Nikon 200mm Lens Sharpness Shoot-out

As promised, here is my Nikon 200mm lens sharpness shoot-out. This started during the discussion of the new 70-200 f/2.8 VR II lens that was recently released by Nikon. Many were wondering how it stacked up against the 70-300 which has a reputation as being one of the sharpest consumer grade lenses at 200mm. This project ended up taking up the better part of a day to shoot, process, examine, etc the images and then write this up. I probably should have my head examined for doing this since this always ends up in arguing and name calling and bickering - lets try to stay above that and take it for what it is - a single test done as well as I could possibly do it without a big buck lab. Please realize I have no point to prove other than finding out which gives the best result and how they stack up against the other. I really don't care which lens wins and have made an attempt to completely eliminate any bias from the tests. I am not attacking any lens. People tend to take offense if the lens they have purchased with their hard earned dollars does not fare as well as another lens, sometimes a less expensive lens. Here we are simply trying to figure out which lens is sharpest in overlaying apertures.

Test procedure:

All photos were taken with a D700 at ISO 200 and were flashed with bounce flash from an SB-900 flash.
Shooting distance was 15 feet to the sensor plane (yes the camera had to be moved somewhat to accomplish this due to the different physical size of the lenses)
The center of the White House on the back of a $20 bill was used on a good condition bill (my trusty $5 bill of the old style is gone)
The rig was tripod mounted and all shots were taken with the tripod locked, using mirror lockup.
All photos were processed in Adobe Camera RAW using the Camera Neutral profile with no lens correction and default capture sharpening which is identical for every image.
No other sharpening was applied.
Lenses used were the Nikon 200mm f/2, Nikon 70-200mm f/2.8 VR II, Nikon 70-300 f/4.5-5.6 VR, Nikon 200-400 f/4 VR
No filters were used on any of the lenses other than the required clear filter in the back of the 200-400
All lenses were carefully calibrated for focus using the moire interference pattern method
All images below are 100% crops. Below is a shot of the scene to give scale. As you can see these crops represent a tiny part of the total frame and that should be taken into consideration when looking at the samples below: Results

f/5.6 Center: The 200mm prime is clearly the sharpest lens as expected. The 70-200 and 70-300 are virtually indistinguishable in detail although the 70-200 due to it's design has substantially less magnification at a 200mm setting at 15 feet. It is amazing that the 70-300, despite being nearly wide open from an aperture perspective holds its own with lenses that are significantly stopped down. The 70-200 is however slightly more contrasty. The 200-400 is a bit softer than the other lenses
f/5.6 Corner: As expected, all of the lenses soften up some in the corners but detail holds up notably well in both the 200 prime and the new 70-200. The 200-400 also fares pretty well. The 70-300 loses the most in the corner at this nearly wide open aperture but that is not a surprise since the other lenses are stopped down from 1 to three stops here.
f/8 Center: results are similar to those gotten at f/5.6 in the center but the 200 prime is simply awesome. Both the 70-200 and the 70-300 give exceptional results. The 200-400 continues to be softer than the others.
f/8 Corner: The prime continues to lead the pack. The three zooms are in a virtual deadlock - the 200-400 is starting to come into it's own by giving very good edge performance at f/8. The 70-300 no longer exhibits the softness seen wide open.
f/11 Center:The 70-200 and 200-400 are notably softer than the prime and the 70-300. the prime continues to excel.
f/11 Corner: Like the f/8 corner shots, the three zooms are very close and surprisingly good but they can't touch the prime.
I opted not to go to f/16 partly due to the additional work and partly due to the fact that diffraction effects could start to cloud the results.

Other Observations:

The most obvious other observation is that the magnification of the 70-200, even at 15 feet is significantly less than the other three lenses which are virtually identical. So I must retract my earlier statement that I thought some of that was a red herring. It clearly is not. I will attempt to determine if at infinity focus the 70-200 is really a 200mm in the future. Another observation not obvious in what I have shown above is that the 200-400 loses significantly more light in the corners than any of the others at 200mm. 200mm is clearly not the 200-400mm lens's strength. It does much better at longer focal lengths.

