1:2 photo - Nikkor 300mm+17E TC versus Sigma 180mm lens

[LVF]

The following is long reading, but for those that own the colored corrected APO MACRO Sigma 180mm f/2.8 EX lens or those who are thinking of buying this lens, there is a very interesting result that you should know about this lens, if you do not already know.

On July 7th, I posted the 1:2 photo resolving power of the Sigma APO MACRO 180mm f/2.8EX DG OS HSM lens at 16 inches working distance. On July 16th, I posted the 1:2 photo resolving power of the Nikkor AF-S 300mm f/4E PF ED VR with an attached Nikon 17E II teleconverter at 48 inches working distance. Both photo systems mounted of the Nikon D500 camera. By "working distance", I mean the distance from in-front of the lens to the resolution chart. By 1:2 photo, I mean 2 times the Nikon D500 camera sensor size, which gives a photo size equal to 1.85x1.24 inches.

I have been requested to compare the above two photographic systems for taking a 1:2 photograph, one being close-up and the other 3 times further away from the subject being photographed. I prefer the further distance, but I decided to compare, in detail, how good the two systems are at making 1:2 photographs.

Note - To be more correct, the Sigma was 1:2, but the Nikkor was 1:2.2 which produced a 2.04x1.36 inch photo compare to a 1.85x1.24 inch photo for the Sigma. I am being picky because when I compare the two systems side by side, the Nikkor photo will appear slightly smaller and some people may wonder why the difference in size.

To make the comparison on equal grounds, I decide to take new photographs of my resolution chart with each system, rather than relying on past photographs. The resolution chart photographed is described in the June 9th post.

Here is a photo of the center section of the resolution chart that is photographed at 1:2 magnification. I added dimensions for this posting to show how small the width of the bars are; 0.004 inches (0.1 mm) is equal to the diameter of the average human hair.



I did the following steps to make the comparison as equal as possible. I examined each systems previous post to determine each systems sharpest f stop; it turns out both were the sharpest a f/8, so I used f/8 with both. It should be noted that Robert (RobertOToole) has posted that the Sigma 180mm lens is sharpest at f/5.6.

I mounted the two systems on a tripod and used live view and manually focused both systems. I used the Nikon MC-36 cable release to release the shutter a few seconds after removing my fingers off the lens focus ring. I set the camera at ISO 1600 inorder to use a fast shutter speed:

For the Sigma lens with OS off - f/8 at 1/1000 second,
and
For the Nikkor/TC lenses with VR off- f/8 at 1/640 second.

Man, the Sigma lens is big and heavy compared to the Nikkor lens and teleconverter. I had to mount the Sigma lens directly to the tripod, whereas, I mounted the D500 camera on the tripod when using the Nikkor/TC lenses. I compare their sizes at the end of the post.

I took three photographs with each system, being extremely careful to get the sharpest focus with each photograph.

I took raw Nef photographs. I opened the Nef files in DxO Optics Pro and exported the Nef files as Dng files: I did zero processing in DxO, especially no sharpening.

I open the Dng files in Camera Raw CS6. I processed the two system Dng files exactly the same in CR CS6: set exposure to +2.5ev, black slider to -100, shadow slider to -60, highlights to +100, whites to +8, clarity to +30, sharpening to 0, and noise reduction to 0. I did no other processing in CR CS6. I exported the processed files as Tiff files to folders on my desktop.

I opened the Tiff files in Photoshop CS6. In Photoshop, I examined the three photos from each system at 200% magnification to determine which of the three were the sharpest. I used the sharpest of the three from each system for the comparison. Other than selecting the sharpest photo, I did zero processing in Photoshop. I used Photoshop for "file>save for web" to meet the file size requirements of this forum.

Here is the 1:2 photo taken with the Nikkor 300mm f/4E PF lens plus the Nikon 17E II teleconverter mounted on the Nikon D500 camera at f/8, 1/640 sec., ISO 1600, at 48 inches working distance:




To view the photo closer, I cropped the 5568x3712 pixel photo file to 512x512 pixels:



The set of bars at #6, have a width of 0.004 inches (0.1 mm), the average thickness of the human hair. Considering this photograph was taken 48 inches from the resolution chart, the Nikon system is amazing.

If you look closely at the above 512x512px crop, you will see clusters of various colors in the numbers and on the edges of some of the bars. The photographic chart used to take the resolution photographs, was printed on glossly photoraphic paper using an Epson R3000 printer. The blacks in the photo were made up of 8 colored inks, variations of Cyan, Yellow, and Magenta inks.

