Camera Labs

Even before someone else has tested how good this lens is, I've decided to take the plunge and get a good complement for my Nikkor 14-24/2.8 zoom (with a Nikkor AF-S 50/1.4G as lens-in-the-middle).
I’ve been fiddling around with enough long fixed focals (Nikkor 85/1.8, Nikkor 105/2.8 VR macro, Sigma 150/2.8 macro, Nikkor 180/2.8, Sigma 300/4.0, Sigma 400/5.6, Tamron 500/8.0) to value the range and flexibility between 70-200mm highly for my photography style and to find image stabilization invaluable for my hand-holding approach.
So I just waited for Nikon to update their well respected 70-200/2.8 VR with improved corner performance on the long end (for future FX use) and the latest version of VR to get one.
I promise you to take this lens to my usual (and not so usual) tests to find out how it performs and give you a first-hand impression of Nikons latest and (hopefully) greatest in professional zooms.
If the lens performs as expected I’ll certainly going to thin out the line of long lenses and get some cash back for the hefty investment. As I already get myself rid of the 85/1.8 and the 180/2.8 the next probable candidates are the 150mm (only need one macro lens and the stabilized micro Nikkor is in the end more valuable) and the 300mm if the new zoom also performs well with a tele-converter. I'm thinking of the 1.4x (=> 100-280/4.0) or even the 1.7x (=> 120-340/4.8 ) depending on the resolving power of the zoom. The 400mm and 500mm lenses will stay with me for those special occasions. Unless perhaps an updated Nikkor AF-S 80-400/4-5.6 VR II comes along

As the delivery date is still some way up the calendar you will have to make do with some insights on the technical data (yeah, famous shrink-factor will be there) comparing it to the technical data from other manufacturers and similar Nikon-zooms for now.
I'll also chalk up a list of tests that I'd like to perform.

As a lot of discussions are to be expected I have split all comments so far out of this thread in their own Nikon AF-S VR 70-200mm 2.8G ED II review comments & Qs thread. So if you have any questions or comments, feel free to join over there.

That way this review-threads remains pretty straight forward and easier to read.
---
Dieser Thread ist verriegelt. Aber wenn es irgendwelche Fragen, Kommentare und Anregungen zu meinem Review gibt, hier gibt's die Möglichkeit auch für unsere deutschsprachigen Forums-Besucher!

===Testing Agenda===

Now, what will I be testing?

- As all test results are null and void if the focus is not optimal I’ll first test for front- and back-focus and report whether I had to use micro AF adjustment to get optimal AF-results from this lens. This also gives a first impression about the amount of longitudinal CAs as shots from the micro-focus test-chart clearly exhibit the telltale greenish and magenta colorations if the lens is not corrected properly. This test has to be repeated at various focal lengths as front-/back-focus may depend on it (the otherwise capable Tamron 70-200/2.8 had this nasty property).

- Next up is confirmation of AF-quality, -reliability and repeatability in some real life shots at f/2.8.
This establishes the baseline for all the other IQ-related findings and I hope that the lens+body combo proves to be a capable AF-performer

- Some handhold shots at longer exposure times will yield a first impression of the efficiency of the VRII image stabilization. This is very important to me as I almost never lug a tripod around. But I doubt that I have time for statistics with a few hundred shots on a test-target (see my effort on this issue with the Sigma 120-400 here).

- I'll also perform some tests with the close-up lens Canon 500D which should bring this zoom up to a magnification of around 1:2.3. The problem with repeatable tests here is that I have no test-target with enough resolution to prove my impressions (you need a 2500 dpi target at least!). So some images will have to suffice in this department. Those tests will also confirm (or disprove) my calculations of shrink-factors and magnifications that you can achieve with this lens.

- Then there will be the mandatory Siemens-star shots to judge sharpness, contrast, astigmatism and CAs at 70mm, 105mm, 150mm and 200mm at f/2.8, f/4.0, f/5.6 and f/8.0 from a tripod with AF. This is why I love Siemens-stars as a test-target because you can test four different parameters of a lens in one go and also have a good visual impression of the quality that is easily comparable across different lenses. I'll also do some cross-checking with LiveView-aided MF to confirm quality of AF – just to be absolutely sure.

- All shots will be done in RAW with conversions to jpg with CaptureNX 2 at normal sharpening settings. With some conversions I'll disable the automatic CA removal to judge the CA performance of the lens.

