Article Index

Resolution

Please note that MTF50 results for APS-C and Full-Frame cameras are not cross-comparable despite the same normalized [0:1] range used to report results for both types of cameras.

 

Leica Summicro-R 35mm f/2 showed pretty solid overall performance and performed more or less consistently on different cameras used in the tests. Based on the MTF50 charts below, the lens exhibits outstanding center performance throughout the aperture range and can handle low and high resolution sensors quite adequately. The main difference is at the borders, specifically at wider apertures, where the lens struggles to squeeze out the maximum resolution. f/2 seems to be the weakest point on both APS-C as well as FF cameras, but while f/2.8 brings up enough detail on APS-C sized 400D and lower resolution 5D, 5DMk2 puts much more stress on the lens and so f/2.8 does not recover as well on this camera. From f/4 quality the lens produces very good results across all camera bodies.


Canon 400D (10Mp)

Canon 5D (12Mp)

Canon 5DMk2 (21Mp)

 

Interestingly, when I look at the chart crops that compare image corners at f/2 and f/8,I can't see that much of difference in quality between images made with a 10Mp 400D and 12Mp 5D. There is a little bit of difference between f/2 and f/8 obviously, but both cameras crank out very similar results at f/2. On the other hand, the 21Mp Canon 5DMl2 clearly shows softer results at f/2, not only compared to f/8 taken with the same camera, but also to f/2 taken with the lower resolution bodies. I'd expect this kind of softness to show up in real life results, which we will examine below.

 

Canon 400D (10Mp)

Canon 5D (12Mp)

Canon 5DMk2 (21Mp)

 

The crops below demonstrate what you can expect from the lens in real life situation and under normal lightning conditions. All shots were made with a 21Mp Canon 5DMk2, using focus bracketing around infinity. As expected, center resolution is superb and I can't see any significant difference between f/2 and f/8 results. Border results however do show difference, with f/2 crops showing noticeable softness in both left and right corners of the image. The drop in detail, while unfortunate, is likely going to be tolerable to many users - applying a little bit of sharpening improves results, making the quality variance between the image center and borders less drastic.

At this point, an interesting question to ask is whether the softness seen around image corners is due to field curvature or due to the optical formula itself. Since I took a set of focus bracketed shots to test the lens for image quality, I also managed to compare results with varied focusing position. Basically, I was hoping to find two sets of images shot at f/2 with one showing best resolution in the center and the other one showing best resolution around corners. On the other hand, if the image with best resolution in the center was also the image with best resolution around corners, then field curvature did not contribute anything significant degradation in quality. After staring at two dozen or so bracketed pictures I can safely say that if Leica Summicron-R 35mm f/2 does have field curvature, it is so minimal that it does not affect visual image quality significantly.

 

Center

f/2, Canon 5DMk2

f/8, Canon 5DMk2

L.Corner

f/2, Canon 5DMk2

f/8, Canon 5DMk2

R.Corner

f/2, Canon 5DMk2

f/8, Canon 5DMk2


Color Reproduction

Chromatic aberration (lateral) was moderate to low with Leica Summicron-R 35mm f/2. Center CA was benign on all cameras used for testing. Border CA was significantly higher, on average hitting 1px on all cameras at f/2 and then gradually dropping towards more manageable levels by f/4 and beyond. In practical terms, this means that you may find some CA at wider apertures, but not consistently so. Even with the lateral CA present, the levels are unlikely to cause major problems and therefore requiring extensive color correction. The lens also showed minor degree of axial (longitudinal) chromatic aberration at f/2, which we will discuss in the DOF section. For a visual example, check out the picture of a blooming tree in the sample gallery and note white coronas around flowers against darker background.


Canon 400D (10Mp)

Canon 5D (12Mp)

Canon 5DMk2 (21Mp)

 

Other then some color fringing around high contrast areas, Leica Summicron-R 35mm f/2 showed good color reproduction. Palette remained well saturated with colors staying vibrant and spunky, although one notable exception was wider apertures and close focus, which we will discuss in the next section. Images carried good amount of global contrast. Tonal reproduction was fairly good from f/4 on and I was consistently surprised at how much detail the lens managed to squeeze into its RAW files - you just need to know how to bring that detail up in post-processing, however at f/2, lower resolution seemed to intefere and compressed micro-contrast a little bit particularly around corners.

