Cyra, check out the Zeiss thread where Philippe has posted some nice comparison shots with 25/2,35/1.4 and 21. He has pronounced the 25/2 as his most favorite ZE.
I just looked again at Lloyd Chamber's tiled roof shots with the 25/2 and 24G and Leica M's and I see the same thing happening in my mid day shots and night time shots with the 25/2. Basically, the lens seems to have higher global contrast in this kind of light where there are bright highlights. If you meter to not blow out the highlights but expose them to the right, you see the foliage ending up darker than reality and why I always noticed I needed to use the fill slider in LR to lighten up the shadows and even lighter darks.
You need contrasty light like midday sunshine or dark street with street lamps type of lighting to see this happening. The 25/2 seemed more contrasty than my 21 on the landscape shots. Just wonder what others see.
Wayne, consider this: Even with the less contrasty lenses that seem to brighten dark areas (shall we call it "micro flare?") to be within the sensor's dynamic range you are still losing the same amount of detail due to reduced contrast in these ares. Either you lose detail to the slight veiling-haze, or to the limits of DR. At least with Zeiss, there exists the possibility that future sensors with more DR will be able to take advantage of its capabilities.
Dan, there is no veiling haze in my ZE 21 shots, just less global contrast. At least that is what it looks to me and I was just wanting other opinions on whether they see that too or I am just seeing things. Whether a lens is more contrasty without veiling haze then it depends on the lighting situation whether it helps or hurts.
When it comes to the ZE21 I have appreciated how it tames down the highlights of a southwest canyon shot so that there is more color and contrast and not just blowing out like most Canon lenses do in the same situation.Yes, the darks were a little darker but not like this 25/2 which also seems to brighten the highlights.
This reminds of the difference between the Canon 85L and the N 85/1.4 which John Black mentioned in his review of the N 85. That in outdoor shooting the 85L seems to always want to blow out the sky and thus to save the sky you expose at a darker setting. Whereas the N 85 doesn't as is so much better to use in outdoor shooting settings.
I assume its possible for lens makers to add different global contrast curves to their lenses via their special lens coatings. Is that right?
wayne seltzer wrote:
Dan, there is no veiling haze in my ZE 21 shots, just less global contrast.
I don't think veiling haze is a good description of randomly scattered light that origins from relatively large bright areas and scatters widely. It looks more like a blend with a layer of very light gray with added noise. You can try adding a layer in photoshop and fill it with a gray level around 240( add some noise to emulate the photon shot noise in the scattered light). Adjust the opacity of the layer around 0 - 1 - 2 -3 percent and observe that it affects global contrast, but not like a veiling haze.
Regarding lenses with "natural contrast", in my opinion that is only a matter of certain tone curves adjusted to a certain dynamic range of an average monitor, and suited to our habit of what we expect too see of contrast in an image. In my opinion all images are very coarse approximations trying to map the reality into something presentable. During history, we have developed tone curves and habits suited to certain output media and certain recorded contrast levels. Recording a different contrast level only needs a different tone curve to get the same perceived contrast in the presentation.
Alundeb and Toothwalker,
Do you see contrast differences between my ZE 21 landscape shot and the 25/2 as far as global contrast?
Unfortunately, the 25/2 was shot f8 at 1/200 and the ZE 21 was shot at f8 1/320. Both were processed with WB set at Daylight. How would compare the contrasts? I just tried slightly lightening the 25/2 shot so that its midtones matched the luminosity of the 21 shot. If I then look at the RGB values for the white cars in the distance they are slightly higher on the 25/2 shot and with slight more CA. The green bush in the foreground in the 25/2 shot is darker than the 21. The 25/2 shot with daylight WB seems to render the dead grasses with too much blue and not as warm and nice as the 21.What differences do you see between the 25/2 and 21?
Or do you think they look the same?
Wayne, for a comparison of global contrast, I would correct for vignetting and exposure differences first, even as small as < 0.1 Ev.