To help you to see these colored ink spots, I cropped the 5568x3712px photo to 256x256px and resized it up to 1024x1024px:




The printing of the bars and numbers was not perfect, so you see these clusters of colored ink spots, primarily cyan and magenta, with a few brownish yellow spots. These colors you see are apparently not caused by chromatic aberration. Look at the numbers 5 and 6, there are clusters of cyan on the left and right sides of the numbers which is not possible with chromatic aberration. Also, this photo was taken with reflected light not refractive light, so there should not be chromatic aberration.

Here are some findings which states that chromatic aberration is nearly non-existent for this Nikkor 300mm f/4E PF lens:

1) Quoting Thom Hogan http://www.dslrbodies.com/lenses/nikon- ... -s-vr.html

"Where the E (300mm f/4E PF) starts to shine compared to the lens it replaces is in the other optical parameters. Chromatic aberration, both longitudinal and latitudinal, is very well corrected. Yes, some small amount of latitudinal chromatic aberration is present, but it’s at levels where I might not even try to correct it. On the longitudinal side, I can’t find enough to even comment on in actual use, which is excellent performance for a prime lens."

2) Quoting Andrew Alexander, SLRgear Review, October 5, 2015, imaging-resource.com:

"Chromatic Aberration. The test results for the Nikon 300mm ƒ/4E PF with regard to chromatic aberration were very impressive - the lens is highly resistant, on both of our Nikon D7000 and D800e test bodies, across all apertures."

3) Quoting Gary Wolstenholme, 6 Mar 2015, ephotozine.com:

"Levels of chromatic aberrations are very well controlled, remaining comfortably below 0.5 pixel widths towards the edges of the frame. This low level of fringing should be pretty much impossible to spot in normal image taking, even with areas of high contrast towards the edges of the frame."

These clusters are made up of individual ink spots that are 2 picoliters in size, or 0.0005 inches (0.01mm) in diameter. Obviously, any individual ink spots on the edges of the bars or numbers are going to be a blur.

Moving on, here is the 1:2 photo taken with the Sigma APO MACRO 180mm f/2.8 EX lens mounted on the Nikon D500 camera at f/8, 1/1000 sec., ISO 1600, at 16 inches working distance:



To view the photo closer, I cropped the 5568x3712 pixel photo file to 512x512 pixels:



Now for the comparison between the two systems. I combined the two 512x512px photos side by side in one file using Photoshop for ease of comparison:



Comparing the two systems 512x512px crops, shows that the Sigma 180mm lens is sharper than the Nikkor 300mm lens with the 17E II TC. This result was expected since the Sigma lens is a colored corrected MACRO lens, it should be sharper. Note - the Nikkor photo is slightly smaller than the Sigma photo because it is a 1:2.2 photo whereas the Sigma is a 1:2 photos, as discussed above.

But can Photoshop improve the Nikkor 300mm photograph to make it exceptable and comparable to the Sigma 180mm lens? For those that maybe interested, I discuss this at the end of this comparison.

NOTE - the following was not expected, and was not planned to be in this comparison write-up, but I believe it could be significant to some photographers using the Sigma APO MACRO 180mm lens.

I took the Sigma photo and cropped the 5568x3712 pixels to 256x256 pixels and resized it to 1024x1024 pixels:



Here is what really surprised me, it was totally unexpected,

"WHERE ARE ALL OF THE CLUSTERS OF COLOR INK?"

There are a few small spots of cyan, few spots of magenta, and a couple yellowish spots, but most all colored clusters are gone. Compare the numbers "5" and "6", all the cyan clusters are gone and replaced with black clusters. The Sigma APO MACRO 180mm lens is so well color corrected that it removed practically all the color clusters and replaced them with black. However, there is the possibility that the color elimination problem is with my copy of the lens and not all other copies.

The significance of this finding (if true) is that, if I am photographing a very colorful insect, bug, butterfly, with the Sigma APO MACRO 180mm lens, I may not record all the color spots on these critters, some colored spots may appear black!

But what if the Nikkor 300mm f/4E PF lens is seeing colors that are not there?

I took my Bausch & Lomb 7 times magnifier and placed it directly on the resolution chart surface and look at the charts' smallest bars and numbers. No good, the bars and numbers are to small to distinguish any colors. WHAT IRONY, I can easily make out the smallest details and colors in the resolution chart with the front of my 300mm f/4E PF lens "4 FEET" away from the resolution chart, but I cannot see these colors with my Bausch & Lomb 7x magnifier sitting on top of the chart! This Nikkor lens is really sharp!

