The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Another (not so) quick comparison: Zeiss 100mm f/2 ZF versus Leica 100mm f/2.8 APO

Stuart Richardson

Active member
Well, Bondo was kind enough to lend me his brand new 100mm ZF macro lens for me to try out on my D3 and F6. I am greatly indebted to him for this kindness. It was uncommonly generous, and I am very grateful that he allowed me to try out this superb lens. Before I proceed, I want to specifically note that these are my results, within my own system and workflow. I do not want to be regarded as the authority on any differences there might be between these two lenses. These are simply the differences as I see them in my experience.

I should also note that both these lenses are spectacularly good, but they are different. The Leica is a slower lens, apochromatically corrected, and of course over 4000 dollars new. Used prices run from about 1300 for a user 3-cam version to 2000+ for a ROM version in mint condition. The Zeiss lens is approximately 1580 dollars new, and includes a hood. It is a stop faster, but does not have apochromatic correction. Build quality on both these lenses is excellent. I would have to give the edge to the Leica based on the superior feel of the aperture ring, slightly smoother focus and built in hood. Both are extremely rigid even when at full extension -- there is no play in the lens body of either lens, which is a good sign. Both have nice, largely circular irises at all apertures. The Leica has 7 blades, and the Zeiss 8. My only criticism of the Zeiss build is that its focus and aperture turn the wrong way ;). It mounts backwards too. Oddly, all Nikon mount lenses seem to share this defect. The Leica is thinner (E60 versus E67), slightly longer, and about 100g heavier. Despite these differences, these lenses are similar in application and in use. They are both 100mm manual focus macro lenses optimized for very high optical performance at all distances. Used correctly, neither of these lenses will let a photographer down -- they are cream of the crop. The question is which is creamier? In short, for me the Elmarit, but reasonable people could choose the Zeiss as well.

I compared these lenses on two different systems. I did a digital comparison using the D3 and DMR, and a film comparison using an R6 and an F6 on Rollei 25 Pan film. I scanned the film results as 6300 dpi on an Imacon 646. One interesting corollary of the test was how the film and digital results compared. In short, both the DMR and the D3 offered significantly greater clarity over even ISO 25 film developed in a fine grain developer (xtol). In any case, the results of the film test closely mirrored the results of the digital test, so I am not going to bother posting the film results.

Sharpness: These two lenses are remarkably close, at least when focused with similar image areas. In the test where I photographed the wall, both lenses held superb resolution from the center all the way across the frame, even wide open. The Leica had a miniscule edge at f/2.8, particularly in the edges, but even this is debatable. When I tested at “infinity”, just the lenses at infinity shooting across the East River, the Leica clearly won, but this is entirely because the DMR is a 1.37 crop, so a 137mm lens rather than the 100mm lens on the D3.

Vignetting: This is an issue. The Leica doesn’t have any; the Zeiss does. The vignetting at f/2 in the Zeiss lens is quite visible, and it remains visible at f/2.8. It disappears by f/4. Luckily, unless you are shooting film, this is easily corrected. Even wide open, while visible, the vignetting is not atrocious.

Chromatic Aberration: In my opinion, the largest problem with the Zeiss lens is its chromatic aberration. In my experience, it shows up readily in images shot at f/2. It will show up as magenta or green, depending on the state of focus of the edge. It seems that in focus areas go magenta, and out of focus areas go green. This is just my general experience. Stopping down to f/2.8 makes it far less apparent, and it is usually gone by f/4. The Leica’s apochromatic correction really helps it here, and I was not able to induce chromatic aberration in any of the shots I took.

Bokeh: Each lens has excellent bokeh, and I don’t see either exhibiting much harshness in their out of focus areas. If I had to give an edge, I would give it to the Zeiss. The Leica has the slightest tendency to give ring bokeh (circles where the edges are brighter than the center), while this is less apparent in the Zeiss. Both look great on this front.

