Minggu, 12 Agustus 2012

M43 Super Wide Angle Zoom Lens Comparison


Panasonic 7-14 mm vs Olympus 9-18 mm Using Panasonic GH2 and Olympus EM5
Author AndrewS  August 2012
Left to right: Panasonic 14-45mm, Panasonic 7-14mm, Olympus 9-18mm, Olympus 12-50mm.
Introduction  The Micro Four Thirds system is growing strongly partly due to the number and quality of lenses available. Panasonic and Olympus have taken a quite different approach to the design brief  for a super wide angle zoom leading to an interesting comparison.  My experience over many years has been that standard test procedures using charts in controlled conditions only tell half the story about lens performance.  So I bought both lenses and both the GH2 and EM5  camera bodies for comparison and testing over a period of two months and many hundreds of photographs in a wide variety of conditions, including test charts and real world situations. 
On the left Olympus EM5 with 9-18mm extended for use. On the right Panasonic GH2 with 7-14mm.
Panasonic-Olympus lens-body compatibility  My understanding of the M43 concept is that in principle all lenses conforming to the standard would be compatible and work properly on all bodies. However in practice this is not entirely the case. To help myself understand this matter more clearly I have somewhat arbitrarily defined three levels of lens/body compatibility:
Level 1 covers the primary exposure parameters of ISO, Shutter Speed and Lens Aperture. This level also includes autofocus performance and reliability.
Level 2 deals with in camera correction of chromatic aberration and possibly other aberrations, corner shading correction and distortion correction. Panasonic cameras correct distortion and CA but not necessarily shading in RAW and JPG files but only with Panasonic lenses. Olympus cameras do not correct CA with either Olympus or Panasonic lenses. They may correct distortion to some extent with some lenses.
Level 3 includes a range of issues such as banding with certain body/lens combinations, electronic lockups and any other unexpected faults, failures or performance issues resulting from mixing maufacturers.
My testing has shown that lens-body compatibility issues are a major factor in the overall performance and functional reliability of each lens. Compatibility issues can, and in my experience do, occur at each of the three levels.