I would prefer to judge from images of the full scene after this sequence of processing:
1) Correct for exposure (use central, uniform grey/white area)
2) Correct for vignetting
3) Crop the 21mm image to the FOV of the 25mm image.
4) Resize both images to the same pixel dimensions.
wayne seltzer wrote:
Alundeb and Toothwalker,
Do you see contrast differences between my ZE 21 landscape shot and the 25/2 as far as global contrast?
Unfortunately, the 25/2 was shot f8 at 1/200 and the ZE 21 was shot at f8 1/320. Both were processed with WB set at Daylight. How would compare the contrasts? I just tried slightly lightening the 25/2 shot so that its midtones matched the luminosity of the 21 shot. If I then look at the RGB values for the white cars in the distance they are slightly higher on the 25/2 shot and with slight more CA. The green bush in the foreground in the 25/2 shot is darker than the 21. The 25/2 shot with daylight WB seems to render the dead grasses with too much blue and not as warm and nice as the 21.What differences do you see between the 25/2 and 21?
Or do you think they look the same?...Show more →
It is difficult to say. The 25/2 shot is darker, on average. If I equalize the mean pixel values of the two images, they compare like this:
The transmission characteristics of a lens are independent of light intensity, but there can be a wavelength dependence resulting in a different color balance. In that case one could see subtle effects resembling a global contrast issue if the bright and dark areas have different colors. However, if that is the case the balance can shift in favor of the other lens with other subjects.
Toothwalker;
With many raw-converters (and also built-in jpg engines in the camera) you get an "adaptive blackpoint" setting. With Nikons you unfortunately get it for "free" whether you want it or not, since the firmware scales the blackpoint to clipping (within a certain limit of course).
This can in many cases give the unfortunate (and very counter-intuitive!) effect that a lens that has LOWER global contrast is perceived as having HIGHER global contrast - since the midtones is what leaves the biggest visual impression in an image. If a lens has mroe veiling flare, the camera/converters drags that blackpoint down to "zero" - and leaves you with an over-inflated midtone contrast.
Alundeb:
Veiling flare is indeed the "wide-area" diffuse/random scatter in a lens. I think it's a good denomer, since it says exactly what is is: a "veil" of light that covers most of the image frame.
In PS you can mimic this by adding a layer of very dark gray (35 in 0-255 gamma2.2?), add mild noise to that layer, and set opacity to "screen". On top of this you put an adjustment layer, either levels or curves, and drag the BP back down. The (re)distribution of noise and contrast in the shadows is actually pretty interesting...
Even more interesting: make the gray layer tinted. Fill the gray layer with 40 blue, 20 green and 30 red, and see what happens when you pull the blackpoints down again. You get that trademark "old Sigma color".
Here you have the "colour" of a lens - it has NOTHING to do with the colour you see when looking in through the lens. If you take f.ex the yellowish colour Sigma was (is still, wrongly) known for, then you can actually model their AR layers and get the "sigma colour" of old (I don't see it in the newer lenses) from almost any good lens... If you WB for the bright parts of the image, and pull the blackpoints of each channel down to black again, you get yellowish lower midtones.
The clarity of Zeiss comes from a very wide-band AR coating, and careful attention to scatter performance. They have less blue scatter and veil than almost anything out there.
theSuede wrote:
Veiling flare is indeed the "wide-area" diffuse/random scatter in a lens. I think it's a good denomer, since it says exactly what is is: a "veil" of light that covers most of the image frame.
.
Technically this is of course something I accept. But since it does not look like flare until the amount is very large or you inspect the shadows very closely, it does not effectively describe what we see in slightly reduced contrast. Like "No, it is not veiling haze I see. The shadows are just a litte less dark". Since it is difficult to "sell" the explanation with that description, I try to walk around it.
And to be precise: you said "veilig flare", and Toothwalker said "veiling glare". The term i mentioned was "veiling haze". That one is often associated with heavy SA. When "veiling haze" is mentioned, I think may people on this board immediately get a picture of a wide aperture Planar at close distance.
