D5200 liveview vs. regular mode - why do shots turn out differently?

[PaulWog]

I can be shooting in manual mode, aperture priority mode, auto mode, etc, and shots generally come out differently. In liveview, I appear to get a better exposure, and possibly color/white balancing (though it may just be better exposure). I'm not sure if the raw files are pre-edited when liveview is used, or what the case is, but I've tried setting my camera up on a tripod, and I've taken identical shots between liveview and regular mode... and the shots come out different.

Is this common? I'm curious what the reason might be. I know there's a difference in the focusing system when I switch to liveview. In fact, liveview seems superior in most ways, minus the fact that it takes forever between shots, focuses slower (though more accurately on average), and doesn't allow simultaneous viewfinder usage.

I just did a quick test (I'll post photos if necessary): I did some self face shots. The shots were done with the same settings in manual mode. White balance was set to automatic, metering was set to matrix. My face came out slightly more exposed (just a tiny tiny little bit, maybe a third stop more exposed) on the shot through the viewfinder. On the liveview shot of my face, the contrast and colors are much more vibrant and nice with more detail. This is all in the raw file (no jpegs at all). I have a feeling it has something to do with the white balance settings, and maybe more processing and better decision-making on the automatic settings (white balance, maybe the matrix metering as well) go into the liveview mode.

side-note: I'm really disliking that liveview gets nearly 100% accuracy with my 35mm 1.8G on focus, however when using the viewfinder my accuracy rate is severely reduced to the point at which it's not much of a walk-around lens at the moment. I have yet to fully test the lens to see if there's focusing issues, but it seems to be fine when I run test shots in static situations (note that I never blame out-of-focus shots on shutters below 1/200 of a second... I just came home with 1/1000 of a second shots, a series of them, out of focus... so I've got to test the lens more... but the liveview shots turned out well, not that I did many). My 16-85 is generally quite accurate though (quite reliable). /end-of side-note rant

 

[weepete]

Sounds to me like you are missing the focal point you want when using the viewfinder. I Don't mean to be cheeky but maybe you need glasses? I'm finding harder to hit those focal points through the viewfinder as my vision is a wee bit worse now than it was when I was in my 20's

 

[PaulWog]

Sounds to me like you are missing the focal point you want when using the viewfinder. I Don't mean to be cheeky but maybe you need glasses? I'm finding harder to hit those focal points through the viewfinder as my vision is a wee bit worse now than it was when I was in my 20's

That isn't the issue. I have 20/20 vision. I can see very clearly. I haven't ruled out the possibility of user error (since when I do notice the focusing problems I didn't keep track of exactly what I was doing), but it has gotten to the point where when I go out I do feel nervous about using my 35mm 1.8G and not about my 16-85mm lens. When I review the images in ViewNX 2, the focus points are dead on the correct target, but the focus point is not in focus. This isn't always the case, but today it was. Lighting was adequate (1/1000 of a second at EV -1.0 and ISO 250, f2.2, ended up with proper exposure in aperture priority).

 

[weepete]

Hmm, weird one then. The only time I've experienced a similar thing was when I couldn't get some sharp photos with my 1.8 I misjudged the minimal focus distance and was too close to my subject. I notice that your dof is quite shallow, what's the distance to your subject like and how is it at smaller apetures?

If you can get sharp shots sometimes it kinda suggests to user error rather than a lens issue. Its weird also that you are getting different exposures in different modes, given that all other things are equal the same settings in manual should prodce the same results as the same settings in auto so it sounds like something in your camera is changing.

Can you post some examples?

 

[SCraig]

Your camera behaves completely different between "Normal" mode and "Live View" mode. When in Live View the mirror is up constantly so the camera is more like a mirrorless in that respect. When the mirror is up more light reaches the autofocus module since it doesn't have to pass through the semi-opaque mirror. While this shouldn't have been a factor in the last group of shots when you said you had plenty of light, it still could well be for whatever reason.

Also, when going between normal and live view modes your camera may be switching between phase detection autofocus and contrast detection autofocus. I'm not very familiar with the D5200 so I'm not real sure about that though.

The other thing is that at f/2.2 you aren't going to have a lot of depth of field. If you miss focus at all, or look at other portions of the image other than the point of focus, it is entirely possible that they are blurred due to lack of DOF.

