Yes, its another David vs. Goliath. I’ve been meaning to do a comparison along these lines for some time, but time has proven elusive. The Taliban have been doing the only thing they know well (murdering innocent people) and I’ve therefore been a bit distracted. Thankfully the insanity of the last few weeks in the capital have subsided and I can think about more than the security of the compound I live in!
Everyone knows the Sony Zeiss 55mm f1.8 FE is the most technically accomplished 50mm ‘ish’ lens this side of a Zeiss Otus and I have been seriously impressed (see quick review here). Readers of my blog will also know that I was quite dazzled by the performance of the 30 year old Canon FDn 50mm F1.4 lens when I tested it (see here). The Canon 35mm f2.8 FDn also did rather well when compared to the Sony Zeiss 35mm f2.8 FE Sonnar. The FE lenses clearly come with certain advantages, including autofocus, EXIF data and lens profiles, but should these things not matter to you, it starts to become very interesting indeed.
General Impressions: The Canon and Novoflex adapter are about the same combined weight as the Sony Zeiss 55mm Sonnar, although the latter is a bit bulkier. From my point of view, the focus action is not something that I see as better on one or the other. I know the FE lens has a fly by wire system with no hard infinity stops, but as far as I am concerned it works very well and feels just great. The presence of AF and effective MF (or DMF mode), to me, make the 55mm the more useful, flexible and more ‘natural’ in the context of a system, but for the dedicated landscape and scenic shooter most of this goes right out of the window and in the process about $700+ comes flying back in, assuming an approximate price of about $80 for a really clean 50mm f1.4 FDn and about $800 or so for the Sony Zeiss 55.
Now for some boring technical notes before the test photos are revealed: All images were shot hand held on the 24 mp Sony A7. and I have very steady hands. Really. Not that it matters much at the shutter speeds used. I shot multiple frames and I can confirm that they are all entirely representative. Why the A7 and not the A7? That’s pretty simple: I wanted to see if what differences make an appearance at the lower pixel count and at a later date (soon) I will so further testing with the A7R to see how the additional resolution and pixel density affects things. I shot at ISO 200 to give me the speed needed for crisp frames, but for reasons unknown and previously noted with the adapted FDn lenses, exposures were a bit lower on the FDn shots. I therefore lifted them by 0.6 stops in post to match them. Yes, yes, yes, this means more noise and (in theory) a slight loss in resolution, but I am confident the sensible amongst you know that this will have didli squat impact on what we are about to see. This is also a polite way of saying that if you want to pick holes in my simple methodology, you are not sensible. The FDn frames have had no vignetting corrections, so you will note a difference there, but who cares about that? This is all correctable for those with the inclination. All crops are 100% unless stated otherwise. P.S Sorry for the railing in the lower right of some of the frames. I just did not want to lean forwards and attract the attention of the AK-47 tooting maniacs at the end of the street on the right.
Firstly, the whole frames at f5.6 – a nice representative middle aperture. You can see that the 55mm has a bit more contrast and sparkle. This is typical of what you see across the Sony Zeiss line vs the Canon FDn lenses. It can be very welcome, but not always. It is definitely a harder rendering, but how it presents is definitely down to post processing and can definitely be attenuated if you wish.
Now for whole f2 frames. The vignetting in the uncorrected FDn shot is clear.
Click below for next page.