I finally received my 800/6.3S PF and 1.4X TC, so I have everything I wanted to fully test a lot of the options folk might want to consider for 400mm, 800mm, and beyond. Specifically, the lenses are 100-400S, 400/2.8S TC, 800/5.6E, and 800/6.3S PF, along with the two Z mount teleconverters and the F-mount TC800-1.25E, TC-14E III, and TC-20E III. (I decided the TC-17E II wouldn’t be worth including.) I could’ve tested plenty of other lenses that can get to 400mm or beyond, but wanted to limit this test to lenses that are at least 400mm natively and can stretch to 800 or more.

There are so many components that go into a thorough lens test. This set of tests only assesses the subjective and relative quality of the center of the image. It is only qualitative, not quantitative. As much as I’d like to produce specific resolution numbers, I don’t have a 400′ long hallway, which is what I would need to test lenses at 1600mm with my resolution testing chart. And the test doesn’t test the corners, or compare corners, or measure contrast, or measure chromatic aberrations, or …

There are some additional observations and inferences that I add related to what I saw while doing the tests. Hopefully those will be informative, as well.

I was reluctant to include the actual test chart images like I did in the last post, but decided I would so you can see what I saw that made me make the conclusions I’ve made. But my real goal is just to share my thoughts about each lens’ performance and then offer my summary list of what lens does best at the different focal lengths and how they compare. For the first time ever, I added my own very subjective ratings, which are in the Detailed Commentary section. To aid in readability, I’ll put the more technical stuff toward the bottom.

Commentary on Each Lens’ Test Results

What Is the Best Lens at Each Focal Length? How Do They Compare with One Another?

Detailed Commentary and Ratings

(When “making up” a grade, I give most credit to resolution, using the integers 0-9. Contrast makes up the tenths, from x.0-x.9. A 9.9 is a perfect score.)

400mm – There is just no competition in this race. As good as the 100-400S is, it’s set against one of the absolute best lenses I’ve ever seen. The 400/2.8S TC is very very sharp and contrasty wide open, but at its best aperture, f/3.5, it is slightly sharper still and significantly higher in contrast!

I’d rate the 100-400S an 8.4, and the 400/2.8S TC a 9.9!

560mm – Among the three contenders, there’s a bit of a surprise here. The 400/2.8S TC + 1.4x is clearly the best wide open, and, though the built-in TC improves a lot by it’s best aperture of f/5.6, it’s still not as good as the 400+1.4x at its best f/4.5. Makes me wonder why I keep hearing Nikon reps say the 400/2.8S TC + 1.4x TC is “not recommended.”

I’d rate the 100-400S a 7.4, the 400/2.8S TC + BI an 8.6 due to similar resolution, but slightly lower contrast, and the 400/2.8S TC + 1.4x an 8.8. For this rating, I’m assuming the resolution of a native 600mm lens will be even better. If/when I get a chance to test one I’ll review these ratings.

786/800mm – Looking at the test results, you’d immediately say the 800/6.3S PF is the sharpest both wide open and at f/7.1, beating the 800/5.6E. And it’s true, the 800/6.3S overall is exceptional wide open and even better at it’s best aperture, f/7.1. But if you look more carefully, the resolution of the 800/5.6E is actually better. What you’re seeing that makes the 800/6.3S PF look better is it’s excellent contrast. But the 800/5.6E is very slightly sharper, with much lower contrast. (If you add some contrast, sharpening, and/or clarity to the 800/5.6E lens images, it becomes much more obvious that it holds a slight, but definite edge.)

The 400/2.8S TC has two stories to tell, both good, but not at the same level of performance as the native 800mm lenses. The lens reaches 784mm using two (!) teleconverters with surprisingly decent results. And the lens with just the 2x TC is a very good 800mm competitor. Between the two, the 400+2x is sharper and higher in contrast wide open. The dual TC setup gets closer by their shared best aperture of f/8, but the 400+2x is still slightly better.

And given that the 100-400S enters this race already suffering terribly from diffraction (f/11), it’s no surprise that its resolution and contrast can’t compete with the faster lenses. But that’s not to say it’s unusable. With added contrast, sharpening, and/or clarity, it can definitely produce an acceptably sharp and usable image at 800mm.

Based on my grading scale, the 800/5.6E is a 9.2, due to its low contrast. The 800/6.3S PF is an 8.9, much better in contrast, but just slightly less sharp. The 400/2.8S TC + 2x is only a little behind at 8.7, with similar resolution to the 800/6.3S PF, but with slightly lower yet still great contrast. The 400/2.8S TC + BI + 1.4x would be maybe 8.4. I would put the 100-400S + 2x at about 7.1, being a step lower in resolution and with very poor contrast. (And remember, I’m a very tough grader! And the price and convenience of that lens means it will be in my kit nearly all of the time.)