So far, I have not answered the question "is the Nikkor 300mm lens seeing colors that are not there?"

Then I had an idea, I'll photograph the chart with my AF-S VR Micro-Nikkor 105mm f/2.8G IF ED lens and see if this lens sees the colors.

I took 3 - 1:2 photographs with the Nikkor 105mm f/2.8G lens at f/8, 1/1250 sec., ISO 1600, with the front of the lens 10 inches from the resolution chart. I used live view, manually focused, processed the Nef files exactly the same in DxO, and Camera raw CS6, and selected the sharpest of the 3 photos in Photoshop CS6.

Here is the best 1:2 photo taken with the Nikkor 105mm f/2.8G at f/8, 1/1250 sec,, iso 1600, at 10 inches from the chart



To view the photo closer, I cropped the 5568x3712 pixel photo file to 512x512pixels:



Looking close-up, I could discern what look like colors. I wanted to see the colors, so I took the Nikkor 105mm photo and cropped the 5568x3712 pixels to 256x256 pixels and resized it to 1024x1024 pixels:



The Nikkor 105mm lens recorded purples, greens, cyan, magenta, yellow, and some yellowish browns. Some of these colors seen by the Nikkor 105mm are replaced by black color in the Sigma photo. From this Nikkor 105mm photo, I conclude that the colors are in the photographic print.

So, I must assume until proven otherwise, the Sigma APO MACRO 180mm f/2.8 ED lens removes some colors that are not formed be chromatic aberration, inaddition to removing chromatic aberration colors.

For those that own the Sigma APO MACRO 180mm f/2.8 ED lens please chime in, I would like to here your opinion. Especially, from Robert (RobertOToole), who has owned several of the Sigma macro lenses.

Now, off topic from the comparison. The following is for my benefit, but it may be of interest to some who might be interested in the Nikon system.

I ask myself, is the Nikkor 300mm f/4E PF+17E II system 1:2 photographs taken at 4 feet, good enough to be comparable to the Sigma 180mm lens 1:2 photographs taken at 16 inches?

The above unsharpened photographs shows that the Sigma 180mm lens is sharper than the Nikon system. So I asked myself, "can Photoshop CS6 improve the Nikkor 1:2 photographs enough to made it comparable with the Sigma system?"

Here are my findings.

I opened the best Nikkor 300mm 1:2 Tiff photograph in Camera Raw CS6, and produced several modified photographs by varying the clarity, sharpening, and sharpening detail. I varied the clarity by +50 and +100, sharpening be 25, 50, and 100, and sharpening detail by 25, 50, and 100. This combination produced 18 modified photographs.

I viewed these 18 photos in Photoshop to find the sharpest, cleanest, and clearest photograph that was not over sharpened or made to crispy.

After spending a lot of time comparing these 18 photographs, I was left with 4 photos that were processed with clarity = 50 and 100, sharpening = 25, and sharpening detail = 25 and 50.

I took these 4 Photoshop processed photos and cropped each 5568x3712 pixel photo to 256x256 pixels and placed them in one photograph for easy comparison. I also included the original, unsharpened 256x256 pixel crops for the Nikkor system and Sigma system. Here is the combined photo:



Top center is the original, unsharpened 256x256px cropped photo taken by the Nikkor 300mm+17E II lens. Bottom center is the original, unsharpened 256x256px cropped photo taken by the Sigma 180mm lens.

The 4 outer photos are the Photoshop processed photos of the 256x256px cropped photo taken by the Nikkor 300mm+17E II lens. They show that with minor processing in Photoshop, I can get good enough 1:2 close-up at 48 inches with the Nikon system.

But wait, the Sigma owners are saying, you can also process the Sigma photo in Photoshop to make it better, and right they are. Here are the Photoshop processed photos of the Sigma 180mm1:2 photos:



You can get some really nice sharp photos with the Sigma APO MACRO 180mm f/2.8 EX lens with a little sharpening in Photoshop.

Here is a side by side comparison of the Photoshop processed 1:2 photos for the Nikon system and Sigma system for minimum and maximum chosen processing parameters.



The Sigma APO MACRO 180mm lens is the clear winner, Photoshop sharpened or not Photoshop sharpened. However, the Nikkor 300mm+17E II TC combo looks good when minimally processed in Photoshop.