- Comparisons will be made with the micro-Nikkor AF-S 105mm f/2.8 VR ED, the Sigma 150mm APO macro and the Nikkor AF-S 18-200mm f/3.5-5.6 DX to give you an impression of the differences in image quality between those lenses. That may sound a bit unfair, as we all know that fixed focal macro lenses belong to the sharpest you can get. On the other hand the 11x DX-zoom is not really expected to be a match for the new 70-200mm. Well, there may be surprises, and I really want to see the differences that many other reviews will only show in numbers. Because how an MTF50-value of 1821 and an MTF50-value of 1583 translates into visible IQ-differences is no easy feat to imagine.

- And if the lens proves to be worthy (which I hope/expect), a test with a tele-converter will follow at a later date. The “old” Nikon 70-200 was said to harmonize pretty good with 1.4x and even 1.7x converters. So this looks like a good way to recapture some of the huge investment in this lens. A TC is also much easier to carry than an additional 300/4.0 lens or a xx-400mm zoom. But the acid-test will be how an enlarged image compares to the original with TC. Because if there is no discernible difference between both approaches, why would you then shell out another 300 EUR and fiddle with a TC that invariably reduces your max aperture and produces a smaller field of view than the original lens? A TC really comes to shine, when the lens you combine it with comfortably out resolves the sensor. Only in this case will the lens+TC combo show more details than an uprezzed original. I'll show you some images where you can see for yourself.

Phew, quite some work to do...

===Disclaimer===

Well, I have to get one point out of the way first:

I will only test the lens on my D300 as I have not (yet) made the transition to an FX-body. Does this mean that I'm now going Ken Rockwell style (see my comment on his comparison between Nikon, Canon and Leica M9 here)? Well no! I will not be extrapolating how this lens performs on a D3x (or future D700x) and I will not be showing you FX corner performance. I know that the latter point will disappoint many of you, as the “old” version of 70-200/2.8 came under fire after people put the lens to their newly acquired FX bodies and were less than totally satisfied with its corner performance. But that's the way it is for the foreseeable future in my reviews.
If you watch the extreme corner performance on the Nikon D300 you cover the 14mm radius performance. This comfortably covers the height of an FX sensor. But of the total area of an FX sensor this still leaves around 40% uncovered. See the following image:


And mind you, I'm not talking about performance on a real D700 or D3x as the AA-filter and pixel-pitch of these bodies are all different from a D300 and this has an influence on the real-life resolution that those cameras will be capable of when combined with the new 70-200/2.8 zoom!

===preview===

Let's have a look at the theoretical performance of the new 70-200 in comparison to the current zoom and the older Nikkor AF(-S) 80-200/2.8D. Following are the MTF charts from Nikon-central, first line at the wide end, second at 200mm:
NEW AF-S 70-200/2.8G VRII.......current AF-S 70-200/2.8G VR.......discont. AF-S 80-200/2.8D.......older AF 80-200/2.8D




The read lines mark the contrast at 10 linepairs/mm (watch out: Nikon writes "lines/mm), the blue lines are for 30 lp/mm.
There is also the discontinued AF-S 80-200/2.8D. From the outside it looks more similar to the new zoom than any of the other predecessors and was said to have better IQ than the older, still for sale AF 80-200mm design. And the MTF-charts clearly confirm this.

For those using any of these lenses on a DX-body (as I will do) the diagrams are only relevant up to 14mm. Now you can easily see why the current 70-200/VR has an excellent reputation before Nikon introduced their FX bodies. (Seems nobody was using it on a film-body )
You can also see that the still available AF 80-200mm is no match for both the current and the new 70-200mm at 200mm even on DX.
An interesting zoom is the discontinued AF-S 80-200mm. As you can see the MTF chart at 80mm looks better than its successor, the current 70-200! There you have the reason, why Nikon pulled this version: It would have hindered the sales of the current zoom.

On the other hand one characteristic of the current 70-200 zoom is clear: past the 14mm image radius the meridional (or tangential) and the sagittal contrast differ widely, esp. at the long end. What does this mean? It tells us that although some structures outside the 14mm circle are well resolved others have a much lower contrast/resolution depending on the orientation of the structure. You can see the result clearly in the following example shot with a Tamron! 70-200mm at 135mm in one corner:

You could say that the sagittal contrast represents the best case and the meridonal contrast the worst case of a lens.

With the new 70-200 the sagittal and meridional contrast is much closer. This should lead to a more harmonious reproduction of structures independent of orientation. But one has to concede that the sagittal contrast (indicated by the continuous red or blue line) of the new 70-200mm is sometimes clearly lower than of the current 70-200mm.

