Longitudinal Chromatic Aberration 1X Macro Lens Test

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

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Lou Jost
Posts: 5933
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

For those of us who use Nikon lenses on non-Nikon bodies, these results without the automatic correction are the relevant ones and are extremely useful!!!!

Doppelgänger
Posts: 15
Joined: Sun Feb 11, 2018 5:20 pm

Re: Longitudinal Chromatic Aberration 1X Macro Lens Test

Post by Doppelgänger »

RobertOToole wrote:LoCA 1X Test

With a big pile of macro lenses at my disposal this weekend was my last chance to run a LoCA test before the 105 and 200 have to go back to Nikon. To keep things simple I did not include any Canon or Sony lenses this time but I plan on adding the Sony 90mm and a few Canon macro lenses at a later date.

Image

Lenses tested:

Sigma 50mm F2.8 EX DG Macro
Sigma 70mm F2.8 EX DG Macro
Rodenstock 75mm f/4 APO-Rodagon D
Schneider Macro Varon 85mm f/4.5
Nikon Scanner Nikkor ED 100mm
Sigma 105mm f/2.8 EX OS Macro
Nikon AF-S VR Micro-NIKKOR 105mm f/2.8G IF-ED
Voigtlander 125mm f/2.5 APO-Lanthar
Sigma 150mm f/2.8 EX OS APO Macro
Sigma 180mm f/2.8 APO EX OS Macro
Sigma 180mm f/3.5 APO EX Macro
Nikon AF Micro-NIKKOR 200mm f/4D IF-ED Lens

Test procedure:

Nikon D810 clamped on to a PRC-3 carrier on a PRL-12 rail with the target at 30 degrees on a tilt head lit by a single SB-R200 wireless flash. Camera and lens were moved on the rail to focus and to accommodate the changes in focal length. The flash and target never moved during the test. All files shot in RAW and converted in ACR with all lens correction off.
Again, much respect for your efforts (I've scrolled through several of your tests, do you ever sleep? ;))

While these tests are revealing in some sense, again I believe performing comparisons on realistic subjects gives a more comprehensive understanding of the optic.

RobertOToole wrote:Image

Voigtlander 125mm f/2.5 APO-Lanthar top row vs Sigma 150mm f/2.8 EX OS APO Macro at the bottom
I am not sure if you're familiar with the CV (Voigtlander) 125, but it's actually at its worst (CA-wise) @ 1:1 :)

Still, while there's some CA up this close, are you noticing the dramatic difference in sharpness, wide-open, between the CV and the Sigma?

The Sigma is blurry at 2.8 and barely is in focus by 5.6.
In fact, it appears the AF of the Sigma has back-focused on the rear lettering, rather than the mid-lettering as intended.
In comparison the CV 125 is razor-sharp, where it's supposed to be, even wide-open.
(Remember: you can correct CA in post, but you can't correct lack of sharpness ...)

Also, here again, there is no is no way to evaluate bokeh and rendering in this test, where the CV especially shines and the Sigma is ordinary.

RobertOToole wrote:Image

Sigma 180mm f/2.8 APO EX OS Macro vs Sigma 180mm f/3.5 APO EX Macro vs
Nikon AF Micro-NIKKOR 200mm f/4D IF-ED Lens
This one is interesting to me :)

Both Sigmas defeat the venerable Nikon in CA.

However, the Sigma 180 f/2.8 and Nikkor 200 appear sharper than the Sigma 180 f/3.5, while the f/3.5 appears to have better CA than either.

I also notice you're also comparing the Sigma 180 f/2.8 @ f/4 to the other two @ f/5.6 ...
and you're comparing the Sigma 180 f/2.8 @ f/5.6 to the other two @ f/8.
At equal apertures, the Sigma 180 f/2.8 is by far the sharper lens than the 180 f/3.5 (but about equal to the Nikkor).

And, here again, minor CA can be corrected in post but lack of sharpness not so much.

RobertOToole wrote:Any questions or comments just reply to the thread.

BTW anyone with any good ideas for a nice and easy Bokeh test, let me know!

Robert
RobertOToole.com
Closeuphotography.com

As mentioned before, I think a bokeh/rendering test would be much more useful.
While your CA tests are useful too, the sharpness (esp, at wide apertures) is being totally overlooked.

The entire beauty of macro images is contained in extreme sharpness matched against subtle bokeh/color rendering, so omitting these considerations from a test is quite a big hole.

The Sigma 150 has nice CA correction but it is unusably-soft wide open compared to the CV 125 and its bigger brother the 180 f/2.8.
Same story with the elder Sigma 180 f/3.5/Nikkor 200 compared to the Sigma 180 f/2.8.

If you attempted to render a natural stack in the field (@ f/ 2.8 in perfect light), with a Voigtlander 125, and then tried this with a Sigma 150, you'd understand why the former has a 2-decade following but the Sigma not so much.
The sharpness/bokeh/subtle color transitions are part of what make macro lenses special, which are not taken into account here.

Every lens has its strengths and weaknesses, as well as its sweet spot, which is where taking the time to understand each lens is important also. Really getting to know a lens, and then using it where it shines, is part of the story.

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

Re: Longitudinal Chromatic Aberration 1X Macro Lens Test

Post by RobertOToole »

Doppelgänger wrote:
Again, much respect for your efforts (I've scrolled through several of your tests, do you ever sleep? Wink)
Thanks for your comments, positive and negative.
Doppelgänger wrote: While these tests are revealing in some sense, again I believe performing comparisons on realistic subjects gives a more comprehensive understanding of the optic.
Remember these findings are with my lenses (the nikons were borrowed) on my set-up, so your results may vary even with the same optics. With lenses as good as these, lets be honest, they will all make very fine photos. Over the last year or so I have moves away from nature subjects since they like to curl, warp, bend as you photograph them ruining the test in the process. Also man made targets like wafers with more contrast and fine detail show chromatic aberrations and softness differences more readily than a leaf or butterfly wing.
Doppelgänger wrote: I am not sure if you're familiar with the CV (Voigtlander) 125, but it's actually at its worst (CA-wise) @ 1:1 Smile

Still, while there's some CA up this close, are you noticing the dramatic difference in sharpness, wide-open, between the CV and the Sigma?

The Sigma is blurry at 2.8 and barely is in focus by 5.6.
In fact, it appears the AF of the Sigma has back-focused on the rear lettering, rather than the mid-lettering as intended.
In comparison the CV 125 is razor-sharp, where it's supposed to be, even wide-open.
(Remember: you can correct CA in post, but you can't correct lack of sharpness ...)
The lens are made for difference markets and have difference strengths and weaknesses to be sure. One has AF and OS and costs about $1000 USD. The other is out of production is MF and no OS ( BTW I tend to turn off OS on my lenses anyway) and costs $1700-2000 last time I checked. So the Sigma will be better for some photographers and the APO-Lanthar is better and preferred by others.
Doppelgänger wrote: Also, here again, there is no is no way to evaluate bokeh and rendering in this test, where the CV especially shines and the Sigma is ordinary.
That would be hard you are right, especially since I didnt test for Boken or rendering.
Doppelgänger wrote: This one is interesting to me Smile

Both Sigmas defeat the venerable Nikon in CA.

However, the Sigma 180 f/2.8 and Nikkor 200 appear sharper than the Sigma 180 f/3.5, while the f/3.5 appears to have better CA than either.

I also notice you're also comparing the Sigma 180 f/2.8 @ f/4 to the other two @ f/5.6 ...
and you're comparing the Sigma 180 f/2.8 @ f/5.6 to the other two @ f/8.
At equal apertures, the Sigma 180 f/2.8 is by far the sharper lens than the 180 f/3.5 (but about equal to the Nikkor).
The lenses were shot wide open to f/8 or f/11 and the sharpest images were selected and compared.
Doppelgänger wrote: And, here again, minor CA can be corrected in post but lack of sharpness not so much.

As mentioned before, I think a bokeh/rendering test would be much more useful.
While your CA tests are useful too, the sharpness (esp, at wide apertures) is being totally overlooked.
One reason is that I do not shoot wide open. Also none of the lenses was sharpest wide-open. Thats enough reasons for me. You are welcome to run your own test for wide open sharpess, rendering and Bokeh. Be sure to share the results.
Doppelgänger wrote: The entire beauty of macro images is contained in extreme sharpness matched against subtle bokeh/color rendering, so omitting these considerations from a test is quite a big hole.
It would be a big hole if this was a bokeh / rendering test but it is not.
Doppelgänger wrote: The Sigma 150 has nice CA correction but it is unusably-soft wide open compared to the CV 125 and its bigger brother the 180 f/2.8.
Same story with the elder Sigma 180 f/3.5/Nikkor 200 compared to the Sigma 180 f/2.8.
I don’t think the images shot with the Sigma at f/2.8 that are unusable but lens samples can and do vary.
Doppelgänger wrote: If you attempted to render a natural stack in the field (@ f/ 2.8 in perfect light), with a Voigtlander 125, and then tried this with a Sigma 150, you'd understand why the former has a 2-decade following but the Sigma not so much.
The sharpness/bokeh/subtle color transitions are part of what make macro lenses special, which are not taken into account here.
I see micro-contrast, transition, sublime atmospheric tests posted online all the time but I don’t think I was ever able to read or watch one through until the end so I can’t comment.
Doppelgänger wrote: Every lens has its strengths and weaknesses, as well as its sweet spot, which is where taking the time to understand each lens is important also. Really getting to know a lens, and then using it where it shines, is part of the story.
Agreed.

