• Welcome to Focus on Photography Forum!
    Come join the fun, make new friends and get access to hidden forums, resources, galleries and more.
    We encourage you to sign up and join our community.

Focus stacking RF 100mm macro anomalies

Kurtru

POTN Refugee
Joined
20 Nov 2023
Posts
29
Likes
115
Location
Salem Oregon
Name
Kurt
Image Editing
No
Recently got a RF 10omm Macro lens and trying my hand at focus stacking pictures of mushrooms. I am seeing lots of anomalies in the resulting stacks. I have included a couple of crops of the issue.

Could this be the result of the "focus shift" issue talked bout with this lens, which of course I didn't see until searching for an answer to this problem.

I have tried different apertures, more or less images to create the stack, and different software (DPP, Affinity 2, Helicon trial).

Grateful for any light anyone can shed on this.

R62_8574_cr.jpg

R62_0547_cr.jpgR62_1251_cr.jpg
 
Last edited by a moderator:
What these show are classic stacking halos. They often arise when there is a substantial distance (front to back) between one element and whatever is behind it. Depth map stacking algorithms are particularly vulnerable to this. It has nothing to do with the lens, the aperture, etc.

Please look at this recent thread, https://focus-on-photography-forum....-you-use-for-focus-stacking.4367/#post-131049, and in particular, read the article linked in post #14. Stacking is technically complex, so this is not a simple matter to deal with.
 
As Paddler4 says, this comes with the territory, the deeper the stack the more the issue.

Did you use the DPP depth composite to stack? That works quite well, normally. Example with the 180L
fungi Odd Black Fungi A02_20-11-23-0001-100 by Lester W, on Flickr

I don't have the RF 100 but have used the 100 USM classic and 180L to stack using focus bracketing, there is an issue due to the change in focal length with deep stacks that you don't see with a fixed magnification and moving the kit on a rail, but the above is not it.

Have you tried adding any accessories to your RF 100 (tubes, closeup lens) to get more magnification?
I would be interested to see how it performs.
 
I wouldn't get bogged down at this time, given what you are photographing, with the issue of a rail vs. focusing with the lens barrel. It won't matter at this scale. If you are interested in this anyway and want some counterintuitive findings, check out
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
. It's long but informative.

I'd also forget about which lens you are using. You are using a high quality lens. The choice among high quality lenses is irrelevant to this problem.

As the article I linked explained, the primary cause of this is the physical subject: there is no focal point at which both the top surface and the bottom surface that is adjacent (in 2D) are both in focus, so there is no clean slice for depth mapping to work with. There are other contributing factors too, but that's the biggest one. In the case of Zerene, the most effective solution is to paint over the halos with material from a PMax composite because the PMax algorithm is less susceptible to halos. That will generally get rid of enough of the halos to allow clean prints. I have never used Helicon, but it has three stacking algorithms, so it might also have a retouching option like Zerene's.

You will get blurry areas if you have too few images--that is, if your images are focused too far apart so that some areas are not in sharp focus in at least one. However, that problem isn't normally limited to edges. And once you have enough in-focus images, adding more won't help.

I don't mean to discourage you. On the contrary, I find macro photography enjoyable and rewarding, and I encourage people to dive in. In fact, I've taught a few classes on focus stacking to help novices get going. However, it's much more technically demanding than some other forms of photography, so you have to be prepared to spend some time learning.
 
On the Zerene I find painting to DMap from Pmax problematic because of the inherently higher contrast of the Pmax algorithm.

I tend to paint from the appropriate original frame to deal with stacking artefacts, just how it is for me.
 
I never paint from PMax to Dmap. I only do the reverse. I use the smallest retouching brush that will cover the halo.
 
As Paddler4 says, this comes with the territory, the deeper the stack the more the issue.

