Panasonic Lumix G Vario 12-32mm f/3.5-5.6 ASPH Mega O.I.S. - Review / Lens Test Report - Analysis
Lens Reviews -
Page 2 of 3
The image distortions are auto-corrected in the Micro-Four-Thirds system either directly by the camera (JPEGs) or by most RAW-converters. The "exposed" distortion is therefore moderate especially for a standard kit lens. It ranges from a moderate (~2.1%) barrel distortion at 12mm to negligible(~0.05%) pincushion distortion at 32mm. While you may spot the barrel distortion at times (e.g. during architecture photography), this is usually nothing to worry about.
However, when looking behind the scenes (e.g. via RawTherapee) the non-corrected raw images reveal a different situation with an extreme (~6.3%) barrel distortion at 12mm (on the positive side this is still less than for the Lumix G X 14-42mm f/3.5-5.6 PZ).
If you move your mouse cursor over the image you can switch to the corresponding non-corrected results.
The tiny lens has a fairly typical vignetting characteristic. At 12mm and f/3.5 the amount of vignetting is ~1.15EV which will be noticeable in field images. Nevertheless, the situation improves significantly when stopping down and the light falloff is only slight from ~f/5 onward.
Note: with activated vignetting correction, this is nothing to worry about in any case although you'll have to accept a slightly increased corner (sensor) noise. Expect a slightly higher vignetting in "true" RAW mode (without distortion correction).
Given the manufacturer's own resolution figures, we were, of course, curios how the Lumix G 12-32mm f/3.5-5.6 ASPH Mega OIS performed in our lab ... unfortunately somewhat worse than expected really. This originates at least partially in automatic distortion correction. The associated image interpolation decreases border and corner resolution.
The MTF curves of the lens feel a bit odd to newcomers. Due to the small MFT sensor the peak resolution of the system is usually reached in the f/2.8 to f/4 range. Now, because the lens has an aperture range of max f/3.5-5.6 this results in a decreasing center resolution the more you stop down - simply because of the increasing dampening impact by diffraction. This is a physical effect that cannot be compensated via the optical design. Anyway, at 12mm f/3.5 is center quality is excellent and still very good at f/5.6-8. The borders as well as corners are good but certainly not tack sharp and there's a more pronounced drop in the very last square-mm of the (sensor-)corners. Unsurprisingly the border/corner results improve somewhat at 18mm and 25mm (because of less distortion correction). The center quality decreases slightly the more you zoom out. At 32mm the image quality is fairly even across and varies around the good to very good mark here.
The centering quality of the tested sample was about average. The field curvature is only slight.
Below is a simplified summary of the formal findings. The chart shows line widths
per picture height (LW/PH) which can be taken as a measure for sharpness.
If you want to know more about the MTF50 figures you may check out the corresponding
Chromatic Aberrations (CAs)
There's only a slight amount of lateral chromatic aberrations (color shadows at harsh contrast transitions) with an average CA pixel width of around 1px at the image borders.