Lou Jost
Posts: 5933
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Doppelganger, you've stated several times that CA can be corrected in post, while lack of sharpness cannot. However, the type of CA that Robert's tests were designed to reveal, LoCA, is not readily correctable in post. That's why Robert's tests here, which are aimed just at that aspect of the image, are especially useful.

Also, sharpness can indeed be increased in post to a certain extent.

I surely agree that this is just one aspect of an image, and that other aspects also need to be taken into account when choosing a lens for a particular job. I am sure Robert was not implying that LoCA is the only criterion for lens choice.

ray_parkhurst
Posts: 3402
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: Longitudinal Chromatic Aberration 1X Macro Lens Test

Post by ray_parkhurst »

Doppelgänger wrote: The entire beauty of macro images is contained in extreme sharpness matched against subtle bokeh/color rendering, so omitting these considerations from a test is quite a big hole.
While this describes an aesthetic subset of macro imaging, it's certainly not the entirety. For me, macro encompasses a broad range of applications: coins; bugs and flowers ("every flower has a bug"), IC Failure Analysis; Phonograph Stylus evaluation; Product Marketing images; Iridology. Many of these are more technical than aesthetic, so sharpness and color accuracy is of prime importance. It is these applications where a test of LoCA is so important, since it cannot be corrected in post processing.

billjanes1
Posts: 91
Joined: Fri Dec 30, 2016 1:59 pm
Location: Lake Forest, IL, USA

Post by billjanes1 »

Lou Jost wrote:Doppelganger, you've stated several times that CA can be corrected in post, while lack of sharpness cannot. However, the type of CA that Robert's tests were designed to reveal, LoCA, is not readily correctable in post. That's why Robert's tests here, which are aimed just at that aspect of the image, are especially useful.

Also, sharpness can indeed be increased in post to a certain extent.
Robert did indicate that LoCA can be removed by stacking with Zerene (and presumably with other stacking algorithms), which is interesting and previously unknown to me. Since stacking is often used with higher magnification, LoCA is less problematic in this situation.

One limitation of his test series is that he tested at 1:1, and many times we use our "macro" lenses at 1:2 or less. Jim Kasson tested at 1:2 and got better results with the Nikon 105 f/2.8 VR.

Bill

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

Post by rjlittlefield »

billjanes1 wrote:... LoCA can be removed by stacking with Zerene (and presumably with other stacking algorithms)...
That statement is too strong. LoCA is often reduced by focus stacking, because it shows most strongly in slightly OOF regions that are mostly selected out so they do not appear in the in-focus slab. But color casts can still remain, either through information leakage by PMax or through retention of OOF regions in the completed stack.

See http://www.photomacrography.net/forum/v ... 9228#99228 for an example of single-frame versus completed stack (PMax).

--Rik

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

Post by RobertOToole »

billjanes1 wrote:
Robert did indicate that LoCA can be removed by stacking with Zerene (and presumably with other stacking algorithms), which is interesting and previously unknown to me. Since stacking is often used with higher magnification, LoCA is less problematic in this situation.

One limitation of his test series is that he tested at 1:1, and many times we use our "macro" lenses at 1:2 or less. Jim Kasson tested at 1:2 and got better results with the Nikon 105 f/2.8 VR.

Bill
Good points Bill.

I would like to run some tests at 1:2 and even 1:3 since that is where I personally shoot most but I haven't been able to find the time. I started at 1x since I thought it would be more popular and it is tougher test for lenses, as you mention.

All the best,

Robert

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

Post by RobertOToole »

rjlittlefield wrote:
billjanes1 wrote:... LoCA can be removed by stacking with Zerene (and presumably with other stacking algorithms)...
That statement is too strong. LoCA is often reduced by focus stacking, because it shows most strongly in slightly OOF regions that are mostly selected out so they do not appear in the in-focus slab. But color casts can still remain, either through information leakage by PMax or through retention of OOF regions in the completed stack.

See http://www.photomacrography.net/forum/v ... 9228#99228 for an example of single-frame versus completed stack (PMax).

--Rik
Thanks for the note Rik, I like your clear explanation.

Will include a similar note in the future, thanks

Robert

Lou Jost
Posts: 5933
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

I often have LoCA causing colored areas in my stacks, most notably in photos of layers of white butterfly scales on the edges of wings. That's why I prefer these complex 3-d wings as resolution targets rather than flat wafers. With perfectly flat subjects, stacking may indeed eliminate most LoCa.

