ZE 50/2 and 100/2 Makro-Planar charts

After looking carefully at the recently-uploaded ISO test charts for these lenses, I noticed that both of these lenses seem to have some contrast issues in the center at all f-numbers, which seems very strange to me. For example:


http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=115&Camera=453&Sample=0&am p;FLIComp=0&APIComp=3&LensComp=727&Cam eraComp=453&SampleComp=0&FLI=0&API=5


and


http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=107&Camera=453&Sample=0&am p;FLI=0&API=2&LensComp=728&CameraComp= 453&SampleComp=0&FLIComp=0&APIComp=3


Notice how in both cases, the edges on the center target (concentric circles) have sharp contrast on both the EF 50/1.4 and EF 100/2.8 macro, but the ZE lenses have a kind of halo effect. Also notice that stopping down further doesn't reduce it.


Any thoughts about this? Many owners of these ZE MPs are quite vocal about how much better their lenses are compared to the respective EF lenses; and if one looked in the corners, I suppose they'd be right. But I must wonder what is going on in the image center. It looks vastly inferior. And please don't start spouting nebulous buzzwords about "microcontrast" and "Zeiss look." I want to talk about possible explanations for the chart results.

Looks like flare. Perhaps Canon

Maybe the studio lighting has changed since previous shots, causing flare?


Anyway - the center of the ZE 2/100 shots have a slight magenta cast @ f/2 and f/2.2, suggesting that the focus is narrowly off.


Except for the soft spot at the center, both Zeiss Makro-Planars soundly trounce the comparable Canon lenses regarding image quality on these test shots.


I own the Makro-Planar 2/50, and can attest to its superb image quality. I previously owned ZE Planar 1.4/50, and was rather disappointed at the performance, regarding contrast wide open and especially the downright horrible bokeh at full aperture approaching minimum focus distance. I upgraded to MP 2/50, and have gotten everything I want from a stellar 50mm. :)

Although there is a possibility of imperfect focus, having seen the care with which Bryan has performed all the testing on all the other lenses, I doubt that this remotely likely. A magenta cast is more often the result of axial chromatic aberration, and its visibility at the fastest apertures in these lenses is consistent with such a cause. And since axial CA can be present even at the point of sharpest focus, I think it is reasonable to conclude that misfocus is not the reason for its appearance here.


Furthermore, to say "except for the soft spot in the center..." isn't very objective. One could just as easily have correctly concluded, "except for softness in the corners, the Canon lenses soundly trounce the Zeiss Makro-Planars." One could even say that because the test charts show a loss of contrast in the center--and therefore the most critical region--of the image, and that this phenomenon is present at ALL f-numbers, that this issue is more detrimental to the overall imaging quality of these lenses. That is to say, it won't matter if the corners are sharp if the center isn't. But I am withholding that judgment until I have a better understanding of what I am seeing--i.e., whether it is a result of testing methodology, or if it is a significant aberration of the lens that will be visible in real-world shooting conditions.


Moreover, if this phenomenon is indeed some kind of flare, why are we only seeing it in the very center of the image, and not anywhere else? The ZE 50/1.4 Planar and the 21/2.8 Distagon don't show the same, either.


If the results I've seen had been from any other lens, I probably wouldn't have bothered to question it, as I consider Bryan's testing to be fairly rigorous and sufficiently precise. The fact that I DO find it odd, and am seeking an explanation for it, means that even I acknowledge the reputation that Zeiss lenses have. However I will not go so far as to defend the performance if its prestige is not consistent with the observed results.

Quote:

---

Originally Posted by wickerprints

I acknowledge the reputation that Zeiss lenses have.

---

My Zeiss lenses are certainly excellent. But then, mine are microscope objective lenses and those are made at the Zeiss production facility in Jena, Germany. FWIW,Zeiss dSLR lenses are OEM'd by Cosina in Japan.

I believe the slight magenta cast of axial CA is due to a slight misfocus. But - that misfocus would have to be so slight that it would not be the reason for the center softness.


I see the 2/50 and 2/100 charts have been pulled - I guess we will get an explanation for this phenomenon later.





For a color cast indicating that the focus is way off, affecting the result - check most of the Nikon lens reviews - like the Nikkor 135 f/2.

Quote:

---

Originally Posted by Trondster

For a color cast indicating that the focus is way off, affecting the result - check most of the Nikon lens reviews - like the Nikkor 135 f/2.

