I'm sure holes will be poked in this, but FWIW here are some corner comparisons. Shot with 5D, tripod, MLU, etc. Focus was on the center of the frame (the building), so some of the corner softness may be due to field curvature. Processed through DxO with lens corrections turned off (only matters for the 16-35 mk II.) I can't remember if sharpening was turned off or set to default. Crops are of upper left corner. Left capture is with the 16-35 mk II, right is with the Zeiss 21mm Distagon.
Don't know if this is helpful, but figured I'd post.
-Tim
Hougaard Malan wrote:
I'm still undecided about upgrading from my 16-35mm II to the 18mm or 21mm.
Did any of you guys have the 16-35mm II prior to the 21mm? How does the IQ compare between the two? Specifically corner performance... Any input is appreciated!
Poor, poor little Canon boy.... a law was just proposed in France to forbid spanking of children, but little Mark2 just got spanked at ALL apertures by the mean Zeiss!
Valorin wrote:
Diffraction has little to do with the camera. The only difference is that with higher resolution you perceive the effects of diffraction earlier. However, the diffraction remains the same, f/16 on a 12MP and 24MP sensor will yield the same diffraction-caused softness (in print of course). The reason people complain about diffraction on higher resolution cameras is that they look at their images at 100% on the computer, and mentally compare those results with 100% images from lower resolution cameras. It's not wrong to do that though, it doesn't make a lot of sense having a high resolution sensor where you're not making use of that resolution, unless you have a burning need for that extra DOF....Show more →
Well, I should have been more clear. Ok, you're right going by what I had said, thank you for the correction. What I meant was the diffraction is more apparent with increased pixel density. As far as lenses go, diffraction is a product of any lens. Which is why pinholes look the way they do no matter what camera they're used on. The lower the pixel density the higher the apparent diffraction limit. Make more sense.....?
belsha wrote:
Poor, poor little Canon boy.... a law was just proposed in France to forbid spanking of children, but little Mark2 just got spanked at ALL apertures by the mean Zeiss!
At 21mm........ I would hope so! I don't understand why everyone is making a big deal about this. Guess what, I bet the Zeiss 25 f/2.8 is better at 25mm, the 28 f/2 is better at 28mm, and the 35 f/2 is better at 35mm. The point is, a prime is being compared to a zoom. The advantage of a zoom is that you have all FL's at the same time. The advantage of a prime is it's supposed to be the best at it's stated FL.
Hmm, I have this lens and it's A LOT sharper than this at full resolution, on 5D Mk2. It's possible that you have camera shake or f/16 is suffering from diffraction?
When it comes to ultra-wides, you get enough perspective distortion to get the "sweeping" look for landscapes, and then you can just compose with your feet. I've never understood the ~16-35 zoom lenses for landscapes.
try composing with your feet at the edge of the Grand Canyon. i'll hold your gear while you get on your antigravity platform. shooting with a prime is nearly always a compromise between FOV and what i want. nonetheless, i will use primes when they offer clear advantages over the zoom. in this case, the Zeiss lenses are a lot sharper at working apertures than most of their competition and that has value over compositional flexibility. there are times when compositional flexibility is more important and the 17-35/2.8 comes out even though it is slightly soft in the corners at all useful apertures.
Herb....
thrice wrote:
When it comes to ultra-wides, you get enough perspective distortion to get the "sweeping" look for landscapes, and then you can just compose with your feet. I've never understood the ~16-35 zoom lenses for landscapes.
HerbChong wrote:
try composing with your feet at the edge of the Grand Canyon. i'll hold your gear while you get on your antigravity platform. shooting with a prime is nearly always a compromise between FOV and what i want. nonetheless, i will use primes when they offer clear advantages over the zoom. in this case, the Zeiss lenses are a lot sharper at working apertures than most of their competition and that has value over compositional flexibility. there are times when compositional flexibility is more important and the 17-35/2.8 comes out even though it is slightly soft in the corners at all useful apertures.
