Sigma vs Nikon vs Voigtlander CA test @ 1X

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

RobertOToole
Posts: 2627
Joined: Thu Jan 17, 2013 9:34 pm
Location: United States
Contact:

Sigma vs Nikon vs Voigtlander CA test @ 1X

Post by RobertOToole »

Updated this post with new results:http://www.photomacrography.net/forum/v ... 441#219441

This is a follow up to a sharpness test I posted last week:

Sigma 150 vs Voigtlander 125 APO-Lanthar Sharpness Test @ 1X

http://www.photomacrography.net/forum/v ... hp?t=35261

Some of the results of that test showed more lateral CAs in the corners than expected. So I ran a quick test today at 1X to check again.

Rodenstock 75mm f/4 APO-Rodagon D
Schneider Macro Varon 85mm f/4.5
Scanner-Nikkor ED lens
Nikon AF-S VR Micro-NIKKOR 105mm f/2.8G
Sigma 105mm f/2.8 EX DG OS Macro
Sigma 150mm f/2.8 EX DG OS APO
Voigtlander 125mm f/2.5 APO-Lanthar


200% corner crops

Nikon D810 1X Single SB-R200 Nikon flash ISO 64 vertical MM-11 stand
Images converted from RAW files in ACR with all lens corrections turned off and all defaults zeroed out.

This is the un-cropped image of a laser printed paper to show the 200% sample area.

Image

Rodenstock 75mm f/4 APO-Rodagon D
Image

Schneider Macro Varon 85mm f/4.5
Image

Scanner Nikkor ED
Image

Nikon AF-S VR Micro-NIKKOR 105mm f/2.8G
Image
Sigma 105mm f/2.8 EX DG OS Macro
Image
Sigma 150mm f/2.8 EX DG OS APO
Image
Voigtlander 125mm f/2.5 APO-Lanthar
Image

Rodenstock 75mm f/4 APO-Rodagon D
Schneider Macro Varon 85mm f/4.5
Scanner Nikkor ED


The first 3 lenses were basically free from lateral CAs and show impressive correction!

Nikon AF-S VR Micro-NIKKOR 105mm f/2.8G

The 105 VR was the worst of the lenses here with pronounced purple and green CAs but that is the case in the tests online so it shouldn't be a surprise, but I was still surprised. Stabilized lenses sometimes show element de-centering when the VR is turned off as it was here, this could be a contributing factor.

The corner sharpness was the weakest of this group also BTW.

Sigma 105mm f/2.8 EX DG OS Macro

The Sigma 105 shows slight lateral CAs. This level usually will not show up in actual photos of 3D subjects. I know since I use this lens all the time.

Sigma 150mm f/2.8 EX DG OS APO

The Sigma 150 very slight traces of red and blue CAs, less than most other lenses labeled APO and usually impossible to see with in-the-field subjects. This is one of my most used lenses.

Voigtlander 125mm f/2.5 APO-Lanthar

The APO-lanthar CA results match the sharpness test. There are CAs in the images with this sample. Its sharp but not as well color corrected as the Sigmas, but better than the 105 Nikkor. Not out of line for a lens designed more than 15 years ago!

Notes:

These are all my own lenses, except for the 105 VR that is on loan from Nikon and the APO-lanthar that is on an extended loan thanks to a friend.

My plan was to also test for Bokeh Fringing, or LoCAs (longitudinal CAs) but I ran out of time, but I do plan to run that at a later date.

All of the lenses were shot wide open except for the two Sigmas, the Nikon and the Voigtlander, these were shot at f/4.

Questions or comments? Leave a reply below.

Robert

www.Closeuphotography.com
Last edited by RobertOToole on Sun Sep 24, 2017 10:26 pm, edited 1 time in total.

mjkzz
Posts: 1681
Joined: Wed Jul 01, 2015 3:38 pm
Location: California/Shenzhen
Contact:

Post by mjkzz »

wow, thanks!

RobertOToole
Posts: 2627
Joined: Thu Jan 17, 2013 9:34 pm
Location: United States
Contact:

Post by RobertOToole »

mjkzz wrote:wow, thanks!
Glad you like it.