---

The numbers on the chart are pretty sharp - sharp enough that I'd say focus cannot be "way off," as you suggest. Rather, I suggest instead that the purple cast is truelongitudinal (axial)CA - the fact that stopping down progressively improves the fringing is consistent with that suggestion. In fact, a slight defocusing would actuallyimprovelongitudinalCA, rather than making it worse. Also note that this is a flat chart - we're not talking about 'bokeh fringing' here. As a side note, technically CA applies only in the plane of sharp focus, since by definition an aberration affects only what is in focus. But as the term is commonly applied, I'd call the fringing on both the Nikkor and the Zeiss lenses longitudinal CA, and not due to missed focus.

Hi Wicker...


I too, always look to Bryan

Quote:

---

Originally Posted by wickerprints

And since axial CA can be present even at
the point of sharpest focus, I think it is reasonable to conclude that
misfocus is not the reason for its appearance here.

---

I think this makes sense, but in some sense does not contradict what Trondster was saying. Some colors *are* out of focus, not due to operator error but because
the focal plane is not flat in all colors.

Quote:

---

Originally Posted by neuroanatomist

My Zeiss lenses are certainly excellent. But then, mine are microscope objective lenses and those are made at the Zeiss production facility in Jena, Germany.

---

Mine is an eyepiece, also Zeiss Jena. (Actually, I don't know where it is made, but it's branded "Zeiss Jena"). It is a tiny eyepiece with a narrow field of view. And it is awesome, really.


Never been tempted by Zeiss camera lenses, though. (Okay, well... not much [:)])

Quote:

---

Originally Posted by neuroanatomist

The numbers on the chart are pretty sharp - sharp enough that I'd say
focus cannot be "way off," as you suggest. Rather, I suggest instead
that the purple cast is truelongitudinal (axial)CA - the fact that
stopping down progressively improves the fringing is consistent with
that suggestion. In fact, a slight defocusing would actuallyimprovelongitudinalCA,
rather than making it worse. Also note that this is a flat chart -
we're not talking about 'bokeh fringing' here. As a side note,
technically CA applies only in the plane of sharp focus, since by
definition an aberration affects only what is in focus. But as the term
is commonly applied, I'd call the fringing on both the Nikkor and the
Zeiss lenses longitudinal CA, and not due to missed focus.<span class="field-item-description"]

---

I definitely agree that the cast is due to longitudinal CA - that is precisely my point.


I am very curious as to how a slight defocusing would improve longitudinal CA, as longitudinal CA typically manifests itself as a magenta cast on one side of the focus plane and a greenish cast on the other side, while the focus plane itself (if the lens is constructed correctly) is without any color cast. I can see this for myself in my own pictures with the Zeiss ZE 2/50, where there is a color cast on objects just outside of the focus plane. If the focus had been spot on, we would not have seen any noticeable longitudinal CA.
If the focus was even more off, the cast would not be as visible, but then the whole image would be really blurry.


A good example of longitudinal CA would be in Photozones review of the Sony mount Zeiss ZA Sonnar T* 135/1.8, where the lines in front of the focus plane are purple/magenta and the ones behind are green. There are also examples in Photozones Nikkor 135/2 review, where crops of pictures just behind and in front of the focus plane show greenish and purple casts.


I agree that describing the focus as being "way off" was a bad choice of words - what I meant was "enough out of focus that it would affect the visible result in the 100% crop shots" - or at least "way off for an image taken for a lens review site", and I still believe that the focus is off in the Nikkor 135/2 review, and that the magenta cast is a very clear indication of just that.


I'd rather call the fringing on the Nikkor lenses longitudinal CA due to missed focus.


The very slight CA on the Zeiss picture could be a very slight focus miss, and/or be due to a slightly curved focus plane or a slight lens defect. The Zeiss lenses are known for pronounced longitudinal CA (in my humble opinion, it is their only weakness), and thus a focus miss would be extra visible. However - the very slight color cast on the Zeiss picture (which now has been pulled) was so slight as to be negligible.





As well - Bokeh fringing, Bokeh CA, Longitudinal CA, Axial CA - are all words for the same phenomenon. You would be able to see "bokeh fringing" or "Longitudinal CA" all over a flat chart, if the entire chart was out of focus - as I believe we see in the crops for the Nikkor 135/2.

Quote:

---

Originally Posted by Trondster

Bokeh fringing, Bokeh CA, Longitudinal CA, Axial CA - are all words for the same phenomenon.

---

I guess my point was that those terms are all used interchangeably - but that doesn't mean that such use is correct. There are lots of people who say or write the word, 'irregardless,' but that doesn't make it a word, nor does the common use of the word 'data' as a singular noun (e.g., the data is correct) make that usage correct - data are plural (except when used as the proper name of a Star Trek character, of course ;).


The distinction becomes important in other types of lens applications, for example in confocal microscopy where out-of-focus light is eliminated by a pinhole in the optical path, but axial CA can still be present if the lens is not sufficiently corrected. That's why the lenses on my confocals are fluorite or planar apochromatic lenses corrected for SA and CA in at least 3 and usually 4 colors - (and many of them cost well into the 5-figure range).

Quote:

---

Originally Posted by Trondster

I'd rather call the fringing on the Nikkor lenses longitudinal CA due to missed focus....You would be able to see "bokeh fringing" or "Longitudinal CA" all over a flat chart, if theentirechart was out of focus - as I believe we see in the crops for the Nikkor 135/2.

---

The thing I have trouble reconciling is that despite the purple fringing, the chart looks focused and sharp. Other Nikon fast primes (50/1.4, 85/1.4, 85/1.8) exhibit similar fringing - not just the 135/2. So there are a few possibilities:

• Bryan is intentionally trying to make these lenses look bad.
• Bryan is not very good at focusing on his test chart.
• There is some unidentified step in the testing workflow that is introducing an artifact into the results.
• The fringing is a 'feature' of some Nikon lenses.

I don't buy the first two at all. The third one is certainly possible. The fourth one is also possible - bear in mind that pure black on pure white illuminated by four SB-900 or 580EX II strobes generates more contrast than most 'real-world' shots.

The poor results regarding several Nikkor lenses are easily explained. All the Nikkor images were batch shot and hastily uploaded into the database, to provide a large database of Nikon-related images. There was probably not time to check every image and reshoot if necessary.


When the time comes to write the reviews proper, the ISO crops will quite certainly be reshot for the ones being out of focus or imperfect in any other way.


And no, they are not completely blurry, they are nudged just that little bit out of focus, causing a color cast. Check my example links above for the Photozone reviews - the lines with color cast closest to the focus plane are not necessarily that blurry - a bit blurry, but with quite the color.


Try it yourself - shoot a perfectly flat black and white chart with an 85L or similar, shooting using manual focus and focus stepping - you'll probably end up with both green, black and purple charts.

Since the aforementioned charts have been taken down as of this writing, I am withholding further speculation on the matter until further clarification is provided.


As to the chromatic aberrations in the Nikon charts, we have already discussed this issue to some extent in other threads. At this time I do not have an explanation for the purple fringing, but I would be highly doubtful that this could be caused by flawed or sloppy methodology; and I think it is a bit unbecoming to suspect that as the cause, without any evidence to support such a claim. That is to say, if you are going to accuse Bryan of "hasty uploading" of test results on his own site, you better have a way to back it up, because I find that accusation to be distasteful.


In any case, what we DO know is that there must be some kind of automatic lateral CA correction taking place, because NONE of the Nikkor results show any lateral color whatsoever. That is an obvious impossibility. So the likelihood of these lenses needing retesting is high, in my opinion--regardless of any axial color issues.


Finally, I hope that everyone can step back and refrain from jumping to conclusions about the test charts or make sweeping generalizations about how some brands are better than others, and try to be a bit more objective about the results themselves.

..And the charts are back up - with significantly sharper results in the center. I'd say that the Makro-Planar 2/50 still soundly trounce the comparable EF lenses in the corners at f/2.


It depends on the priorities, I guess - EF 50/1.2L is optimized for performance wide open, EF 50/1.4 has a large aperture and fantastic performance stopped down, at an attractive price point, and Makro-Planar 2/50 has very good performance at the cost of being slower, more expensive and manual focus only, and with an insanely long focus throw.


Makro-Planar has in my opinion two weaknesses - neither of which are shown here: It has a lot of LoCA, and the focus plane is curved - I have not performed any systematic testing, but it seems that this effect is more pronounced at longer distances - it seems the lens is optimized to excel at close range.


So - which is "better"? It depends on your usage - they all have strengths and weaknesses and excel in their own areas - what is important for you? :)