Yeah I spose, gotta get that shot quickly lest the canyon disappear.
What I'm saying is I can change lenses in roughly 10s total time... a bag of primes is worth it for landscapers, especially who print big and have to compete with those shooting with medium/large format cameras.
Incidentally my 19/28/35 combo (slr mounts) took up roughly the same amount of room (volume) as a 16-35 II. The downside was weight. I want to avoid bringing up a system argument but all my lenses for RF mount (25/35/50/90) take up the same volume as that lens
tsdevine wrote:
I'm sure holes will be poked in this, but FWIW here are some corner comparisons. Shot with 5D, tripod, MLU, etc. Focus was on the center of the frame (the building), so some of the corner softness may be due to field curvature. Processed through DxO with lens corrections turned off (only matters for the 16-35 mk II.) I can't remember if sharpening was turned off or set to default. Crops are of upper left corner. Left capture is with the 16-35 mk II, right is with the Zeiss 21mm Distagon.
Jeffrey; when you are shooting vistas and subjects with less foreground I suggest you open up a little and stop down no more than f8/f11 - I think you will be pleasantly surprised to see there is more to squeeze out.
trajan wrote:
How is the field curvature on the 16-35? Wouldn't that affect the extreme corner sharpness?
--trajan
Field curvature is pretty significant with my copy of the 16-35 II and it certainly can affect corner sharpness. It's possible to focus for the corners, but depending on the aperture and subject, the center of the image will be soft (back focused).
rscheffler wrote:
Field curvature is pretty significant with my copy of the 16-35 II and it certainly can affect corner sharpness. It's possible to focus for the corners, but depending on the aperture and subject, the center of the image will be soft (back focused).
Ron
Field curvature should only be visible wide open on a wide-angle lens. But here the Canon's images are sub-par up to f11 and truly catastrophic up to f5.6. I guess even at f5.6 the DOF is pretty large on a 21mm lens (it should range from about 1.5m to infinity).
i use the best tool for the job and with restrictions on where one can stand, the lens sometimes has to make up for those restrictions. i know many places where the best place to stand is about 10 ft past the edge of the cliff. either i crop more than i want or i zoom a bit to cut off the branch in the way. the ZF 21 is a very sharp lens compared to many of the mainstream competition but it's not sharper than my ZF 25 at the apertures i work at.
Herb....
thrice wrote:
Yeah I spose, gotta get that shot quickly lest the canyon disappear.
What I'm saying is I can change lenses in roughly 10s total time... a bag of primes is worth it for landscapers, especially who print big and have to compete with those shooting with medium/large format cameras.
dancam wrote:
The point is, a prime is being compared to a zoom. The advantage of a zoom is that you have all FL's at the same time. The advantage of a prime is it's supposed to be the best at it's stated FL.
The point is that the Nikkor 14-24 has been showed to be impeccable, nearly on par with the Distagon, at all f-stops and focal lengths, in the center as on the edges. The Canon 16-35 MkII performance, if this test is true, is just TERRIBLE in the corners. No point whatever to buy a 1Ds MKII or 5DMkII to use it with a lens that bad, all those pixels are clearly wasted.
I pesonnally use a Zuiko 21 3.5. It costs 200$ and fits into my shirt pocket and is better than the Canon (at least on 5D). For me no need to zoom for ultrawide, 21mm is all I need.
That's amazing! I have the 1DsIII and a 16-35II. I've made some very fine images, and stopped down at that. Keep in mine that it is impossible to see the differences on the web, but I don't think this image suffers from Canon glass. It may have been better with the Zeiss, but the differences are small.
Lovely shot, Jeffrey! Positively lovely. Actually, the fact that 100% crops of the 16-35 look blurry probably doesn't matter a bit in real life, where nobody prints 100% crops of corners anyhow.
More important matters IMHO are rendition of colour and contrast, CA, distortion. All of which only if they are "bad" enough that they can be seen on a print.
