Review: PhotoZone - 35L II on 5DS R

[expatinasia]

I'd hate to see what forums would have looked like in the 90's.

It was not that long ago, he says..... Basically, they were the same, just much, much, much quieter. People in the whole had still not grasped the online forum thing and the few that were around were much quieter. Most from that day have closed down, as many webmasters gave up.

 

[hendrik-sg]

in My opinion Lenstip is much better, they are pysicians and have a look at many more possible aberrations than photozone. What i do not like at photozone as well is, that they jump the camera body without notifications. The 11-24 is tested on 5Ds and the 16-35 4.0 on 5Diii, so no comparision is possible. Lenstip at least are giving recommendations, how to compare between the bodies, even between the brands.

Next point is that there should be tested with raw files, as to measure lens performance, we do not want to have sharpening algorithms incuded in the test, especially as they can simulate (by guessing information which may be there or not, and wih may appear as noise or artefacts) a higer resolution which is not valid, even beyond the diffraction limit. In such aspects i would better trust lenstip.

Sure, lensrentals are the only testers who have access to pools of lenses and can measure sample variation, but they do resolution tests only and give no informaton about other aberrations, as lenstip do.

BUT: the best is to combine the statistic information of lensrentals with the detailed examination of lenstip

 

[Larsskv]

Gotta go with Dilbert on this one, he is annoying and irritating and often seems wrong, he wasn't rude or being a dick.

+1.

I agree with Dilbert this time. I really do appreciate photozone.de, and I have always thought of them as objective and fair. Shure, they're tests are not perfect, but we can't expect everybody to test 10 copies and every lens on multiple bodies.

Dpreview and DXO on the other hand... lets just say that I don't take advise from anyone I in general disagree with.

 

[Loswr]

You are right, I was wrong at including "Lens Tip" there - I did that because someone else had suggested the Lens Tip was as good as Lens Rentals (may fault for assuming they were right - silly me!) however Lens Tips is a lot similar to Photozone as they only use 1 lens too:

Even when admitting you were wrong trying to excuse/justify your mistake, you are wrong about what you were wrong about. Pathetic.

 

[chromophore]

in My opinion Lenstip is much better, they are pysicians and have a look at many more possible aberrations than photozone. What i do not like at photozone as well is, that they jump the camera body without notifications. The 11-24 is tested on 5Ds and the 16-35 4.0 on 5Diii, so no comparision is possible. Lenstip at least are giving recommendations, how to compare between the bodies, even between the brands.

Next point is that there should be tested with raw files, as to measure lens performance, we do not want to have sharpening algorithms incuded in the test, especially as they can simulate (by guessing information which may be there or not, and wih may appear as noise or artefacts) a higer resolution which is not valid, even beyond the diffraction limit. In such aspects i would better trust lenstip.

Sure, lensrentals are the only testers who have access to pools of lenses and can measure sample variation, but they do resolution tests only and give no informaton about other aberrations, as lenstip do.

BUT: the best is to combine the statistic information of lensrentals with the detailed examination of lenstip

But LensRentals optical bench testing can show you things about the lens aberrations that you can't necessarily see in a resolution chart, too. Astigmatism is one clear example. The MTF chart also shows how the lens behaves as a function of spatial frequency, and again how these vary with image height, which is not always how Imatest results are reported.

People such as "dilbert" who criticize MTF charts at infinity focus as not being representative of how a lens performs on a camera body in real-world conditions only reveal that they don't understand how to read MTF charts. That's not to say that optical bench testing is the end-all and be-all of how a lens performs, no; not by any means. If I want to understand the nature and extent of color aberrations, for example, I will look at other sources of information. But not all review sites have information that I trust, because sometimes what I see being reported doesn't make sense when I gather all the information together.

The bottom line is that it's all splitting hairs anyway. Virtually all lenses produced today are excellent, and people should be happy with their performance. I would be more concerned about a lens that fails to focus consistently if it offers AF, or a lens that is easily damaged or not put together well, because that's an issue of value. But these have very little to do with whether a $5000 Zeiss lens or a $700 Sigma lens or a $1700 Canon lens are of equivalent optical performance. I just wish photographers would not use review sites as justification that one lens is "absolutely" better than another.

 

[Bdube]

in My opinion Lenstip is much better, they are pysicians and have a look at many more possible aberrations than photozone. What i do not like at photozone as well is, that they jump the camera body without notifications. The 11-24 is tested on 5Ds and the 16-35 4.0 on 5Diii, so no comparision is possible. Lenstip at least are giving recommendations, how to compare between the bodies, even between the brands.

Next point is that there should be tested with raw files, as to measure lens performance, we do not want to have sharpening algorithms incuded in the test, especially as they can simulate (by guessing information which may be there or not, and wih may appear as noise or artefacts) a higer resolution which is not valid, even beyond the diffraction limit. In such aspects i would better trust lenstip.

Sure, lensrentals are the only testers who have access to pools of lenses and can measure sample variation, but they do resolution tests only and give no informaton about other aberrations, as lenstip do.

