Test: Lumix 14mm f/2.5 for Micro 4/3

[enricosavazzi]

In the neverending quest for a true wideangle lens for UV imaging, it is the turn of the Lumix 14mm f/2.5 for Micro 4/3. Tested on full spectrum Olympus E-PL6 at 800 ISO, Baader U, diffused daylight (overcast, not really direct sunlight). The same custom WB was used in all images, no post-processing.

 

This lens is equivalent to a 28mm on full frame, and therefore qualifies as a true wideangle. It also allows fast AF in UV. The optical scheme uses 6 elements in 5 groups, which means only one cemented group of two lenses. This, and the use of relatively thin optical elements (of which three are aspheric), suggested that it might be usable in UV imaging.

 

Results are mixed. With a tripod, UV imaging is indeed possible, but even at f/5.6 the corner and edge resolution is quite bad. It is much better at the center. Exposure at f/5.6 with Baader U was around 2.5s, compared with 0.4s with the CoastalOpt 60mm Apo at f/5.6 and Baader U. This means the Lumix is over 2 stops less transparent in UV than the CoastalOpt (not necessarily across the whole UVA, though). False color is also quite monochromatic with the Lumix, suggesting only a marginal transparency of the Lumix 14mm with a cutoff wavelength around 380-390 nm.

 

Lumix 14mm, whole frame reduced:

 

Lumix 14mm, center 1:1 crop

 

Lumix 14mm, near corner 1:1 crop

 

Lumix 14mm, central area, reduced to 50%, cropped to fit the FOV of the CoastalOpt 60mm:

 

CoastalOpt 60 mm, whole frame, reduced:

[Guest]

The Coastal lens looks nice and crisp compared to the Lumix, so much less purple bleed as well. Also, I'm surprised that in both of the images the glass windows appear at least somewhat transparent.

[Andy Perrin]

Most windows don't seem to block UVA totally. They aren't designed for it, as far as I'm aware, since UV is only a tiny fraction of sunlight so there is no heating issue as there is with infrared. So it's just the natural fact that most glasses don't pass UV terribly well past 350nm that attenuates the light.

[enricosavazzi]

Until last year we used to live in our own wood house, built in 1979. The double-glass windows were not glazed, and largely transparent even around 320 nm.

[Andy Perrin]

Hah, well certainly none of my local windows pass that much, but obviously YMMV. I saw one very dramatic example (I think this may be some kind of plastic window?):

[Andrea B.]

Enrico, thank you for the review of the Lumix 14/2.5. Although many lenses we consider 'good' lenses are not so great in the corners, this one seens particularly bad! And seems to also have some chromatic aberration blur?

 

I get very similar UV results with the Lumix 14-45 zoom which came with my long gone Panasonic G. The 14-45 can pass the very high UV/violet and does permit some auto-focus in good light.

 

Going up against the CO60 is a very tough comparison for any lens. :D

[OlDoinyo]

The double-glass windows were not glazed.

 

???

 

Interesting that you get good autofocus, but there is certainly a load of CA in those images. I wonder what the spectral bandpass is--any better than the Enna 28 or Tamron 21 on 35mm rigs?

[Dmitry]

Lumix 14/2.5 have huge UV focus shift. Its visible on my spectrum test images.

[Andrea B.]

But easy to refocus with Live View?

 

Edit: added a question mark. That was meant to be a question !!!