Final Conclusion:

As stated, the original reason for the test was to see how the consumer grade 70-300 stacks up and as can be seen, it stacks up admirably and at 200mm it is at worst tied for second with the new 70-200 behind the awesome 200mm f/2 prime lens. It does give up a lot in light gathering due to being an f/5.3 lens at 200mm and it's AF is much slower and it is not that sharp at 300mm but at 200mm it is quite impressive. The 200mm prime which I was fortunate enough to test at the last second is simply an amazing lens in all regards. It's big, it's sinfully expensive, but it is one bad a$$ lens that smokes anything I've ever used at that focal length. Even the f/2 shots which I didn't post here are impressive being about equal to the best shot of the other lenses in their sweet spot. Subjectively, the best aperture for all of these lenses from center to edge is around f/8.

Hopefully this has been useful. As I said this was a lot of work and I did my very best to eliminate all the variables but invariably there are some even if you don't know about them. I would put any of the lenses on my camera and be confident that I would get sharp shots - even the 200-400 at 200mm - remember these shots have not been output sharpened and the crops represent a very small part of the frame with some incredibly fine detail rendered from a significant distance.

E.J. an outstanding job as usual. Thorough and complete. I really appreciate you taking time to do this. I've been considering the 70-200 AFSII and It seems to perform acceptably.

I didnt expect to see your test so soon, thanks for doing this. It would seem that the 70-300vr performs nicely from 70-200,pretty big bang for your buck at around $500. The performace of the 200-400 is surprising, but it is what it is.

Thanks again.

EJ, thanks for posting this, very concise! I'll confess there are days I miss the size of my previous 70-200/2.8 VR I (and it did produce some great shots). The version I had was actually light-grey in color and needless-to-say it got some occasional odd looks by folks in the crowd who were familiar with Nikkor lenses..i.e. "did I paint it? (nope!)" For what it's worth, Marianne Oelund also did a 200mm shootout recently, but at greater distances: http://tiny.cc/xqQEn.

So next question, when you do your 400mm shootout you'll be moving up to a new $50 bill, correct (well considering your $5 is worn out)

--DJ

Wow!!!
Nice job and VERY thorough.
Thanks

Very informative. Thanks for the work on this EJ.

Thanks, E.J. Just picked up a used 70-300 VR for my wife and you've boosted my confidence in it.

Thanks for your time and report, E.J.

Thanks for sharing E.J. The 70-300 VR was definitely a very good performer back when I had one. If I ever start shooting Nikon again, I'd probably pick one up.

E.J. - Thanks also for your effort, MUCH appreciated. To me the most surprising info. was how well the 70-300 performed and how poorly ( compared to those you used ) the 200-400 did, INTERESTING. Jeff A.

Thanks for the effort and the time you spent. It is appreciated

E.J., very nice. I am curious why you chose to use your D700 rather than you D3x.

Thanks for the work, E.J. One of the benefits when my wife switched to Canon early in the year was I inherited her 70-300VR

Larry Rosier wrote:E.J., very nice. I am curious why you chose to use your D700 rather than you D3x.

Mainly because the D3x is already into diffraction at f/11 which for photography isn't a problem but when comparing views on a per pixel level as I have done here, I didn't want that clouding my ability to see any differences.

Thank you, E.J., for the test !!

I knew that the 70-300 f4.5-5. VR, unlike many other folks' opinions, is one of the best-bang-for-the-buck lenses out there. I just didn't expect my 70-300 can out-perform my 200-400 @ 200, considering the latter actually cost 10 times more

DChan wrote:I just didn't expect my 70-300 can out-perform my 200-400 @ 200, considering the latter actually cost 10 times more

Yeah but the 200-400 blows away the 70-300 at 300mm

Thanks for your review EJ. Very informative as usual.

Did you notice any differences in sharpness among the corners as Lloyd Chambers did? I'm wondering if Lloyd's is an isolated incidence or if this is a common defect with this lens.

http://www.diglloyd.com/diglloyd/2009-1 ... n70_200VR2

My 70-200 definitely does not have that problem. I have used it extensively already and that would have been noticed right away.

Looks good EJ. I already like seeing the results of these. Very well done

Thanks for all of your work on this