BUT: the best is to combine the statistic information of lensrentals with the detailed examination of lenstip

But LensRentals optical bench testing can show you things about the lens aberrations that you can't necessarily see in a resolution chart, too. Astigmatism is one clear example. The MTF chart also shows how the lens behaves as a function of spatial frequency, and again how these vary with image height, which is not always how Imatest results are reported.

People such as "dilbert" who criticize MTF charts at infinity focus as not being representative of how a lens performs on a camera body in real-world conditions only reveal that they don't understand how to read MTF charts. That's not to say that optical bench testing is the end-all and be-all of how a lens performs, no; not by any means. If I want to understand the nature and extent of color aberrations, for example, I will look at other sources of information. But not all review sites have information that I trust, because sometimes what I see being reported doesn't make sense when I gather all the information together.

The bottom line is that it's all splitting hairs anyway. Virtually all lenses produced today are excellent, and people should be happy with their performance. I would be more concerned about a lens that fails to focus consistently if it offers AF, or a lens that is easily damaged or not put together well, because that's an issue of value. But these have very little to do with whether a $5000 Zeiss lens or a $700 Sigma lens or a $1700 Canon lens are of equivalent optical performance. I just wish photographers would not use review sites as justification that one lens is "absolutely" better than another.

You can't discern astigmatism from a large aperture plot of MTF vs spatial frequency. In fact, I would say it is quite the opposite. You also can't do it from small aperture MTF charts unless you know about the lateral color.

Regards,
Brandon
Olaf Optical Testing, Photozone.de

 

[Loswr]

MTF charts represent theoretical performance. They provide a guide as to what sort of performance to expect from a lens but you need testing on a body to establish actual performance (e.g. what everyone with imatest does.)

If someone gives you a dollar, will you go buy a clue? :

Granted, it's probably my fault – I wouldn't expect a bowling ball to grasp the difference between the theoretical MTF charts published by most lens manufacturers and empirically determined MTFs measured by LensRentals (and by Zeiss for their own lenses), so why should you understand it?

 

[Mt Spokane Photography]

...
Not true, Lens Rentals has been doing lens testing for some time now with an optical bench. The optical bench takes measurements at infinity focus. All of those fancy MTF curves they've been showing for the last two years or so have been taken at infinity focus.
...

I was waiting for someone to post that.

It seems you are doing selective reading (from the Lens Rentals page):

The weaknesses of the optical bench tend to be almost the opposite of the weaknesses of Computerized Target Testing.

For testing purposes, I'll be using both Imatest and an optical bench.

For macro lenses, physically large lenses, and lenses with very long focal lengths, the optical bench is not particularly useful and Imatest will continue to be our test of choice.

The bottom line, though, is we should read lab tests realizing what is being tested.

i.e. there is no single test of lens that is better than anything else, i.e.

It's all trade offs.

When you see tests from Photozone and others using a 5DsR to test a lens, you're getting an idea of how that lens performs on that camera. When you see Lens Rentals test a lens using an optical bench you're seeing test results that have no bearing on actual shooting with a camera.

Different folks might see value in different approaches but neither is "a one test has all the answers."

I try to read lots of reviews because you do get different viewpoints. However, it is going to be misleading to compare results of lens tests on different cameras. The lens testers warn of this, but some persist.

 

[Mt Spokane Photography]

You can't discern astigmatism from a large aperture plot of MTF vs spatial frequency. In fact, I would say it is quite the opposite. You also can't do it from small aperture MTF charts unless you know about the lateral color.

Regards,
Brandon
Olaf Optical Testing, Photozone.de

Brandon, tell us more about how to discern astigmatism. I know Roger has some good information, but it has long since rolled of my memory chart, which is not all that long anymore.

 

[Viggo]

For me the deciding factor is which lens will give the most keepers with the best IQ, and after owning 5 Art lenses and now the 35 L II there is absolutely NO contest whatsoever. The 35 L II is just out of this world. It's been proven in tests also, better sharpness and contrast, better colors , less ca and no distortion. And it delivers with spot on AF every time. These are facts.

 

[Bdube]

You can't discern astigmatism from a large aperture plot of MTF vs spatial frequency. In fact, I would say it is quite the opposite. You also can't do it from small aperture MTF charts unless you know about the lateral color.

Regards,
Brandon
Olaf Optical Testing, Photozone.de

Brandon, tell us more about how to discern astigmatism. I know Roger has some good information, but it has long since rolled of my memory chart, which is not all that long anymore.

If you have a lens you know is very well corrected for lateral color (e.g. Zeiss 100/2 makro-planar), you make consider a difference in the tangential and sagittal MTF at closed apertures - perhaps f/8 - to be the astigmatism. At larger apertures, aberrations such as oblique spherical aberration and misalignment-related coma will cloud the view.

If you have a lens - perhaps a 24mm f/1.4, which is much less well corrected for astigmatism, you will have no such luck at any aperture.

However, one may also use monochromatic measurements of the MTF, which removes any lateral color, to tell the astigmatism as well.