My Nikon 14-24 (adapter): a bust

So, perhaps Canon isn't the only one capable of producing bad copies, though I have yet to see any mention of any Nikon 14-24 problems. Leave it to me! I am using this on my Canon 5D2. My first generation Canon 16-35 performed (past tense--water situation) better at the corners (at 16mm). I've seen one copy of 16-35 Mark II, and it is just as hideous as the first generation to 20mm. I'm not sure why I've seen better reviews of the Mark II, but I realize I've only seen one. At the same time, Canon's horrible track record speaks for itself. Anyway, moving onto my 14-24. All pictures below are raw images straight out of the camera.
First up, I think I've read about some natural polarization tendencies lenses have around 14mm. I want to confirm if this phenomenon is to be expected, of if it could be a lens flaw. If you can look past this sweet shot , the far left edge is essentially west, and due north is directly over the peak on the right. How the heck can the left edge be brighter? I could understand the far right being lighter, then fading to dark, but not back to light. Note that I get the same effect no matter which direction I center the camera towards. Perhaps more curiously is that Lightroom automatically adjusted this on all but these sky shots.
Exhibit B: This is looking up at a portion of the sky that would be above where the last image left off. This is at 16mm, though 24mm produced similar results.
Same section of sky using my friend's 16-35 Mark II. Virtually identical effect, so I'm guessing this is some kind of phenomenon that can be explained. Both of our lenses on his 1Ds2 had the same effects.
Full scene at 14mm for reference for below shots
100% crop of center section. This was somewhere between f/11 and f/14 with emphasis on the background. Sharp enough; note the details in the smaller talus on the far slope.
Extreme right-hand section. Yes, the left side is just as bad.

I'll be sending it back to Nikon after the fall season for tuning this brick. Thoughts?

I'd suggest trying it on a Nikon full frame camera if at all possible, to rule out issues with your adapter. You'll probably need to give Nikon a sample in any case and if you give them a Canon file...........

I agree with Chris. Test the lens without the adapter. Your images don't look like they were made on anything like the 14-24mm I have.

--Gib

Wow... there's something really wrong. Mine shines!
By all means try it on a Nikon body.

Maybe I'm wrong but an adapter works "like a tiny extension tube". Could be it?

But having more or less the same effect with two different lenses on two different bodies is at least od!

Thanks, gents. I can't imagine the adapter having any effect as it is just a hole and shouldn't affect image quality. Still, you're right; It's definitely worth a shot to definitively rule out. I'll have to go check out a Nikon body at a local camera store.

I would still be curious to hear any insight on the polarization issue.

If you look at the 16:9 site which reviews adapted lenses they often refer to the quality of adapters used. You think it would be a simple machining job, but it does need to be very precisely dimensioned to hold the lens at the correct distance from the sensor as well as perpendicular and centred, plus the machining to match the bayonet mounting of the body and the lens has to hold it nice and firm so the lens can't move excessively.

See here:

http://www.16-9.net/lens_tests/compatible.html

in particular this paragraph:

"It cannot be stressed too strongly that all the arguments in favour of adapted lenses are nullified by using a badly matched adaptor (or adapter). Symptoms of bad matching include blurring in one half of the image circle, atypical colour fringing, failure to acquire infinity focus, and overall resolution loss. The jury is still out on whether adaptors of different thicknesses should be used for lenses in the same system with different focal lengths – for instance, demanding a thinner Contax-mount adaptor for the 15mm Distagon than the 35mm Distagon. Sample variation affects adaptors, too. If you're suffering any of the above symptoms, try a different lens on the same adaptor – then a different adaptor on the same lens. When you find a match that works, stick with it!"

I think your corner problems are a bad adapter. Every lens is designed for an optimal distance between the rear of the lens and the focal plane. They will work if this is wrong, but will become progressively worse around the periphery of the image. I used to run into this sort of problem all the time with adapters for my astronomical CCD cameras; I had to have a different adapter for each telescope and one for each lens type (Canon EF, Nikon F, etc.). The precision required for optimal performance can be on the order of 1/10 of a millimeter.

To check any lens it should never have a filter of any type or adapter. I think if you would put that on a Nikon body it would have a totally different out come.

To repair that lens, go buy a Nikon D300, D700, D3 or D3x body.

Hello,

Not adding much to what others have said, but I don't completely trust the adapters out there.

There was a very trusted member at the old RG forums (who now co-runs the forums at getdpi.com) who had a very sharp copy of the Oly 21/3.5. I'd seen his samples - they were great. I purchased his lens and could not get results even close to his with 3 different adapters.

These adapters need to be exactly spot on - if they are not the benefits of a good lens are lost.

That being said - I am about to list a d3x adapter for this lens on the b/s board. Give me a shout if your are interested......

Oh, on the polarization: The sky is naturally polarized. Its most easily visible at high altitudes, and it frequently shows up in wide-angle shots like this.

Agree with the others. I think it's most likely an adapter problem.

Thank you all for the feedback. It has been valuable. I went to a store today and tried it out on a D700. A D3X would've been a better test with my application for comparison, but they didn't have any in stock. I'll go back when they do to try one. I took some indoor and outdoor shots at each main focal length and f/11-22. While the results don't look nearly as bad as the shots as I've taken using my camera, I wasn't blown away with edge sharpness, either, though to be honest, I may not have had it focused exactly right. I was a little out of my element shooting inside a camera store and outside looking out the storefront to a parking lot, city street and trees. Perhaps on my upcoming fall trip I'll run into someone with an FX body. I suspect that you're all probably right being an adapter issue. I just wasn't expecting it to affect corner sharpness only.

DavidRamey wrote:To repair that lens, go buy a Nikon D300, D700, D3 or D3x body.

That's not a terrible suggestion! I've seen it before, but man, the font for the shutter speed (mainly) Nikon uses is crazy! I can hardly read it. I guess you get used to it and all the other differences, though. I have always liked the sounds of Nikon shutters!

jhapeman wrote:Oh, on the polarization: The sky is naturally polarized. Its most easily visible at high altitudes, and it frequently shows up in wide-angle shots like this.

Thank you very much for this. I have just never seen this effect with my Canon 16-35 at 16mm in the past. This is reassuring to hear this.

Steve Fines wrote:I am about to list a d3x adapter for this lens on the b/s board. Give me a shout if your are interested......

Undoubtedly the best adapter for this lens! However, I'm not fond of the D3X or 1Ds3 for landscapes, at least as it relates to hiking/backpacking with them. The things are beasts. To me, the 5D2 and D700 are the ultimate landscape tools.

Thanks again, everyone!

Darren, at f8 and up that lens start to loose it's sharpness. So don't expect stellar results at f11/22.
Not beying an MTF addicted these graphs explain what I'm saying:
http://www.photozone.de/nikon_ff/447-ni ... 1424_28_ff

I'd agree take some shots at f5 to f8, preferably with a more distant scene and tripod mounted if at all possible. Wide angle lenses can vary in performance from infinity to nearby objects even if they in focus for both.