Some technical tests with the 60-600 vs 120-300/2.8 and 300/4 +TCs

[sk66]

All images were taken with a tripod and in live view using CDAF with evenly overcast skies. First is the focus breathing at MFD and 600mm as compared to the 300/4 +2x

300/4 +2x


60-600


I measured the MFD at 8.25ft to the focal plane, but that requires a manual override to get the last few inches. And these were not taken at exactly the same focal plane distance due to differences in tripod foot placement. My best measure is that the 60-600 image was taken at 8.5ft and the recorded FOV was 7". That puts the focal length right at 525mm which is a pretty notable hit from 600mm. The chances of ever using 600mm from 8ft away is pretty slim, but it is still short of a true 600mm until right about 28-30ft. That's probably acceptable for most things, but still significant. FWIW, this is a little better than the 150-600mm offerings from the few reports/tests I can find.
Also worth noting is that the 60-600mm is nearly perfectly parfocal; the focus does not shift notably when zooming from/to either extreme.

However, the 60-600 @ f/6.3 image is much sharper that the 300/4 +2x @ f/8
60-600 100% crop

300/4 +2x 100%crop


I didn't test to see if/when the 300/4 +2x might catch up in resolution as it's a lens I don't use and plan on selling. But my best guess is that it won't be able too due to diffraction on the D850.
I also noted that 300/4 f/8 image also required one stop higher ISO than the 60-600 f/6.3 image did. That indicates to me that the 60-600 f/6.3 aperture is accurate (the other 1/3 stop appears to be due to composition and resulted in slightly different exposures).

Next was some resolution tests compared to the 120-300/2.8 + 2x TC... This is what I was really concerned with because if there were no gains here then I really didn't get a lot of benefit from this purchase. Luckily, that's not the case. Note that the resolution chart is not to scale (ignore the LP/mm scale). The images were taken at ~ 30ft where the 60-600 is at a true 600mm.

The first thing to note is that the 60-600 image is slightly tighter than the 120-300 image... that means either the 120-300 has more significant focus breathing, it's never truly a 300mm, or the Sigma 2x is slightly less than a true 2x (I haven't tested any of that).


continued...

 

[sk66]

Next are the 100% crops starting with the 60-600 @ f/6.3... it's resolving pretty much to the limits of the print/printer.


the 120-300 +2x @ f/5.6 is notably softer with less contrast (whew!)


continued...

 

[sk66]

the 120-300 +2x @ f/8 shows a notable improvement, but it is still lagging behind the 60-600.


the 120-300 +2x @ f/9 actually shows a slight reduction in IQ if anything, which indicates no more improvements are possible due to diffraction... which I did verify.


Now I'm really impressed with this lens overall!

 

[chris malcolm]

i can't comment on the lenses, but congrats for some well devised lens resolution experiments!

 

[sk66]

My best estimate at visually interpreting the MTF chart places the recorded resolution at ~ 50 lp/mm @ 50% contrast; maybe a little higher. I think that's in-line with reports because this was taken as a jpeg. Even though I have my jpeg settings set for reduced camera processing (negative sharpening/contrast/etc) some sharpening and contrast was still applied which will increase the results. Probably reducing the results a little is the slight underexposure and indirect/dull lighting. Either way, I think the 40 lp/mm from a raw file is very probable from my copy of the lens (as tested/reported by LensTip.com).

(I don't own Imatest in order to take a direct measurement)

 

[sk66]

Removed due to being in error.

 

[sk66]

I made a quick video showing the focus breathing from 525mm (@MFD) to 600mm (@infinity). I had to use a .25" aperture mask (f/94) in order to make it clearly visible... the shallow DOF at 600mm/8.5ft makes it extremely hard to see even at the lens' min aperture. The aperture mask is in the threaded filter position which is why there is a little vignetting at 525mm.

