On DPR, a poster asked what macro lens to get and got this reply:
I would definitely go with the Z 105/2.8. It’s better than the F mount 105/2.8 and probably even the best macro lens ever made.
This is a bit like saying that the 27mm f/5 Bio-Summagon is the best landscape lens ever made, or that the 85mm f/0.8 Apo-Sonnarcron is the best portrait lens ever made. The possible use cases for portrait and landscape lenses make picking just one best lens impossible. I think the range of macro use cases is even larger than the range of landscape or portrait uses.
Here are some of the dimensions:
Reproduction ratio. There are lots of definitions of what constitutes macro photography, and I’m going with a fairly — small c — catholic one here: between 1:10 and 10:1 reproduction ratio (the number before the colon is the size of the image on the sensor, and the number after the colon is the size of the object in the real world). No single lens is going to do more than a journeyman job across that hundred to one range. Macro lenses are often optimized for ranges like between 1:1.6 and 1: 3 (you could reverse the lens and probably get good performance at between 1:1.6:1 and 3:1). Many macro lenses made for industrial use (let’s call them process lenses) are optimized for a single reproduction ratio. The Rodenstock 105/5.6 HR Digaron macro lens has a adjustment ring that lets you optimize it for reproduction ratios between 1:3 and 3:1, which is a broad range, but you need to manually set that ring for good results.
Some macro lenses have floating elements to control aberrations as you focus. Be very careful when using those lenses in combination with extension tubes or reversed, which will defeat the optimizations designed into the lenses.
Reproduction ratio affects the maximum f-stop you should choose. If you’re going to use a lens at 1:3, f/5.6 probably fast enough, and diffraction probably won’t affect the results too much. If 3:1 is what you’re after, a faster lens would probably be sharper because of the effects of diffraction.
Subject depth. Lack of field curvature is not important for many macro applications, but is very important for things like copying flat art or digitizing negatives. Field flatness can interact with reproduction ratio. Case in point: the Fuji 120mm f/5.6 GF lens needs a 45mm extension ring to get to 1:1, but suffers from a lot of field curvature there. I happily used that lens at 1:1 for a couple of years before I tried to use it for copy work and got terrible results.
Focus bracketing. There are three ways to do focus bracketing
- Focus the lens.
- Move the sensor plane, but leave the lens fixed.
- Move the whole camera and lens assembly.
If you want to do #1, and have the bracketing take place automatically, you need a lens that the camera can control that way, which favors a native lens.
If you want to do #2, you need a view camera arrangement like the Cambo Actus. The Actus and the Cambo Ultima support the attachment of a stepper motor to the rear focus assembly to facilitate automation of this.
You can do #3 with just about any lens.
Turnkey solution or not. There are lots of macro lenses out there that use mounting arrangements that are unfamiliar to many photographers, and that have no focusing mechanism. Some of those lenses are spectacular for some uses. There are many — probably the vast majority — who have no interest in adapting those lenses to their cameras, because such adaptation would require them to MacGyver a way to attach and focus the lens, or purchase a bellows arrangement.
Autofocus. Personally, I consider autofocus to be useless for macro photography, but I know that there are many who disagree with me on that point.
Tolerance for aberrations. Especially when used outside of their optimum reproduction ratios, and especially off-axis, many macro lenses suffer from longitudinal chromatic aberration, lateral chromatic aberration, coma, and astigmatism. For some uses, these defects are highly important. For others, they are not a consideration al all.
All these variables complicate picking the right macro lens for a particular task, and make it impossible to say that any one lens is the best macro lens for all tasks.
Hi Jim, thank you for your fantastic work in your blog. Regarding the macro lens, I would need your expertise. Like you, I shoot (scan) negatives with the Fuji GFX 100s. Mostly color & BW negatives 35mm, 120s, 4×5 & 8×10 inches. I shoot the 35mm format with a single shot in multi-color shift mode. The 120 & 4×5 inch negatives most in two shots each in pixel & color shift mode and the 8×10 inch negatives in 4 shots each in pixel & color shift mode. I have already tested the Rodenstock 105mm HR Digaron where I was very satisfied. However, I had also heard very good things about the Schneider 120mm Apo-Digitar Macro. But did not have the opportunity to test it until now. So I’m not sure if the Schneider 120mm is better for my application since I only work above 1:1 on the 35mm and am under 1:1 range on the 120, 4×5 & 8×10. For me it is also important that the lenses are as distortion free as possible and that the sharpness is as equal as possible over the whole image area and that the lens has little chromatic aberration as possible. What do you think Jim, which of these two lenses would be better for my way of working. Thanks in advance for your time and assistance. Best, Nathanael
I’ve never tested the SK 120 Apo-Digitar Macro, just the Apo-Symmars. I find the Apo-Symmars just about as good as the Digaron, but you need three or four of the Apo-Symmars to do your job, and you’ve already got the Rodie, so I’d say go with what you’ve got.
I have not found distortion in either the SK or Rodie lenses that we’re talking about, but I haven’t looked for it, either. I do know that the edges of the negative carrier appear to be straight.
Hi Jim, thanks so much for your time. It helped me a lot and i will stay with the Rodie:)
Glenn Whorrall says
Hi Jim. I’m prepping for a project where we are photographing (and filming) the inner workings of some antique watches, often at 1:1
I have access to a cognisys focus rail but found focus stacking moving the whole camera and assembly (your option 3 described above) to give inconsistent results.
Stacking using the camera focus is becoming a more common feature in camera bodies but requires specific lenses, and as you’ve found with the Fuji 100S there aren’t many suitable lenses at 1:1 or beyond.
I hadn’t considered stacking by moving the sensor plane. I have access to a linhoff m679cs and some rodenstock apo-sironar digital lenses, plus the cognisys rail.
Have you focus stacked using this method? Any advice welcome or any other set ups you think I should consider?
I have the Cambo part to connect the rear standard of the Actus to the Cognisys stepper motor, but I’ve not used it. My current plan is to accomplish that function through a collection of Swebo parts.
I now know that the part I have to move the rear standard under the control of the Cognisys controller doesn’t work with my version of the Actus. That explains why I couldn’t figure out how to hook it up.
Informative post, thanks. I’m using an Hasselblad 907x with the HC 120mm Macro to scan 35mm and 645 negatives. I chose the HC Macro because it’s relatively affordable used and because it can do a 1:1 ratio, but I read your comments about field curvature and wonder if that’s an issue. Is there a simple test I can run to check?
The lens is also an older design so I suspect the XCD Macro may perform a little bit better, and the field curve problems might be less at the 2:1 ratio. So far the setup performs great but sometime I might rent the XCD lens to test or consider trying a GFX camera.
I just found Roger Cicala’s posts on how to test for curvature, so I’m reading up…
Try something like this: