Makten wrote:
Bear in mind that you are using a camera with extremely high pixel density, equal to 48 megapixels on FF. That might show "weaknesses" that won't ever be resolved on the cameras these lenses were intended for. Resolution is only one parameter of image quality, and in my opinion it's the least important one.
That's a good point. I was going to say the same thing myself but my brain came up with 24mp equivalence which I thought wrong but was too lazy to think about for more than that brief fleeting moment.
Bifurcator wrote:
Many users here have complained of LoCA and PF with the ZF/ZE/etc. versions.
wiseguy010 wrote:
Many users? I didn't see them. Yes there are some examples of LoCA, but certainly not many.
My impression is that you are working to a predefined conclusion with your tests.
But still interesting.
Yeah, OK, maybe not "many" but if you read through all the threads where it's mentioned it's more than "a few".
And yeah, I guess I am operating from predetermined conclusions. I see how much sharper it is, how much cleaner the colors are, and how much better it's correcting for the various wavelengths. And I want to show exactly by how much - as well as saying so too. I don't think that's necessarily a bad thing though as long as the samples I show are honest - which they are. Right, even a n00b like me does that - ignores most of the commentary and goes straight to the samples.
PS: I realize you weren't saying it was "necessarily bad" but just thought I'd clarify.
crazeazn wrote:
awesome comparo!
Thanks man!
AhamB wrote:
But you never see the corners of the lens with your MFT camera, and that's where the LaCA occurs... (esp. at f/2.8, where some slight field curvature or astigmatism may be aggravating things). In the image center you won't see LaCA per definition, so no wonder you didn't see it with your GH1.
True that. I think I said as much myself. On the other hand I'm going to stick my neck out and claim there is none! And additionally challenge anyone to show otherwise!
edwardkaraa wrote:
I have used the Contax several years on Canon bodies. It is easily my most favorite macro lens. It has a drawing style particularly suited for macro shots, and has an extended DOF effect that Zeiss designs in some lenses, and as mentioned in Dr. Nasse document, it sacrifices some sharpness at the point of focus in order to extend the area of acceptable sharpness beyond the usual limits.
That's interesting! I wanna read that. Where was that? link?
But off hand I'm initially anyway, not really understanding how sacrificing sharpness can increase sharpness.
Bifurcator wrote:
But off hand I'm initially anyway, not really understanding how sacrificing sharpness can increase sharpness.
Increasing the area of acceptable sharpness -- not increasing the maximum sharpness at the plane of focus.
I think this is the article by Dr. Nasse that Edward is referring to, but I don't remember in which section he talks about increasing DOF at the expense of maximum sharpness.
AhamB wrote:
...but I don't remember in which section he talks about increasing DOF at the expense of maximum sharpness.
That's probably just spherical aberration; something you usually want to get rid of but that can be beneficial sometimes. Portraits comes to mind, where both a softer look and extended DOF might be suitable.
To be honest, I can't remember in which part is it mentioned, but this is something I knew since Contax film days, and I was happy to see it confirmed in Dr. Nasse's essay, but please don't ask me to indicate the exact location. It's like looking for a needle in a haystack
Bifurcator wrote:
Don't you have that backwards? Many users here have complained of LoCA and PF with the ZF/ZE/etc. versions. Check out the Zeiss discussion threads. The AEG C/Y version has none - like, at all... I bought two of them and neither one had any.
Denoir is right: the new MP100/2.0 outperforms the old MP100/2.8. I am not sure about the lateral color though.
First, let us agree on some terms. I will use my own abbreviations:
LCA - longitudinal chromatic aberration. To be judged in the plane of (best) focus.
TCA - transverse chromatic aberration. To be judged in the plane of focus.
DCF - defocus color fringing. To be judged in out-of-focus parts of the image.
I tried to obtain signs of chromatic aberration with the MP 100/2.0 and the 5D Mk2 today, but did not succeed. Maybe I need to find more extreme contrasts. However, it is clear that the slightest misfocus may create fringes that resemble chromatic aberration.
