Flash Value lock

cab.in.boston

I'll have to remember to process the two photos and post them, but here's the story. I was shooting my newborn son using my D300 with SB-600 mounted on-camera, bounced off the ceiling. Using TTL BL mode (IIRC), and the exposure settings were the same in both shots - probably something like 1/60 @ f/4 or so, but I don't remember off the top of my head. In the first, I simply focused and shot, flash fired, but the photo was fairly dark. In the second, I used FV lock and re-shot, and the photo was much brighter. Composition was the same, with my son in the center laying on Mom.

So my question:

I don't understand why they were different. My understanding of FV lock is that you use it when you want to get the flash exposure settings with subject centered, lock them in, then recompose and move the subject out of the center, since the light meter is in the center of the frame and would be too dark if the subject isn't centered. But if the subject is in the middle of the frame, shouldn't the result be very close to the same using FV lock or normal TTL flash?

cab.in.boston

Here's the illustration of my question, if anyone has any insight.

Both photos at ISO200, f/5, 1/60s, using 35mm f/1.8 lens on D300 with SB-600 aboard, pointed straight up at the ceiling, in TTL mode. I used single point focus in the middle for both. One taken "normally," the other using FV lock. I don't understand why the light is so different.

Normal, no FV lock:


FV lock:

time2smile

My understanding of FV is its to meter multiple remote flashes, along with your composition feature...and no preflash during the actual photo...
Neither case my photos always seen to look better when I us it also.

Qarik

what exposure mode were you using? if you were using matrix metering then I agree the TTL flash should have been roughly the same output. I you had been using spot metering then things could change dramatically depending on what the spot was on.

cab.in.boston

Qarik wrote: »

what exposure mode were you using? if you were using matrix metering then I agree the TTL flash should have been roughly the same output. I you had been using spot metering then things could change dramatically depending on what the spot was on.

I was using matrix metering, and was in manual, IIRC. I tried this with my D90 and the two photos look much more similar than these two did with the D300. I'm wondering if something is awry with my D300's flash metering...

ETA: However, I thought that the metering mode didn't particularly matter for the flash output level. It was my understanding that the flash metering used only the center of the frame, no matter what metering mode the camera is in for the exposure. That's why there is an FV lock in the first place, so you can get the flash value by centering your subject, lock it in, then recompose and the exposure will use the preset flash value rather than re-metering the (now empty) center of the frame and firing too much flash.

Dan7312

Maybe try a more controled experiment, in the pictures your subject appears to be leaning slightly differently and is in the center of the image so that could effect the flash eval.

Try the comparison with a static scene with the camera on a tripod, just to eliminate the variable a moving subject adds.

cab.in.boston wrote: »

I was using matrix metering, and was in manual, IIRC. I tried this with my D90 and the two photos look much more similar than these two did with the D300. I'm wondering if something is awry with my D300's flash metering...

ETA: However, I thought that the metering mode didn't particularly matter for the flash output level. It was my understanding that the flash metering used only the center of the frame, no matter what metering mode the camera is in for the exposure. That's why there is an FV lock in the first place, so you can get the flash value by centering your subject, lock it in, then recompose and the exposure will use the preset flash value rather than re-metering the (now empty) center of the frame and firing too much flash.

adbsgicom

Based on my understanding of Canon, the TTL in its normal mode has a secret algorithm that makes its calculations on that which the preflash affects. So, that is not necessarily just the center, regardless of the focusing. Perhaps the FE-Lock changes that and locks based on the focus point? I've noticed some similar shifts in the Canon system from time to time, but I haven't done a rigorous experiment to get a bead on what is really going on.

Dan7312

I did a test with my 7D and a gray card and could not see or measure any difference between using FEL and not using FEL. The flash was set to use ettl and the M on the camera with f7.1 1/60 sec. All the exposure was due to the flash, not ambient.

The photos are here if you want to look at the Exif

http://www.danalphotos.com/Other/temp/9344418_jYfFd#908962533_FBazz



Here is the FEL based exposure




and here it is again with no FEL

cab.in.boston

This was on a tripod, ISO200, f/4, 1/60, manual mode, matrix metering, TTL flash, single point focus in the center, right on the doll's eyes.

Normal flash:


Using FV lock:


Not a drastic difference, but the FV lock photo is brighter. The histogram basically just shifts horizontally a bit, but is otherwise the same shot to shot.

Normal:


Using FV lock:


I could try it stopped down more and/or faster shutter so that ambient doesn't come into play. Regarding the earlier comment, I have found that the difference I've noticed is more pronounced when shooting live subjects than static scenes.