, the closest subject is over exposed. I will do some testing on my own to verify my findings, and submit what I find. Thanks for the post Russ- Doug

Excellent tutorial, the information is logical and well presented. Thank you for taking the time to explain. (I also never knew about the 12% gray calibration.)

Thanks. very helpful.

Richard Luse (DaddySS) on September 17, 2013

Thanks Russ, very informative once again and very thorough. Much appreciated.