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

Post by RobertOToole »

Lou Jost wrote:I often have LoCA causing colored areas in my stacks, most notably in photos of layers of white butterfly scales on the edges of wings. That's why I prefer these complex 3-d wings as resolution targets rather than flat wafers. With perfectly flat subjects, stacking may indeed eliminate most LoCa.
Hi Lou,

Quick notes on flat wafer tests.

In most cases, forget about 50x, but in most cases I am able to get the subject stage and sensor aligned so I can use a single frame with a flat-field lens.

I run stacks from way out, to in focus, to way out again just in case I need 2 or 3 frames to get sharp corners but with scanner lenses for example, one frame is usually enough.

This saves me lots of processing time.

In the Scanner Nikkor test it was easy to grab an OOF frame from the stack to show LoCAs, well there were no LoCAs, so in the case of the scanner Nikkor, the APO (or Super-APO) correction.

Robert

billjanes1
Posts: 91
Joined: Fri Dec 30, 2016 1:59 pm
Location: Lake Forest, IL, USA

Post by billjanes1 »

RobertOToole wrote:
billjanes1 wrote:
Robert did indicate that LoCA can be removed by stacking with Zerene (and presumably with other stacking algorithms), which is interesting and previously unknown to me. Since stacking is often used with higher magnification, LoCA is less problematic in this situation.

One limitation of his test series is that he tested at 1:1, and many times we use our "macro" lenses at 1:2 or less. Jim Kasson tested at 1:2 and got better results with the Nikon 105 f/2.8 VR.

Bill
Good points Bill.

I would like to run some tests at 1:2 and even 1:3 since that is where I personally shoot most but I haven't been able to find the time. I started at 1x since I thought it would be more popular and it is tougher test for lenses, as you mention.

All the best,

Robert

Robert,

I certainly do appreciate your work and recognize the amount of effort involved. Time is limited and one can't test everything. One risk of posting one's tests online is that no matter what you do, someone will criticize and request more or a different technique, sometimes tactfully and occasionally not so politely. Since your posts are well received, this indicates that you must be doing something right. Keep up the good work.

Regards,

Bill

Lou Jost
Posts: 5933
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Hi Robert and Bill, I hope you didn't read my comment as criticism. On the contrary, as I said in my earlier comment on this thread, these LoCA tests are rare and valuable and deeply appreciated!

I brought up the wafers just because of the discussion about stacking making LoCA go away. I think stacking might make LoCA go away for such flat surfaces as a wafer, but LoCA may still show up on stacks of subjects with complex overhanging layers. The single slanted frames presented here nicely show which lenses are likely to be free of this problem.

ray_parkhurst
Posts: 3402
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

Most of the time I don't want to shoot stacks of coins, at least for full-coin pics. Any remedy stacking can give is not useful for single images. Interestingly, for copper coins that I usually shoot, there is so much color that CA is lost in the overall color balance of the shot. In my 80mm shootout I did not even attempt to assess the LoCA of the lenses because the colors obscure the effect. The opposite is true of Silver coins. These are very unforgiving of CA. Perhaps I will use a silver coin in the future and include LoCA analysis, as this has been very elucidating.

Doppelgänger
Posts: 15
Joined: Sun Feb 11, 2018 5:20 pm

Re: Longitudinal Chromatic Aberration 1X Macro Lens Test

Post by Doppelgänger »

RobertOToole wrote:One reason is that I do not shoot wide open. Also none of the lenses was sharpest wide-open. Thats enough reasons for me. You are welcome to run your own test for wide open sharpess, rendering and Bokeh. Be sure to share the results.
It's true most lenses are not at their sharpest wide-open.

However, some lenses are very sharp wide-open, and incredibly-sharp stopped-down just a to f/4, while others enter into their own at f/8, while still others never really reach an acceptable level of sharpness at any aperture.

What makes the Voigtlander 125 such a great macro lens is the fact it is unusually-sharp wide-open and razor-sharp by f/4 (just look at it compared to the Sigma).

When you add to this exceptional bokeh and color-rendering qualities, along with something like 600-degrees of focus throw, you have a lens that is capable of delivering terrific images, which really separate the subject from the background (which is beautifully-rendered), especially when light stacks are employed.

It's an effect that a lens which is blurry wide-open, with ordinary bokeh, will never be able to achieve.

When a lens that is soft at f/2.8, and can't even get to a decent level of sharpness until f/8, you're going to find yourself at an aperture that will be including a lot of unwanted background distraction. The effect and presentation will be much different.

A test of black letters on white paper can never tell the full story that needs to be told on a lens comparison. Yet in even this test, the difference in sharpness wide open between these lenses is literally night and day.

Image

Post Reply Previous topicNext topic