OT: The tinted scatter as "lens color" is very interesting. It should easliy translate into a succesful recipe for correcting "lens color" or even (WB amplification tinted) read noise color.
Phillipe, good to see you got your lens after all. Congratulations! So, how does it compare to the 35/1.4 and 21/2.8? Rendering, colours, contrast... is it a worthy lens in between these two?
thanks for the extended test, German. The inifinity corners look good in your images. But: the last tip of the corner can't be judged, because there is sky and no structure.
It would be good to place the buildings further into the corner, so that structure is covering the whole corner.
also in some images the magnification does not show the exact close corner, but a part slightly up or into the image, where there is no problem anyway. People here are talking about softness only in the last mm right in the corner. It shows in your sample 3 (100% of the ground cover) and 6 (100% of the ground with leave).
In the "Ampliación 20-2-2.012" only for curiosity I repeat the same shots with a Canon 24/1, 4 L II that I'm reviewing. The sample was the same, a video camera that is on the top roof, I waited at the same time and tried to make a similar frame. (24 mm. vs 25 mm.)
As can be seen in 100% crops the center is the same, in the half corner (near the APSC) the Zeiss is sharper and in the extreme corner (no angle) the Canon has more clearly. In the absence of the all review process of the Canon I can not speak yet but the transmission of color is significantly different. I put the two full files for download for which it is interested.
In this momen I just put the new Samyang 25/1, 4 on the 5D II, will see "as sing the Korea lens"
"cyra"
In the crop that you are post is crearly that the extrema angle is blurred but must be evaluated that this crop is amplified +/- 25% from the original 100%. I think that on a image at 100% the deficiency is more low.
G. Pierre -- Are you saying that you are testing the ZE 25/2.0 and EF 24/1.4L II on APS-C sensor?
Isn't the concern about the 25's outer edges and corners on full frame imaging? The concensus already seems to be that the Zeiss is very sharp in the central area, which is what the APS-C covers.
Looking forward to your examples comparing the Samyang, Zeiss and Canon. But I would prefer to see full frame examples.
Gunzorro wrote:
G. Pierre -- Are you saying that you are testing the ZE 25/2.0 and EF 24/1.4L II on APS-C sensor?
Isn't the concern about the 25's outer edges and corners on full frame imaging? The concensus already seems to be that the Zeiss is very sharp in the central area, which is what the APS-C covers.
Looking forward to your examples comparing the Samyang, Zeiss and Canon. But I would prefer to see full frame examples.
I did not say that I've done the test on APSc sensor camera, I have referred to the area of the frame near of the corner APSc equivalent. If you look at the top of the page indicates the camera model.
Toothwalker wrote:
It is difficult to say. The 25/2 shot is darker, on average. If I equalize the mean pixel values of the two images, they compare like this:
The transmission characteristics of a lens are independent of light intensity, but there can be a wavelength dependence resulting in a different color balance. In that case one could see subtle effects resembling a global contrast issue if the bright and dark areas have different colors. However, if that is the case the balance can shift in favor of the other lens with other subjects.
Thanks you Toothwalker, Alundeb, and the Suede for all your interesting information above. I wanted to respond quicker than this but I had a personal tragedy last week of having to put our well loved 12 yr black lab down due to unsuccessful surgery to untwist his intestine which had become twisted and bloated.(GSV). I did not know that large chested dogs and cows can have this GSV happen to them. It happened so suddenly and I will really miss him.
When I tried to equalize the exposures between the two shots (ZE 21 and ZF 25/2) I got something similar to above and still notice that white cars in the 25 shot look more bright and harsh than the 21 shot and then some of the bushes and trees look slightly darker. Also, the color of the grass in the foreground of the 25/2 is rendered with a more bluish tone and doesn't look as natural as the 21 color rendering. This was brought up by Philip_PJ earlier. It is as if this lens really amplifies the blue channel in things more than it should some times.
I was hoping those that have the 25/2 would do some more color comparisons with 25/2 and 21 or other zeiss lenses.
... and sold the 21/2.8 in an instant :-o 