I think that most people who complain about 'LoCa' mean DCF. Both lenses behave very similarly in this regard. Here is some background blur:
Straight from DPP with no sharpening whatsoever. The old 100/2.8 performs quite well, considering that this is the extreme corner, but pales in comparison with the new design. Note: There is no apparent TCA with either lens.
LaCA is usually associated with wide angle lenses and affects the corners most. The MP 100 is not wide angle so I would be kinda surprised to find any, Of course if it was there I might not know it being at a 2:1 crop factor with this µ4/3. I've put my copies through some very extreme test conditions and it's just not there.
TCA is associated with asymmetrical designs. Those include retrofocus wideangle lenses and telephoto lenses. Many old telephoto designs (from before manufacturers started to employ low-dispersion glass types) are horrible in this regard.
You sure you're noting thinking of the Japanese versions - either the MMJ or the AEJ ? There are actually some differences between them and the German glass works. Although I read people arguing that point on both sides whenever the topic arises. I've also tested that.
There is no MM version of the MP100/2.8, and the only differences between AEG and AEJ are the inscriptions on the lens and coatings. For what it's worth: the crops above are AEG.
Edit: Bifurcator, I forgot to mention that you should equip the MP100/2.8 with metal hood #5.
Differences between the AEJ and the AEG... Uh, I read officially where the actual glass was manufactured by different foundries (probably with some stock carryover) and probably using different source materials (sand). If you test two of them side by side there are some humanly distinguishable differences.
On the corner samples, thanks!! What camera is that so I know just how extreme those corners are. Thanks! For centers however the two MP100/2.8 s I tried out preformed the ZF 100/2.0 up to f/4. After f/4 there were no distinguishable differences - on the GH1's sensor. This slight superiority became quite evident at the 100/2's MFD for detail in flower pistol, and various surface details.
Thanks again for the info man! it's appreciated!
EDIT: I originally had my wires crossed and said that the C/Y MP-100/2.8 had tested better than the "ZF 100mm f/2.0" lenses. In fact I meant that it had tested better than the older C/Y 100mm f/2.0 lenses. Oops! Sorry for the mix-up guys!
Toothwalker wrote:
I think that most people who complain about 'LoCa' mean DCF. Both lenses behave very similarly in this regard.
They look very similar indeed in your sample, but the C/Y mixes in more yellow in the DCF. In other circumstances (closer to the focal plane), you'll see the C/Y produce blue foreground fringing and yellow in the background. I believe that the Z* 100/2 produces mainly magenta foreground fringing and more cyan/green in the background than yellow/green.
Bifurcator wrote:
Differences between the AEJ and the AEG... Uh, I read officially where the actual glass was manufactured by different foundries (probably with some stock carryover) and probably using different source materials (sand). If you test two of them side by side there are some humanly distinguishable differences.
On the corner samples, thanks!! What camera is that so I know just how extreme those corners are. Thanks! For centers however the two MP100/2.8 s I tried out preformed the ZF 100/2.0 up to f/4. After f/4 there were no distinguishable differences - on the GH1's sensor. This slight superiority became quite evident at the 100/2's MFD for detail in flower pistol, and various surface details. ...Show more →
The camera is the 5D Mk2.
Surely humanly distinguishable differences between G and J occur. Hell, this forum abounds in wildly varying observations and conclusions even for a single lens. The human factor in the tests themselves cannot be ignored.
I did not survey the entire parameter space, but according to my observations the 100/2.0 outperforms the 100/2.8 in the center and in the corner, also at large apertures, at infinity and at 1:2.
AhamB wrote:
They look very similar indeed in your sample, but the C/Y mixes in more yellow in the DCF. In other circumstances (closer to the focal plane), you'll see the C/Y produce blue foreground fringing and yellow in the background. I believe that the Z* 100/2 produces mainly magenta foreground fringing and more cyan/green in the background than yellow/green.