I can't say that the results I get from Summicron-R 35mm f/2 are unique and I'd rate them as A or A- if you want to nitpick on the issue of chromatic aberration. Combination of high center resolution, good tonal reproduction and color can give you very lively images. You might be able to get a nice semi-pop (is that really a word?) to your subject once in a while if you know what you're doing, but generally, I don't feel that getting a nice 3D impression is as easy as with say Summilux-R 80mm f/1.4 or some of the Zeiss lenses like Planar T* 100mm. But don't despair - even without the unique 3D-ness, your images will not look flat, that's for sure.


DOF & Bokeh

I discussed this elsewhere, but generally speaking, I don't like using wide-angles when shooting for scenery that requries good levels of separation between the foreground and background. It's not that it is impossible to get a good DOF with wide-angles, it's just too much trouble - you need to get really, really close to the subject for the optical properties to kick in and render out of focus areas into something that can be called bokeh (I'm exagerrating a little bit here, but you get the point). And so if 'get into your face' scenario is not an option for you, then you should not expect any miracles from Summicron-R 35mm f/2. But let's walk through this one step at a time.

To begin, here is the customary ruler shot, done at f/2 and 30cm. The lens does show a fairly thin DOF here. Surprised? You should not be, since the lens is so close to the subject. Transitions into OOF are fast and furious so to speak, meaning that the loss of detail is significant even with small movement away from the plane of focus. Note however, that the image is not terribly sharp to begin with, even dead in the center where it is focused. Also note color fringing around the necklace clip, which is slightly out of focus - the longitudinal chromatic aberration is minimal, but seems to be standard at f/2 in most cases.

 

ISO 400, 1/200, f/2, 35mm (Canon 5DMk2)

 

Moving on to the real-life scenario. The two shots below were also made at close to the minimum focusing distance, to demonstrate how the lens renders bokeh under normal circumstances. The scenario here is close to ideal - a static object to focus on and distant background, which should get almost completely blurred (recall that due to the optical properties, the farther away the background is, the heavier it will be blurred). And so at f/2, Summicron  renders the background into an almost-shapeless mast. There's only a hint to the building and trees, everything else lost any definition. There's a nice feel of separation here and the highlights are rendered quite uniformly and I'd even say pleasantly. But, stop down the lens to f/8 and most definition around background is back. The objects are still too coarse and you cannot distinguish small details, but you can clearly see most shapes. Highlights are still uniform, but are no longer as pleasant, more neutral I'd say.


DOF @ 0.35m

ISO 100, 1/3200, f/2, 35mm (Canon 5DMk2)

ISO 100, 1/160, f/8, 35mm (Canon 5DMk2)

 

However, once you move away from the subject or change the distance to the background, you would end up with a completely different rendering in OOF areas. The image below was taken also at f/2, but the distance to the subject in focus is about 60cm and the background objects are also located at much closer distance then what we've seen in the pictures above. The first thing to note is that resolution up close is not on par with what you can get at longer focusing distances (i.e. infinity) - it's not bad, to be precise, but not jaw dropping either. Not a critical item IMO. The background rendering is also different - there is not enough blur here to render the OOF area significantly and so you can easily distinguish all objects. OOF highlights still retain their neutral coloring and edging, and there are no visible double contouring around OOF objects. But the overall results are not particularly special, and that's what I meant when saying that it takes an effort to bring up the best in DOF with wide angles - Summicron-R 35mm not being an exception here. Bottom line is that the lens offers a very short minimum focusing distance and so if you manage to take advantage of it in your sessions, then by all means, you can get a smooth and in many cases pleasantly looking bokeh.

 

ISO 400, 1/100, f/2, 35mm (Canon 5D)
ISO 400, 1/100, f/2, 35mm (Canon 5D)