Why do these two pictures look different? (shot on D300)

[gabriel_afana]

I just got a 17-55mm f/2.8 lens today :-)

<BR><BR>

So I was out playing around shooting pictures at my office and I noticed something that I dont understand. I was

taking several pictures outside on a balcony and they looked ok. Then I closed down the aperture and I noticed the

pictures were getting darker. I was shooting in aperture priority so I would imagine that the exposure of the picture

would look the same regardless of the aperture....no?

<BR><BR>

Well I tried shooting on manual and I discovered that at f/22, the meter was off by 1 stop (under exposed). Is this a

settings thing, a lens thing, or a I-dont-know-what-the-hell-im-doing thing?

<BR><BR>

 

Shutter: 1/2500 Aperture: f/2.8<BR>

<img src="http://web1.plurlife.com/temp/1.JPG">

<BR><BR><BR>

 

 

Shutter: 1/30 Aperture: f/22 <BR>

<img src="http://web1.plurlife.com/temp/2.JPG">

<BR><BR>

 

Ideas?

[sam_ellis]

The top picture has a white car driving by while the bottom one has a dark van. Naturally, this is throwing the camera's meter off. Just kidding of course!

What type of metering were you using? It could be that the camera read the scene differently between the two pictures, maybe adjusting for the sky in one and the pavement in the next. Just my thoughts.

[berg_na]

The first shot was taken in center weighted metering, and the second was shot with matrix metering. You should repeat the experiment using the same metering mode.

[mark_sirota1]

Also, I think you can set a minimum shutter speed for aperture priority, which may have been set to 1/30th.

[mark_sirota1]

Oh, never mind, I must have been thinking of the flash sync speed, which could be adjusted to 1/30th. Assuming you weren't using your on-board flash for this, that's not it.

[johnw63]

Justing figuring in my head ... does f22 at 1/30th = f2.8 at 1/2500 ? I get approx f2.8 at 1/2000 of a second. but f2.8 isn't a full stop down from f4. It just seems to me that it's under exposed because the shutter is too fast at 1/2500 th.

[johnw63]

Ooops. Never mind. The question was why the CAMERA chose the wrong settings. If you DID use center weighted on the top one, and the center was on the blacktop, then it makes sense. Matrix metering takes in the brighter open sky above as well as the parking lot. Center weighted only gets, well, the center.

[gabriel_afana]

Sorry everybody, that was my mistake. While shooting I tried it in all 3 weighing modes and I uploaded the wrong two.

 

I replace both photos above...those are both the exact same metering...exact same everything. You can see that the camera still didn't expose the picture the same. In manual mode, I discovered that at f/22, the camera was about 1 stop under-exposed compared to what it did at f/2.8

[rolubich]

Wide open nominal aperture can be a little different from actual wide open aperture, you should try to redo the same test comparing f/4 and f/22.

[johnw63]

Hmmmm. If it is a brand new lens, and not a cheap one at that, I would expect more consistency. If it was a store, perhaps you could exchange it, or test another example, just to rule out a defective lens. I'll be curious to read what others have to say.

[gabriel_afana]

These photos were taken with a Nikon 17-55mm f/2.8 (taken at 17mm)

[lex_jenkins]

Puzzling. The EXIF data shows those two photos were taken a few seconds apart. Otherwise the data is virtually identical, including metering modes. The steps are appropriate for EV 12 lighting: 1/2000 @ f/2.8, 1/30 @ f/22.

 

Try repeating everything, as consistently as possible. Be sure the scene illumination is absolutely consistent. If possible, meter a simpler target with no moving objects, to minimize the variables. While it's unlikely that a reflection from a moving vehicle would throw off the metering this much (matrix metering, wide angle), it's best to eliminate all the variables.

 

I've encountered one - and only one - Nikkor that showed this kind of variation between shots, a used 20-35/2.8 AF Nikkor that was almost certainly defective. And even it showed only a 1/3-2/3 variation in exposure accuracy between apertures. I spent an hour with that darned lens, making sure I'd eliminated all the possible variables, because I'd never seen that sort of inconsistency before even in my manual AI and AI-S lenses.

[gabriel_afana]

I took about 50 photos like this....testing all different combinations of metering modes and shooting mode (M, A, S...etc). All yielded the exact same result.

 

The lighting was exactly the same. There was no wind to possibly create variations of shadows on the parking lot or anything.

 

Think this could be a lens thing? Its on a Nikkor 17-55mm f/2.8 that I just got (bought it used).

[gabriel_afana]

Hmmm....you think Auto WB could have anything to do with this? I tihnk I might try tomorrow with fixed wb.

[mats nilson photography]

Old lenses can have sticky apertures if they haven't had too much exercise lately. Exept, that would produce the reverse

result. I had an old lens that did that, giving grossly overexposed images when I set a small aperture, since it wouldn't close

all the way. So that shouldn't be it.

 

Nor can it be the WB - both images seem to be the same in that respect.

 

Very puzzling..

[tri-x1]

I had an old 37-70 zoom I was using on a D100 that sometimes made lighter images than other lenses at the same settings. It had a sticky aperture.