• 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Forums > Back > In defense of the sh*t lens
#1
I have followed PZ for a number of years but now I feel that something needs to be done with the way lenses are tested.



I read with great interest the performance of the Nikkor 18-35mm first tested the D200 then the D3x and recently on the D700. All using the same sample!



There is no way on earth that this lens is sh*t as some members have described it. I have 25 Nikon lenses and have tested the sh*t lens against:

18-55 II

18-55 VR

18-70 DX

24-85 f/3.5-4.5

24 f2.8 AI-s

28 f2.8 (AFD)

35 f2 AI



The 18-35 is as good or better than every lens above. And it's not just about sharpness. The colors are better!



So, I would like to ask those who actually have this lens what you think about it?



Also, exactly how did PZ get such results at 35mm?



According to Bjorn Rorslett the 18-35 is as good as the 17-35 at 35mm.



If I had 500€ just laying around doing nothing I would seriously think of picking up two of these lenses.
  Reply
#2
I have read many, many posts like this, often by owners of the 18-35, but even if I directly asked some of them, not one offered to send his or her unit, which they all believed performed better than we measured, over here to recheck the lens' performance.



Regarding Bjorn: he tested the lens in the early days of digital photography on both the F5 and the D1. Read: on film and on a 2.7 MP DX sensor. Sorry, but that doesn't tell us anything about what to expect on a 16 MP DX sensor or even FX.



If anyone feels the results we measured don't do the lens justice and he or she owns a significantly better copy than we had available, please be so kind and loan us your unit so we can talk about facts instead of hurt feelings.



-- Markus
Editor
opticallimits.com

  Reply
#3
"I have read many, many posts like this, ..., which they all believed performed better than we measured".



What does this tell you?



I have not read onereport other than PZ giving the 18-35 such a bad review.



PZ's own Lens Performance This is most likely spam content has the 18-35 rated higher than: 18-70, 18-135 and 18-200.



Apart from Bjorn, Thom Hogan rates this lens highly. You probably would point out that TH also tested this lens "only" on the D1 and D100.



How that unmentionable person - Ken R?



You can like him or hate him but he has "real world" comparisons between this lens and a bunch of others.



For example, here:

http://www.kenrockwell.com/nikon/comparisons/24mm/



And it was tested on a d3!



KR even goes further to say "this plastic 18-35mm is actually a little bit sharper on my D3 than the 17-35mm f/2.8 AF-S and the 20-35mm AF-D"



Pure fantasy? Perhaps. But he made the statement based on Real images at Real distances. Correct me if I am wrong, but someone mentioned to me that PZ's resolution testing is based on minimum or near-minimum distances. If this is the case, any conclusions are meaningless.





"Sorry, but that doesn't tell us anything about what to expect on a 16 MP DX sensor or even FX."



I made my tests on a D300.

According to PZ, the 18-35mm @35mm and f4.5 (D7000) gives: 2145(center); 1175(border); 1073(extreme)

vs 18-55 VR at f5 2588(center); 2131(border); 1929(extreme)



This is simply wrong. At 35mm the 18-35 is visibly better wide open on my tests against every lens tested against, including the 35mm f/2 at f4!







"we can talk about facts instead of hurt feelings".

Testing and concluding using one and only one sample does not lead to facts.



Hurt feelings? Are you kidding me?

Look, I would like, and I'm sure we would all like PZ to be a reliable site for information on photographic equipment, in particular, lenses!



"please be so kind and loan us your unit"

Granted, you have some problems with getting various samples.



I can not lend you my copy since it is my most used lens. However, I would like to contribute to PZ by lending some others that I do not use so frequently. For example, my trusty 75-150mm Series E, the outstanding 50-135mm f/3.5 and a few others.



And finally, some photos with the 18-35. It's not about resolutions. It's all about the light!

[Image: pz1.jpg]

[Image: pz2.jpg]

[Image: pz3.jpg]

[Image: pz4.jpg]
  Reply
#4
[quote name='Studor13' timestamp='1320739945' post='12747']

What does this tell you?[/quote]



Among other things, that owners and users tend to justify their purchases. The worse the score here, the more they do. Happened with the AF-D 85/1.4 before (and quite a few others), but that doesn't change any of the scores we measured here.



What does it tell you that there are quite a few responses in threads over at dpr and elsewhere that confirm our findings?



[quote name='Studor13' timestamp='1320739945' post='12747']

I have not read onereport other than PZ giving the 18-35 such a bad review.

[/quote]



Right. And especially no other review where the lens was tested on either the D7000 or the D3x. Most of the other reviews you mention tested the lens on less demanding sensors. Including all of the reviews and reviewers you mentioned. Like Ken, where...



[quote name='Studor13' timestamp='1320739945' post='12747']

it was tested on a d3!

[/quote]



...which has even lower pixel density than a D70. Or D100.



[quote name='Studor13' timestamp='1320739945' post='12747']

Correct me if I am wrong, but someone mentioned to me that PZ's resolution testing is based on minimum or near-minimum distances. If this is the case, any conclusions are meaningless.[/quote]



No, the tests are done at 60x (DX) or 40x (FX) the focal length. This is far from minimum distance. In fact it's closer to infinity than MFD on the lens' distance scale.



But it has become a favourite argument of some who try to impeach our credibility. And usually is never an issue with reviews, where lenses score better than the AF-D 18-35.





[quote name='Studor13' timestamp='1320739945' post='12747']

Testing and concluding using one and only one sample does not lead to facts.