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

===facts from the catalog===



Now, while waiting for Nikon to deliver on their latest and hopefully greatest 70-200mm f/2.8 VRII zoom, let's have a look at the technical data and make some comparison to similar lenses.

Size: 87x206mm = medium size 9mm shorter than the current 70-200mm Nikkor but still 21mm longer than the Sigma. The maximum width is the same as with the current zoom, but around the zoom ring the new version is clearly wider than the current one. This might have to do with the improved corner performance and/or the improved VR. As my Sigma 300/4.0 has approx. the same size I have a good feeling that this size is still acceptable. Btw. the 80-200mm Nikkor is only 187mm long.
Weight: 1,540 g = The heaviest of all current 70-200mm zoom, 70g heavier than the current version (but still lighter than the discontinued AF-S 80-200/2.8 ). Compare this to the Tamron with 1150g which is the lightest in the group. Being used to schlepping my 1200g Sigma I'm pretty sure that you begin to notice the extra 350g after a while
Optics: 21 elements in 16 groups = same # of elements but one group more than the current version. All those glass/air surfaces with their tendency to reflect some light Tamron has a 18/13 design. But only the new Nikkor has nano-crystal coating. There is also a tendency to use more special glass: The new Nikkor has 7 ED-elements vs. 5 ED-elements in the current version. By the way: it is good engineering practice to not use a special glasses as the front element, because these glasses tend to be softer and thus less scratch resistant. See the new design (left, first) compared to the current version (right/second):

Closest focus distance/max.magnification: 1.4m / 1:8.3 This is the worst of the group. The current Nikkor goes to 1:5.6, the Sigma to 1:3.5 and the Tamron to 1:3.1
Filter-thread: 77mm = standard
IS: YES, VRII = very good The current version has VRI, both have "normal" and "active" modes, the active mode being optimized for stabilized shooting, the normal mode optimized for stabilizing the image while viewing through the viewfinder. All other alternatives have NO stabilization at all.
AF: AF-S with SWM (silent wave motor), so does work on D60/3000/5000-bodies , manual-focus override by turning the focus ring There is a new AF switch position: "A/m". This will make inadvertent manual override of auto-focus harder. Very practical, when you hold the lens somewhere in front and accidentally move the focus ring after the camera has locked on target.
Covers full frame/FX or smaller = normal
Comes with an nice looking lens-bag
Price: around 2100 EUR new (incl. 19% VAT) = expensive The current version is around 1700 EUR, the Tamron at 600 EUR and the Sigma at 660 EUR. So this is a huge premium for the Nikkor(s) but the optics and the image stabilization make them superior lenses.
The lens-caps are standard Nikon's The others still play catch-up
Distance information is relayed to the camera, so the Nikon body can do all the advanced exposure-related stuff with this lens But this is the same with the other zooms too.
Aperture ring = no, just like all Nikon G-lenses
Lens-shade included and revertible for transport. It has a small lock to prevent it from getting loose and subsequently lost But never try to place the lens upside-down on its lens-hood: it will immediately topple over
The foot of the Tripod-collar can be removed without dismounting the lens , easy to turn camera to portrait-mode
Limiter switch: Yes: infinity to 5m The current version has one too (infinity to 2.5m). I'm not sure about the Sigma and the Tamron.
Sealing: Yes Unlike the Tamron and the Sigma

Motivation:
I was tired of carrying around two/three lenses, which did not cover the 70-200mm range and only one of which was stabilized.

Alternatives:
- The current Version 1 Nikkor AF-S 70-200/2.8: An excellent lens but only first generation VR and with weakness in the FX corners. Not cheap in itself.
- The Sigma 70-200/2.8: not tested by myself, but should be an ok zoom
- The Nikkor AF 80-200/2.8: Now 3 generations behind, smaller and lighter than the other Nikons. You can see from the MTF charts that the optical design is dated.
- The Tamron 70-200/2.8: pretty good potential optically and the lightest in the group but marred by changing back-/front-focus (see my review)
- A collection of fixed focals, e.g. Nikkor AF 85/1.8D + micro-Nikkor 105/2.8 VR + Sigma 150/2.8 macro or Nikkor AF 180/2.8D: Well you get macro and larger aperture on the short end, but only one lens is stabilized, both the 85mm and the 180mm Nikkors are past their prime and you now need 3 lenses and still lack the flexibility of the zoom. Price- and weight-wise you end up almost the same but you have more to carry around

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

Now it's time for the first shots. And as it's dark and raining outside you will have to make do with my (un)famous Siemens-star target.
I changed the testing schedule a little bit to bring the first qualified image to you: I skipped the check for front-/back-focus and used MF in live-view to guaranty optimal sharpness. I consider this cheating, as you normally use AF to focus and if the AF is no good you will not reach optimal sharpness in most of your shots.
But first tests with AF showed just a small degradation of sharpness in comparison to live-view aided MF. So I'll come back later with results concerning focus-quality.