I am satisfied that I will get good enough 1:2 photographs with the Nikon system at 48 inches working distance. I will not have to kneel on the ground, what a relief!

For those that are curious as to how big the Sigma APO MACRO 180mm lens is compared to the the Nikkor 300mm f/4E PF + 17E II TC lenses, here is a photo of them side by side, I threw in the Nikkor 105mm f/2.8G also:



The Sigma 180mm lens weighs in at 3.6 lbs (1.63 Kg), and length 8.0 inches (200mm) vs. 2.2 lbs (1 Kg) and 7 inches (178mm) for the Nikkor 300mm+Nikon 17E II TC.

At times I will use the Sigma 180mm lens but when I do, the lens will probably be on a tripod.

Well, I thought this post ends a long series of posts showing the studies I have been doing to find a photographic system that will take good enough close-up photographs at a distance that keeps me from kneeling on the ground.

However, I am still puzzled by why my Sigma 180mm lens replaced some colors with black color, colors that my Nikkor lenses see. Now I plan to try to create a color graphic in Photoshop, that can be used to test the Sigma 180mm lens, Nikkor 300mm lens, and Nikkor 105mm lens. A test to see if colors are correctly recorded in 1:2 photographs.

Until that post, I think I will take a break.

Leon

[Lou Jost]

Thanks for the exhaustive tests. This is a nice confirmation of the quality of the Sigma 180. I am completely puzzled by the desaturated colors, though. It will be interesting to track down the reason for that.

[LVF]

Thanks Lou

I too am puzzled. I looked at the larger bars in the photos and found the same results, desaturation of colors. The only explanation I can think of is these colors are extremely small and may look like chromatic aberration to the lens. But why replace the colors with black?

Leon

[Lou Jost]

A lens doesn't replace colors, there must be some other explanation. I assume your RAW files are not getting any automatic correction based on a lens profile, judging from your workflow description, so I have no idea what the answer is!

[rjlittlefield]

For what it's worth, here's an animated comparison of the Nikkor and Sigma images, created by just cropping the images that LVF posted. Clearly there's a lot more going on than just desaturating the colors. The black stuff is moving around all over the place too. In fact there are hardly any common elements between the two, when it comes down to details of dot placement. If these really are from photographing the same piece of paper, then the whole process has to be suspect.



--Rik

[Lou Jost]

Could there be a large exposure difference, such that the desaturated image is just much more exposed than the Nikon image? This would wash out the colors and make islands out of some of the irregular chunks of black seen in the Nikon image.

[rjlittlefield]

Beats me. If I wanted to create those two images from the same original target, I would have no idea how to do it. Look at the top right section of the 6 -- one of the images has a big white hole in the middle of it, and the other one doesn't. Or the top left section, where one of the images has a big bold chunk of black in the middle of the stroke, and the other one has nothing like it.



--Rik

[Lou Jost]

Yes, you're right. I give up. We'll have to wait for Leon to solve the mystery.

[Grahame]

Hi folks,
This is from the OP

"For the Sigma lens with OS off - f/8 at 1/1000 second,
and
For the Nikkor/TC lenses with VR off- f/8 at 1/640 second. "

I was puzzled as to why when I first read it.
Is it enough to cause the difference ??

[Lou Jost]

We know Nikon reports effective apertures when using Nikon lenses, but maybe this doesn't work reliably for third-party lenses? and maybe this somehow leads to confusion about proper exposure?

[LVF]

Just got back from 3 day trip to Spokane, Washington, granddaughter's wedding.

Rik and Lou. The two photographs are of the same printed resolution chart, with the chart located in exact location (did not move the mounted chart), and under same Ott light located in the same position relative to the chart, and at the exposures given in the above post. I took the photos with the two lenses about 2 hours apart.


Could the fact that one lens was at 16 inches from chart and the other lens at 48 inches from the chart, have anything to do with the difference? Does reflective color light rays travel at an angle so that the two systems are seeing different perspectives of the charts reflected light rays?

In fact, the more I think about this, the following could be the answer. The photo sensitive particles in the three layers in the glossy color print paper are distributed at many different angles so the light rays are reflected at many angles. So it is highly possible that different viewing distances will see different micro differences in color, and in sharpness, one position sees micro holes another position sees color or black at the same position in the paper. Thus producing the observed results. Makes sense to me.