Planned on running a LoCA or bokeh CA test but I ran out of time!

I will find time this week and update the post.

Robert

Chris S.
Site Admin
Posts: 4037
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Post by Chris S. »

Well done, Robert! :D

To my eye, your laser-printed test target is revealing and easy to evaluate. (Maybe a training effect, as laser printed paper, pasted to a microscope slide, is my go-to target for lens testing at 20x and below.) Laser printing is cheap, widely available, and flat—would be my candidate for “universal test subject”—or as close to universal as seems likely in the real world.

I’m unpleasantly surprised at the amount of chromatic aberration you’re demonstrating. It seems that the Sigma and Voigtlander lenses should strike “APO” from their names. Your Schneider Macro Varon looks wonderful.

A few thoughts:

Comparing your Nikon micro-Nikkor 105mm f/2.8 (current generation "AF-S VR") against my micro-Nikkor 105 f/2.8 (one generation prior "AF-D"): What a difference! I’ve read that the newer lens is optically poorer, but not guessed this magnitude. My lens exhibits essentially no CA, and much higher resolution at 1:1. While there exists the issue of sample variation, one would expect your test lens to be a decent specimen, since it was lent you by Nikon for testing.
All of the lenses were shot wide open except for the two Sigmas, the Nikon and the Voigtlander, these were shot at f/4.
An understandable choice. But might it be better to shoot an aperture series for each lens, then compare each lens at its best aperture? For my use, this would more closely model lens performance under most conditions. Also, am willing to grant some forgiveness to a lens that renders well when stopped down two or three stops, but provides a bright viewfinder and snappy focus wide open.
Rodenstock 75mm f/4 APO-Rodagon D
Schneider Macro Varon 85mm f/4.5
Scanner Nikkor ED

The first 3 lenses were basically free from lateral CAs and show impressive correction!
My eye agrees that against the other lenses you tested, these three appear comparatively free from CA. This said (knowing that individuals vary in tolerance of CA), only your Schneider Macro Varon appears free of CA to my eyes. I see bothersome CA in the two others.
200% corner crops
Might you describe in more detail how you produced your crops? I’d like to post a test complementary to yours, and would try to include crops as comparable with yours as possible.

Cheers,

--Chris S.

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

Robert, Chris,

Thanks for an interesting comparison. My Nikon 105/2.8 VR doesn't seem to be up to the level of these other lenses.

I had the older Nikon 105/2.8 D version and sold it long ago to a friend when I got the VR version. I recall the D did seem to be slightly better in sharpness. Over at The-Digital-Picture.com lens compare site, the D shows more CA and maybe very slightly better sharpness. I'm wondering if the Nikon D3 body is removing the CA from the VR in these lens evaluations?

https://www.the-digital-picture.com/Rev ... &APIComp=2

Comparing with the Sigmas is not directly plausible because they are evaluated on a Canon body, but comparing the Sigma 100 vs. 150 favors the 100 slightly.

The Nikon 105 VR looks so bad in sharpness and CA, agree maybe it's a bad copy.

Best,

Mike

RobertOToole
Posts: 2627
Joined: Thu Jan 17, 2013 9:34 pm
Location: United States
Contact:

Post by RobertOToole »

Chris S. wrote:Well done, Robert! :D
.......I’m unpleasantly surprised at the amount of chromatic aberration you’re demonstrating. It seems that the Sigma and Voigtlander lenses should strike “APO” from their names. Your Schneider Macro Varon looks wonderful.
Honestly this is a unforgiving high contrast test, the majority of macro photographers would never notice the CAs using the Sigma 150 or the Voigtlander with 3D subjects in the field. I think the CAs would be very hard to spot with normal use.

Generally speaking I think the CAs that show up in this test (other than the Nikon 105) are very mild compared to almost all consumer market, non industrial specialized lenses. For example the Zeiss 100mm F/2 MAKRO Planar T has really noticeable CAs, close to the same level as the Nikon 105 VR.