Did you use the DPP depth composite to stack? That works quite well, normally. Example with the 180L
fungi Odd Black Fungi A02_20-11-23-0001-100 by Lester W, on Flickr

I don't have the RF 100 but have used the 100 USM classic and 180L to stack using focus bracketing, there is an issue due to the change in focal length with deep stacks that you don't see with a fixed magnification and moving the kit on a rail, but the above is not it.

Have you tried adding any accessories to your RF 100 (tubes, closeup lens) to get more magnification?
I would be interested to see how it performs.

Hello Lester,
Thank you for your reply. I did use DPP, along with the Helicon software (trial) and Affinity 2. I used DPP exclusively for many years (DXO PL 8 now), never knew it had this capability.

In answer to your questions as to whether I have used tubes with it the answer is no, or more precisely, not yet.
 
I wouldn't get bogged down at this time, given what you are photographing, with the issue of a rail vs. focusing with the lens barrel. It won't matter at this scale. If you are interested in this anyway and want some counterintuitive findings, check out
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
. It's long but informative.

I'd also forget about which lens you are using. You are using a high quality lens. The choice among high quality lenses is irrelevant to this problem.

As the article I linked explained, the primary cause of this is the physical subject: there is no focal point at which both the top surface and the bottom surface that is adjacent (in 2D) are both in focus, so there is no clean slice for depth mapping to work with. There are other contributing factors too, but that's the biggest one. In the case of Zerene, the most effective solution is to paint over the halos with material from a PMax composite because the PMax algorithm is less susceptible to halos. That will generally get rid of enough of the halos to allow clean prints. I have never used Helicon, but it has three stacking algorithms, so it might also have a retouching option like Zerene's.

You will get blurry areas if you have too few images--that is, if your images are focused too far apart so that some areas are not in sharp focus in at least one. However, that problem isn't normally limited to edges. And once you have enough in-focus images, adding more won't help.

I don't mean to discourage you. On the contrary, I find macro photography enjoyable and rewarding, and I encourage people to dive in. In fact, I've taught a few classes on focus stacking to help novices get going. However, it's much more technically demanding than some other forms of photography, so you have to be prepared to spend some time learning.

Paddler,

Thank you for your informative replies, and for taking the time to do so, very much appreciated.

You explanations have been most helpful. The link was great, and I found his whole series of videos which is now bookmarked. This will be very helpful in my macro learning process. I've had a macro lens for a very long time but never got very deep into it, although I have always enjoyed using it. But macro has really piqued my interest lately. I have gotten some focus stacks that I am quite happy with.

Looking forward to learning more about close up and macro photography. Close up and macro seems like the perfect antidote to the "I don't even know what to shoot" doldrums that sometimes strikes.

By the way, what do you paddle?
 
I used to paddle whitewater (kayak), but age, a change in location, and a back injury changed that. Now I paddle a sea kayak.

I've thought more about this and have a couple of additional thoughts:

--You are going to find halos of this sort with any depth-map stacking algorithm, which includes most that you will encounter. That's one of several reasons I find Zerene so helpful: it gives you tools for dealing with it.

--Lester said "the deeper the stack, the more the issue". I think that's not quite correct. What matters is the distance (front to back) between elements that are adjacent in the 2-dimensional photo. That's why the problem occurs on edges. If those front-to-back distances between adjacent elements are small, you can have a very deep stack with no problems.

I mentioned that Zerene has two stacking algorithms, DMap (depth mapping, similar to what you will find elsewhere) and PMax. PMax has a number of advantages: it's better at preserving the finest detail, and it's far less vulnerable to halos. However, it doesn't maintain colors as well and in general often produces a less pleasing result, particularly for things like flowers. That's the reason I use a DMap composite as my main image and use a PMax only to paint over problems. Zerene lets you touch up from any single image or from another composite. So I put the PMax composite in the source window, put the DMap in the target window, and paint with a very small brush to replace the halo areas with material from the PMax composite. This usually works very well.

I don't know enough about Helicon to know whether it offers similar functions.

Good luck!

For fun, here are a few stacks done in Zerene.




 
Last edited:
Back
Top Bottom