Now here it is, the star in all its glorious quality shot at 200mm, f/2.8, 1/250 sec, ISO 200, -1EV, mounted on a tripod with mirror-up, VR off, incandescent light, target at 2m distance. Post-processing included developing the RAW with "normal" settings (concerning sharpness, contrast, saturation), adjusting the white-balance and correcting for the under-exposure. I also disable the automatic CA removal to judge the CA performance of the lens.



I uploaded the full-res jpg as it came straight out of CaptureNX2 V2.2.3. So if you click through the image to the flickr-page and then on "all sizes" you can pixel-peep until your eyes pop.

One thing I can tell you: This lens is extremely sharp! Remember though that I shot with my D300 which has a higher pixel-density than the mighty D3x, but corner-sharpness on FX cannot be judged from my samples!

Other apertures and focal lengths will follow in due course accompanied by 100% crops to make comparisons easier. But for tonight that's it, folks!

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

Two other image this time shot outside at 70mm resp. 200mm, f/2.8, 1/30 sec handheld, VR=ON, ISO 320 resp. 200, -0EV. Processed as before again with CA removal switched OFF. You can click through to the full resolution version.



As the sign slightly tilts away from the observer the upper part of the sign is not optimally sharp. I used single-point AF this time at the middle between both signs. There seems to be a little front-focus.
What you can also judge from the image is the quality of the bokeh with a busy background and VR=ON (some people say VR influences the background quality). Remember though that this is only 70mm at f/2.8. So you cannot expect the same degree of background isolation as at 200mm which the second shot nicely shows.

I did some testshots of the new zoom with Canon 500D close-up filter here. They are at 200mm and 70mm and also show what sort of dof you can expect when shooting at 0.4x magnification and large apertures: it's only millimeters
All shots were live-view aided MF but still i was not exactly focussed on the line of text in the middle. But don't worry - just look at what's sharp (e.g. that very small red thread above the "s" of the word "this". You have to view the larger sizes to see it.

Now, what are the results, can the Canon 500D close-up filter heal the "deficiency" of the grand new professional 70-200 VRII?*
Well, at 200mm there is bad border sharpness at large apertures at least on flat objects (should be field curvature - not yet tested). But from f/5.6 and f/8 and onwards it is very usable on DX-bodies. Cannot speak for FX-bodies, but field curvature is not really a problem when shooting nature or other three-dimensional objects...

But all in all the Canon 500D will be my constant companion when walking around with the new 70-200 zoom and the combo makes for a nice instrument to capture things at 1:2.4 - 1:7.1 magnification with image stabilization!
Highly recommended

---
*You won't believe the raging debate about the "unexpectedly" large shrink-factor of this lens when focusing at MFD. I told you all about it already on September 23rd, see here. So Camera Labs were the first to break the shocking news to you

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

Here's a 100% crop from a focus test-chart shot from 2.5m distance at 200mm focal length and f/2.8.



Again developed with CaptureNX2 w/o automatic CA-removal.
I can see only slight longitudinal CA (greenish background=left, redish foreground=right).

As the war on the web about the (for some people) unexpectedly grave "shrink-factor" rages on, I thought to bring some facts to the table.

Many people were caught with their knickers in a twist, when they bought the new 70-200 VRII and only found out later about the meager max. magnification of 1:8.3 at an MFD of 1.4m*. That was caused by the inevitable shrink-factor of IF (internal-focus) designs that reached a new low of 0.66x with the VRII. For your comparison: the old VRI had a max. magnification of 1:6.1 with AF (1:5.6 only with MF) and a shrink-factor of 0.91x.

Now, how "grave" is the situation? Well, you get 26% less max. magnification with the new VRII as opposed to the "old" VRI lens (in AF). So should you need a magnification of 1:6.1 you need to crop 26% (linear, i.e. you lose 45% of your pixels), mount a 1.4x TC (and lose 1 F-stop) or rely on some other magnification increasing tool (with some other side effects) - or you buy the VRI. That's too bad.

But: This problem only occurs at a focal length close to 200mm AND minimum focusing distance (MFD). If you need e.g. 1:10 or even smaller magnifications you just get a slightly shorter working-distance from the VRII than from the VRI. That means you have to take a step closer to your subject if you're already close to 200mm focal length (which changes perspective a little) OR you just zoom in.