Quoting Wikipedia,

"Structure

All photographic papers consist of a light-sensitive emulsion, consisting of silver halide salts suspended in a colloidal material - usually gelatin- coated onto a paper, resin coated paper or polyester support. In black-and-white papers, the emulsion is normally sensitised to blue and green light, but is insensitive to wavelengths longer than 600 nm in order to facilitate handling under red or orange safelighting.[2] In Chromogenic colour papers, the emulsion layers are sensitive to red, green and blue light, respectively producing cyan, magenta and yellow dye during processing."

Obviously, the two lenses resolve the same chart differently because of the difference in position of the lenses. They see color light rays and sharpness differently, because the light rays travel at an angle off the print paper surface..

I can repeat the tests if you think something is wrong with these photographs, let me know.

I have been wondering if the inclusion of the 17E II teleconverter affects the results of the Nikkor 300mm lens. Looking back at my previous posts, I was reminded that the use of 68mm of Kenko extension tubes with the Nikkor 300mm f/4E PF lens produces a 1:2 photo at 48 inches. Because there are no additional optics added when using the extension tubes, I may get different results.

Therefore, I plan to photograph the same resolution chart with the extension tubes and Nikkor 300mm lens under the exact conditions used for the above. Probably will get a third puzzling result to muddy the mystery.

It was a 12 hour trip back to Sequim from Spokane (360 miles plus a ferry), do not travel on highway 2 to Seattle on a Sunday, 3 stoplights in very small town of Sultan caused miles and miles of cars backup, moved 3 to 4 feet then stopped, thousands of times. Going to rest-up, so results of future tests will be in a few days.

Leon

[rjlittlefield]

The photo sensitive particles in the three layers in the glossy color print paper are distributed at many different angles so the light rays are reflected at many angles.

In earlier threads, you wrote about resolution targets that were printed on an Epson R3000 inkjet printer.

Was the target for this thread printed on something different?

--Rik

[LVF]

Rik & Lou

This resolution chart was printed on a Epson R3000 printer which use 8 inks (5 colors and 3 blacks). It prints black with these 5 color inks. I am thinking about printing the chart with just the 3 black inks which I can do with this printer.

Rik, lets remember dimensions. The white hole you refer as a big white hole, is on the order of 0.0004 inches (00.01mm). The Sigma lens is sharp enough at 16 inches to see those inkless white holes. The Nikkor lens is sharp, but at 4 feet, its not that sharp. The Nikkor lens cannot resolve those micro white holes at 4 feet, and the 300mm will not focus any closer than 4 feet. Look at the graphic showing a comparison between the Sigma lens and Nikkor lens, the Sigma lens sees several micro white holes in the 0.004 inch wide bars and printed numbers, but the Nikkor lens cannot resolve these micro holes.

Just got back from the YMCA, I am tired, so no photographs today. Lets keep the dialogue going. This has resulted in quite an interesting discussion.

Leon

[JohnKoerner]

Honestly, this should be a no-brainer.

According to LenScore, the $1,900 Nikkor 300mm f/4E PF isn't even as good as the $1,700 Sigma 180 'APO' Macro (loosely-defined as such), as-is, let alone with a TC 1.7.

Both the budget Nikkor and the Sigma rate in the 900s, quality-wise. The Nikon has an advantage in reach, being 300mm, while the Sigma has an advantage in macro (which, as its label suggests, it ought to).

Why don't you rate the Sigma 180mm with an extender, and see how it performs at 300mm, compared to the Nikkor PF at its native 300mm? The Sigma will fail too.

Or, if you want the best of both worlds, try the $5,400 Nikkor 300 f/2.8 VR II (which rates 400 points higher than the Sigma), at 1:2, with an extender, and let me know how your tests come out

This Nikkor will beat the Sigma at 1:2 and it will beat the pants off of the Sigma at 300mm too.

I owned the Sigma 180 (ahem) 'Apo' Macro and it is not a true Apo lens ... whereas the Nikkor 300 VR II most assuredly is.

[Lou Jost]

...the vastly-superior Nikkor 300mm f/2.8 VR II will beat the Sigma with an extender, at 1:2...

John, is that particular conclusion based on your actual experience? Theoretically, a dedicated macro lens optimized for m=0.5 could beat even an excellent long internally-focusing telephoto lens on extension tubes. The longer lens is probably optimized for m<<0.1 and many internally focusing long lenses are not designed for extension tubes ( Nikon warns against using ext tubes even on their dedicated macro 200mm IF lens).

Of course, your real-life experience trumps such theoretical generalities, if you did that experiment.