Why such different quality in these 2 shots?

[abitfishy]

Hi all,

I'm finding now and again that 2 similar shots with different settings can be miles apart in terms of 'quality' and I have no idea why. I'm hoping I can post a few shots here and get opinions. All EXIF data should be intact.

I just happened to take these 2 quick photos of my dog (yes, I know she's the cutest little button!), and was amazed at how different they are. Only differences that I can see from exif are 1/80 & 2.8 vs 1/60 and 5.6. TTL flash (SB600) was used on both situations. I'm assuming based on the settings that it would have used more flash in the second shot, and I know TTL isn't always perfect, but I would have expected it a bit better than it turned out.

Am I missing something?

Expert opinion (and any not so expert!) would be appreciated.

[abitfishy]

OK. I thought I worked out what it was. I had BL turned off and thought it might have simply been giving too much light to the whole scene because of it. So I redid the photos the same but some with BL (backlight, balanced fill, whatever you want to call it) turned off, some with BL turned on. Same results with the second pics with each setting, so it was only the f5.6 photos. Both f2.8 photos were ok.

The flash did fire with both settings.

[ricktas]

If you are not shooting in manual mode, the difference can be caused by the metering. If you have metering set to spot or partial, then the camera determines the exposure based on what meter readings it gets. Now if you meter off a darker area of the scene, the camera will try and lighten that, thus lightening everything. The meter system works by trying to evaluate the scene based on what information it gets, then it tries to get a 'correct' exposure. Conversely, if you meter off a bright part of the scene, the metering will adjust settings to darken the result.

I am guessing here, but I reckon, you have metered off two different areas, one in each photo, and the camera has thus made the adjustments it thinks are needed to get a correct exposure, and in this instance it has done just that, resulting in the varying results

[abitfishy]

Thanks for that Rick. I understand exactly where you are coming from. Is that still likely to be the case since it is set to Matrix metering? (Or whatever Nikon currently calls it).

[ricktas]

Thanks for that Rick. I understand exactly where you are coming from. Is that still likely to be the case since it is set to Matrix metering? (Or whatever Nikon currently calls it).

Yeah. It Matrix metering selects various parts of the frame to grab a reading from, if those all happen to be a fraction dark, the result will be that the camera adjusts settings to get a 'correct' exposure. If control over the result is of ultimate importance, then using full manual is the way to go, cause you as the photographer, then control the exposure and result, rather than relying on a micro computer to work out what it thinks you want.

[arthurking83]

The way I see this comparison is:

in the first shot @ 1/80 and f/2.8, the ambient light is probably the main lighting source, and the flash is simply helping get an exposure as set by either the operator or the camera.. just trying to get a good exposure.. the light from the flash is less important.
in the second shot, to produce the same level of ambient light, the shutter should have been more like 1/20, to get the same ambient light exposure compared to the first shot(alternatively the shutter speed could have been more like 1/250s in the first shot).
So, because the ambient light is providing less of the overall lighting compared to the first shot, the flashes output becomes more important.
Flash light output is not as 'nice' as ambient light generally is.

looks to me as though there was a window behind the couch providing some outdoor ambient light, which is warmer that the harsh lighting provided by flash output.

I'm far from expert when it comes to flash output and how to correctly control it, and the Nikon CLS system generally works well for me tho.. so I'm happy to maximise it's abilities! :D

[arthurking83]

Also! I noticed in the exif data that you had ADL set to Auto.

ADL is a strange feature, that is not 100% reliable in producing easily reproducible results using the same settings.
From what I've seen of it, it can sometimes help, but it can also not help as well.
I used the term not help as opposed to hurt or hinder the process of capturing a good exposure, because I've never seen it do any damage.. it's just that some images don't come out exactly as I'd have expected them too with ADL set to on in camera. Luckily if you shoot in NEF format, you can turn that setting off using Nikon software.

D90 may work better in ADL mode(compared to the D300 that is) if Nikon have tweaked it to perform more consistently.. on the D300, I found that it was consistently inconsistent :D

[abitfishy]

Thanks all.

I too am not sure about whether or not to use ADL, or the settings that alter contrast/brightness etc within the camera (can't for the life of me think of what they are called at this moment). I also just came across a post somewhere on a web site that mentioned someone finding inconsistent results with matrix metering, ADL and the SB600. I might go for centre weighted, turn off ADL and see how I go. Would this be appropriate for outdoor portraits and a few shots in a bikini? (her, not me!). I'm doing some later in the week with a HonL traveller8 diffuser most likely with the SB600 off camera too. Light should be good so will only be using for fill.

[arthurking83]

Picture Controls, are a strange bunch of settings to work with.

Most programs do not recognise them in the resultant NEF image. Nikon's software does!
So if you use ViewNX to initially view your images, and those images are NEFs ViewNX allows you to swap between different settings, as you would in the camera, and this can help you to achieve a consistent look to your images, in the initial capture.
Other programs can only emulate these Nikon camera settings.
If you shoot in jpg mode the Picture Control settings are applied to the jpgs, and can't be undone. Shoot raw(NEF) and change your mind later, where changing your mind = apply different Picture Control looks to your images.

ADL is a bit strange in that, some progrmas do not recognise the boost in shadow details it gives an NEF image, but the camera supposedly adjusts the camera's exposure to try to maintain a well balanced exposure in the final image. This is where it falls down for me(in not being consistent). In a group of three continuous shots, it may try to keep a check in highlight details(not blow them) then it may not, then it may again.. or not.. or do... etc? There's no method to the madness! BUT, it does bring back shadow detail pretty well.. where you can then adjust accordingly in PP.

[MarkChap]

My non expert opinion on this

Would seem that it has a lot to do with the framing of the shot as well, in the first you are closer than in the second thus eliminating a lot of the darker coloured lounge.
Combined with 2 stops less aperture, and remembering that flash exposure is controlled by aperture, not shutter speed, has caused the camera in the second to "see" a lot more of the darker background and thus it has said - "aha, lots of dark in this one, I need more oompf from the flash"

Just my opinion, for what it is worth