Nikon Z 180-600 initial critical testing.

[sk66]

I received my 180-600 this morning and I've done some critical testing comparing it to my Sigma 60-600... I've taken a lot of excellent images with that lens and I was quite impressed with it when I first got it and tested it (initial review, technical testing).
Did a zeiss siemens star test with both on the Z9, checking for decentering, and resolution at 15ft. I then used a Normen Koren test chart at 50ft with both lenses. I would say both lenses performed better on the Z9 than the 60-600 did on the D850. And I would put them as being comparable in resolution; neither showed evidence of decentering.

This is the 180-600 showing no notable offset blur/decentering.


I'm guestimating the 180-600's MTF50 at right around 50 lp/mm, and in excess of 100lp/mm MTF10. That is for a raw file demosaiced in LR, and with all edits disabled (no default sharpening/noise reduction/etc) which is a little better than the Sigma 60-600 did on the D850 way back when (similar results, but camera processed jpegs). The reason it is a guestimate is because I did a visual determination and didn't use Imatest/ImageJ to do a direct measurement. It is also limited by the resolution of the printer that I used at that time to make the chart (6yrs ago)... which doesn't seem that great now.

Due to forum limitations I'm choosing to share heavily zoomed screen shots.
This is the 200lp/mm limit of the test chart taken from 50ft; it's at 800% zoom on my computer where the image pixels are becoming very apparent. The settings used are shown (wide open at 600mm). The chart was only at half the distance it should have been, so the l/mm measurements are 50% of scale.





Next is the center of the siemens star projected on my computer monitor taken from 15ft; this time in focus, also shown at 800% zoom. What's interesting here is that you are actually seeing the RGB leds that make up the pixels on my monitor. I cannot see them with the naked eye. The monitor is a MBP liquid retina XDR with a pixel density of 230ppi; that puts the RGB sub pixels at about 690dpi. I was a bit shocked by this, but the Sigma also did just as well. This was also taken wide open at 600mm with a 1/1250 SS as above, but the ISO was 3600 because it was taken indoors... I guess I should have increased my monitor brightness to give it a bit better chance. It does seem to show a slight bit of LaCA (the red/blue fringes on opposite sides), but nothing major (the 60-600 was nearly identical). There might be some color moire as well; I'm not sure.




All of the testing was done on a tripod, with a locked down gimbal head, VR on, 5 second timer, and pinpoint AF. I certainly won't be expecting to get the same kind of results consistently in the field; but it won't be the equipment's fault.

I'm not testing it against my 400/2.8, 120-300/2.8, or any other lens... that's not why I bought it. It is a bit of a tough call because it doesn't seem to outperform my Sigma 60-600 (@600) notably. But it's over a pound lighter, has full VR integration , and has function buttons/integration I want (recall focus). It's also cheaper than the Sigma 60-600 was 8 years ago and has less focus breathing.

So far I'm happy; at least I know I got a decent copy. Now I need to get some time in the field with it... As long as the AF isn't particularly slow or something similar, I think I'll be keeping it.

FWIW, the Sigma 60-600 MTF50 tests a little better than both the 500pf and the 200-500 at max FL, it's also at least equivalent to the Z 100-400 at the long end/max aperture... so keeping up with that isn't bad.

 

[Bristolian]

Steven, thank you very much for sharing this information. This lens could well be the reason that I move to mirrorless.

 

[Sootchucker]

Thanks Steven, very interesting.

Well something weirds happening with my copy. As I posted over on the Nikon Z thread, I wasn't impressed with mine when I got it and images at 600mm were noticeably soft and much less acuity than my 500mm PF and my 100-400Z (with and without the 1.4 Teleconverter). Well, before I boxed it up to return to Nikon for my money back, I thought I'd give it one last try, and was very surprised how much sharper they were then when I first tested it. Unlike Steven's controlled shots, all mine were handheld and some in pretty atrocious light. These are much better. Not show stopping images by any means, but I think I'll need to give it a stay of execution ?

All at 600mm AUTO ISO, and on the Z9 body.

 

[sk66]

The mirrorless cameras are very dependent on lens aperture for light/AF; and there is a reason the Nikon Z's only autofocus at the selected aperture down to f/5.6.

With mirrorless, as the lens aperture gets smaller the amount of light (and contrast) available for AF is reduced... this is bad. As the aperture gets smaller, the virtual images used for PDAF (and the image for subject rec) are more in focus initially; which speeds up AF and makes PDAF even possible... this is good. And, as the aperture gets smaller the angular offset of the virtual images used for PDAF becomes smaller, so the phase offset becomes smaller and the phase shift is quicker; this makes PDAF both quicker and less accurate.

The difference between f/5.6 and f/6.3 is minimal so I don't think that really explain the differences you saw; especially compared to the zoom with TC's (that makes little sense to me). The only thing I can really say is that this lens is certainly working with some handicaps when it comes to low light/low contrast scenes compared to the faster lenses. It's not a lens I would choose for low light/indoors.

These are much better. Not show stopping images by any means, but I think I'll need to give it a stay of execution ?

They look decent to me.

I think it is reasonable to test the equipment as you intend to use it to see if it works well for you.
But that kind of testing introduces the potential for a wide variety of errors, and little way to know which caused an issue. It also creates very little consistency between shots/setups... the smallest changes can make a difference; even the WB setting set (probably very small though). Subject detection and what you point the camera at adds a whole bunch of other unknown variables into the equation.

For my testing I eliminated a lot of the variables to try to see the len's full potential. But I will probably never use the lens exactly that way in the field, so I can't reasonably expect to consistently get the same level of quality.