Overall: If I had to choose between the two without regard to price or camera system, I would choose the Leica. I would happily sacrifice the extra stop to not have to worry about chromatic aberration or vignetting. I also prefer the ergonomics of the Leica better -- its slimmer body, larger focus ring, built-in hood and properly turning focus and aperture rings. Users concerned with speed and bokeh can easily choose the Zeiss without reservation -- though it has some vignetting and aberration wide open, it is completely usable at f/2. By f/2.8, it is nearly at the performance of the Leica, and by f/4, they look remarkably similar. Bottom line, choose the one that fits your camera body...you are not going to find better 100mm lenses. Just to throw a monkey wrench in the works, I compared the 105mm f/2 DC nikkor at the same time, and while it was clearly bettered by the Leica and Zeiss, it still performed superbly well. If you need or want autofocus, particularly for portraiture, this lens cannot be ignored. In my last test, the AF seemed to be malfunctioning. I performed this test with live view and contrast detect AF, and the lens did much better. Though it is still behind the Zeiss and Leica, it is not very far behind at all. I have not had the chance to try the 105mm f/2.8 VR lens, though I am sure this would also compare well.


On to the images.
At the famous wall. Here I was unable to perfectly match the image sizes of the D3 and the DMR. Because the DMR is a 1.37 crop, I had to move back to get the same scene in the viewfinder, but I wound up literally hitting the wall. The Leica image is still slightly tighter. This gives it a slight advantage here, though it is at a slight disadvantage in terms of megapixels, so I think it more or less cancels out.

Please note, it may be easier to open these images in new tabs or windows in order to compare them side to side more easily.
Leica at f/2.8


And the Zeiss at f/2 (wide open)



Zeiss at f/2.8


Leica crop:


Nikon f/2 crop:


Nikon f/2.8 crop


And the Leica crop at left, f/2.8:


The Nikon crop at left, f/2:

and at f/2.8

I will refrain from posting the f/5.6 crops, as both lenses do so well as to make them identical. The vignetting of the Zeiss disappears.
 

Stuart Richardson

Active member
The next series demonstrates the bokeh and the chromatic aberration. Please do not regard the sharpness, as these were taken handheld and care was not given to exact framing or precise focus. The intent is to demonstrate bokeh and chromatic aberration.

Leica at f/2.8


http://www.stuartrichardson.com/r9f6-test/100mm/r9-100-f2p8-bt-center.jpg[/img]

Nikon at f/2



The vignetting and chromatic aberration are both visible in this shot. Note the color fringing on both the razor wire and the chainlink fence. This diminishes at f/2.8, but is still visible. At f/5.6, it looks like this:





This is the Leica at f/5.6

 

Stuart Richardson

Active member
The final series is a view at infinity. Please note the difference in the crop factor advantages the DMR heavily in this comparison. This is just to give you an idea how superb both lenses do at infinity, even wide open.

R9 at f/2.8:


center crop:

right crop:


Now the Zeiss at f/2...note the vignetting compared to the Leica or even itself at smaller apertures. The vignetting does not diminish the sharpness, however.





By comparison, here is the Zeiss at f/5.6

Definition improves slightly, but it is not dramatic:
center crop:

 

Stuart Richardson

Active member
Ok, I think that is all I can handle right now. Please, I am open to answer any and all questions and constructive criticism about this, but please don't complain to me about the results or any minor discrepancies -- I spent hours and hours on this, so I just wanted to share my findings.
 
Last edited:

Steen

Senior Subscriber Member
This is the ideal comparison for my purpose. Often we want to exclusively compare glass by using both lenses on the same camera, or we want to compare cameras by shooting both camera bodies with the same lens.
In this case it's a comparison of entire systems, the R-mount system and the F-mount system, and thus reflects the realistic choice. To my eyes the R-mount system looks like the winner. Which is of course to be expected considering the price difference.
Personally I suspect it has also, at least to some extent, to do with the different sensors, the Leica with its Kodak CCD sensor without Anti-Aliasing filter, and the Nikon with its Nikon (+ Sony ?) CMOS sensor with Anti-Aliasing filter ? On the other hand you said you found similar differences with film, so I dunno ?
Anyway, what a relief it is that we now also have an F-mount system that holds up so well with the expensive high-level R-mount system. Choices are good.
Stuart, thank you for your efforts with all this !
/Steen
 

Greg Seitz

New member
Stuart,

Nice work on the comparison and appreciate all the effort you put into it. It's real interesting to see the differences. To my eyes, the exposure looks to be somewhat different between the two - the Nikon appears somewhat brighter relative to the Leica. Were the exposures matched when shooting or is this the default metering from each camera?

Thanks,

Greg
 

s.agar

Member
Thanks for your efforts.
The exposures seem to be different, even if Manual mode may have been used. I have not checked for the availability of any EXIF data, but even if manual exposures are used, I suspect differences of ISO speeds from the indicated values.

Although it depends on the person, most, including myself, may tend to favor the Leica photos right away, w/o even checking the other detail aspects.

Although it may be a lot of additional work, is it possible to adjust the digital photos with PS to approximately similar values.

On the other hand, I also prefer the Leica alternative, considering the other points.

I have never used the Leica DMR (I have the M8, and SLR/n-Kodak sensor anyway) and was really surprised to see the DMR perform so well. I had never expected that, and now was thinking about a very appealing deal that I had rejected (including the R9, DMR and 5 lenses package) and felt bad.

Best regards

Seyhun
 

Stuart Richardson

Active member
Thank you all for the comments. Yes, the exposures are different -- they are the default for each camera. I used aperture priority and matrix metering for both cameras, and all were shot at the base ISO (100 for the DMR, 200 for the D3). The Leica lens/camera combination seems to have higher inherent contrast...whether this is the lens, or the sensor or both, I am not aware. It is my experience that the D3 tends to expose on the brighter side, and the DMR (after the firmware updates) tends to be a bit darker. I feel that correct exposure is somewhere in between.

I tried not to alter the image in any way after processing, and I think this favors the DMR. Because it seems to have higher inherent contrast (either from the lens or sensor algorithms) and because the lack of its anti-alias filter helps with sharpness. I opened them both with ACR...using Nikon capture for the Nikon and FlexColor for the DMR might yield different results. I tried to minimize the variables.

I think that the DMR and the D3 resolve very similarly...I don't think either has an edge at base ISO...the differences are more the lenses than the sensors. Unfortunately, it is not possible to test the same lenses on both cameras, so the differences remain elusive.

It is possible to conclude that both are superb image makers at their base ISO. The Nikon has the heavy-advantage in real world use though -- the huge rear LCD, much better battery life, dramatically better high ISO, autofocus, live view, faster frame rate....these are just a few of the many advantages that the Nikon has. The DMR only has its performance at base ISO, better manual focus viewfinder, and simple method of operation.

The Nikon seems to have a "base smoothness" and the Leica seems to have a "base roughness" that I think demonstrates the difference between systems with and without an anti-alias filter. The Leica has slightly rougher, sharper looking detail at 100%, but I don't think this is actual detail in most cases...it is rougher, but it is not resolution. If you go into lightroom and do the compare mode at 200% with the sharpening turned off, you will see that the Nikon has a tiny bit more true detail at least in the photo I am looking at (both lenses at f/2.8). The difference is that at the finest level, the Nikon is smoothed off, rather than rough.

Anyway, here is the classic blind taste test. These are the two cameras, more or less equalized for exposure, sharpening is equal, and the Nikon had some vignetting correction. Both lenses at f/2.8. Otherwise I did not modify them. If you can tell the difference from perspective, please do not say. If you can tell the difference another way, I would like to hear it. Even if you can, is there any SIGNIFICANT difference between the two that is not correctible?
I will post the crops, and if people can or cannot tell the difference, I will post the whole photos, which could give away the game, given that the perspectives are slightly different.
A center crop:

B center crop:


A edge:


B edge:
 

Steen

Senior Subscriber Member
Oops, great challenge :thumbup: now they look so close !
I do have my personal guess, and I have sent my guess by email to the Judge at the High Court (Stuart).
Whether I'm right or wrong I will tell you when the truth is finally revealed :)
 

Greg Seitz

New member
In the center crop, looks like A is the Leica to me. It seems to have a bit more sharpness and the reds are much more saturated. In the edge crop, they look very very close. I'd have to guess Leica for both :confused:, due to the bit of aliasing I see on the label in both shots.

Greg
 

Stuart Richardson

Active member
Whoops, Greg, I believe you are right there. So many little fiddly letters to write...I missed one! Here is the "B" left edge.




But anyway, you were both write. The first one is the Leica. But I think the difference other than the red saturation is very minor. The Leica seems to have more microcontrast, which gives it the appearance of higher sharpness. The little edges and details are whiter or blacker, hence appearing sharper. In terms of real detail, I believe they are practically identical. The difference could even be one being in slightly better focus than the other. That is one thing that I clearly found -- even the slightest misfocus will make any difference academic -- focus needs to be absolutely perfect for the differences to be visible. For this reason, the better viewfinder in the R9 helps, as does the focus confirmation in the D3 (or live view for that matter).

Also, keep in mind the sharpening in these files is identical...I could crank up the sharpness in the Nikon and achieve similar results as the Leica.
 

Steen

Senior Subscriber Member
I think it has to be said that if you copy the last two "left edge" shots to the harddisk and switch back and forth between them, the Nikon/Zeiss picture actually looks quite bad in comparison with the Leica.

Either this sample of the ZF lens is simply soft at the edges (or maybe just in the left side).

Or (which of course I hope) the images just weren't captured exactly 90° rectangular to the wall. Actually the initial full pictures could very well indicate something like that if we look at the doorstep line running right under the door and also the ground line ?
But in that case why would the Leica be within Depth Of Field with both center and left edge, and the Nikon not so ?

Well maybe, just maybe, Leicas 1.33x cropped sensor causes this. In order to get the same composition with the 1.33x camera, one has to move further back from the subject, which would result in slightly more depth of field ? Am I just desperately trying to defend the ZF here :eek:

I don't know if this minor difference in hardware and thus camera-to-subject distance is enough to explain the very visible difference in sharpness in the edge images, but of course I hope so. Because otherwise the Zeiss lens seems to have a problem with soft edges or at least a soft left side.

/Steen
 

GrahamWelland

Subscriber & Workshop Member
In fairness to the Nikon/Zeiss, if you'd used Nikon Capture or Capture NX to open the files I think you'd definitely see an improvement in sharpness with the Zeiss images, if only because of the automatic CA reduction.

I think I'd argue that ACR is probably equally bad at getting the best from either image :p
 

Steen

Senior Subscriber Member
(...) I think I'd argue that ACR is probably equally bad at getting the best from either image :p
Well, at least in that case the two systems are even with regards to postprocessing :D

Anyway, I dont think it would change things much. The problem is that there is a substantial difference between the center (or actually the center of the right half) and the left edge in the Nikon/Zeiss image and not so in the Leica image.
I suspect it might be because the image was taken with a slightly twisted angle of the wall and the right side of the Nikon frame being only just within Depth Of Field while the left side being outside DOF ?
With the D3 FX ("full frame") sensor the camera-to-subject distance is closer and thus the focus plane is more narrow than that of the DMR.
I'm just repeating myself here, hoping that this is the explanation of what we see. After all Stuart has said that in his judgement the lens looks like an exemplary performer across the frame with regards to sharpness.

Btw. are you saying that Nikon Capture can be updated to read the D3 / D300 NEF files ? I would be interested in this, but I thought that this software was discontinued and no longer supported or updated ?
/Steen
 

robsteve

Subscriber
My guess is A is the Leica and B the Zeiss. Not bad considering the Leica was wide open and the Zeiss stopped down a stop. Shooting both at F4 may give Leica the advantage of being stopped down a bit.

Robert
 

cmb_

Subscriber & Workshop Member
> Vignetting: This is an issue. The Leica doesn’t have any; the Zeiss does. The vignetting at f/2 in the Zeiss lens is quite visible, and it remains visible at f/2.8. It disappears by f/4. Luckily, unless you are shooting film, this is easily corrected. Even wide open, while visible, the vignetting is not atrocious. <

Stuart - Does the Leica have ROM contacts? If so, what little vignetting that occurs with the R 100 will be corrected with the DMR. Rob Stevens demonstrated this some time ago. E. Puts lists the vignetting at 0.7 stops and virtually none when stopped down to f/5.6.
 

Stuart Richardson

Active member
cmb -- the 100mm APO does have ROM contacts, so that may be in play here. That said, I think the same relationship as visible on film, but I will check again.

Rob -- you are correct. To be honest though, I don't think the Leica is at an undue advantage here...usually a slower lens wide open is better than a fast lens stopped down one. For example, the 50mm summicron M was better at f/2 than the 50mm f/1.4 Pre-asph. In the same way, the 75/2 is better than the 75/1.4 at f/2. I know that both of these are different designs, but in general, I think the faster the lens, the greater disadvantage it has, even stopped down slightly.

Steen -- you may be right about the distance from the wall affecting things. The Nikon was closer, so there would be less apparent depth of field. Also, the tripods were not (and could not) be in exactly the same orientation, so it is possible that the camera was pointed more parallel with the Leica than the Zeiss, causing the zeiss to be disadvantaged. Basically though, I think these differences are so minor as to be irrelevant. You would only see the difference on a tripod with extremely careful focus on an object that was perfectly flat. You even needed to download them and switch back and forth quickly to make the difference appear great. Both of these lenses really do a spectacular job. If you need the detail that badly, you are probably better served by just going to medium format.
 
D

diglloyd

Guest
> Vignetting: This is an issue. The Leica doesn’t have any; the Zeiss does. The vignetting at f/2 in the Zeiss lens is quite visible, and it remains visible at f/2.8. It disappears by f/4. Luckily, unless you are shooting film, this is easily corrected. Even wide open, while visible, the vignetting is not atrocious. <
Ummm...I prefer vignetting wide open not just with the 100/2 but with lenses in general. I think many subjects benefit from vignetting; I don't consider it a flaw other than maybe (in theory) increasing noise in the corners shooting at high ISOs. And of course, one should only compare f/2.8 to f/2.8, f/2 is a "bonus".

Regarding magenta/green color fringing, this is a characteristic of nearly all non-APO lenses. See my blog example with the 100/2 which shows the effect clearly; magenta in front, green to the rear. If present, it indicates the subject matter is out of focus.

Even the Leica APO lenses show some magenta/green discoloration (I recently tested the 180/2.8 APO for this issue along with several others). The only lenses completely free of the effect that I've found so far are the Coastal Optics 60/4 APO and the Voigtlander 180/4 APO. Of course, at f/4, one must compare the alternatives also at f/4. A subject for a future article.
 

toddbee

Member
I agree i kind of like the vignetting at wide apertures too. I think it actually helps draw your attention the middle of the frame. just my opinion
 

Stuart Richardson

Active member
It is fine to prefer it...I like it in a lot of images too, but the fact remains that it is an optical aberration, and as such I noted it. It's also extremely easy to add after the fact, though less so to remove (and will lead to increased noise in the vignetted areas of the image if you need to correct the vignetting). It's like spherical aberration -- it can be gorgeous in a portrait lens, but less desirable in a general purpose lens. Either way, I was nitpicking, even at f/2 it is very minor in the 100/2.
 
Top