16-35 vs. 18-35

[beryllium10]

Chip, Frank and Keith (pluton) - thanks for the further feedback.  This is a difficult scene and definitely the most challenging/worst performance by the lens while I had it to work with.  Focus should have been on the edge of the blue building in the background, hopefully this is not a misfire.  Frank - yes, good things are being said about the Tamron 15-30 mm 2.8, and Tokina makes a number of potential alternatives.  The Tokina 11-16 mm 2.8 DX is a brilliant performer on the D7000, so I should definitely take a look at their offerings.  Good thing I have time to do this at a leisurely pace, as buying a lens shortly after the D810 would be excessive.  Pluton - thanks also for your strong endorsement of the 17-35.  I think it is entirely possible that this particular one, a rental, has been handled unkindly in the past.  I will keep an eye out for other copies that show up second-hand locally and see if I can make some comparisons.

Cheers again,  John

[Frank Fremerey]

One last thing: I had my 12-24 Tokina (first edition) together with my D70 as kindof 18-36 crop lens (I do not believe in equiva'lens' and will post visual evidence as soon as I find time, possibly in Scotland).

!!!: This lens served as a great 18-24 on my D3 also, because the image circkle was much bigger than necessary. I feel the "PRO" Tokinas are really well made. Mechanically the 12-24 was much more rugged than the 12-24DX Nikkor and optically they played in the same league. Someone even claimed this lens was a co-development by Nikon and Tokina.

[simsurace]

I used to have the Tokina 11-16. I played around with it when I switched to FX. Of course you could get an image at 16mm but it had very weird distortion and the corners were just abysmal. It was usable for fun but was no comparison to any dedicated FX lens. There is a reason why that part of the image circle is cropped off.

But even on DX, the results I could get were not in the same league as from an FX lens on FX, the jump in format size was very obvious for me in th wide angle territory and less obvious for longer focal lengths, where it can be difficult to tell images from FX and DX apart. That said, the 11-16 is a very good lens on DX and I believe the new one is even better.

[MFloyd]

Hereby a picture and a 100% crop of the extreme left upper corner taken with the Nikkor 16-35mm f/4 at 22mm f/6.3 1/500 ISO 100.  The 100% crop is an uncorrected NEF file, contrary to the full frame picture. Personally, I find the "corner" performance very good.

[Erik Lund]

That is indeed very good corner performance! Fully acceptable.

[MFloyd]

Thank you Erik.  Here another example for the Nikkor 16-35mm f/4, this time at the widest angle (16mm) but at f/11; again full frame and 100% crop of the lower left and right corner; most off the time I'm using this lens at f/8-11 to maximise the depth of field, without entering into the diffraction danger area:

[simsurace]

Maybe your copy is indeed better than mine, maybe it's the fact that we are using them on different sensors (16mp vs 24mp), but your samples look much cleaner in the corners than e.g. my tree sample where there is excessive smearing and double-images (it is wide open at f/4 though, but the 18-35 is much cleaner at the same aperture). I must say that my 16-35 seemed pretty flawless on the D700, but the D600 revealed its weaknesses. I would expect that my copy would be even weaker on a D800.

Anyway, good to know that there are good copies and satisfied owners out there

[MFloyd]

Hi Simone, I just took some existing pictures out of my catalog. If I have some time, I will take some more "on-purpose" pictures at full opening.  The D4s has been replaced by the D5; the highest pixel camera I have is the D610; no D8xx on hand ....  I know that D810's and the like are sensitive to motion / vibration blur.  Would it also be the case for sharpening fall-off ?

[Erik Lund]

....  I know that D810's and the like are sensitive to motion / vibration blur.,,,,,,,

The D810 has a very well dampned mirror/shutter action, so only issue would be the high MP and motion blur - not so much the case for a wide angle,,,

Again the image is perfectly sharp and well defined in the corner of the 16-35mm

[simsurace]

Hi Simone, I just took some existing pictures out of my catalog. If I have some time, I will take some more "on-purpose" pictures at full opening.  The D4s has been replaced by the D5; the highest pixel camera I have is the D610; no D8xx on hand ....  I know that D810's and the like are sensitive to motion / vibration blur.  Would it also be the case for sharpening fall-off ?

I needed to be more precise. Of course any lens will resolve more on the denser sensor, even if a poorly resolving lens gains very little. The biggest boost in resolution is commonly seen in the center, where I believe both the 18-35 and 16-35 are limited by the 24mp sensor. Hence I did not post center crops of my samples, the lenses are virtually indistinguishable if you look at the center. The fall-off of sharpness from center to corner will be more pronounced on the denser sensor. So while the fall-off was barely visible on a 12mp sensor, it is apparent on my 24mp sensor and I expect it to be even more pronounced on the D800 or D810. It does not matter much whether the sharpness fall-off I observe in my 16-35 is due to a defect or is by design (I don't believe it is by design, there are too many people whose findings are different from mine); the 24mp sensor is simply more revealing of the limits of my 16-35 copy.

The vibration is a different issue and not what I meant. I made all my test shots with MUP and IR remote release on a tripod to minimize camera shake.

[BEZ]

Simone,
Your choice was understandable because you chose between two lens samples.

But for the benefit of the discussion, I stated earlier my 16-35 has excellent corner performance  .....just to add I use the lens with my D800's. I would expect similar results to MFloyd's sample images.

Cheers

[simsurace]

Simone,
Your choice was understandable because you chose between two lens samples.

But for the benefit of the discussion, I stated earlier my 16-35 has excellent corner performance  .....just to add I use the lens with my D800's. I would expect similar results to MFloyd's sample images.

Cheers

Good to know!

[Jyda]

FWIW, I've had two of the 16-35. The first one was quite mushy in the corners but my second one is much better.

[John Geerts]

I'm still pleased with my one copy of the 17-35, 9.5 years later. Bjorn R observed sample variation issues way back when the lens was introduced in 1999/2000.  Mine has proved useful from D200 and D2xs to D3 and surprisingly, D800. 

Yes, works indeed great on the D800E, but also on the Df and the D600.  On the D700 it didn't work so well, somehow. Not sure why.

[pluton]

Yes, works indeed great on the D800E, but also on the Df and the D600.  On the D700 it didn't work so well, somehow. Not sure why.

I always felt the 17-35 worked fine on my D3 cameras, with supposedly the same imager as the D700.  Oh well....another mystery of the universe.