1000mm – The 800/5.6E wins by default. I’ll give it a 9.2, due to its much improved performance at f/11.

1120mm – We’re totally in diffraction territory now, and it shows with lower contrast and sharpness across the group. The 800/5.6E is so low in contrast, it’s just not possible to determine exactly how much resolution it is producing. Thus the 800/6.3S PF is certainly the winner due to its sharpness and decent contrast, the 400/2.8S TC + BI + 2x second place with lower resolution and contrast, and then 800/5.6E + 1.4x due to extremely low contrast. As far as ratings go, I’ll give them 8.7, 8.4, and 7.2 respectively.

1600mm – Even more in the diffraction zone, the 800/6.3S PF + 2x beats the 800/5.6E + 2x handily, though its contrast is being drained by the small apertures. Not sure I’d use the 800/5.6E with the 2x TC, but the 800/6.3S PF is very usable at 1600mm. I’d give the 800/6.3S + 2x an 8.2 and the 800/5.6E + 2x, 7.0.

Focus breathing – (See paragraph 3 below for a definition and explanation of focus breathing.) If one assumes that the marked focal lengths are correct (@ infinity), and that the largest image among the group is from the lens with no focus breathing (not necessarily a great assumption – see below) and can thus be used as an anchor for the rest, then the functional focal lengths can be measured based on their image sizes. Since the test target includes measured scales, that’s actually pretty easy to do. The 800/5.6E lens has the largest image scale, so it will be used as the anchor and assumed functional focal length of 800mm. The rest of the lenses’ functional focal lengths at 23′ are therefore:

Should I take off points because of the significant focus breathing? E.g., the 400/2.8S TC + BI + 2x image at “1120mm” is only the slightest bit larger than the 1000mm above. I didn’t explicitly punish lenses due to focus breathing, but they hurt themselves in the resolution test being compared to the other lenses that have more focal length to magnify the target.

Just for fun, let’s dive a bit deeper into my assumption above that the 800/5.6E is the one to use as an anchor at its listed 800mm focal length. A simple lens actually increases its “focal length” as you focus closer, since the lens has to move farther away from the focal plane to bring the subject into focus, and thus magnifies more. Most modern lens designs, even “primes”, end up changing (reducing) their effective or functional focal length (they “zoom” out a bit) as they focus instead of physically lengthening the lens, as is necessary to be an Internal Focus (IF) lens. When the functional focal length becomes different (typically smaller in IF lens designs) with close focus, it’s called “focus breathing”. It hasn’t bothered still photographers too much, but videographers are quite irritated it by it since framing changes as they rack focus, a common occurrence in video. So while it might be possible the 800/5.6E is actually even higher than 800mm when focused close, since it is an IF lens that would be unlikely. In practice, the 800/5.6E lens retains fairly constant framing regardless of focus distance; it does not demonstrate focus breathing in either direction. Given that a simple 800mm lens would increase the image size as it focuses closer, and with the 800/5.6E neither the image scale nor the physical lens length changes significantly during focusing, it is evident the 800/5.6E lens design reduces the focal length at about the same rate that the focus distance decreases, offsetting one another, and keeping the functional focal length approximately the same (800mm). Thus it is an adequate anchor to compare the rest.

Diffraction – I don’t often test lenses below f/5.6, and this test contained numerous sets of images well below the diffraction limiting aperture of the Z9 of f/6.3. I was blown away by the effects of diffraction on these lenses, some of which are absolutely best in class. There’s just nothing a lens designer can do to overcome physics!

What does diffraction look like? It’s hard to describe fully. But to attempt an explanation, it’s as if with each decrease in aperture size (bigger f-stop number), a little more fog rolls into your image, reducing contrast and resolution with each step.

How much difference does it make? I made a video, but my web software doesn’t appear to be willing to let me use it. 🙁 It’s evident in the images I’ve posted in this review when you compare any of the image pairs where the right image is taken at f/7.1 or below. It’s especially evident the smaller the right image’s aperture is.

Test Images

Here are .png screen captures of what I was looking at in Capture One at 400%. (Sorry I had Exposure Warning turned on so there are some small red areas in the images. I don’t see any in important areas.) I hope the images show the same detail I saw. There’s a fair chance that Photocrati / WordPress / etc. may downsize them or do other things to them. Can’t control that. 🙁

When you see color (besides the red exposure warnings LOL) that’s a good sign. As are interference patterns. It means the lens is producing more sharpness than the sensor can handle. The measurement scales are a frequent place where you’ll see color until the longer focal lengths start to actually resolve them. The radial lines and the large “gray” box area in the lower left are both areas that are resolved at the longer focal lengths, but create interference patterns of false detail on sharp lenses below the focal lengths that can actually resolve them on the sensor. A super quick way to immediately notice performance differences (between images taken at the same focal length) is to recognize sharper lenses as those giving more colorful or contrasty interference patterns.

Here is the wide open (and best aperture) test image from the 100-400S @ 400mm:

Here is the wide open (and best aperture) test image from the 100-400S w/1.4x TC @ 560mm:

Here are the wide open and best test images from the 100-400S w/2x TC @ 800mm:

Here are the wide open and best test images from the 400/2.8S TC @ 400mm:

Here are the wide open and best test images from the 400/2.8S TC w/BI TC @ 560mm:

Here are the wide open and best test images from the 400/2.8S TC w/1.4x TC @ 560mm:

Here are the wide open and best test images from the 400/2.8S TC w/BI TC + 1.4x TC @ 784mm:

Here are the wide open and best test images from the 400/2.8S TC w/2x TC @ 800mm:

Here are the wide open and best test images from the 400/2.8S TC w/BI TC + 2x TC @ 1120mm:

Here are the wide open and best test images from the 800/6.3S PF @ 800mm:

Here are the wide open and best test images from the 800/6.3S PF w/1.4x TC @ 1120mm:

Here are the wide open and best test images from the 800/6.3S PF w/2x TC @ 1600mm:

Here are the wide open and best test images from the 800/5.6E @ 800mm:

Here are the wide open and best test images from the 800/5.6E w/1.25x TC @ 1000mm:

Here are the wide open and best test images from the 800/5.6E w/1.4x TC @ 1120mm:

Here are the wide open and best test images from the 800/5.6E w/2x TC @ 1600mm:

And, finally, here are the images I used for the focus breathing comments and calculations above:

About My Testing Method

Here is my test methodology. After the last test, I decided I needed to find a better target with enough detail that would be too fine to resolve, even at 1600mm from a fairly close range. I found a microscope testing target that had good variable components (radial lines of varying frequency), so I purchased that and taped it to the back of another target that is bright white foam board and perfectly flat. Then I measured the height of the center of a lens mounted on the tripod and placed the target at that same height, perfectly vertical (per a torpedo level), and perpendicular to the camera / lens location (per a drywall T square). (That’s all to ensure the entire target (small as it is) is flat with respect to the lens so there are no focus / depth of field issues.) I chose a distance of 23′, because most of my use of these telephotos will be on small birds, hopefully in that distance range, and because that specific distance was the correct distance for another test I was doing using Reikan Focal on some 300mm lenses for another purpose.

For each lens, I mounted it, aimed the lens at the center of the target, and then tightened everything down. With the longer lenses, I added an additional tripod supporting the front of the lens. (That makes a world of difference in damping vibrations!) I decided to use a standard set of exposures on all lenses, hoping to capture any noticeable differences in transmission, specifically ISO 64, 1/250, @ f/2.8 or equivalent. I used a wired remote shutter release. Shutter, aperture, and ISO were all manual. Focus was AF-S, pinpoint. VR was off, of course. And I always used the back screen. A note for next time: I could’ve (should’ve) cropped the frame to APS-C, since even at 1600mm the target was a small part of the center of the frame, but I didn’t think of it until I was in the middle of the tests. That would’ve saved tons of space and time.

The workflow I followed for every single shot was:

1. Tap shutter to make sure I wasn’t in Playback mode.
2. Set exposure. (Either initial wide open exposure, or stop down 1/3 stop, lengthen exposure by 1/3 stop.)
3. Make sure exposure was set right.
4. Tap the magnify button to maximum magnification.
5. Hit the AF button to focus.
6. Ensure the focus looks perfect.
7. Wait for the vibrations to stop.
8. Wait for a few more seconds.
9. On remote, activate shutter for 3-5 photos. (To mitigate any possibility of motion / lack of sharpness.)
10. Tap playback button to verify exposure, sharpness, etc.
11. Wash, rinse, repeat. 150x!

After all of the tests were done, I uploaded the nearly 700 photos to my Mac and then culled them with FastRawViewer at 200% to detect and remove any unsharp shots, and delete all of the rest of the (unneeded) duplicates. As it turned out, out of about 150 different sets of photos, only once was there ANY difference among them that needed the backup photos. Testament to a solid tripod, no mechanical shutter, and decent technique.

After culling, I brought the finalists into Capture One for assessment and comparison, all done at 400%. (It’s impossible to see small detail differences on a 5K 27″ display.) There were NO adjustments made to the images.