Let's complain about lenses!

2»

Comments

  • WestEndFotoWestEndFoto Posts: 3,742Member
    I don't think that the distortion on Nikon lenses is an issue.

    People complain because after correction, it is potentially less sharp, which is true.

    But the end result is that these lenses are proving to be sharp out to the edges and even the corners. Whatever problems distortion is causing is more than offset with something else. The only competing mount is the new Canon mount. Not the old Canon, Nikon or the new Sony.

    Would you guys rather get to "10" by adding 21 and subtracting 11. Or would you rather get to "7" by adding 11 and subtracting 4.

    I don't know about you guys, but I would rather get to 10. I am not going to choose "7" just because 4 is being subtracted and not 11.
  • ggbutcherggbutcher Posts: 390Member
    I don't use LR, but I've read enough from folks I'd trust to know that Adobe has indeed hard-coded the application of the Nikon Z lens-provided corrections in their raw processing softwares. The information is in the raw metadata, in MakerNote fields that have not yet been reverse-engineered by others. You can get at the data by using the Adobe DNG Converter to make DNGs from the NEFs, as the DNG format is well-documented. I'd surmise that Nikon and Adobe struck some sort of agreement: access to the documentation to use the lens-provided correction data if it is hard-coded into the processing of Nikon Z raws.

    If you want uncorrected images, probably need to use another raw processor. I know the open-source alternatives, RawTherapee, darktable, PhotoFlow, don't apply lens corrections unless you turn it on, and instead of the lens-provided data they use lensfun data (lensfun is an open source lens correction programming library). I use raw processing software I wrote, it doesn't do lens correction by default, and I've observed for the S 24-70 f4 the difference between corrected and uncorrected distortion is marked.
  • PistnbrokePistnbroke Posts: 2,443Member
    Well the Z 24-70 f4 only use 19mp on a Z7...hopeless God help the Z6 with 24MP and a LPF
  • tc88tc88 Posts: 537Member
    edited August 2020

    Would you guys rather get to "10" by adding 21 and subtracting 11. Or would you rather get to "7" by adding 11 and subtracting 4.

    I don't know about you guys, but I would rather get to 10. I am not going to choose "7" just because 4 is being subtracted and not 11.

    Well, that's assuming the numbers you quoted make sense. Pistnbroke piqued my interest in looking more at the DXO numbers and the end numbers for Z mount lens are bad. In fact it's quite shocking. Maybe it's because Sony's A7r4 is 60mp, but still ...

    I think I read a comment that the reason the Z lenses do so bad on DXO is because DXO have turned off the corrections.

    I can see it's difficult to differentiate wavey lines. Still if the correction is applied, I doubt those lines can be made clear since you can't create information out of no information. Maybe it's an artifact of the benchmark. But until there is better way of measuring the lens, the current methodology seems to be considered the most objective way of measuring.
    Post edited by tc88 on
  • WestEndFotoWestEndFoto Posts: 3,742Member
    I don't really buy into DXO. Frankly, I think the best guide is Nikon's MTF charts. But remember, the MTF charts are not comparable across manufacturers.

    I also like the Imatest system. A few years ago I looked into buying it until I determined that it needed an upgrade to test sensors with more than about 45mp of resolution. I may re-connect with them and see if they have fixed that.

    Photography Life uses Imatest. You will notice a disparity between the Imatest results and the MTF charts for non-centre sharpness. That is because Photography Life is not correcting for field curvature. If you focus in the mid-frame, real world results will look more like Nikon's MTF charts than Photography Life's implementation of Imatest.

    Thom also uses Imatest, but also does the work to extensively test in the field. I like Thom's reviews because of his approach.

    So based on my research and the research of others that I trust (Thom, Roger Circala, Photography Life, Steve Perry), I am pretty confident that the sharpness across the frame for Nikon's S line leaves nothing to complain about.
  • tc88tc88 Posts: 537Member
    edited August 2020

    I think the best guide is Nikon's MTF charts.

    I also like the Imatest system.

    That's good point. MTF is very important info.

    Now regarding lmatest, I thought DXO's method is pretty much equivalent to Imatest in theory? In fact, I thought all of them just measure how many lines they can differentiate? Someone can probably enlighten me.
    Post edited by tc88 on
  • mhedgesmhedges Posts: 2,881Member
    edited August 2020

    Well the Z 24-70 f4 only use 19mp on a Z7...hopeless God help the Z6 with 24MP and a LPF

    Sorry but I just don't buy that at all. I think you are reading way too much into the DXO tests. If that was true then people would be up in arms about it, and from everything I have seen the reaction to that lens (and just about all the Z lenses) has been overwhelmingly positive.

    Did the F mount 24-70 2.8's only give 19 mp on D850? Because the Z 24-70 F4 is better than those lenses from everything I have read.
    Post edited by mhedges on
  • WestEndFotoWestEndFoto Posts: 3,742Member
    I have never been able to get a truly transparent explanation on what DXO does. This disqualifies it in my view.

    Imatest is similar to an MTF chart as it measures contrast.

    Imatest is measuring a system though as it relies on a camera sensor. As long as your sensor exceeds the sharpness of your lens, you willl get a useful result. As I mentioned before, however, that was an issue and may still be an issue.

    The gold standard is an optical bench which does not rely on a camera sensor. Roger Circala has one so I pay attention to his results. All the lens manufacturers will also have one, or maybe several, to among other uses, produce their MTF charts.
  • mhedgesmhedges Posts: 2,881Member
    Very true but I'm not sure it is possible to use optical benches with modern focus by wire electronic aperture lenses. How do you focus and set the aperture?
  • WestEndFotoWestEndFoto Posts: 3,742Member
    That is beyond my ken but I am sure there is a way.
  • PistnbrokePistnbroke Posts: 2,443Member
    I use DXO for comparison rather than definitive figures. Finding a database will as many lenses as DXO is difficult or impossible
  • KnockKnockKnockKnock Posts: 398Member
    Hey, you can change the thread title? :-P

    nice
    D7100, D60, 35mm f/1.8 DX, 50mm f/1.4, 18-105mm DX, 18-55mm VR II, Sony RX-100 ii
  • Ton14Ton14 Posts: 698Member
    edited August 2020
    @KnockKnock or back on topic.

    Well, the 24-70mm f/2.8 S has a number of 26 M-pix. This Perceptual Megapixel quantifies the sharpness and replaces the MTF scores. This lens has achieved the highest score and it is the best 24-70mm lens DxO have tested to date !!.

    Compared with the Nikon AF-S NIKKOR 24-70mm f/2.8E ED VR on the Nikon D850 it is better. Even this Nikon AF-S NIKKOR 24-70mm f/2.8E ED VR on the Nikon D850 compared with the NIKKOR 24-70mm f/4 S on the Nikon Z7, the S lens is better and in real life it is outstanding on the Nikon Z6.

    Nothing to complain at all that I start to build my Nikon Z system about hardly 2 years ago.
    Post edited by Ton14 on
    User Ton changed to Ton14, Google sign in did not work anymore
  • photobunnyphotobunny Posts: 646Member

    Hey, you can change the thread title? :-P

    nice

    Yup, I think we got one post till the topic changed so I kinda just went with the flow.
  • photobunnyphotobunny Posts: 646Member
    Here's a wee fuss: What on Earth are Nikon doing with their Z lenses. Other than the 85 S, I have had nothing new to buy for months on end. The 70-200 is nice and all, but where's the 100-400?

    20-85 are well covered by 5 primes, 2-4 zooms. But past that we need the wart.
Sign In or Register to comment.