3 Way Holiday Camera Shootout Part 5, Picture Quality

G6 with 14-140mm lens

D5200 with 18-200mm lens

 

Technical Image Quality vs Picture Quality   There are several internet sites which publish data about camera image quality using technical terminology. A well known one is DXO Mark (dxomark.com). DXO reports overall image quality, Color Depth, Dynamic Range and Low Light ISO. DXO also reports a range of other technical measurements such as signal to noise ratios at various ISO levels. These measurements are made in highly controlled conditions in a laboratory style environment using test targets designed to reveal specific imaging characteristics.  I have no doubt these results form a useful part of an overall evaluation of the capability of a camera.  For want of a better term, I refer to the figures produced by this type of testing as "Technical Image Quality".

But when we use a camera it is never in a highly controlled environment. Real world use introduces many variables which intervene between the technical capability of a sensor and the final picture. I call this final result "Picture Quality".

Factors Influencing Picture Quality   Some of the factors which might and in fact often do influence actual output picture quality include:

* Camera Movement. Some users can hold a camera steady, others don't do so well. Some camera/lens kits are easier to hold steady than others.

Some cameras have Auto ISO algorithms which keep shutter speeds in the acceptable range for hand held use, others allow shutter speed to fall to levels which can cause unsharpness.

Some users watch their shutter speed readout and increase ISO setting if the shutter speed falls too low.

Some cameras/lenses have some kind of image stabiliser which can help steady the camera/lens unit.

* Lens quality.

* Exposure. Some cameras have exposure metering systems which allow highlights to become overexposed (blown out), others protect highlights assiduously.

* Color rendition. Cameras differ in the way they render colors.

* Focus accuracy.

* Aperture and depth of field. A camera with a large sensor will have to use a smaller lens aperture to achieve the same depth of field as a camera with a smaller sensor. This in turn requires a higher ISO setting and/or slower shutter speed both of which could have a negative impact on picture quality.

Real World test procedure  I made hundreds of photos of a wide range of subjects, shooting the same subject in the same light at the same time with the three kits. I used both the Auto setting and one of the P,A,S,M settings, usually Aperture priority. I used Auto ISO most of the time as that is what I and I suspect most users do. I noticed that each camera typically used a different Auto ISO setting and therefore  a different shutter speed.

Subjects were in bright sun, front lit and back lit, shady bright light, outdoors, indoors, in medium and low level lighting, natural and artificial. I shot people, single and groups,  landscapes and  buildings inside and out. I shot JPG and RAW files. I uploaded selected shots to the computer and compared images side by side on screen. I looked at JPG's, unmodified RAW's and RAW's processed to "best result" in Adobe Camera RAW 8.1.

G6 with 14-140mm lens

D5200 with 18-200mm lens

 

Defining Picture Quality 

* Color rendition    Some people like saturated colors, others prefer a more natural rendition which may be less saturated.  Cameras often differ in the way they place output colors in the spectrum. For instance one camera might render blues slightly cyan, another might render blues slightly violet. The accuracy of each can be measured with suitable test materials but the individual photographer might not be overly concerned with accuracy. The cameras in this test do render colors differently but I had a hard time deciding if any one of those representations was better than another.

* Highlight and shadow detail  It is easy enough to evaluate the amount of highlight and shadow detail in a JPG or RAW file on screen or in print. This is not the same as technical Dynamic Range which is measured in a specific way, generally by photographing a transilluminated step wedge.  I found that although the D5200 has a higher technical Dynamic Range, this was not reflected in actual pictures.

* Resolution/sharpness  I looked at resolution and sharpness in the center, at the edges and corners of test shots. I found that lens quality, focus accuracy and auto selected shutter speed had a big influence on sharpness.

* Noise  I looked at noise levels in each of the comparison photos.

The cameras on test, real world results  If you review the technical image quality data in Part 2 of this comparison series you will see that the D5200 scores best, followed by the G6 with the V2 scoring lowest on DXO Mark ratings.

You might expect the D5200 would easily beat the G6 for picture quality but that is not the case in the many hundreds of photos which I made for this comparison test.  However both the G6 and D5200 made better pictures than the V2 in almost all comparisons.

* Color:  There are differences in color rendition between the three cameras but I could not say one is better than the other. Another observer might have a preference for one or the other. For instance the D5200 typically delivers more saturated colors than the other two.

* Highlight and shadow detail: the D5200 and G6 are both very good and very similar. Even in subjects with a high brightness range I could not proclaim one to be better than the other.  By comparison the V2 suffers from less highlight and shadow detail than the other two when subject brightness range is high.

* Resolution/sharpness:  In the  frame center there was not much difference between the three kits. One might have expected the D5200 with a lot more pixels to excel here but that was not often  the case when actual photos were examined.  At the edges and corners the 18-200mm lens let the D5200 down,  producing consistently softer peripheral resolution than the other two lenses. The V2 in lower light levels often selected a low Auto ISO level presumably to keep noise levels in check. But this also produced a low shutter speed which fairly often led to unsharpness due to camera shake.

 I don't understand why the D5200 did not deliver better sharpness/resolution in the frame center. The 18-200mm lens works well in the center and the camera has more pixels than the others so one might expect it to beat the others easily. I did not have the time, money or motivation to test the D5200 with a better lens.

* Noise: The D5200 produces less noise at high ISO levels than the G6 which in turn is less noisy than the V2, which exhibits noise even at base ISO.  In actual photos the D5200 would typically set the highest Auto ISO level, followed by the G6 then the V2.  As a result the noise superiority of the D5200 was often countered by the higher ISO setting. With the camera on a tripod and with matched ISO settings the D5200 was able to demonstrate it's superiority. V2 pictures were always more noisy than the other two.

Summary My overall ranking for picture quality taking into account all the real world factors is:

1. G6/14-140mm.

2. D5200/18-200mm. With a better lens the D5200 would be the winner.

3. V2/10-100mm

Next: Performance