Chris S. wrote:A few thoughts:

Comparing your Nikon micro-Nikkor 105mm f/2.8 (current generation "AF-S VR") against my micro-Nikkor 105 f/2.8 (one generation prior "AF-D"): What a difference! I’ve read that the newer lens is optically poorer, but not guessed this magnitude. My lens exhibits essentially no CA, and much higher resolution at 1:1. While there exists the issue of sample variation, one would expect your test lens to be a decent specimen, since it was lent you by Nikon for testing.
As I mentioned in the notes in the test I have read that the VR lenses complicate testing since when switched off, the VR floating group can lead to element de-centering. Some of the Sigma lens manuals tell the user to switch off OS before you turn the camera off. If you fail to do that you might here a rattle coming from the lens. I asked the head of Sigma service about it and he says when you switch off OS a little centering fork comes out to stabilize the group. Since I almost never us VR or OS when I shoot (or test) this might be a factor.

Chris S. wrote:...might it be better to shoot an aperture series for each lens, then compare each lens at its best aperture?For my use, this would more closely model lens performance under most conditions.
That is why I decided to shoot at f/4 since it seemed to be the sharpest aperture at least with the sharpness test I ran recently. For that test I did run each aperture from wide open down to f/8 for each lens. There was a drop off in fine detail with the Sigma and Voigtlander at f/5.6 - f/8 at 1X.

Then there is the question of best aperture for sharpness, CAs or corner IQ? I am sure they will be different apertures for each.

Believe it or not the Nikon 105VR (and the Sigma 180 not shown in the test) shows more CAs as you stop down!

The point you bring up is a good one, I don't think many people think about that and it is one of the reasons people need to take care when they read test reports on lensRentals.com. They only shoot wide open with their macro lenses MTF tests (and 1:2 maximum).

Chris S. wrote: My eye agrees that against the other lenses you tested, these three appear comparatively free from CA. This said (knowing that individuals vary in tolerance of CA), only your Schneider Macro Varon appears free of CA to my eyes. I see bothersome CA in the two others.


Really, that is interesting!
Chris S. wrote: Might you describe in more detail how you produced your crops? I’d like to post a test complementary to yours, and would try to include crops as comparable with yours as possible.
Sure.
Open the files
Stack the images into one new file (be sure to label the layers)
Align slightly if needed
Zoom in to 200% view
Create guides to use as a crop reference
Screen capture each layer being careful to label each with the name of the lens

Thats it.

Robert

RobertOToole
Posts: 2627
Joined: Thu Jan 17, 2013 9:34 pm
Location: United States
Contact:

Post by RobertOToole »

mawyatt wrote:Robert, Chris,

Thanks for an interesting comparison. My Nikon 105/2.8 VR doesn't seem to be up to the level of these other lenses.

I had the older Nikon 105/2.8 D version and sold it long ago to a friend when I got the VR version. I recall the D did seem to be slightly better in sharpness. Over at The-Digital-Picture.com lens compare site, the D shows more CA and maybe very slightly better sharpness. I'm wondering if the Nikon D3 body is removing the CA from the VR in these lens evaluations?

https://www.the-digital-picture.com/Rev ... &APIComp=2

Comparing with the Sigmas is not directly plausible because they are evaluated on a Canon body, but comparing the Sigma 100 vs. 150 favors the 100 slightly.

The Nikon 105 VR looks so bad in sharpness and CA, agree maybe it's a bad copy.

Best,

Mike
Hi Mike,

The site says they have all aberration correction turned off for testing
(https://www.the-digital-picture.com/Help/ISO-12233.aspx)

I could very well could be a bad copy. I have shot with the 105 VR before and it did show CAs but I don't remember how bad.

Looking around online, the test on Photozone the 105VR has pronounced lateral CAs. LensTip saw no problem with CAs. On Mark Goodman's site he reports "moderate fringing on the periphery - worsens with increasing magnification"

LensTip.com didn't see a CA issue so maybe its the VR off issue, or it could be sample variation.


