AFS Nikkor 105/2.8 VR CA issue

[MEPER]

I have not yet upgraded to the Z-version so still use the old AFS version. It is known to have a bit of CA when used for macro work or maybe in generel. When used for portraits I have been quite pleased with the results so maybe I have just not observed the CA.

I used it recently for at quick handheld macro shot of a detail on an old drawing instrument.
First the AF really confused me so I switched that off. Then the constant active noisy VR also confused me so switched that off also. Then I could concentrate on the image.
I wonder if it is possible to program the body so VR is only active when release button is pressed half way?
On newer lenses the VR is very silent but on this old lens I have a feeling that it drains the battery rather quickly and I don't like the constant noise.

But with VR off I also have a "feeling" that CA maybe will be a bit reduced compared to VR on. Could there be some technical explanation behind this?
If VR is on then a lens element or group is moving around so it will be a little bit decentered?

I have not made any experiments at all so just a thought.

The attached image was shot with both AF / VR off and I had expected some CA in the highlights but image seems to be quite ok for a quick handheld shot.

[Birna Rørslett]

First of all, are these jpgs generated in camera? If that be the case, CA probably has been removed by software. Secondly, if NEFs are processed by Nikon software, are you sure CA wasn't removed there?

Thirdly, the [spectral] quality of the light also will influence how dominantly CA is manifested.

I don't use Nikon software on a regular basis and was so frustrated with the CA of my AFS 105mm f/2.8 Micro-Nikkor that I sold it off after three months and replaced it with the Voigtländer APO-Lanthar 125mm f/2.5. No AF, but far superior in terms of colour handling.

The 'Z' 105mm f/2.8 MC Nikkor also is a much better performer than the older AFS 105 in terms of having low or negligible CA.

[MEPER]

It is NEF processed in NX Studio. It seems Lateral CA correction is on (default setting). Snip attached.
But I tried both on/off with no visible change in image. Also attached one of the images with CA adjust off. Still "clean" regarding CA?

Then I can ask a question about NX-Studio. Something I have never found out.
The first snip where CA correction is on.......why is check-mark in right corner not "marked"?    .....is this because it was off all the time?
Then when I switched off the CA check mark then the check mark in right corner switched on (this ship is also attached).
I think this is generel very confusing in NX Studio.
My own conclusion is that the software has worked in the first cast with CA-correct on even that the check mark in right corner is not checked and that it is an error in the software. But also strange that if it is an error that it has not been corrected.
The logical meaning would be that the check-mark in the right corner was there to enable / disable all the features checked-on inside this "frame".

My sample of AFS 105/2.8 sample is an older "Made in Japan" version.

Regarding VR on/off it seems that on a Z-body VR is active all the time and not just when shutter is pressed half way.

Image attached is with CA-off.....

[ColinM]

Hi Meper
I remember reading Birna's review of that version of the 105 macros at the time and feeling depressed
(I'd just ordered the lens...)

I had the advantage of never owning any of the previous nikon macros, so never knew what I was missing.

I loved being able to take reasonable macros and the CA only became an issue under certain conditions.
I also found I liked the results of using the 105VR in non-macro mode.

Here, with some CA



Here, using it with a TC1.4 for BIF, the results were better than I dared expect



The shots above were all shot with F mount bodies (D50 I think and D300 the second)
I take it you're using a Z body & FTZ?

I also struggle with NX studio.
I could just about manage with Capture NX, but less so now

[Erik Lund]

I bought the 105mm AF-S 2.8 VR and returned it more or less immediately, I found it had way too much CA, blooming, purple fringing for my usage. I also took the APO Lanthar 125mm f/2.5 in after borrowing Birnas lens, and adapted it to use on Nikon F with a new mount and dedicated CPU. I continued to use also the Micro Nikkor 105mm f/4 Ais for simplicity and IR performance
Now all is replaced with Micro Nikkor  MC 105mm f/2.8 S - Far superior in all aspects!
VR can create double lines in the oof areas,,, so in principle it can mess with CA but not in a helpful way I guess,,,

[MEPER]

I think I will challenge the lens to get the best out of it. Then we will see if I can live without the Z-version. Nikon has made quite many of these. Most "Made in China" it seems. The 2nd hand value as a private sale is not that high anyway. The "Made in Japan" version is probably same 2nd hand price.

Regarding VR I was wondering if the VR lens group was out of optical axis when image was shot. It was my assumption and my reason to propose that maybe CA could be affected by this. Normally we like lens elements to be perfectly aligned?

I found a bit of technical VR information here:
https://www.nikonimgsupport.com/na/NSG_article?articleNo=000050844&configured=1&lang=en_SG

It seems that just before exposure the VR lens group resets to have as little off-axis error as possible and to have room for maximum VR adjustment during the exposure. So how much off-axis there is during exposure depends on how much camera shake there is.

This may also explain why the image in the viewfinder jumps around just after the exposure.

When we see cross section schematics of lenses the VR lens group is not "marked"?

[Erik Lund]

This site has some break down on optical design for many lenses, here link to 105mm AF-S 2.8 VR

https://www.photonstophotos.net/GeneralTopics/Lenses/OpticalBench/OpticalBench.htm#Data/US007218457_Example03P.txt,figureOpacity=0.27,AxisO,MarginalO,OffAxis,Pupils,AxisZ,AxisR

And here, not sure what lens this is,,,

https://www.photonstophotos.net/GeneralTopics/Lenses/OpticalBench/OpticalBench.htm#Nikon/US005751485/Example09.txt,figureOpacity=0,MarginalO,OffAxis,Pupils,AxisZ,AxisR,AxisO

[MEPER]

Thank you!
The two links are identical?

[Erik Lund]

I have changed the first link now 
Basically you choose from the drop down list of lenses, however if you want to link it's best to use the link button,,,

[Akira]

When SIGMA released the latest 105mm macro lens (105mm f2.8 DG DN Macro for mirrorless cameras), Mr. Yamaki, CEO of SIGMA, explained that they employed the optical design that is slower to focus but shows less LoCA.  He compared its optical design with that of 105mm f2.8 EX DG OS HSM (for DSLR) and esplained that the optical design of the latter one was commonly employed for the micro/macro lenses:

https://platinumaps.jp/d/newotani-tokyo?area=1&floor=2F-G6F&gm=99

(around 33:28.  The upper cross sections are focused at infinity, and the lower ones, at the minimal focusing distances)

AF-S VR 105/2.8 and AF-S 60/2.8 seem to share the same optcal design method, which should be why they suffer from LoCA.  It's the inherent problem of the optical design of this type.

[ColinM]

Interesting Akira.
In the early 2000's Sigma started to get a reputation for the quality of their 150mm Macro.
I came very close to getting one.

Once I'd moved to Nikon bodies, I decided to stick with their lenses and unfortunately bought the 105mm model Meper has been talking about

The Sigma 150mm would have been a good choice for me as even 105mm is barely enough working distance for some of the living subjects I was interested in. It's performance as a 150mm wasn't too bad, so it would have been a versatile tool. I guess there will be a few F mount copies going cheap now, as people move to the Z mount.

When SIGMA released the latest 105mm macro lens (105mm f2.8 DG DN Macro for mirrorless cameras), Mr. Yamaki, CEO of SIGMA, explained that they employed the optical design that is slower to focus but shows less LoCA.

In retrospect, you wonder why focus speed should be an overriding criteria for a macro lens.
I remember many people (Birna included?) asking why VR should be a requirement for a lens that would often be used on a tripod. Similarly AF was of some (but limited) benefit. As this period marked when accessible focus stacking tools became available, manual focus or moving the lens became the mode of operation for many.

[Birna Rørslett]

Still many people believe AF is really necessary for a "macro" lens. Photomacrographers know this is not the case and there are good reasons for using manual focus. Sometimes the AF functionality impairs manual focusing and then is a drawback, but fortunately most modern 'macro' lenses do well in manual mode even if they are AF-capable.

I'm not convinced VR is a benefit or a drawback, however if the feature can be easily switched off that point is moot anyway.

A long focus throw and precise focusing are on the other hand highly recommended features.

The Z MC 105/2.8 Nikkor is really good albeit could be smaller in size, at least if your hands are small (like mine). Although manual focusing is "by wire", it is pretty smooth so I'm not bothered when I focus that lens manually. However, compared to the buttery smooth focusing of the APO-Lanthar 65mm f/2, it falls dramatically behind. I reckon virtually all other lenses would fare in  similar fashion.

[Akira]

Interesting Akira.
In the early 2000's Sigma started to get a reputation for the quality of their 150mm Macro.
I came very close to getting one.

Once I'd moved to Nikon bodies, I decided to stick with their lenses and unfortunately bought the 105mm model Meper has been talking about

The Sigma 150mm would have been a good choice for me as even 105mm is barely enough working distance for some of the living subjects I was interested in. It's performance as a 150mm wasn't too bad, so it would have been a versatile tool. I guess there will be a few F mount copies going cheap now, as people move to the Z mount.

In retrospect, you wonder why focus speed should be an overriding criteria for a macro lens.
I remember many people (Birna included?) asking why VR should be a requirement for a lens that would often be used on a tripod. Similarly AF was of some (but limited) benefit. As this period marked when accessible focus stacking tools became available, manual focus or moving the lens became the mode of operation for many.

Armando used the very SIGMA 150mm f2.8 macro (the older version without "OS", the image stabilization), and his images with the lens showed virtually no LoCA, if I remember correctly.

I guess that the people required the faster focusing speed (and maybe VR also) to shoot the closeups of small and agile creatures like butterflies, bees or hummingbirds.  Also, the manufacturer thought the potential buyers of 105mm micro/macro would use the lens more for general purposes for landscapes and portraits as well as closeups.

[Michael Erlewine]

Ditto. I stopped using the 105/2.8 VR years ago and sold it.

LIke many, I find that the Z 105 MC Macro does a fine joh, IMO.

[MEPER]

Here is an example of CA taken with AFS 105/2.8 at f/2.8 at closets distance with VR off and AF off.
CA correction and sharpening off in NX-Studio. 1st image shows "front" blur and 2nd image "far" blur. There are some visible CA.
For very flat objects lens can be ok. If you are just interested what is in focus. E.g. just to document a detail which has nothing to do with "art" photographing.