Treillage, Panasonic 7-14mm on GH2
Note on sample variation  My experience and reading indicates significant sample variation in both lenses and bodies. The findings reported here are valid for the particular samples which landed in my posession and may not be replicated exactly with other samples. However both lenses and bodies tested appear to be good with no obvious faults or defects. The only issue I noted is slight decentering of the Olympus 9-18mm lens as evidenced by slightly more softness in the top corners than the bottom corners  in landscape orientation at the wide end.
Design philosophy  You can see at a glance that the Panasonic and Olympus designers have taken a completely different approach to the SWA lens concept. The Panasonic goes for the highest possible optical performance while the Olympus achieves an astoundingly compact form. I think both succeed rather brilliantly but their success lies in different aspects of the performance spectrum so they are somewhat difficult to compare directly.
Strand Arcade Sydney Olympus EM5 with Panasonic 7-14mm
Specifications  You can read all the details elsewhere but in summary the main points are:
Zoom Range:  Both are 2x zooms but they have a different focal  length range. If you have not used a SWA lens before, you might not think there would be much difference between 7 and 9 mm at the wide end, but in practice the difference is very noticeable indeed, moving the Panasonic 7-14mm lens into Ultra Wide Angle territory, capable of extreme perspectives and requiring considerable care with composition. The other thing to bear in mind is that the 9-18mm range of the Olympus overlaps quite a bit with standard 12-50mm and 14-42mm zooms.
Collapsing Design:  The Olympus collapses from 71mm long in operating configuration to 49mm for storage. The Panasonic is 83mm long including the fixed petal hood.
Nominal Cylinder Volume: This is length x Pi x (half max diameter) squared, approximately representing the space which the lens will require in a camera bag. With front and rear caps in place the Panasonic measures 9.3 x 7.2 cm giving a Nominal Cylinder Volume of  380cc. The Olympus, again with front and rear caps in place and in collapsed configuration is 6.2 x 5.5 cm for a volume of 147cc. So the  Olympus nominal volume is 38%  of the Panasonic.  The Panasonic is by no means a big lens, indeed it is diminutive compared to SWA lenses for APS-C and Full Frame sensors.  But the Olympus is even smaller when collapsed, a remarkable feat of optical engineering. It can be tucked into an otherwise unused corner of a camera bag or nestled below or on top of another lens, taking up very little space.
Filter: With it's large dome shaped front element partly protected by a built in petal shaped lens hood, the Panasonic cannot accept a filter. I would think carefully about exposing this lens to salt spray or heavy dust. The Olympus somehow manages to enable a standard 52mm filter to be fitted, it doesn't even need an ultra slim type.
Mechanical operation  Both lenses zoom smoothly and feel well made. Both change length with zooming but on the Pansonic this occurs inside the fixed lens hood. Presumably the Panasonic/Olympus coordination committee was having a picnic day when someone designed the zoom action because they turn the opposite way !!! When mixing and matching between Olympus and Panasonic lenses I constantly find myself zooming the wrong way as a result.
The collapsing function on the Olympus is easy and smooth to operate, with a lock button.
Autofocus Performance and Reliability  I made several hundred photos using a wide variety of subjects and lighting conditions, using each lens on each camera. I kept notes along the way. In summary the rate of focus error was lowest with brand matched lens/body pairing. The Pana 7-14 on EM5 was quite prone to focus errors, especially in conditions with low light level and/or low subject brightness range. Sometimes the camera signalled difficulty finding focus but at other times it indicated focus acquired when that was not the case. Subjects with numerous glowing light sources in frame were a particular problem. This appears to be an issue for contrast detect AF systems as I have encountered it with the Samsung NX10/11 which usually have highly reliable AF, and seen the issue reported with Sony NEX cameras. The photograph of Strand Arcade made with the Pana 7-14 on EM5 produced several misfocussed frames.
Manual Focus  (No focus by scale)  No problems were experienced with manual focus using AF+MF or MF, both with MF assist. My ongoing complaint with M43 and all mirrorless ILC's to date is they lack any kind of facility for manually setting a chosen distance by scale. I don't understand why this is not provided as an on screen readout since the camera presumably "knows" what is the focussed distance set by the lens at any time. If this facility were available I would use it almost all the time with Wide and Superwide angle lenses which are ideally suited to prefocussing at the hyperfocal distance.  
Olympus 9-18mm on EM5 100% crop from corner
Optical Performance
Resolution: Both lenses deliver very high resolution in the central image area, with little discernible difference between them at any comparable focal length or aperture. In the central area virtually maximum resolution is delivered at the widest aperture available. The story is different at the edges and even more so in the corners. Here the Pana 7-14 shows clearly better resolution at all apertures and focal lengths. Stopping down provides slight benefit for the Oly 9-18 but makes little discernible difference with the Pana 7-14.
Panasonic 7-14mm on GH2 100% crop from corner
Contrast: The Pana 7-14 holds local contrast well, right into the corners, but the Oly 9-18 suffers from loss of contrast towards the corners.
Flare, General, Local, purple: As with most super wide lenses, direct sun on the front element of either lens will produce a variety of unwelcome flare patterns. The Oly 9-18 is prone to local flare towards the corners particularly with subjects having a high brightness range. In these conditions there is considerable local image degradation due to light flaring from bright subject areas into dark areas. The Pana7-14 is subject to a curious phenomenon which I have only encountered on the EM5 thus far. This is one or more large areas of soft  purple flaring in the vicinity of bright light sources in the frame. This has been reported on the dpreview m43 user forum. I am not aware of a cure for this problem.
Chromatic Aberration: This is clearly present toward the corners with both lenses. It is corrected in camera with the Pana7-14 on Panasonic cameras but not corrected in the other lens/body combinations. It can easily be eliminated in ACR 7.1 or LR4.1.
Purple Fringing: This can be very prominent towards the corners with both lenses and is corrected in camera, sometimes incompletely depending on the subject, with the Pana7-14 on Panasonic camera. Purple fringing can generally be eliminated or at least greatly reduced by judicious use of the control sliders in ACR7.1 or LR4.1.  
Drawing (geometric distortion): The Pana 7-14  on Panasonic cameras produces partly corrected barrel distortion at the wide end and incompletely corrected pincushion distortion at the long end. The Oly 9-18 on EM5 shows more obvious barrel distortion at the wide end and just the slightest trace of barrel distortion at the long end.  In each case the distortion pattern is evenly circular (not moustache type) and easily corrected in ACR 7.1 or LR 4.1.
Optical Performance Summary  In a broad central area of the frame both deliver excellent performance, with little to choose between them. Towards the edges and corners the Pana 7-14 has a clear advantage. So overall the Panasonic has the best optical performance and is a stop faster at the long end as well. The actual level of resolution achievable by both lenses is extremely high, enabling images containg a remarkable amount of fine detail.
Conclusion: Two lenses, Two winners  Each of these lenses has clearly been designed to optimise a different range of characteristics and capabilities and each succeeds resoundingly. Your buying decision will depend on which are most important for your own requirements.




Tidak ada komentar:

Posting Komentar