Robert

Chris S.
Site Admin
Posts: 4037
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Post by Chris S. »

RobertOToole wrote:
Chris S. wrote:Well done, Robert! :D
.......I’m unpleasantly surprised at the amount of chromatic aberration you’re demonstrating. It seems that the Sigma and Voigtlander lenses should strike “APO” from their names. Your Schneider Macro Varon looks wonderful.
Honestly this is a unforgiving high contrast test, the majority of macro photographers would never notice the CAs using the Sigma 150 or the Voigtlander with 3D subjects in the field. I think the CAs would be very hard to spot with normal use.
Recall that CA degrades images in two ways: In addition to producing color fringes, CA also smears detail by spreading it out. If a lens has CA, it will smear details even with subjects that hide color fringes. A benefit of your unforgiving test is that it makes obvious an issue that can degrade results even with forgiving subjects.
RobertOToole wrote:. . .I have read that the VR lenses complicate testing since when switched off, the VR floating group can lead to element de-centering. . . . Since I almost never use VR or OS when I shoot (or test) this might be a factor.
I'm not quick to give Nikon a pass for adding a feature to a lens that degrades results in most situations where that lens is used. My Nikon 80-400 zoom has VR, and I appreciate this feature in this lens. For me, VR makes hand-holding very doable down to 1/100 second or so. But on a tripod, or at shutter speeds faster than 1/800, VR absolutely must be turned off, and will ruin the image if this is not done. OK, not a big problem with that lens. And with that lens, I don't see noticeable image degradation with VR turned off.

But a macro lens? In my use, macro images are made with tripod or flash the vast majority of the time. This seems to be true for you as well, and probably most other macro photographers. So adding VR--if VR degrades images when turned off--seems a poor idea.
RobertOToole wrote:
Chris S. wrote: My eye agrees that against the other lenses you tested, these three appear comparatively free from CA. This said (knowing that individuals vary in tolerance of CA), only your Schneider Macro Varon appears free of CA to my eyes. I see bothersome CA in the two others.

Really, that is interesting!
Differences in visual perception are interesting to me as well, and I have a hunch that learning to see as various others do has practical benefits.

Care to try seeing the three "CA-free" lenses as my eyes do?
  • 1. If so, look again at the Macro-Varon image; note that the left and bottom edges of the black box are a neutral gray.

    2. Now look up at the Rodenstock image. Are the left and bottom edges neutral gray? Not to my eye--I see a red or purple cast.

    3. Now look down at the Scanner Nikkor image (this one is harder to see than the Rodenstock). My eye sees a slight warming along the left and bottom edges.
Do you see any of this?

This is admittedly very picky, but is something my eye does naturally. And with your monochrome test subject, any deviation from neutral gray means that details are being smeared in some portion of the spectrum.

Cheers,

--Chris

rjlittlefield
Site Admin
Posts: 23543
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Chris S. wrote:Care to try seeing the three "CA-free" lenses as my eyes do?
  • 1. If so, look again at the Macro-Varon image; note that the left and bottom edges of the black box are a neutral gray.

    2. Now look up at the Rodenstock image. Are the left and bottom edges neutral gray? Not to my eye--I see a red or purple cast.

    3. Now look down at the Scanner Nikkor image (this one is harder to see than the Rodenstock). My eye sees a slight warming along the left and bottom edges.
Do you see any of this?
Yes, but I don't interpret it the same way that you do. I see it as a matter of color contrast rather than chromatic aberration.

Consider that the radial direction in the test area extends from lower right to upper left. Moving outward along that radial line, the bottom of the black area is a bright-to-dark transition, while the left side is the reverse, dark to bright.

It seems to me that any radial CA should have opposite effects in these two regions, like we see in the cases where there are obvious fringes.

But you're reporting the same color shift on both edges.

To my mind, that suggests that you're detecting a slight difference in overall tone from one lens to another, that difference happening to be more obvious when comparing a dark region with a nearby light region that is less saturated.

