Wednesday, December 5, 2012

Shootout - The Sequel

This is a follow-up to the Shootout: Pentax Q vs Sony NEX-7 vs Nikon D600 that was reported previously. Following some feedback provided by readers on the methodology used in the previous shootout, a new test has been performed which hopefully improves upon the comparison.

For details regarding the test and cameras involved, please see the previous post:

The following scene was shot under these conditions:
  • Base ISO, which is ISO 100 for the NEX-7 and D600, and ISO 125 for the Pentax Q.
  • Focal length approximately 50mm equivalent.
  • Aperture of F/5.6.
  • Focus achieved using contrast detect autofocus. 
  • All shot in RAW format and processed to JPG using Lightroom with no corrections, except:
    • White balance corrected in Lightroom.
    • Exported at 1600 pixels wide.


The full test scene as shot on the Sony NEX-7, 50mm equivalent focal length.

The full test scene is shown in the photo above. I won't bother repeating this for the other two cameras, as you can't see much at this scale.

The following images were cropped to show the same view, and resized to 1600 pixels wide. This represents a 100% crop on the Pentax Q, and approximately 62% crop on the NEX-7 and D600. These cropping percentages apply only when image is viewed full size at 1600 pixels wide.

Pentax Q - Cropped to 100%

Sony NEX-7 - Cropped to 62%

Nikon D600 - Cropped to 62%

This is where we get to the meat of the review. Cropped to this scale, differences between the three cameras are more obvious.

We'll start with an assessment of the Pentax Q, the image is blurry and fuzzy. The colour reproduction is not too bad, but everything lacks clarity. A commenter on the previous Shootout review suggested that the Pentax Q may not have focused properly, so this time I tried to achieve good focus. Again I have not been able to shoot a sharp image under these conditions, so either this is the true performance of the sensor, or the autofocus is very unreliable. I did try autofocus multiple times, and also attempted manual focusing without much success.

The Sony NEX-7 and Nikon D600 do much better, as we'd expect. The small text appears slightly sharper in the D600 image compared to the NEX-7, but this advantage is offset by the obvious false colour being introduced into the image with the extreme details (such as the band of vertical lines at marker 6). Let's go for a closer crop for finer comparison, the following images are 100% crops from the D600 and the NEX-7.

Sony NEX-7 - Cropped to 100%

Nikon D600 - Cropped to 100%

These crops further reinforce the previous comments. The D600 shows sharper text but with the disadvantage of false colour in areas of extreme detail. The NEX-7 on the other hand handles those details in a more pleasing manner. This might suggest a weaker anti-aliasing filter on the D600 compared to the NEX-7. It should be noted that the NEX-7 has a higher pixel density than the D600, which may also be a factor in avoiding false colour arising from fine details.

Dynamic Range

The following image tests the dynamic range of the three cameras. Dynamic range refers to the ability of the camera sensor to capture both details in dark areas at the same time as bright areas (i.e. being able to see inside a cave while retaining correct exposure in brightly sun-lit areas).

A simple method was designed to test this performance measure. The room lighting is now set to a very dim level, and a bright light source is introduced into the image, which in this case is a cellphone screen. The image is exposed correctly for the cellphone, leaving the dark areas completely black. An attempt is then made during postprocessing to recover the shadow details.

Details for this image:

  • Base ISO, which is ISO 100 for the NEX-7 and D600, and ISO 125 for the Pentax Q.
  • Focal length approximately 50mm equivalent.
  • Aperture of F/4.5, shutter speed of 0.3s.
  • Focus achieved using contrast detect autofocus. 
  • All shot in RAW format and processed to JPG using Lightroom, with:
    • Shadow Recovery +100
    • Exposure +2.0
    • Colours (white balance) not corrected.
    • Exported at 1600 pixels wide.

Sony NEX-7 - Exposure as Shot
Pentax Q - Exposure +2.0, Shadow Recovery +100

Sony NEX-7 - Exposure +2.0, Shadow Recovery +100

Nikon D600 - Exposure +2.0, Shadow Recovery +100

I won't bother offering too much comment on this test, as the results are very obvious. The Pentax is very noisy and has a purple cast to the shadow areas, no text is readable in the background. The Sony NEX-7 is slightly better, still a fair amount of noise and a green cast. The D600 performs much much better than both other competitors, with fairly good colour reproduction and some minor noise.


Conclusions

Again, we demonstrate that the Pentax Q is not a competitor to APS-C and full frame cameras. Of course, I would never expect it to be able to compete either. This series of tests does push the Pentax Q beyond its limits, and would not represent what the typical usage for this camera should be.

This version 2 shootout is more a comparison of the NEX-7 and the D600, both 24 megapixel cameras with very high expectations, except the D600 is full frame while the NEX-7 is cropped APS-C.

The D600 shows it is king of fine detail, being able to resolve the small text sharper than the NEX-7 does. However, it does also suffer from false colour in the extreme fine details, while the NEX-7 manages to handle it quite nicely.

The dynamic range test shows the D600 outperforming both competitors by several classes and would suggest it would have just as much advantage in real world shooting when both dark and bright areas are present in a single scene.

No comments:

Post a Comment