View: https://youtu.be/puZC55-K0HM

 

[RichardC27]

A
The first thing to note is that the 60-600 image is slightly tighter than the 120-300 image... that means either the 120-300 has more significant focus breathing, it's never truly a 300mm, or the Sigma 2x is slightly less than a true 2x (I haven't tested any of that).
View attachment 140247

continued...

I have read a couple of reviews that state the 120-300mm actually maxes out at about 280-290mm, it's not a true 300mm at the long end. Interesting comparison, I'm very impressed that Sigma have got the 60-600mm so sharp at the long end.

 

[sk66]

I have read a couple of reviews that state the 120-300mm actually maxes out at about 280-290mm, it's not a true 300mm at the long end. Interesting comparison, I'm very impressed that Sigma have got the 60-600mm so sharp at the long end.

Doesn't surprise me that it might only be 280mm... hard to test @ infinity though.
I'm going to test the sharpness with the 1.4x (EX-DG) to see what it can do... but I suspect being at f/8.8 will really impact AF significantly.

 

[sk66]

Tested with the 1.4x DG, which is not supposed to be firmware compatible. The lens reports as f/9 and autofocuses on the D850, although it really struggles focusing on fine details even with the central focus point... probably due to the significant reduction of image contrast. However, the resolution is still quite impressive and contrast can be increased/improved in post.

This was again taken at 30ft as a jpeg with lower picture settings (but some contrast/sharpening was still applied in camera), CDAF, tripod, ISO 1600.

100% crop from the upper right corner of the test chart.


The fine lines at the right edge are hard to discern with the naked eye even at a distance of < 2ft... that's pretty impressive. However, the camera could not focus on these lines using PDAF... it could focus on the lines near the left side of this crop. I did not test AF tracking, but it was quite apparent that it will be significantly degraded... not to mention the significant loss of available focus points, and even the available focus points are (probably) somewhat vignetted/hampered. There's also no room for stopping down any further to increase sharpness/contrast due to diffraction on the D850.

Cropping or using an extension tube (to reduce the focus breathing) are clearly better options than using the TC IMO.

Here is a similar crop from the 600/6.3 test image upscaled to the same dimensions (LR scaling at export). ISO 500.


FWIW, these images have not been edited for color/contrast/sharpening in post; the slight purple tint is present/visible in the print I made of the test target. Additionally, the 600/6.3 image has a 1.5 stop light/ISO advantage in this comparison (on sensor, ~.5 stop in the output image).
It should be noted that the TC used was not the latest "global vision" TC-1401, which is the one that is supposed to be used/compatible with this lens. If I had used the TC-1401 it might have made some small difference, but I highly doubt it would be significant in any way. And I'm not about to spend $350 on one in order to find out...

 

[sk66]

I did some tests with it on my D850 yesterday while out looking for Harriers/SEO's... I put the AF point selection to auto and placed high contrast objects at various locations w/in the scene. Below ~200-300mm zoom (≤ f/5.6) the camera would happily/quickly/accurately use any/all of the 153 focus points. Beyond ~300mm the accuracy/speed began to drop off. And by 600mm it would not use/select points beyond the central ~72 AF point area at all, other than the odd momentary wander. Trying to force it to focus using AF points outside the central area was abysmal, resulting in nothing but hunting and failed focus.
I'm not saying it absolutely can't use the outer AF points at 600mm... in optimal (studio type) test conditions it might have some success.

But in the field I would focus on central compositions. IMO that's not much of an issue because most wildlife/sports/action is central AF/composition. But it's going to struggle more with small/fast/hard to track subjects. It's also worth keeping this in mind because the Nikon AF system will allow you to select any of the AF points; whether they are usable/functional or not.

The reason I chose to use auto AF point selection is because it's one of the few modes where the 153pt system shows/reports the actual focus point used (w/o having to extract/read the detailed EXIF). I also figured that if the AF module can see/use the focus point effectively it would, and if it couldn't it would try to use another point instead... that's exactly what happened; at 600mm the more outer focus points were pretty much completely abandoned/ignored.