--Rik

RobertOToole
Posts: 2627
Joined: Thu Jan 17, 2013 9:34 pm
Location: United States
Contact:

Post by RobertOToole »

Chris S. wrote:
....In my use, macro images are made with tripod or flash the vast majority of the time. This seems to be true for you as well, and probably most other macro photographers.


100% right.
Chris S. wrote: So adding VR--if VR degrades images when turned off--seems a poor idea.


I believe the VR on this lens is used as a selling point, but it would help for portraits with the 105VR.

BTW that idea of VR off leading to de-centering is from a post by one of the technicians from OLAF on the LensRental bog.

Canon's 100 IS is the only lens that I know of, there maybe more out there now, that senses when the lens is being used for close-ups and shifts into a high res VR mode that you can supposedly use for macro work.
Chris S. wrote: Differences in visual perception are interesting to me as well, and I have a hunch that learning to see as various others do has practical benefits.

Do you see any of this?
I need to take another look :D

RobertOToole
Posts: 2627
Joined: Thu Jan 17, 2013 9:34 pm
Location: United States
Contact:

Updated with New Lenses

Post by RobertOToole »

Update, Sun Sept 24

Added some interesting results from a few lenses. Same test as the previous.

200% corner crops

Nikon D810 1X Single SB-R200 Nikon flash ISO 64 vertical MM-11 stand
Images converted from RAW files in ACR with all lens corrections turned off and all defaults zeroed out.

Sigma 70mm F2.8 EX DG Macro
Sigma 180mm f/2.8 APO EX OS Macro
Nikon AF Micro-NIKKOR 200mm f/4D IF-ED



Sigma 70mm F2.8 EX DG Macro at f/2.8

Image

Sigma 70mm F2.8 EX DG Macro at f/4

Image

Sigma 70mm F2.8 EX DG Macro at f/5.6

Image

Sigma 180mm f/2.8 APO EX OS Macro at f/2.8

Image

Sigma 180mm f/2.8 APO EX OS Macro at f/4

Image

Sigma 180mm f/2.8 APO EX OS Macro at f/5.6

Image

Nikon AF Micro-NIKKOR 200mm f/4D IF-ED at f/4

Image

Nikon AF Micro-NIKKOR 200mm f/4D IF-ED at f/5.6

Image

Nikon AF Micro-NIKKOR 200mm f/4D IF-ED at f/8

Image

Whats interesting in these new crops?

As far as lateral CA control, the Sigma 70mm beats both the Sigma 180 and Nikon 200. The Sigma 70 design uses 3 UD glass elements!

The Sigma 180 and Nikon 200 both show an increase in lateral CAs when stopped down. I knew about the Sigma 180, but the Nikon surprised me a little. No Its not a mistake, the CAs get worse when stopped down.

Any questions or comments, just reply below.

Robert
www.RobertOToole.com
www.Closeuphotography.com

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

Robert,

Interesting. The Sigma 70 behaves as expected, or at least from my experience. However, the Sigma 180 and Nikon 200 don't improve with smaller aperture which I didn't expect!!

The Sigma 70 is certainly the softest of the three, so maybe that's the lens design tradeoff.

Best,

Mike

RobertOToole
Posts: 2627
Joined: Thu Jan 17, 2013 9:34 pm
Location: United States
Contact:

Post by RobertOToole »

mawyatt wrote:Robert,

Interesting. The Sigma 70 behaves as expected, or at least from my experience. However, the Sigma 180 and Nikon 200 don't improve with smaller aperture which I didn't expect!!

The Sigma 70 is certainly the softest of the three, so maybe that's the lens design tradeoff.

Best,

Mike
Hi Mike,

I knew about the Sigma 180 since I own one, but when I saw the results from the 200 Nikon, I double checked the EXIF info, I thought I mixed up the shot order. No I was right.

I agree with you, the Sigma 70 is a little soft in the far corner at f/2.8. In my experience where I shoot most, f/5.6 or f/8, the 70mm would be tough to beat. Working distance could be better and that is the only factor that keeps me from shooting with it more often but then again, I have seen these sell for $250 in the box on Ebay, when you consider the price, you can't go wrong.

Robert

Post Reply Previous topicNext topic