The parameter space is large; the most important factors that influence such fringing are object distance, aperture, and amount of defocus. Generally DCF leaps to the eye when there is only a little defocus. Completely blurred backgrounds will not reveal it. Anyway, I will eat Bifurcator's 100/2.8 if his claim of 'no LoCa' is correct (where I assume he meant DCF).
I compared my C/Y 100/2.8 AEJ to my Tokina 90/2.5 at 1:2 yesterday, shooting some printed text on white paper at an angle of ~30 degrees or so, just out of curiosity to see what kind of defocus CA I'd see. Unfortunately I don't have the MP100/2... Maybe I can post some of the results tonight.
In short: it turns out that the C/Y does have some mild magenta/cyan DCF mixed in with the blue/yellow, whereas the Tokina almost only shows blue/yellow-green. I also compared my Mamiya 645 200/2.8 APO to my Zuiko 200/4.
OK, Here's the bokeh and distance IQ wide open between the two lenses. There are 5 scenes. For the Bokeh comparisons I tried to choose scenes which pose bokeh difficulties for lenses in the 100mm range. I took great care in ensuring that there were no human-error anomalies. A tripod was used for all shots - so no camera shake. I waited for totally still winds or skipped the shot all together - so no motion blur. I used 15x MF assist focusing and spent extra time to make sure the ultimate focus was achieved - so no focus differences were introduced. I focused on the exact same scene elements between lenses - again so there were no focus differences. I also tried to frame and compose the scenes as closely as possible. In order to do this and make up for the difference between 100mm and 125mm I moved the tripod about 1/10 of the focus distance. So if the focus point was 10m like in scene 1, 2, and 3 I moved the zeiss up closer about 1 meter. In the 100% crop samples this should actually give the Zeiss an ever-so-slight advantage but as you'll see it didn't even come close to closing any gaps. In scenes 1 and 4 this difference in camera position added a little bit to the bokeh differences but I took extra shots at the same distances and compared to ensure that what you'll see are dang close to the actual differences had the camera positions remained the same. These scenes were selected as best representations from about 10 total scenes shot. <pant pant>
Each scene has the scaled image from each lens followed by the 100% crops from each with one scene per post (4 total images per post). The 100% Crop areas represent the focus subject in each set. None of these have any PP at all. All sliders were set to zero even including color NR and ect. These are direct RAW conversions. All images were shot on the GH1. Where there's a color difference look to the Lanthar image for the actual colors - this lens nails them every time! And lastly don't forget that the Zeiss is an exceptional lens by itself! Besides the Lanthar it's the sharpest lens in my collection and that includes the Canon 300 f/2.8 L which I love and the 4.0L. Enjoy:
Nope, nothing is wrong with it! It's one of two I purchased - both identical! This is just the C/Y MP 100 2.8 on the GH1 with all of the ACR defaults CHANGED to 0. That's just how it is. And believe me, that's actually very very good for my camera at these settings. It's just that in comparison the Lanthar is extreme and doesn't need the ACR defaults. With the ACR defaults (or LR defaults) applied the Zeiss cleans right up!
Oh Dear, now I know what you meant in your opener, the results from the 125/2.5 are so lovely, and with such smooth bokeh.
Who's also got a Z* MP 100/2 to do a similar comparison with it?
Bifurcator wrote:
Nope, nothing is wrong with it! It's one of two I purchased - both identical! This is just the MP 100 on the GH1 with all of the ACR defaults CHANGED to 0. That's just how it is. And believe me, that's actually very very good for my camera at these settings. It's just that in comparison the Lanthar is extreme and doesn't need the ACR defaults. With the ACR defaults (or LR defaults) applied the Zeiss cleans right up!
You must do something wrong with the Zeiss. The Voigtlander is abolutely not sharper than the Zeiss.