To give you an impression of the shrink-factor at more normal magnifications of the new VRII, here's an overview for different nominal focal lengths at a magnification of 1:18**:
- 200mm nominal at 3.2m becomes 160mm effective, shrink-factor = 0.80x
- 150mm nominal at 2.7m becomes 134mm effective, shrink-factor = 0.89x
- 105mm nominal at 2.0m becomes 99mm effective, shrink-factor = 0.94x
- 70mm nominal at 1.4m becomes 70mm effective, shrink-factor = 1x
So for all intents and practical purposes you can correct for the additional shrink of the new VRII for all focal lengths below 170mm by just zooming in. Resulting in no loss of pixels or change in perspective!
Or you just step in another 30-60cm (i.e. 1-2 feet) or so which will change your perspective a little but works even at 200mm nominal focal length.
Oh, and if you wonder, what 1:18 magnification can do for you: On a DX-body this gives you a pretty tight head&neck-shot with just some part of the shoulders, on FX you would also capture part of the chest, but only part

----
This thread is locked. Please post your comments and questions here.
*Nikon published these data from day one. So those who can read have a clear advantage
** all focus distances measured in real life

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

The whole bunch is now up here.
I will select the appropriate 100% crops next week for easier side-by-side comparison. But I can tell you already: the results are great!
Again: These images were developed with CaptureNX2, standard-settings, and automatic CA-removal switched OFF. So you can see the true power of this lens.

I've also extensively compared the new VRII at 150mm with the Sigma 150/2.8 macro lens and at 105mm with the micro-Nikkor 105/2.8 today.
Using CaptureNX2 at standard settings but with automatic CA-removal=OFF and focusing the lenses with live-view aided MF I see the following ranking:
150mm:
- At F/2.8 the VRII is slightly behind the Sigma when viewed at 100%
- At f/5.6 the VRII and the Sigma are very close in resolution and contrast, but the VRII has still some visible CA
105mm:
- At f/2.8 the micro-Nikkor lags the VRII
- At f/5.6 the micro-Nikkor has almost caught up with the VRII

If these first results hold up, that would be quite an achievement as the Sigma is one of the sharpest lenses out there. And don't forget: The Sigma (and the micro-Nikkor) has the benefit of being a fixed focal!

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

********TEASER*******

This is not the final version, but here's a taste of what the new lens is capable of at 200mm:


These are 100% crops, now with CA-removal ON from left f/2.8 to right f/8.0. Taken approximately 10mm off center, which is 66% of the image-radius of a DX-sensor and 46% of a FX-sensor.
Distance was 3m, magnification 1:18.

I seldom see a lens that produces images at f/2.8 as good as at f/8.0
Me happy!

Why is this not the final version? Well, with lenses sooo sharp I'm touching the print-resolution of the test-target. So I have to step further away from the target to have the test-target out-resolve the lens+sensor!

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

Now here's a more challenging proposition. Shot from 10m away at a magnification of a meager 1:50. 100% crop from the center:



Looks much worse than the above, doesn't it? It has to at a magnification that is around 2.5x weaker than the previous one!
Now we're out of the danger zone: The combined system of lens and camera clearly has a lower resolution than the test-chart
What still astonishes me: I can see (almost) no change between the left crop @f/2.8 and the right crop @f/8.0.
CA-removal was ON, all shots live-view aided MF from a tripod with mirror lock-up plus flash. Developed from RAW in CaptureNX2 with standard-settings. Only corrected for white-balance and exposure variations.

Thought you were interested in what you can achieve with this lens in real-life close-up situations.
Capitalizing on what little sun came out today I grabbed the new zoom and my trusty old Canon 500D close-up filter and shot a series of "macro"-shots of which I present three to you here. All images were developed with CaptureNX2, standard settings (incl. automatic CA removal), no extra sharpening, just some exposure correction and a wee little bit of (linear) contrast expansion in the first shot. Here we go - clicking on the original leads you to the flickr-page where you can access the large original saved at 100% quality:

#1 shot at 130mm (1:3.8), f/5.6, 1/125 sec, ISO 200, -1EV:

100% crop:


#2 shot at 135mm (1:3.7), f/11, 1/125 sec, ISO 360, -1EV:

100% crop:


#3 shot at 200mm (1:2.5), f/8.0, 1/250 sec, ISO 200, -1EV:

100% crop:

The dof is so shallow that not all ice crystals are in focus.

All shots handheld, VR ON/normal, AF
And remember: This is not a designated macro!