 

[PaulWog]

Your camera behaves completely different between "Normal" mode and "Live View" mode. When in Live View the mirror is up constantly so the camera is more like a mirrorless in that respect. When the mirror is up more light reaches the autofocus module since it doesn't have to pass through the semi-opaque mirror. While this shouldn't have been a factor in the last group of shots when you said you had plenty of light, it still could well be for whatever reason.

Also, when going between normal and live view modes your camera may be switching between phase detection autofocus and contrast detection autofocus. I'm not very familiar with the D5200 so I'm not real sure about that though.

The other thing is that at f/2.2 you aren't going to have a lot of depth of field. If you miss focus at all, or look at other portions of the image other than the point of focus, it is entirely possible that they are blurred due to lack of DOF.

Thanks for the insight into some of this stuff. It's odd that I get a better matrix balance and white balance (and just better colors in the end and a better picture) when using the live-view.

As for the thin DoF: It's definitely not that. I often get slight back-focus or front-focus, and I'll have a very specific focus point (usually the eye on a head, sometimes the nose). In this case, the entire body was out of focus by a long-shot, to the point at which it was a ridiculous miss-focus (and I at least know my lens wasn't hunting for focus today).

I think I'll go do some more test shots and come back with some raw files/etc rather than just post descriptions.

 

[KmH]

In live view mode, neither the AF module in the bottom of the camera, nor the regular light metering sensor can be used.

The secondary mirror that is behind the main mirror that reflects light down to the AF module is attached to the main mirror, so it to moves out of the light path when live voew is used.
The light metering sensor also uses light reflected from the main mirror to function.

TTL exposure metering uses the image sensor, and Matrix metering is the method used (no options).

From page 99 of the D5200 users's manual:

Shooting in Live View Mode
To prevent light entering via the viewfinder from interfering with exposure, remove the rubber eyecup and cover the viewfinder with the supplied DK-5 eyepiece cap before shooting (page 32).

Although they will not appear in the final picture, distortion may be visible in the monitor if the camera is panned horizontally or an object moves at high speed through frame. Bright light sources may leave after-images in the monitor when the camera is panned. Bright spots may also appear. Flicker and banding visible in the monitor under fluorescent, mercury vapor, or sodium lamps can be reduced using Flicker reduction (page 172), although they may still be visible in the final photograph at some shutter speeds. When shooting in live view mode, avoid pointing the camera at the sun or other strong light sources. Failure to observe this precaution could result in damage to the camera’s internal circuitry.

Live view ends automatically if the lens is removed or when the monitor is closed (closing the monitor does not end live view on televisions or other external displays).

Live view may end automatically to prevent damage to the camera’s internal circuits; exit live view when the camera is not in use. Note that the temperature of the camera’s internal circuits may rise and noise (bright spots, randomly-spaced bright pixels, or fog) may be displayed in the following instances (the camera may also become noticeably warm, but this does not indicate a malfunction):
• The ambient temperature is high
• The camera has been used for extended periods in live view or to record movies
• The camera has been used in continuous release mode for extended periods

If live view does not start when you attempt to start live view, wait for the internal circuits to
cool and then try again.

 

[Derrel]

What about differences in light entering the viewfinder eyepiece with the eye in position behind the eyepiece and with no eye behind the eyepiece when in Live View mode? In some situations, where there's a lot of light around, or sunlight is actually hitting the eyepiece area, meter readings in tripod-based shots can be affected by light that enters the prism, and makes its way down to the viewfinder screen. The light metering in most camera these days is done by reading the viewfinder screen's image.

On higher-end Nikon cameras there has long been an "eyepiece shutter" or "eyepiece blind" built into the prism, which allows for more accurate metering when the camera is used with the photographer NOT close to the camera and shielding the eyepiece from stray light. On lower-end Nikons and most Canon cameras, a simple plastic eyepiece "cap" has long been provided. Hence Nikon's warning , "Shooting in Live View Mode
To prevent light entering via the viewfinder from interfering with exposure, remove the rubber eyecup and cover the viewfinder with the supplied DK-5 eyepiece cap before shooting (page 32)."