[/quote]



Of course it does. Facts based on one sample, but still facts. You may have well reasoned doubts that this single unit was not representative, but the only valid approach to prove our measurements wrong is to measure a second unit. Which, by the way, we very often do if we have doubts about the results we measure. And very often this is NOT mentioned in the review. Examples: I just finished measurements with a second unit of the AF-S 35/1.4 on the D7000. The upcoming review of the AF-S 105 VR on FX and DX is based on several units. Same applies to reviews of the Sigma 85/1.4, Nikon AF-D 85/1.4 and AF-S 85/1.4. The ZF 25 reviews are based on two units, just as most of the Voigtländer SL II reviews. Measurements of the Nikon AF-S 16-85 VR, Tamron 17-50 and 17-50 VC have been confirmed with additional samples after the reviews had been published. Next week, I'll get a second copy of the AF-S 24-70 to confirm my initial measurements on the D7000 (and I expect some improvement on FX here).



In case of the AF-D 18-35, however, a second unit has simply not been offered to us. And sorry, I'm not going to purchase one just for this purpose.



[quote name='Studor13' timestamp='1320739945' post='12747']

Look, I would like, and I'm sure we would all like PZ to be a reliable site for information on photographic equipment, in particular, lenses![/quote]



Sure. That's why I continue to ask for another loaner.



[quote name='Studor13' timestamp='1320739945' post='12747']

Granted, you have some problems with getting various samples.[/quote]



Usually not. Just in this case. For whatever reason.



[quote name='Studor13' timestamp='1320739945' post='12747']

However, I would like to contribute to PZ by lending some others that I do not use so frequently. For example, my trusty 75-150mm Series E, the outstanding 50-135mm f/3.5 and a few others.[/quote]



Thanks for your offer. However, the main focus is on current lenses for now and we unfortunately have only very limited resources to test discontinued lenses.



[quote name='Studor13' timestamp='1320739945' post='12747']

And finally, some photos with the 18-35. It's not about resolutions. It's all about the light!

[/quote]



Of course. And/or the subject. But not for (objective) reviews. The samples you posted, especially downscaled that much, tell us a lot about your skills as a photographer, but close to nothing about the optical quality of the lens used <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/wink.gif' class='bbc_emoticon' alt='Wink' />



-- Markus
Editor
opticallimits.com

  Reply
#5
Those images looks like they're taken in Norway. Are they?
  Reply
#6
Without going backwards and forwards on everything that you said, I would like to add just a couple of notes.



The D3 is still an outstanding camera however you look at it and if you can't tell the difference between the performance of a lens on the screen, you are not going to see it in print.



The one thing that bothers me about PZ's rating is that there is a bias on resolution to give the 1 to 5 stars.



Whilst resolution is important the true value of a lens is what I call the X-factor. For example, the 180mm f/2.8 AFD lens has it. And so does the 50mm f/1.4 AFS.



Here is where we get back to the 18-35mm. In terms of colors, contrast and the X-factor, this lens has it. According to PZ's testing and more importantly recommendations, the 18-55 II, 18-55 VR as well as the 18-70mm DX are all rated higher.



I have these lenses and I can assure you they are simply not in the same class in giving an X-factor. Nikon is not so stupid to continue to make and sell a lens that costs 700 euros (in Slovenia) if it was sh*t, are they?



As for your comment "but close to nothing about the optical quality of the lens used", this is not true. I can tell that things such as CA, vignetting, contrast and color can be seen by the naked eye. And of course sharpness that PZ is totally stuck on.



I can relate to why owners of the 85mm f/1.4 being annoyed with PZ's findings because from what I understand it has a huge amount of the X-factor.



My suggestion for PZ is to reduce the importance of resolution in the rating, especially the corners. When you print - remember this ancient process - the last few millimeters are framed! No one is going to see it or care.



As for how you would measure the X-factor, well that's an entirely different story.
  Reply
#7
Alexander, no, not in Norway.



It's where I live - in Bohinj, Slovenia.
  Reply
#8
Ok. It reminded me a lot of Norway <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/smile.gif' class='bbc_emoticon' alt='Smile' />
  Reply
#9
[quote name='Studor13' timestamp='1320745607' post='12752']

My suggestion for PZ is to reduce the importance of resolution in the rating, especially the corners. When you print - remember this ancient process - the last few millimeters are framed! No one is going to see it or care.[/quote]



I think the problem here is the classic case of seeing things in a particular manner, and not liking the test since it does not represent that. Photozone have chosen to do things in a particular way. It is useful to many people including me, but it can not be relevant to all scenarios. For example, I never print, so what might or might not be visible in a framed print is irrelevant to me. I do crop hard all over the frame, so the Photozone resolution metrics are useful guidance to me. So if you don't think Photozone tests are relevant to your style, you can look elsewhere for different reviews that better represent what you are looking for. While I wouldn't say no if Photozone tests included more detail in other parameters, what they do now I think is good and useful resource.
<a class="bbc_url" href="http://snowporing.deviantart.com/">dA</a> Canon 7D2, 7D, 5D2, 600D, 450D, 300D IR modified, 1D, EF-S 10-18, 15-85, EF 35/2, 85/1.8, 135/2, 70-300L, 100-400L, MP-E65, Zeiss 2/50, Sigma 150 macro, 120-300/2.8, Samyang 8mm fisheye, Olympus E-P1, Panasonic 20/1.7, Sony HX9V, Fuji X100.
  Reply
#10
[quote name='Studor13' timestamp='1320745607' post='12752']



Whilst resolution is important the true value of a lens is what I call the X-factor. For example, the 180mm f/2.8 AFD lens has it. And so does the 50mm f/1.4 AFS.





[/quote]



Yet no one objectively tested amount of Pixie dust in lenses... PZ resources seems to be devoted for more measurable aspects <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/smile.gif' class='bbc_emoticon' alt='Smile' />.



A.
  Reply


Forum Jump:


Users browsing this thread: 2 Guest(s)